INTRODUCTION

TOP ABSTRACT INTRODUCTION THE THEORY OF PFD MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

A plausible hypothesis for the production of force in the actomyosin system is the lever arm model which proposes that, during the energy transducing cycle, the motor domain (MD) of the myosin head binds rigidly to actin in the thin filament, and a hinge within the myosin head allows the light chain domain (LCD) to tilt like an arm flexing at its elbow (Huxley and Kress, 1985; Cooke, 1986; Vibert and Cohen, 1988; Rayment et al., 1993a; Goldman, 1998; Geeves and Holmes, 1999). The crystal structure of chicken skeletal myosin subfragment 1 (S1) (Rayment et al., 1993b) provided strong support for this hypothesis and stimulated many experiments designed to detect tilting motions of myosin domains.

Relative motion between the catalytic and light chain domains during the ATPase cycle of myosin fragments has been detected from electric birefringence (Highsmith and Eden, 1986), scattering of x-rays and neutrons (Wakabayashi et al., 1992; Mendelson et al., 1996), electron paramagnetic resonance (EPR) of spin labels (Adhikari et al., 1997), resonance energy transfer (Suzuki et al., 1998; Shih et al., 2000) and comparison of crystal structures with various bound nucleotide and phosphate analogs (Fisher et al., 1995; Smith and Rayment, 1996; Dominguez et al., 1998; Houdusse et al., 1999). These studies have shown that the MD and LCD rotate relative to each other as expected for a lever arm mechanism. In the transition state between M · ATP and M · ADP · P_i, the myosin head is bent with the LCD tilted toward the ATP binding site. Upon release of P_i from M · ADP · P_i, myosin straightens by rotation of the LCD together with part of the motor domain, termed the converter, relative to the rest of the MD. Based on the orientation of myosin with respect to actin in decorated actin filaments (Rayment et al., 1993a; Milligan and Flicker, 1987), the straightening of S1 tilts the LCD in a direction that would actively move the load (thick filament or cargo) toward the barbed end of actin (Z line in muscle). This "power stroke" is reversed during the hydrolysis step M · ATP to M · ADP · P_i or at an isomerization just before this step (Suzuki et al., 1998). These studies were all conducted on fragments of myosin bearing no mechanical force, and most of them in the absence of actin. Thus, the relationship between the conformational changes of isolated myosin and tilting motions in an active muscle fiber is still unclear. In particular, does the tilting of the light chain accompany force generation, filament sliding or both? Does the MD rotate relative to actin as well? These questions can be addressed only by experiments on systems actually transducing chemical energy to mechanical work.

Structural biological methods have been applied to detect tilting motions of myosin domains in muscle fibers. In low-angle x-ray diffraction patterns of muscle, the intensity and splitting of the 14.3-nm meridional reflection are sensitive indicators and provide strong support for tilting and flexibility of the LCD (Huxley et al., 1983; Lombardi et al., 1995; Dobbie et al., 1998; Linari et al., 2000). Fluorescent probes and extrinsic spin labels have been placed in the motor domain at a highly reactive thiol, Cys⁷⁰⁷, of the rabbit psoas myosin heavy chain (Borejdo et al., 1982; Tanner et al., 1992; Berger et al., 1996; Cooke et al., 1982; Hellen et al., 1995), and at various positions in the RLC (Hambly et al., 1992; Irving et al., 1995, Ling et al., 1996; Allen et al., 1996; Sabido-David et al., 1998; Corrie et al., 1999, Hopkins et al., 2002). Electron microscopy also resolved the MD and LCD (Pollard et al., 1993; Taylor et al., 1999).

These studies have generally shown that both the MD and the LCD have a broad orientation distribution in relaxation and contraction and a narrower distribution in rigor. During contraction, the ordered component of the MD hardly tilts in response to perturbations that alter tension, such as length steps (Cooke, 1981; Berger et al., 1996; Burghardt et al., 1997) or increase of phosphate concentration (Zhao et al., 1995), suggesting that it is rigidly attached to actin. However, some evidence suggests that the MD does rotate during force development (Taylor et al., 1999; Tsaturian et al., 1999). The LCD tilts in response to applied length changes (Irving et al., 1995; Hopkins et al., 1998; Dobbie et al., 1998; Corrie et al., 1999), providing further support for the lever arm hypothesis for contraction of the fully assembled sarcomere.

