Myosin-Specific Adaptations of In vitro Fluorescence Microscopy-Based Motility Assays

Summary

Presented here is a procedure to express and purify myosin 5a followed by a discussion of its characterization, using both ensemble and single molecule in vitro fluorescence microscopy-based assays, and how these methods can be modified for the characterization of nonmuscle myosin 2b.

Abstract

Myosin proteins bind and interact with filamentous actin (F-actin) and are found in organisms across the phylogenetic tree. Their structure and enzymatic properties are adapted for the particular function they execute in cells. Myosin 5a processively walks on F-actin to transport melanosomes and vesicles in cells. Conversely, nonmuscle myosin 2b operates as a bipolar filament containing approximately 30 molecules. It moves F-actin of opposite polarity toward the center of the filament, where the myosin molecules work asynchronously to bind actin, impart a power stroke, and dissociate before repeating the cycle. Nonmuscle myosin 2b, along with its other nonmuscle myosin 2 isoforms, has roles that include cell adhesion, cytokinesis, and tension maintenance. The mechanochemistry of myosins can be studied by performing in vitro motility assays using purified proteins. In the gliding actin filament assay, the myosins are bound to a microscope coverslip surface and translocate fluorescently labeled F-actin, which can be tracked. In the single molecule/ensemble motility assay, however, F-actin is bound to a coverslip and the movement of fluorescently labeled myosin molecules on the F-actin is observed. In this report, the purification of recombinant myosin 5a from Sf9 cells using affinity chromatography is outlined. Following this, we outline two fluorescence microscopy-based assays: the gliding actin filament assay and the inverted motility assay. From these assays, parameters such as actin translocation velocities and single molecule run lengths and velocities can be extracted using the image analysis software. These techniques can also be applied to study the movement of single filaments of the nonmuscle myosin 2 isoforms, discussed herein in the context of nonmuscle myosin 2b. This workflow represents a protocol and a set of quantitative tools that can be used to study the single molecule and ensemble dynamics of nonmuscle myosins.

Introduction

Myosins are motor proteins that exert force on actin filaments using the energy derived from adenosine triphosphate (ATP) hydrolysis¹. Myosins contain a head, neck, and tail domain. The head domain contains the actin-binding region as well as the site of ATP binding and hydrolysis. The neck domains are composed of IQ motifs, which bind to light chains, calmodulin, or calmodulin-like proteins^2,3. The tail region has several functions specific to each class of myosins, including but not limited to the dimerization of two heavy chains, binding of cargo molecules, and regulation of the myosin via autoinhibitory interactions with the head domains¹.

The motile properties of myosin vary greatly between classes. Some of these properties include duty ratio (the fraction of myosin's mechanical cycle in which the myosin is bound to actin) and processivity (the ability of a motor to make multiple steps on its track before detachment)⁴. The over 40 classes of myosins were determined based on sequence analyses^5,6,7,8. The class 2 myosins are classified as "conventional" since they were the first to be studied; all other classes of myosins are, therefore, classified as "unconventional."

Myosin 5a (M5a) is a class 5 myosin and is a processive motor, meaning that it can take multiple steps along actin before dissociating. It has a high duty ratio, indicating that it spends a large part of its mechanical cycle bound to actin^{9,10,11,12,13,14}. In common with other myosins, the heavy chain contains an N-terminal motor domain that includes both an actin-binding and an ATP hydrolysis site followed by a neck region that serves as a lever-arm, with six IQ motifs that bind to essential light chains (ELC) and calmodulin (CaM)¹⁵. The tail region contains α-helical coiled-coils, which dimerize the molecule, followed by a globular tail region for binding cargo. Its kinetics reflect its involvement in the transport of melanosomes in melanocytes and of the endoplasmic reticulum in Purkinje neurons^16,17. M5a is considered the prototypical cargo transport motor¹⁸.

Class 2 myosins, or the conventional myosins, include the myosins that power contraction of skeletal, cardiac, and smooth muscle in addition to the nonmuscle myosin 2 (NM2) isoforms, NM2a, 2b, and 2c¹⁹. The NM2 isoforms are found in the cytoplasm of all cells and have shared roles in cytokinesis, adhesion, tissue morphogenesis, and cell migration^19,20,21,22. This paper discusses conventional myosin protocols in the context of nonmuscle myosin 2b (NM2b)²³. NM2b, in comparison to M5a, has a low duty ratio and is enzymatically slower with a V_max of 0.2 s^{-1 23} compared to M5a's V_max of ≈18 s^{-1 24}. Notably, truncated NM2b constructs with two heads do not readily move processively on actin; rather, each encounter with actin results in a power stroke followed by dissociation of the molecule²⁵.

NM2b contains two myosin heavy chains, each with one globular head domain, one lever-arm (with one ELC and one regulatory light chain (RLC)), and an α-helical coiled-coil rod/tail domain, approximately 1,100 amino acids long, that dimerizes these two heavy chains. The enzymatic activity and structural state of NM2b are regulated by phosphorylation of the RLC²³. Unphosphorylated NM2b, in the presence of ATP and physiological ionic strengths (approximately 150 mM salt), adopts a compact conformation wherein the two heads make participate in an asymmetric interaction and the tail folds back over the heads in two places²³. In this state, the myosin does not interact strongly with actin and has very low enzymatic activity. Upon RLC phosphorylation by calmodulin-dependent myosin light chain kinase (MLCK) or Rho-associated protein kinase, the molecule extends and associates with other myosins through the tail region to form bipolar filaments of approximately 30 myosin molecules²³. The aforementioned phosphorylation of the RLC also leads to increased actin-activated ATPase activity of NM2b by approximately four times^26,27,28. This bipolar filament arrangement, featuring many myosin motors at each end, is optimized for roles in contraction and tension maintenance, where actin filaments with opposing polarities can be moved relative to each other^23,29. Accordingly, NM2b has been shown to act as an ensemble of motors when interacting with actin. The large number of motors within this filament allow NM2b filaments to move processively on actin filaments, making in vitro filament processivity possible to characterize²⁹.

While progress has been made in understanding the role of myosins in the cell, there is a need to understand their individual characteristics at the protein level. To understand actomyosin interactions at a simple protein-protein interaction level, rather than inside of a cell, we can express and purify recombinant myosins for use in in vitro studies. The results of such studies then inform mechanobiologists about the biophysical properties of specific myosins that ultimately drive complex cellular processes^{12,13,14,25,29}. Typically, this is accomplished by adding an affinity tag to a full-length or truncated myosin construct and purifying via affinity chromatography^29,30,31. Additionally, the construct can be engineered to include a genetically encodable fluorophore or a tag for protein labeling with a synthetic fluorophore. By adding such a fluorescent label, single molecule imaging studies can be performed to observe myosin mechanics and kinetics.

Following purification, the myosin can be characterized in several ways. ATPase activity can be measured by colorimetric methods, providing insight into the overall energy consumption and actin affinity of the motor under different conditions³². To learn about the mechanochemistry of its motility, further experiments are required. This paper details two in vitro fluorescence microscopy-based methods that can be used to characterize the motile properties of a purified myosin protein.

The first of these methods is the gliding actin filament assay, which can be used to quantitatively study the ensemble properties of myosin motors, as well as qualitatively study the quality of a batch of purified protein³³. Although this paper discusses the use of total internal reflection fluorescence (TIRF) microscopy for this assay, these experiments can be effectively performed using a wide-field fluorescence microscope equipped with a digital camera, commonly found in many labs³⁴. In this assay, a saturating layer of myosin motors is attached to a coverslip. This can be accomplished using nitrocellulose, antibodies, membranes, SiO₂-derivatized surfaces (such as trimethylchlorosilane), among others^{29,33,35,36,37,38}. Fluorescently labeled actin filaments are passed through the coverslip chamber, upon which the actin binds to the myosin attached to the surface. Upon addition of ATP (and kinases in the study of NM2), the chamber is imaged to observe the translocation of actin filaments by the surface-bound myosins. Tracking software can be used to correlate the velocity and length of each gliding actin filament. Analysis software can also provide a measure of the number of both moving and stationary actin filaments, which can be useful to determine the quality of a given myosin preparation. The proportion of stalled filaments can also be intentionally modulated by surface tethering of actin to other proteins and measured to determine the load dependence of the myosin³⁹. Because each actin filament can be propelled by a large number of available motors, this assay is very reproducible, with the final measured velocity being robust to perturbations such as alterations in the starting myosin concentration or the presence of additional factors in the solution. This means it can be easily modified to study myosin activity under different conditions, such as altered phosphorylation, temperature, ionic strength, solution viscosity, and the effects of load induced by surface tethers. Although factors such as strong-binding myosin "dead heads" incapable of ATP hydrolysis can cause stalled actin filaments, multiple methods exist to mitigate such issues and allow for accurate measurements. The kinetic properties of myosin vary greatly across classes and, depending on the specific myosin used, the speed of actin filament gliding in this assay can vary from under 20 nm/s (myosin 9)^40,41, and up to 60,000 nm/s (Characean myosin 11)⁴².

The second assay inverts the geometry of the gliding actin filament assay¹². Here, the actin filaments are attached to the coverslip surface and the movement of single molecules of M5a or of individual bipolar filaments of NM2b are visualized. This assay can be used to quantify the run lengths and velocities of single myosin molecules or filaments on actin. A coverslip is coated with a chemical compound that blocks non-specific binding and simultaneously functionalizes the surface, such as biotin-polyethylene glycol (biotin-PEG). The addition of modified avidin derivatives then primes the surface and biotinylated actin is passed through the chamber, resulting in a layer of F-actin stably bound to the bottom of the chamber. Finally, activated and fluorescently labeled myosin (typically 1-100 nM) is flowed through the chamber, which is then imaged to observe myosin movement over the stationary actin filaments.

These modalities represent fast and reproducible methods that can be employed to examine the dynamics of both nonmuscle and muscle myosins. This report outlines the procedures to purify and characterize both M5a and NM2b, representing unconventional and conventional myosins, respectively. This is followed by a discussion of some of the myosin-specific adaptations, which can be performed to achieve successful capturing of motion in the two types of the assay.

Expression and molecular biology
The cDNA for the myosin of interest must be cloned onto a modified pFastBac1 vector that encodes for either a C-terminal FLAG-tag (DYKDDDDK) if expressing M5a-HMM, or an N-terminal FLAG-tag if expressing the full-length molecule of NM2b^{23,43,44,45,46}. C-terminal FLAG-tags on NM2b results in a weakened affinity of the protein for the FLAG-affinity column. In contrast, the N-terminally FLAG-tagged protein usually binds well to the FLAG-affinity column²³. The N-terminally tagged protein retains enzymatic activity, mechanical activity and phosphorylation-dependent regulation²³.

In this paper, a truncated mouse M5a heavy meromyosin (HMM)-like construct with a GFP between the FLAG-tag and the C-terminus of the myosin heavy chain was used. Note that unlike NM2b, M5a-HMM can be successfully tagged and purified with either N- or C-terminal FLAG tags and in both cases the resulting construct will be active. The M5a heavy chain was truncated at amino acid 1090 and contains a three amino acid linker (GCG) between the GFP and the coiled-coil region of the M5a⁴⁷. No additional linker was added between the GFP and FLAG-tag. M5a-HMM was co-expressed with calmodulin. The full-length human NM2b construct was co-expressed with ELC and RLC. The N-termini of the RLC was fused with a GFP via a linker of five amino acids (SGLRS). Directly attached to the FLAG-tag was a HaloTag. Between the HaloTag and the N-terminus of the myosin heavy chain was a linker made of two amino acids (AS).

Both myosin preparations were purified from one liter of Sf9 cell culture infected with baculovirus at a density of approximately 2 x 10⁶ cells/mL. The volumes of the baculovirus for each subunit depended on the virus's multiplicity of infection as determined by the manufacturer's instructions. In the case of M5a, cells were co-infected with two different baculoviruses-one for calmodulin, and one for M5a heavy chain. In the case of the NM2b, cells were co-infected with three different viruses-one for ELC, one for RLC, and one for NM2b heavy chain. For labs working with a diversity of myosins (or other multi-complex proteins), this approach is efficient since it allows for many combinations of heavy and light chains and commonly used light chains such as calmodulin can be co-transfected with many different myosin heavy chains. All cell work was completed in a biosafety cabinet with proper sterile technique to avoid contamination.

For the expression of both M5a and NM2b, the Sf9 cells producing the recombinant myosins were collected 2-3 days post-infection, via centrifugation, and stored at -80 °C. Cell pellets were obtained by centrifuging the co-infected Sf9 cells at 4 °C for 30 min at 2,800 x g. The protein purification process is detailed below.