The angular distributions for relaxed and actively cycling heads may be broad due to dynamic motions from thermal reorientation on the nanosecond to microsecond timescale or sequential population of the states of the enzymatic cycle up to the millisecond timescale. Additional static disorder is expected among the attached myosin heads, due to the incommensurate periodicities of the actin and myosin filaments (Huxley and Brown, 1967).

Time-resolved decay of phosphorescence anisotropy (TPA) and saturation transfer EPR (ST-EPR) are capable of detecting protein rotational dynamics on the microsecond time scale of the cross-bridge motions expected in an active muscle fiber. These techniques have been applied to muscle fibers and myofibrils labeled at Cys⁷⁰⁷ (Thomas et al., 1980; Barnett and Thomas, 1984, 1989; Ludescher and Thomas, 1988; Stein et al., 1990; Berger and Thomas, 1993) and have identified components of motion of the MD at approximately 20 and 300 µs that are unique to active contractions (Stein et al., 1990). However, few reports of ST-EPR or TPA have been carried out with probes bound to sites on the RLC (Thomas et al., 1995; Ramachandran and Thomas, 1999). ST-EPR reports the time scale of motions but cannot independently resolve the amplitude. TPA is somewhat insensitive because the phosphorescence emission is ~10⁵-fold dimmer than fluorescence (Johnson and Garland, 1981; Yoshida and Barisas, 1986).

Here we report the development of a novel extension of fluorescence polarization spectroscopy on ordered samples, such as muscle fibers, to resolve the extent of dynamic motions on the 20- to 500-µs time scale with improved sensitivity. Polarized fluorescence depletion (PFD) has been described for isotropic samples, membranes and cells (Johnson and Garland, 1981; Yoshida and Barisas, 1986; Corin et al., 1987; Londo et al., 1993). Hellen et al. (1995) have used a similar method to detect motions of probes bound to Cys⁷⁰⁷ in muscle fibers.

In the absence of oxygen, fluorescent probes will often populate a long-lived triplet state. Triplet population is not restricted to probes exhibiting phosphorescence emission. Following transient pumping of probes into the triplet state, the probes remaining in the ground state have an orientational distribution that is depleted around the polarization direction of the pumping pulse. This polarized depletion is detected by fluorescence excited by a weak interrogation beam, and it lasts until it is filled in by dynamic rotational motions and decay of the triplet population. The PFD method potentially provides dynamic information describing the extent of motions on the time scale between the fluorescence lifetime (one to a few ns) and the duration (up to a few ms) of the triplet state as well as improved detail about the static probe orientation. As implemented here, using rhodamine as the probe, the PFD technique provides a practical, sensitive method to detect the dynamics of protein rotational motions in the time scale 4 ns t 20 µs.

The PFD technique was used in the present work to determine the orientation and mobility of the myosin LCD in single muscle fibers in relaxation, contraction, and rigor. During contraction, motions of the RLC on the microsecond time scale are sufficient to account for much of the orientational dispersion. Part of the work has previously been reported in abstract form (Bell et al., 2000).

	THE THEORY OF PFD

TOP ABSTRACT INTRODUCTION THE THEORY OF PFD MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Polarized fluorescence depletion is an enhancement of fluorescence polarization that enables characterization of the orientation distribution of fluorescent probes in an ordered system at higher resolution than either fluorescence or phosphorescence polarization. It can provide information on the rate and extent of rotational motions on a time scale comparable with that of the decay of phosphorescence (Johnson and Garland, 1981; Yoshida and Barisas, 1986; Dale, 1987; Corin et al., 1987).

In an ordered biological system such as a biological membrane or a muscle fiber, specifically labeled with an extrinsic fluorophore, an intense pulse of exciting light is applied to the probe. A proportion of the optically excited chromophores undergoes intersystem crossing to populate a long-lived triplet excited state, temporarily leaving a reduced population of probes in the ground state. This pulse excitation of the triplet state is termed transient bleaching or photoselection. Several factors govern the angular distribution of the probes remaining in the ground state for a period, typically several milliseconds, after the photoselection pulse: 1) the angular distribution of the population that was present before the bleaching pulse; 2) the polarization of the bleaching pulse; 3) the extent of bleaching; 4) the rate of return to the ground state of the triplet population; and 5) rotational motions of the probe reflecting those of the labeled proteins. Polarized fluorescence of the ground state population is detected using a weak excitation beam to interrogate the same sample volume.