Protocol

1. Protein purification

1. Cell lysis and protein extraction
   1. Prepare a 1.5x Extraction Buffer based on Table 1. Filter and store at 4 °C.
   2. Begin thawing the cell pellets on ice. While the pellets are thawing, supplement 100 mL of Extraction Buffer with 1.2 mM dithiothreitol (DTT), 5 µg/mL leupeptin, 0.5 µM phenylmethylsulfonyl fluoride (PMSF) and two protease inhibitor tablets. Keep on ice.
   3. Once the pellet has thawed, add 1 mL of the supplemented Extraction Buffer per 10 mL of cell culture. For example, if the cell pellets were formed from 500 mL of cell culture, then add 50 mL of supplemented Extraction Buffer to the pellet.
   4. Sonicate the cell pellets while keeping them on ice. For each pellet, use the following conditions: 5 s ON, 5 s OFF, duration of 5 min, power 4-5.
   5. Collect all the homogenized lysate into a beaker and add ATP (0.1 M stock solution; pH 7.0) such that the final concentration of ATP is 1 mM. Stir for 15 min in a cold room. The ATP dissociates active myosin from actin, allowing it to be separated in the following centrifugation step. It is, therefore, essential to proceed to the next step immediately to minimize the possibility for ATP depletion and rebinding to actin.
   6. Centrifuge the lysates at 48,000 x g for 1 h at 4 °C. While this is occurring, begin washing 1-5 mL of a 50% slurry of Anti-FLAG affinity resin (for a pellet formed from 1 L of cells) with 100 mL phosphate-buffered saline (PBS), according to the manufacturer's instructions. For example, for 5 mL of resin, wash 10 mL of a 50% slurry. In the final wash step, resuspend the resin with 1-5 mL of PBS with enough volume to create a 50% slurry.
   7. Following lysate centrifugation, combine the supernatant with the washed resin slurry and rock gently in the cold room for 1-4 h. While waiting, make the buffers described in Table 1 and keep them on ice.
2. FLAG affinity purification preparation
   1. Centrifuge the solution in step 1.7 at 500 x g for 5 min at 4 °C. The resin will be packed at the bottom of the tube. Without disturbing the resin, remove the supernatant.
   2. Resuspend the resin in 50 mL of Buffer A as detailed in Table 1 and centrifuge at 500 x g for 5 min at 4 °C. Without disturbing the resin, remove the supernatant.
   3. Resuspend the resin in 50 mL of Buffer B as detailed in Table 1 and centrifuge at 500 x g for 5 min at 4 °C. Repeat this step once more and resuspend the resin in 20 mL of Buffer B. Then, mix the resin and the buffer thoroughly by gently inverting the tube by hand approximately 10 times.
3. Protein elution and concentration
   1. Make 30 mL of Elution Buffer as described in Table 1 and let it chill on ice.
   2. Set up the elution column in a cold room. Gently pour the resin slurry into the column. Wash the column with 1-2 column volumes of Buffer B as the resin packs on the bottom, ensuring that the resin does not dry out.
   3. Flow 1 mL of the Elution Buffer through the resin and collect the flow-through in a 1.5 mL tube. Repeat such that 12, 1 mL fractions are collected.
   4. At this point, perform a crude Bradford test on the fractions to qualitatively determine which fractions are the most concentrated⁴⁸. On one row of a 96-well plate, pipette 60 µL 1x Bradford reagent. As fractions are collected, mix 20 µL of each fraction per well. A darker blue coloration indicates the more concentrated fractions.
   5. In a 50 mL tube, collect the remaining protein by gently pipetting the remaining Elution Buffer through the column, to release any remaining myosin bound to the resin in the column flowthrough. This flow-through will be concentrated in the next step. Ensure that the resin is then regenerated for reuse and stored according to the manufacturer's instructions.
   6. Pool the three most concentrated fractions and further concentrate the flow-through in the 50 mL tube as well as the remaining 1 mL fractions using a 100,000 MWCO concentrating tube. Load the pooled sample onto the concentrating tube and centrifuge at 750 x g for 15 min at 4 °C and repeat until all eluted protein has been concentrated to a final volume of approximately 0.5-1 mL.
      ​NOTE: This pore size allows for the retention of the myosin molecules, which have masses several times the molecular weight cutoff. The light chains remain tightly bound to the motor domains during this time course of concentration, as verified by performing SDS-PAGE gel electrophoresis on the final product.
4. Dialysis and flash-freezing
   1. Make 2 L of Dialysis Buffer, as described in Table 1. Load the sample in a dialysis bag or chamber and dialyze overnight in the cold room. Note that the composition of the dialysis buffers differs for NM2b and M5a.
      ​NOTE: In the case of NM2b, the purpose of this dialysis step is to form myosin filaments in the low ionic strength buffer. Sedimentation of these filaments then provides an additional purification step and allows for the concentration of the sample. There will, therefore, be a visible white precipitate in the dialysis chamber the next day. These filaments will be collected by centrifugation and depolymerized in step 5.1. In the case of M5a-HMM, after the overnight dialysis, the protein will be sufficiently pure for the use in subsequent assays. Further purification steps such as gel filtration or ionic exchange chromatography can be performed, if required. For M5a recovery after dialysis, go to step 5.2.
5. Recovering myosin after dialysis
   1. For NM2b, carefully unload the entire sample from the dialysis bag or chamber and centrifuge at 4 °C for 15 min at 49,000 x g to collect the myosin filaments. Discard the supernatant and incrementally add the Storage Buffer to the pellet as described in Table 1 until it has dissolved. Gentle up and down pipetting helps to solubilize the pellet. Normally, this does not require more than 500 µL per tube. After ensuring that the pellet is fully dissolved in the high ionic strength storage buffer, an additional centrifugation step (15 min at 49,000 x g) can be performed to remove unwanted aggregates if required, since the myosin will now be unpolymerized and will remain in the supernatant.
   2. For M5a-HMM, carefully collect the entire sample from the dialysis chamber and centrifuge at 4 °C for 15 min at 49,000 x g in case any unwanted aggregates are present. Take the supernatant.
6. Concentration determination and flash-freezing
   1. To determine the concentration of the product, measure the absorbance using a spectrophotometer at wavelengths 260, 280, 290, and 320 nm. Calculate the concentration in mg/mL (c_mg/mL) with Equation 1, where A₂₈₀ represents the absorption at 280 nm and A₃₂₀ represents the absorption at 320 nm. The resulting concentration in mg/mL can be converted into µM of myosin molecules with Equation 2, where M is the molecular weight of the entire protein (including the heavy chains, light chains, fluorophores, and all tags).
      c_mg/mL = (A₂₈₀ - A₃₂₀) / ε   (1)
      μM molecules = 1000c_mg/mL/M   (2)
      NOTE: If a dilution is necessary, then it must be done in a high ionic strength buffer. The extinction coefficient (ε) can be determined by importing the amino acid sequence of the protein into a program such as ExPASy. Typical yield for the M5a-HMM is approximately 0.5-1 mL of 1-5 mg/mL protein and for the full-length NM2b is 0.5-1 mL of 0.5-2 mg/mL. The extinction coefficient for the M5a-HMM used in this paper was 0.671. The extinction coefficient for the NM2b used in this paper was 0.611.
   2. Store the purified myosin in one of the two ways. Aliquot between 10-20 µL into a thin-walled tube, such as a polymerization chain reaction tube, and drop the tube into a container of liquid nitrogen for flash-freezing. Alternatively, directly pipette between 20-25 µL of myosin into liquid nitrogen and store the frozen beads of protein in sterile cryogenic tubes. In either case, the resulting tubes can be stored in -80 °C or liquid nitrogen for future use.
      ​NOTE: Since both motility assays described below require very small amounts of protein, storage in small aliquots, as described, is economical.

2. Gliding actin filament assay

1. Coverslip preparation
   1. Make a 1% nitrocellulose solution in amyl acetate.
   2. Obtain a tissue culture dish (150 x 25 mm) and add a circular filter paper (125 mm diameter) to the bottom of the dish.
   3. Load eight No. 1.5 thickness 22 mm square coverslips onto a rack and wash with approximately 2-5 mL of 200-proof ethanol followed by 2-5 mL of distilled water (dH₂O). Repeat this washing step, ending with water. Then, dry the coverslips completely using a filtered air-line or N₂-line.
   4. Take one coverslip and slowly pipette 10 µL of the 1% nitrocellulose solution along one edge of the slip. Then, in one smooth motion, smear it across the rest of the coverslip using the side of a smooth-sided 200 µL pipette tip. Place this coverslip on the tissue culture dish with the nitrocellulose side up. Repeat for the remaining coverslips and allow them to dry while preparing the remaining reagents and use coverslips within 24 h after coating.
2. Chamber preparation
   1. Wipe a microscope slide with an optical lens paper to clean off large debris. Cut two pieces of double-sided tape, approximately 2 cm in length.
   2. Place one piece along the middle of the long edge of the microscope slide. Ensure that the edge of the tape aligns with the edge of the slide. Place the second piece of tape roughly 2 mm below the first piece of tape such that the two are parallel and aligned. This creates a flow chamber that can hold approximately 10 µL of solution (see Figure 1).
   3. Take one of the nitrocellulose-coated coverslips from Part 1. Carefully stick the coverslip onto the tape such that the side coated with nitrocellulose is making direct contact with the tape, (see Figure 1). Using a pipette tip, gently press down on the slide-tape interface to ensure that the coverslip has properly adhered to the slide. Cut the excess tape hanging over the edge of the slide with a razor blade.
3. Actin preparation
   1. Make 20 µM F-actin by polymerizing globular actin (G-actin) in polymerization buffer (50 mM KCl, 2 mM MgCl₂, 1 mM DTT, 25 mM MOPS (pH 7.0)) at 4 °C overnight.
   2. Dilute F-actin to 5 µM in motility buffer (20 mM MOPS, 5 mM MgCl₂, 0.1 mM EGTA, 1 mM DTT (pH 7.4)). Label with at least 1.2x molar excess of rhodamine-phalloidin. Leave (covered in aluminum foil) for at least 2 h on ice. This can be used for up to 1-2 months, stored on ice.
4. Performing the myosin 5a gliding actin filament assay
   ​NOTE: In this section, the details of the myosin 5a (HMM) gliding assay are provided.
   1. Prepare the solutions for myosin 5a described in Table 2 and keep them on ice.
   2. Flow in 10 µL of the myosin 5a (50-100 nM) through the flow chamber and wait for 1 min.
   3. Flow in 10 µL of the 1 mg/mL BSA in 50 mM MB with 1 mM DTT ("low salt" buffer). Repeat this wash two more times and wait for 1 min after the third wash. Use the corner of a tissue paper or filter paper to wick the solution through the channel by gently placing the corner of the paper at the flow chamber exit.
   4. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times.
   5. Flow in 10 µL of the black actin solution (5 µM F-actin, 1 µM calmodulin, and 1 mM ATP in 50 mM MB with 1 mM DTT) to eliminate "dead heads", as discussed in the Discussion section.
      1. Pipette the solution with a 1 mL syringe and 27 G needle to shear the actin filaments before introducing the solution to the chamber. Repeat this step two more times and wait for 1 min after the third time. Approximately 20 pipetting events are sufficient.
      2. To perform the "dead head" spin, add a stoichiometric amount of F-actin to myosin in the presence of 1 mM ATP and 1 mM MgCl₂ at a salt concentration of 500 mM. Then ultracentrifuge at 480,000 x g for 15 min at 4 °C. The dead myosin will be in the pellet.
   6. Flow in 50 µL of 50 mM MB with 1 mM DTT and 1 mM ATP to deplete the chamber of free actin filaments.
   7. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times to deplete the chamber of any ATP.
   8. Flow in 10 µL of 20 nM rhodamine actin (Rh-Actin) solution containing 1 mM DTT in 50 mM MB and wait for 1 min to allow rigor binding of actin filaments to the myosin 5a attached to the surface of the coverslip.
   9. Wash with 10 µL of 50 mM MB with 1 mM DTT to wash away Rh-Actin filaments not bound to the surface. Repeat this wash two more times.
   10. Flow in 30 µL of Final Buffer.
   11. Record images on a fluorescence microscope using an excitation wavelength of 561 nm to visualize Rh-Actin. An appropriate exposure time is 200 ms at 1.4 mW laser power for a total acquisition duration of 0.5-1 min.