Relationships involving the steady-state polarized fluorescence intensities before and the diminished intensities after the photobleaching pulse are derived in the Appendix, which shows how the orientation distribution and dynamics can be determined. The bleached population has an orientation distribution that initially corresponds to a convolution of the original steady-state distribution with the angular distribution of bleaching efficacy. Differences of polarized intensities of fluorescence before and after the bleaching pulse yield estimates of the orientation distribution and dynamics of this bleached population. The distribution of the bleached population relaxes toward the steady-state distribution by rotational motions detectable in the time range greater than the fluorescence lifetime (e.g., 4 ns) and up to several lifetimes of the triplet state. Such rotations cause ratios of the polarized difference intensities to decay toward those of the pre-bleach fluorescence. Finally, in the absence of irreversible bleaching, the difference intensities decay to zero due to return of the excited triplet population to the ground state.

Probes useful in PFD should have triplet lifetimes similar to or longer than the protein rotations of interest, but they need not exhibit delayed luminescence (phosphorescence or delayed fluorescence). The fluorescent probe used here, rhodamine, has a fluorescence (singlet excited state) lifetime, _f, of ~4 ns, and exhibits no appreciable phosphorescence under the conditions of the present studies.

The probability of absorption of a photon by a probe is proportional to cos² _a, where _a is the angle between the electric vector of the linearly polarized excitation beam and the probe absorption dipole moment. The probability of detecting a photon through an analyzing polarizer is given by cos² _e, where _e is the angle between the electric vector of the emitted photon transmitted by the analyzer and the probe emission dipole moment. Thus, in a standard steady-state fluorescence polarization experiment, the polarized fluorescence intensity is given by:

(1)

where I₀ represents the total fluorescence intensity, independent of polarization, E denotes the polarization of the excitation beam, E' that of the detected component of the emitted beam, and a and e denote unit vectors representing the absorption and emission dipoles for a single chromophore. The brackets,  , denote an ensemble average taken by integrating the probability density functions for a and e over all spherical angles.

Using measurements of _EI_E' with excitation and emission polarizations parallel and perpendicular to the muscle fiber axis, F, and optical paths perpendicular to that axis, three order parameters describing the probe orientation distribution are obtainable: P_2d, P₂ and P₄ (Dale et al., 1999; Hopkins et al., 2002; see Appendix). The order parameters are coefficients of a series expansion describing the orientation distributions using the Legendre polynomials as basis functions (Dale et al., 1999; see Appendix). P_2d describes the extent of subnanosecond rotational motions (wobble) of the probe transition dipoles about an axis, c, defined relative to the protein structure and P₂ and P₄ describe the orientation distribution of c relative to F. Both static disorder of c and dynamic disorder caused by motions that are slower than the probe's fluorescence lifetime, _f, contribute to determining P₂ and P₄. The effect of fast probe motions, however, has been `factored out' of P₂ and P₄ (Appendix Eqs. D.15 and D.16). The description used here makes the assumption that the absorption and emission dipoles are collinear, as approximately applies for rhodamine (Chen and Bowman, 1965; Penzkofer and Wiedmann, 1980; Hopkins et al., 1998), but the case of probes with non-collinear dipoles can be analyzed in similar terms (Londo et al., 1993; Dale et al., 1999).

In a PFD experiment, additional orientation and dynamic information becomes available because correlation between three photons (photoselection, interrogation and emission) combine to determine the polarization. As shown in the Appendix, if the extent of transient photobleaching is low, the difference, , between intensities measured before and after a transient photoselection pulse is given by an expression similar to Eq. 1,

(2)

where B is a scale factor proportional to the extent of bleaching, `E is the polarization vector of the photoselection pulse and b is the orientation of the absorption dipole at the instant of bleaching (Appendix, Fig. A.1).

Let order parameters P_2p and P_4p describe rotational motions of c on an intermediate (microsecond) time scale longer than the fluorescence lifetime (_f) but still much shorter than the time, _p, between application of the bleaching pulse and measurement of the remaining fluorescence (_{f   p}). These order parameters are analogous to P_2d for fast wobble (  _f). Just as P_2d is found from measurements of _EI_E', P_2p and P_4p can be estimated from measurements of _`E,EE' with various combinations of polarizer orientations and optical paths. P_2p and P_4p are determined by the extent of rotational motions of c (defined above) on the intermediate time scale about a slower-moving axis, p, also defined within the protein. Defining order parameters, P_2s, P_4s and P_6s, for the static orientation distribution of p with the microsecond wobble parameters "factored out," then:

(3)

The Appendix describes how P_2p, P_4p, P_2s, P_4s and P_6s can be determined from experimentally measured intensities (_EI_E') and intensity differences (_`E,EE'), thereby separating microsecond (_{f   p}) wobble from static disorder and/or any motions of p on a slower time scale than that of triplet decay. Both static disorder of p and dynamic disorder caused by motions that are slower than _p contribute to determining P_2s, P_4s and P_6s. Thus the PFD method extends the time scale of motions detectable by fluorescence polarization and potentially provides extra resolution (up to P_6s) of the orientation distribution static on that time scale.

The above description applies to a situation in which the intensity of the photoselection pulse is low enough that the probability of bleaching a particular probe molecule is B cos² _b, where _b is the angle between the polarization of the photoselection pulse and the probe absorption dipole moment, B is the depth of bleaching, and Eq. 2 is applicable. Most of the experiments presented in this paper were analyzed under this low-bleach assumption. Unlike in a standard (non-bleaching) fluorescence experiment, however, the photoselection pulse can be intense enough that, for probes at orientations close to its polarization vector, bleaching becomes partially saturated and no longer proportional to intensity (Axelrod et al., 1976; Hellen and Burghardt, 1994). The full expression for the angular dependence of the probability of bleaching, due to a linearly polarized beam of uniform intensity, is 1  exp(B(`E · b)<sup>2</sup>) (Dale, 1987), rather than B(`E · b)². The results of experiments using a range of bleach amplitudes can be used to check the validity of assuming the simpler low-bleach angular dependence, B(`E · b)². Expressions that apply to the high-bleach regime and used for this validity check are listed in the Appendix. In principle, higher-rank order parameters (e.g. P_8s) can also be obtained with deep bleaching, but measurement uncertainties limit the practicability of extending the analysis beyond the 6^th rank.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION THE THEORY OF PFD MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Chemicals and solutions

Solution compositions were as described in Hopkins et al. (1998). Nucleotides and rabbit skeletal muscle troponin were obtained from Sigma (St. Louis, MO). The 6-isomer of iodoacetamidotetramethylrhodamine (6-IATR) was kindly provided by Dr. John Corrie, synthesized as previously described (Corrie and Craik, 1994). Chicken gizzard wild-type regulatory light chain was expressed in Escherichia coli, purified, and labeled at its native cysteine residue (108) with the 6-IATR (Sabido-David et al., 1998). Troponin C was prepared as described by Dobrowolski et al. (1991) with modifications.

Muscle fiber preparation and RLC exchange

In preparation for the experiment, 4-mm segments of single muscle fibers were dissected from glycerinated bundles of rabbit psoas muscle (Goldman et al., 1984). Fiber ends were held in aluminum foil T-clips, and the sarcomeres inside and within 100 µm of the T-clips were cross-linked by glutaraldehyde (Allen et al., 1996). The fiber segment was then mounted in the experimental apparatus in 5 mM MgATP relaxing solution at 11°C, and activated briefly (as described below) to test integrity of the fiber. Length and cross-sectional area were measured (Goldman and Simmons, 1984).

Regulatory light chain, monofunctionally labeled with rhodamine at Cys¹⁰⁸, was exchanged for native RLC as described previously (Ling et al., 1996). Briefly, the fiber was incubated for 2 min each in relaxing solution containing 0.1 mM MgATP ("0.1 Rel") at 10°C, rigor solution at 10°C, exchange solution at 10°C, and then for 30 min at 30°C in exchange solution containing 0.5 mg/ml labeled RLC and to which 3 mM DTT had been freshly added. The fiber then was cooled quickly to 10°C and relaxed in relaxing solution containing 5 mM MgATP ("5 Rel"). Troponin and troponin C extracted during RLC exchange were replenished by incubation for 40 min at 10°C in 5 Rel containing 0.5 mg/ml troponin then for 10 min in 5 Rel containing 0.5 mg/ml troponin C. Following the exchange procedure, active tension at 11°C was 0.94 ± 0.28 (mean ± S.D., n = 6) of that before exchange, consistent with previous results (Ling et al., 1996, Allen et al., 1996). Due to a thermal gradient, the solution trough in which fiber mechanical and spectroscopic parameters were assayed was about one degree higher than the other troughs.

Experimental apparatus

Optical system

View larger version (19K): [in this window] [in a new window]   FIGURE 1   Experimental apparatus. Schematic diagram of the optics for pulsed photoselection and interrogation of fluorescence depletion. The sample is a solution of rhodamine, a PVA film containing rhodamine or a muscle fiber held in a trough containing an aqueous solution. The acousto-optic (AO) deflector, shutter, Pockels cell, rotating mirror and photomultiplier gating are under computer control.

Electronics and data collection

Experimental protocol

At the beginning of the experiment, the Pockels cell driving voltages were tuned to optimize extinction through vertical and horizontal polarizers, thereby compensating for any drift of the optical retardation. Spectroscopic reference samples were then assayed as a measure and verification of instrument parameters. The reference samples and their analysis are described later.

A muscle fiber segment was mounted into the experimental apparatus in relaxing solution, tested and measured as described earlier and exchanged with labeled RLC. Spectroscopic trials were then performed in each of the relaxed, rigor, and Ca²⁺-activated conditions. Several locations along the middle third of the fiber length were assayed for each condition. The ordering of the conditions was randomized to minimize systematic effects of fiber rundown on the spectroscopic data for each condition.

Prior to each rigor contraction, the fiber was incubated for 2 min in 0.1 Rel. After PFD signals were recorded for the rigor condition, the fiber was relaxed in 5 Rel. Isometric activation was preceded by incubation for 2 min in pre-activating solution containing 0.1 mM EGTA. The fiber was transferred to activating solution containing ~30 µM free Ca²⁺, and PFD signals were recorded. The fiber was then relaxed in 5 Rel.

The triplet lifetime is highly sensitive to quenching by oxygen in the solutions (Calhoun et al., 1983), so the relaxing, rigor, and activating solutions were maintained under a stream of argon gas. On the day of each experiment, an oxygen scavenging system of glucose, glucose oxidase and catalase (Calhoun et al., 1983) was added to these solutions. Jets of argon gas were also directed at the fiber solution trough to reduce the surrounding oxygen tension and to eliminate condensation on the optical windows of the trough.

At the end of each experiment, spectroscopic reference samples were again assayed to measure and verify instrument parameters.

Spectroscopic reference samples

Three fluorescent reference samples were fashioned to provide predictable physical and optical properties. The samples were made by sandwiching viscous or solid fluorescent material between two isosceles 45°  90°  45° prisms to form a 10-mm cube (Hopkins et al., 1998). The cubes could be placed at the position of the fiber in the experimental setup with faces perpendicular to all of the optical beams. Diffusion of oxygen in the reference samples was inhibited by the viscosity of the solution or the polymer.

A random, viscous solution of IATR was made by diluting a 10 mM stock of IATR in dimethyl formamide into glycerol to 100 µM final concentration. Argon gas was gently bubbled through the solution to mix the dye and to displace dissolved oxygen. The isosceles prisms were held together by strips of double-sided adhesive tape with a gap forming a chamber on the diagonal plane between the prisms. The probe solution was introduced into this gap and the chamber was then sealed with nail polish.

Rigid samples were made from IATR in a polyvinyl-alcohol (PVA) film matrix. PVA powder was dissolved at 1 g/ml in H₂O at 80°C, then maintained at 40°C while IATR stock (10 mM in DMF) was gently stirred in to 20 µM final concentration. Drops of warm PVA-IATR solution were put onto glass slides, which were cured under vacuum to form a ~150-µm-thick film. This film was reannealed at 80°C for 4 h, trimmed to 10 × 5 mm and fixed between two of the 10-mm isosceles prisms with optical grade epoxy (type 302, Epo-tek Corp., Billerica, MA).

Rigid, partially oriented fluorescent samples were made by slowly stretching PVA strips ~5-fold at 80°C using a motor-driven leadscrew. A stretched sample was oriented in the optical cube so that when positioned in the experimental setup, the direction of stretch was oriented along the same axis (z) as the muscle fibers. Starting with a somewhat thicker film for stretched samples yielded a final thickness and fluorescence intensity very similar to that of the isotropic (unstretched) PVA sample.