       NOTE: Ensure that the acquisition rate is scaled appropriately to the speed of the moving filaments. An important consideration before collecting data for use with tracking programs is the acquisition frame rate. Subpixel movements between frames will result in an overestimate of the velocity, and movements of several hundred nanometers are required to obtain accurate values. An optimal acquisition rate features actin gliding for at least one pixel distance between frames. In the case of the TIRF microscope used for the imaging here, this threshold translates to 130 nm; therefore, a myosin expected to travel 1 µm/s must be imaged at a rate of 5 frames/s (0.2 s interval) to achieve 200 nm of movement while a myosin expected to travel 10 nm/s requires 0.05 frames/s (20 s intervals). Data can therefore be downsampled at this stage if necessary (see Discussion for more details).
5. Performing the nonmuscle myosin 2b gliding actin filament assay
   ​NOTE: In this section, the details of the full-length nonmuscle myosin 2b gliding assay are provided. The nonmuscle myosin 2b gliding actin filament assay protocol is different from the myosin 5a protocol at certain steps. Ensure that the correct buffers are used for each of these steps. For example, the NM2b assay requires attachment of myosin to the coverslip in high salt buffer while the M5a can be attached to the coverslip in high or low salt buffers. Additionally, the M5a gliding actin filament assay uses a lower concentration of myosin to mitigate the frequency of actin filaments breaking apart during acquisition.
   1. Prepare the solutions for NM2b as described in Table 2 and keep them on ice.
   2. Flow in 10 µL of the nonmuscle myosin 2b (0.2 µM) in 500 mM Motility Buffer (MB) ("high salt" buffer) and 1 mM dithiothreitol (DTT) through the flow chamber and wait for 1 min.
      NOTE: The high salt buffers dissociate myosin filaments and allow for the attachment of single myosin molecules to the surface, as nonmuscle myosin 2b can polymerize into filaments at ionic concentration <150 mM.
   3. Flow in 10 µL of the 1 mg/mL bovine serum albumin (BSA) in 500 mM MB with 1 mM DTT ("high salt" buffer) as described in Table 2. Repeat this wash two more times and wait for 1 min after the third wash. Use the corner of a tissue paper or filter paper to wick the solution through the channel.
   4. Wash with 10 µL of 500 mM MB with 1 mM DTT as described in Table 2. Repeat this wash two more times.
   5. Flow in 10 µL of the black actin solution as described in Table 2 to eliminate "dead heads," as discussed further in the Discussion section. The black actin solution contains 5 µM of unlabeled F-actin, 1-10 nM MLCK, 1 mM ATP, 0.2 mM CaCl₂, 1 µM CaM, and 1 mM DTT in 50 mM NaCl motility buffer to phosphorylate the nonmuscle myosin 2b on the surface of the chamber.
      1. Pipette the solution with a 1 mL syringe and 27 G needle to shear the actin filaments before introducing the solution to the chamber. Repeat this step two more times and wait for 1 min after the third time. Approximately, 20 pipetting events are sufficient.
   6. Flow in 50 µL of 50 mM MB with 1 mM DTT and 1 mM ATP to deplete the chamber of free actin filaments.
   7. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times to deplete the chamber of any ATP.
   8. Flow in 10 µL of 20 nM Rh-Actin solution containing 1 mM DTT in 50 mM MB and wait for 1 min to allow rigor binding of actin filaments to the nonmuscle myosin 2b attached to the surface of the coverslip.
   9. Wash with 10 µL of 50 mM MB with 1 mM DTT to wash away Rh-Actin filaments not bound to the surface. Repeat this wash two more times.
   10. Flow in 30 µL of Final Buffer. For nonmuscle myosin 2b gliding actin filament assay, the Final Buffer also includes calmodulin, CaCl₂, and myosin light chain kinase to provide full phosphorylation of the nonmuscle myosin 2b during video imaging. 0.7% methylcellulose can also be included in the Final Buffer if actin filaments are only loosely bound or are not bound to the surface. This is discussed further in the Discussion section.
   11. Record images on a fluorescence microscope using an excitation wavelength of 561 nm. An appropriate exposure time is 200 ms at 1.4 mW laser power for a total acquisition duration of 0.5 -3 min.
       NOTE: Ensure that the acquisition rate is scaled appropriately to the speed of the moving filaments. An important consideration before collecting data for use with tracking programs is the acquisition frame rate. Subpixel movements between frames will result in an overestimate of the velocity, and movements of several hundred nanometers are required to obtain accurate values. An optimal acquisition rate features actin gliding for at least one pixel distance between frames. In the case of the TIRF microscope used for the imaging here, this threshold translates to 130 nm; therefore, a myosin expected to travel 1 µm/s must be imaged at a rate of 5 frames/s (0.2 s interval) to achieve 200 nm of movement while a myosin expected to travel 10 nm/s requires 0.05 frames/s (20 s intervals). Data can therefore be down-sampled at this stage, if necessary (see Discussion for more details).

3. Single molecule TIRF assay

1. Coverslip preparation
   1. Divide the stock powder into 10 mg aliquots (in 1.5 mL tubes) of methoxy-Peg-silane (mPEG) and 10 mg aliquots of biotin-Peg-silane (bPEG). Store at -20 °C in a sealed, moisture-free container and use within 6 months.
   2. Load eight No. 1.5H (high precision) thickness 22-mm square coverslips onto a rack and wash with 2-5 mL of 200-proof ethanol followed by 2-5 mL of distilled water. Repeat this washing step, ending with water. Then, dry the coverslips completely using an air-line or N₂ and plasma-clean with argon for 3 min.
   3. Place the clean coverslips on filter paper (90 mm) in a tissue culture dish (100 x 20 mm) and incubate in a 70 °C oven while performing the following steps.
      ​NOTE: The plasma cleaning can be replaced with other chemical cleaning methods⁴⁹.
   4. Prepare 80% ethanol solution with dH₂O and adjust the pH to 2.0 using HCl. Add 1 mL of this to a 10 mg aliquot of mPEG and 1 mL to a 10 mg aliquot of bPEG. Vortex to dissolve, which should not take more than 30 s.
   5. Take 100 µL of the bPEG solution and add 900 µL of 80% ethanol (pH 2.0). This solution is 1 mg/mL bPEG. Then, make a solution of both the PEGs as follows, mixing thoroughly.
      1. 200 µL of 10 mg/mL mPEG (final concentration: 2 mg/mL).
      2. 10 µL of the 1 mg/mL bPEG (final concentration: 10 µg/mL).
      3. 790 µL of the 80% ethanol (pH 2.0) solution.
   6. Take the coverslips out of the oven. Carefully dispense 100 µL of the PEG solution onto the center of each coverslip, ensuring that only the top surface is wet. Then, place the slips back in the oven and incubate for 20 to 30 min.
   7. When the coverslips begin to take on a holey appearance, with small circles apparent across the surface, remove them from the oven.
   8. Wash each coverslip with 100% ethanol, dry with an air-line, and place back in the oven. Incubate only for the time required to create chambers in step 2.
2. Chamber preparation
   1. Clean a microscope slide for use in making the chamber. Cut two pieces of double-sided tape, approximately 2 cm in length.
   2. Place one piece along the middle of the long edge of the microscope slide. Ensure that the edge of the tape aligns with the edge of the slide. Place the second piece of tape roughly 2 mm below the first piece of tape such that the two are parallel and aligned.
   3. Take one of the functionalized coverslips from the oven (created in 3.1). Carefully stick the coverslip onto the tape such that the side coated with PEG is face down and making direct contact with the tape, as shown in Figure 1. Using a pipette tip, gently press down on the slide-tape interface to ensure that the coverslip has properly adhered to the slide.
   4. Cut the excess tape hanging over the slide with a razor blade. These chambers can be used immediately or placed pairwise into a 50 mL tube and stored in a -80 °C freezer for future use. It is important to store immediately or the surface will degrade.
3. Performing the myosin 5a TIRF microscopy assay
   1. Prepare the solutions for myosin 5a inverted motility assay described in Table 3 and keep them on ice.
   2. Wash the chamber with 10 µL of 50 mM MB with 1 mM DTT.
   3. Flow in 10 µL of the 1 mg/mL BSA in 50 mM MB with 1 mM DTT. Repeat this wash two more times and wait for 1 min after the third wash. Use the corner of a tissue paper or filter paper to wick the solution through the channel.
   4. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times.
   5. Flow in 10 µL of the NeutrAvidin solution in 50 mM MB with 1 mM DTT and wait for 1 min.
   6. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times.
   7. Flow in 10 µL of biotinylated rhodamine actin (bRh-Actin) containing 1 mM DTT in 50 mM MB and wait for 1 min. For this step, use a large-bored pipette tip and avoid pipetting up and down to minimize shearing of the fluorescent actin filaments to ensure that long actin filaments can be attached to the surface (20-30 µm or longer). An effective alternative is cutting the cone of a standard pipette tip (with an opening of ≈1-1.5 mm).
   8. Wash with 10 µL of 50 mM MB with 1 mM DTT. Repeat this wash two more times.
   9. Flow in 30 µL of Final Buffer with 10 nM myosin 5a added, then immediately load onto the TIRF microscope and record after finding the optimum focus for TIRF imaging modality. Exposure times between 100-200 ms are appropriate at 1.4 mW laser power for the actin and GFP-labeled myosin. An appropriate acquisition time for velocity analysis is 3 min.
4. Performing the nonmuscle myosin 2b TIRF microscopy assay
   ​NOTE: In this section, the details of the nonmuscle myosin 2b TIRF assay using polymerized and phosphorylated filaments are provided. Detailed protocol (sections 4.1-4.3) for phosphorylating and polymerizing the nonmuscle myosin-2b in a tube is included.
   1. To phosphorylate the purified NM2b, make a 10x kinase mix with the following conditions: 2 mM CaCl₂, 1 µM CaM, 1-10 nM MLCK, and 0.1 mM ATP. This can be brought to volume with 500 mM MB with 10 mM DTT. Add the 10x kinase mix to the myosin at a volumetric ratio of 1:10 and allow this to incubate for 20-30 min at room temperature. Typically, the myosin concentration for this step is 1 µM.
   2. To polymerize the phosphorylated myosin into filaments, lower the salt concentration of the NM2b to 150 mM NaCl. To do so, make a 1x motility buffer (1x MB) with no salt by diluting the 4x MB four times in dH₂O. This 1x MB can be used to lower the salt concentration because the NM2b was frozen in a 500 mM salt buffer.
   3. For every 3 µL of stock NM2b, add 7 µL of 1x MB to lower the salt concentration to 150 mM NaCl and incubate on ice for 20-30 min to form NM2b filaments.
      ​NOTE: The order in sections 4.1-4.3 is not crucial as long as the NM2b is phosphorylated and the final salt concentration is 150 mM. Incubation on the order of 30 min-1 h allows enough time for complete phosphorylation and polymerization.
   4. Prepare the solutions for nonmuscle myosin 2b inverted motility assay described in Table 3 and keep them on ice.
   5. Wash the chamber with 10 µL of 150 mM MB with 1 mM DTT.
   6. Flow in 10 µL of the 1 mg/mL BSA in 150 mM MB with 1 mM DTT Repeat this wash two more times and wait for 1 min after the third wash. Use the corner of a tissue paper or filter paper to wick the solution through the channel.
   7. Wash with 10 µL of 150 mM MB with 1 mM DTT. Repeat this wash two more times.
   8. Flow in 10 µL of the NeutrAvidin solution in 150 mM MB with 1 mM DTT and wait for 1 min.
   9. Wash with 10 µL of 150 mM MB with 1 mM DTT. Repeat this wash two more times.
   10. Flow in 10 µL of bRh-Actin and wait for 1 min. For this step, use a large-bored pipette tip and avoid pipetting up and down to minimize shearing of the fluorescent actin filaments, to ensure that long actin filaments can be attached to the surface (20-30 µm or longer). An effective alternative is cutting the cone of a standard pipette tip.
   11. Wash with 10 µL of 150 mM MB with 1 mM DTT. Repeat this wash two more times.
   12. Flow in 10 µL of the nonmuscle myosin 2b solution (approximately 30 nM) and wait for 1 min.
   13. Wash with 10 µL of 150 MB with 1 mM DTT. Repeat this wash two more times.
   14. Flow in 30 µL of Final Buffer, then immediately load onto the TIRF microscope and record after finding the optimum focus for TIRF imaging modality. Exposure times between 100-200 ms are appropriate at 1.4 mW laser power for the actin and GFP-labeled myosin. An appropriate acquisition time for velocity analysis is 3 min.