Protocol for polarized fluorescence depletion

At the beginning of each polarized fluorescence depletion (PFD) trial, the mechanical shutter was opened and steady-state fluorescence intensity data for both emission polarizations were recorded for 200 or 400 µs at a sampling rate of 2 µs. During this period, polarization of the excitation was alternated between parallel and perpendicular to the muscle fiber axis every 20 µs (Fig. 2). Excitation intensity was set to elicit strong fluorescence without causing significant population of the triplet state or irreversible bleaching. This is the "interrogation intensity," as discussed further in the Results section. The photomultipliers (PMTs) were gated off (at 50 µs in Fig. 2) and then the AO modulator increased the intensity (~500-fold) for 12 µs to transiently populate the probe triplet state. The AO modulator then returned the excitation to the interrogation level. During the photoselection pulse, the highly attenuated PMTs responded slightly to the prompt sample fluorescence, giving rise to the small upward deflections ending at time 0 in Fig. 2. The photoselection pulse was polarized parallel to the fiber axis.

View larger version (42K): [in this window] [in a new window]   FIGURE 2   Unprocessed data from a single polarized fluorescence depletion (PFD) trial for x (in-line) illumination, covering the period before and after the bleach event. The sample is a single muscle fiber in the rigor condition, into which 6-ATR-labeled myosin regulatory light chain had been exchanged. (A) At the beginning of the trial, steady-state intensity data were collected for each polarization of fluorescence (noisy red and blue traces), in response to alternating excitation polarization (magenta trace). For the bleach, excitation (green trace) was modulated to high intensity for a 12-µs period ending at time 0, then returned to a lower interrogation setting. Smooth red and blue lines: the average of steady pre-bleach fluorescence data, followed by a single exponential fitted to the post-bleach recovery, corresponding to the I (red) and I (blue) components of fluorescence. (B) The same as A, but during the subsequent trial in which alternation of the excitation polarization was 180° out of phase with respect to that in A.

After the photoselection pulse, the PMTs were switched back on and fluorescence was recorded for 1.6 ms while the excitation polarization was alternated again every 20 µs. The laser beam was then shuttered off and the data sampling clock was stopped. The sample was kept in the dark for approximately 10 triplet lifetimes to allow complete return to the ground state before the next trial. During that dark time, the data sampling clock was briefly triggered to acquire several data points as a zero-light reference, and the solenoid mirror was switched between x and y illumination in preparation for the next trial.

After each bleaching pulse, fluorescence detected at the interrogation level was diminished, and it recovered at the rate of triplet decay. In Fig. 2, lines representing the average of the constant fluorescence data before the bleach, and single exponential decays fitted to the recovery, are superimposed on the modulated PMT signals.

In order to gather polarized fluorescence data from both the parallel and perpendicular interrogation polarizations immediately after the bleach, each PFD trial was repeated with the excitation polarization switching 180° out of phase from that of the preceding trial (Fig. 2 B). Four individual trials with the input direction and polarization phase at (x, 0°), (y, 0°), (y, 180°), and (x, 180°) were completed as a set in that order (Fig. 3 A). Each complete sweep of the recording oscilloscope contained 10 such sets corresponding to 40 photoselection pulses and attendant recording of modulated polarized fluorescence. Recordings from each combination of input direction and polarization phase were averaged together later to improve signal-to-noise ratio. The duration of recording of the 10 sets was 17 seconds for muscle fiber data and 68 s for PVA reference samples. The reference samples were given longer dark times, appropriate to their longer triplet lifetimes.

View larger version (33K): [in this window] [in a new window]   FIGURE 3   (A) Unprocessed fluorescence intensities representing parallel and perpendicularly polarized excitation and emission, as indicated on the right, from four PFD trials with illumination by the x, y, y, and x paths in succession. Bleach events are signified by arrows. The sample was a rigor muscle fiber as in Fig. 2. During y-illumination, I, I, and I are superimposed, indicating that yQ is close to zero. (B) Averaged, uncorrected fluorescence intensity data from a complete sweep of 40 PFD trials, covering the period before and after the bleach events. Smooth red and blue lines: the average of steady pre-bleach fluorescence data, followed by a single exponential fitted to the post-bleach recoveries, corresponding to the four components of polarized fluorescence as indicated on the right.

Processing of fluorescence intensity data

Time-resolved intensity data

Difference intensities

Instrumental correction factors

Analysis of fluorescence intensity and difference data

Calculation of polarization ratios and correlation functions

	RESULTS

TOP ABSTRACT INTRODUCTION THE THEORY OF PFD MATERIALS AND METHODS RESULTS DISCUSSION APPENDIX REFERENCES

Time-resolved intensities and polarization ratios

Reference samples

View larger version (19K): [in this window]