4. Image analysis

1. Image analysis for gliding actin filament assay
   NOTE: The images can be analyzed using the software and manuals linked in the List of Materials. It is important to note that the program described here requires TIFF-stacks for analysis. The process for analyzing the gliding actin filament assay is as follows⁵⁰.
   1. Upload raw movie stacks into a specified folder structure and input the top-most directory of the movie folders into the program.
      ​NOTE: The program analyzes the files throughout this directory and subdirectories, treating the lowest-level directories as replicates. Average statistics for each group of replicates will be produced. In this case, a single movie was used for each myosin. When characterizing a novel myosin or investigating a new experimental condition, it is recommended to analyze movies from three field-of-views (FOV) per chamber for a total of three chambers and to repeat this workflow for three preparations of the myosin being investigated.
   2. Use the script "stack2tifs" in conjunction with the user-inputted frame rate to convert each TIFF stack into a folder containing a series of individual TIFF files and a corresponding metadata.txt file containing the start time of each frame. For data not in the TIFF stack format, a conversion must first be applied using software such as those listed in the Table of Materials.
      NOTE: This script is the part of the software package. The information of the script can be found here:  "https://github.com/turalaksel/FASTrack/blob/master/bin/stack2tifs"
   3. Use the -px parameter, which is the pixel size (in nm) during acquisition. In this case, the pixel size is 130 nm. Use the -xmax and -ymax parameters for scaling the axes for the scatter plot outputs. These correspond to the longest plotted filament length and the maximum plotted velocity (in nm/s).
      ​NOTE: These are estimated values and can be set to higher-than-expected values to ensure data are contained in the plot. Following analysis, the raw data can also be exported for use in other statistical or graphing software for viewing and analysis.
   4. Use the -minv parameter, which is a minimum velocity cutoff parameter, to define the filaments that are not moving, and can, therefore, be excluded from the analysis. For a slow myosin such as NM2b, this parameter must be low (in this example, 5 nm/s) to avoid cutting out true gliding movements. For a fast myosin such as M5a, this parameter can be higher (in this example, 100 nm/s) to apply a more stringent filter, while retaining the true gliding speed distribution.
   5. Usethe -pt cutoff parameter to identify smooth movement. For each sampling window, a value is calculated equivalent to 100 x Velocity Standard Deviation/Mean velocity. Tracks with higher values than the cutoff, have more variable velocities and are excluded from further analysis. In this example, a cutoff value of 33 was used. Tracks with higher values have more variable velocities and are excluded from further analysis.
   6. Use -maxd to set a maximum allowed linkage distance between frames. This is a calculated frame-to-frame distance moved by the centroid of the filament in units of nm. It can be useful for excluding sporadic movements or incorrect linkage between filaments. In the examples here, the parameter was left on the default value of 2,000 nm.
2. Image analysis for TIRF microscopy assay
   NOTE: The process for analyzing the single molecule TIRF assay on the imaging software specifically listed in the Table of Materials is as follows²⁹.
   1. Click and drag the recorded microscopy video to the software's workspace to open it⁵¹. Then, split the acquisition channels. Click on Image > Color > Split Channels.
      NOTE: In the event of appreciable stage drift during the acquisition, images must be stabilized to correct instrumental drift on the imaging plane. In this case, no compensation for Z-axis drift was used as the microscope used to obtain this data stabilizes the Z-focus well. To stabilize the image on the image analysis program, install the appropriate stabilizer plug-in that is linked on the List of Materials. The image stabilizer assumes fixed positions for the objects in the image and uses a rolling average of the previous frames as a reference. The recommended procedure is therefore to begin with the channel containing images of labeled actin, since this is in a fixed position.
   2. Click on Plugins, then find Image Stabilizer; ensure that Translation is selected and keep the default settings. Check the box next to Log Transformation Coefficients. Applying this Log step allows for the calculated shift parameters to be applied to the other channel in the next step. Allow for the process to complete.
   3. Then, open the channel with labeled myosin and apply the stabilization by clicking on Plugins > Image Stabilizer Log Applier. If images of actin cannot be acquired during the same acquisition due to a requirement for higher rate imaging in a single channel, drift stabilize the stack of images by selecting a region that contains static objects such as a biotinylated fiducial marker or fluorophores bound non-specifically to the biotin-PEG surface. This region can be cropped from the original stack and stabilized, followed by application of the resulting shift values to the original stack.
      ​NOTE: In practice, the drift observed in motility experiments will be negligible relative to the motion of myosins which move at several hundred nm/s, but for the slowest myosins this becomes an important consideration.
   4. Then, open TrackMate, click on Plugins; then, in its dropdown menu click on Tracking and finally on TrackMate. At this point, the image analysis is subject to optimization based on the parameters of the fluorophore and assay conditions. However, ideal starting parameters are as follows.
      1. Calibration settings: keep all of the default values.
      2. Detector: LoG detector.
      3. Estimated blob diameter: 0.5-1.0 micron.
      4. Threshold: 25-200. (This can be determined by clicking on Preview after choosing a number to see whether the detected spots match up to the movie and adjusting appropriately.)
      5. Initial thresholding: not set.
      6. View: HyperStack Displayer.
      7. Set filters on spots: not set.
      8. Tracker: Simple LAP tracker.
         ​NOTE: These depend on frame rate and myosin velocity and must be large enough to connect subsequent positions while excluding unwanted connections between different particles.
      9. Linking max distance: 1.0 micron.
      10. Gap-closing max distance: 1.0 micron.
      11. Gap-closing max frame gap: 1.
      12. Set filters on tracks: Track Displacement (>0.39-to include only spots moving more than 3 pixels), Spots in tracks (>3-to include only tracks with at least 3 spots). Other filters such as Minimal Velocity may be introduced to exclude spots that stall for long periods. The results of filtering must be checked by visual inspection of tracks to ensure that spurious tracks (i.e., myosin movement in the background that is not along an actin track) are removed while retaining the tracks associated with actin.
   5. Once the Display Options screen comes up, click on Analysis for the relevant outputs. Save the three tables produced (Track Statistics, Links in Track Statistics, and Spots in Track Statistics). The Track Statistics table will contain the velocity and displacement data that can then be subsequently analyzed to characterize a novel protein or the effects of a certain experimental condition, for example.

Representative Results

The purification of myosin can be evaluated by performing reducing sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel-electrophoresis as shown in Figure 2. While this figure represents the final, post-dialyzed myosin, SDS-PAGE can be performed on aliquots from the various stages of the purification procedure to identify any products lost to the supernatant. Myosin 5a HMM has a band in the 120-130 kDa range and the full-length nonmuscle myosin 2b has a band in the 200-230 kDa range, corresponding to heavy chains^29,44. The myosin 5a also has a band near the 17 kDa mark, marking calmodulin. The nonmuscle myosin 2b has a band at approximately 17 kDa, denoting the ELC. Because a GFP-tagged RLC is present in this NM2b preparation, the RLC appears at approximately 47 kDa; however, an unlabeled RLC will be present at approximately 20 kDa if not tagged with a GFP.

The gliding actin filament assay shown in Video 1 and Figure 3 represents the characteristics of an ideal and trackable movie. This gliding actin filament assay features the smooth movement of labeled actin filaments. The black actin wash ensures that the dead myosin heads are removed from the measurement field, further contributing to the overall smooth movement of the actin filaments. The fluorescently labeled filaments are short enough that a single filament does not cross over on itself, which is more optimal for the tracking program. Actin filaments that are too long will cross over other filaments, which can present difficulties to the gliding actin filament assay tracking program. This problem can be avoided by pipetting up and down 10-20 times to shear the actin filaments before loading onto the coverslip.

In the case of NM2b, the use of methylcellulose can significantly improve the quality of the recorded movies as it reduces the diffusion of the actin away from the imaging surface. This is not necessary for M5a because its higher duty ratio allows for a stronger attachment of the actin to the myosin-coated surface. If methylcellulose is used, wicking the solution through the chamber is necessary to ensure the solution flows through. As shown in Video 2, when all other conditions are identical except for the exclusion of methylcellulose, the actin filaments do not remain as closely associated with the myosin-coated surface.

Conversely, the goal for the inverted motility assay shown in Video 3 and Figure 4 is to introduce surface-tethered fluorescent actin filaments upon which myosin movement can be observed. An important requirement of the inverted assay is to ensure that the myosin movement is consistently observed across the FOV, as shown. The use of a mixture of DTT, glucose, catalase, and glucose oxidase can minimize photobleaching to allow for longer measurements⁵². Furthermore, if the acquisition rate for the assay is low, shuttering the illumination light off between acquiring frames can help with excessive photobleaching. Shuttering of the excitation light can be done via a mechanical shutter, or an acousto-optic tunable filter (AOTF).

Figure 1: Preparation of functionalized flow-cell chambers. (A) Begin with a cleaned microscope slide, two pieces of double-sided tape cut to approximately 2 cm, and a functionalized coverslip. (B) Add the tape to the center of the microscope slide. (C) Attach the coverslip to the tape with the coating (i.e., nitrocellulose) facing down and gently press on the overlapping regions with the tape using a plastic pipette tip to ensure that the coverslip has adhered to the chamber. Please click here to view a larger version of this figure.

Figure 2: SDS polyacrylamide gel electrophoresis of expressed NM2b and M5a-HMM. (A) A representative SDS PAGE gel image for a full-length NM2b heavy chain (≈230 kDa) and GFP-RLC (≈47 kDa) and ELC (≈17 kDa). Gel image reproduced and modified from Melli et al. (2018)²⁹. (B) A representative SDS PAGE gel image for an M5a-HMM-like heavy chain (≈120 kDa) and calmodulin (≈17 kDa). Note that the gel in this image does not have a GFP inserted to the C-terminal end. A GFP inserted in the myosin heavy chain increases the molecular weight by ≈27 kDa. Gel image reproduced and modified was originally published in the Journal of Biological Chemistry⁴⁴. Please click here to view a larger version of this figure.

Figure 3: Gliding actin filament assay results acquired via TIRF illumination. (A) Example frame from a movie showing translocation of rhodamine-phalloidin labeled actin filaments (in red) on 0.2 µM NM2b in the presence of 0.7% methylcellulose at 30 °C. Scale bar = 10 µm. (B) Filament tracking image output from the FASTrack program for NM2b for the same FOV as shown in (A) Scale bar = 10 µm. (C) Representative histogram of the acto-NM2b gliding velocity, showing that this sample of NM2b can generate an actin gliding velocity of 77 ± 15 nm/s (mean ± standard deviation; number of tracks = 550). (D) Example frame from a movie showing translocation of rhodamine-phalloidin labeled actin filaments (in red) on 75 nM M5a-HMM. Scale bar = 10 µm. (E) Filament tracking image output by the FASTrack program for M5a-HMM for the same FOV as shown in (D) Scale bar = 10 µm. (F) A representative histogram of the acto-M5a-HMM gliding velocity, showing that this sample of M5a can generate an actin gliding velocity of 515 ± 165 nm/s (mean ± standard deviation; number of tracks = 25098). Please click here to view a larger version of this figure.

Figure 4: Inverted assay results acquired via TIRF illumination. (A) A representative FOV from a two-channel merged image stack showing the movement of NM2b filaments (displayed in green) on biotinylated actin filaments labeled with AF647-phalloidin (displayed in blue) at 30 °C. Polymerized filaments of recombinantly expressed and purified NM2b co-expressed with ELC and GFP-RLC are observed as the green, elongated particles in the FOV. Scale bar = 10 µm. (B) Representative histogram of the velocity of NM2b filaments. Analysis was performed using the image analysis software described in the Table of Materials. Single NM2b filaments have a velocity of 84 ± 22 nm/s (mean ± standard deviation; number of particles tracked = 133), when moving along single actin filaments. (C) Example kymograph of the NM2b filament motion along a single actin filament. Note that some of the regions of the particle shows a "rotation" of the NM2b filament, along the actin filament, which most likely represents the time when one side of the bipolar NM2b filament detaches from the actin filament, as shown previously in Melli et al.²⁹. (D) A representative FOV of the single molecule movement M5a-HMM (displayed in green) on biotinylated actin filaments labeled with rhodamine-phalloidin (displayed in red). Scale bar = 10 µm. (E) Representative histogram of run length of M5a-HMM, fit to a single exponential. Analysis was performed using the image analysis software described in the List of Materials. The characteristic run length is 1.3 µm with a 95% confidence interval of 1.23-1.42 µm in this example. (F) Representative histogram of single molecule M5a-HMM velocity on single actin filaments. Analyzed data output from image analysis shows a mean velocity of 668 ± 258 nm/s (mean ± standard deviation; number of particles tracked = 684). (G) Example kymograph of single molecules of M5a-HMM motion along a single actin filament. Please click here to view a larger version of this figure.

Video 1: Comparison of NM2b and M5a-HMM gliding actin filament assay. The NM2b gliding actin filament assay was performed in the presence of methylcellulose (left; video panel A) and the M5a-HMM in the absence of methylcellulose (right; video panel B) at 30 °C. Note that the time stamp advances faster in the NM2b video panel, compared to the M5a-HMM video panel to show the movement of the rhodamine-labeled actin filaments (red) that is approximately the same. This is since the actual actin translocation velocity of NM2b is close to 7 times slower than that for M5a-HMM (77 nm/s, versus 515 nm/s, respectively, extracted from the Gaussian peak fit to the histogram in Figure 3). Scale bar = 10 µm in both video panels. NM2b data acquired at 0.33 frames per second with 200 ms exposure. M5a-HMM data acquired at 5 frames per second with 200 ms exposure (continuous) and subsequently down-sampled to 1 frame per second. Timestamps were added using the plugin described in the List of Materials. Please click here to download this video.

Video 2: Gliding actin filament assay of NM2b in the absence of methylcellulose. When all other conditions are the same except for the absence of methylcellulose, the actin filaments sometimes do not stick well to the coverslip coated with 0.2 µM NM2b, leading to lower-quality movies with actin filaments "flopping" close to the surface of the NM2b coated coverslip. Scale bar = 10 µm. This can be resolved by introducing methylcellulose to show the smooth motion of the actin filaments, as shown in the left video panel of Video 1 (NM2b gliding actin filament assay). Another alternative is to increase the NM2b concentration to ≈1 µM. This movie was acquired at 0.33 frames per second with 200 ms exposure. Please click here to download this video.

Video 3: Comparison of NM2b and M5a-HMM inverted motility assay. The NM2b inverted motility assay was performed in the presence of methylcellulose and recorded at a rate of 0.33 frames per second with the use of a shutter (left; video panel A), Video panel C shows the same FOV as A, but with particles are identified and tracked using image analysis software. Similarly, the inverted motility assay for M5a-HMM in the absence of methylcellulose was recorded at a rate of 5 frames per second (right; video panel B). Video panel D shows the same FOV as B, but with particles identified and tracked using image analysis software. Scale bars = 10 µm in all video panels. The two lasers were toggled back and forth with the use of a single camera for acquisition. Please click here to download this video.

Buffer Name	Composition	Step(s) Used	Comments
M5a Extraction Buffer	0.3 M NaCl	1.1	Keep on ice.
	15 mM MOPS, pH 7.2
	15 mM MgCl2
	1.5 mM EGTA
	4.5 mM NaN3
NM2b Extraction Buffer	0.5 M NaCl	1.1	Keep on ice.
	15 mM MOPS, pH 7.2
	15 mM MgCl2
	1.5 mM EGTA
	4.5 mM NaN3
Buffer A	0.5 M NaCl	2.2	Keep on ice.
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	3 mM NaN3
	1 mM ATP
	1 mM DTT
	5 mM MgCl2
Buffer B	0.5 M NaCl	2.3	Keep on ice.
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	3 mM NaN3
	1 mM DTT
Elution Buffer	0.5 M NaCl	3.1	Keep on ice.
	0.5 mg/mL FLAG peptide
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	3 mM NaN3
	pH 7.2
M5a Dialysis Buffer	500 mM KCl	4.1	Use cold dH2O to bring to volume.
	10 mM MgCl2
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	1 mM DTT
NM2b Dialysis Buffer	25 mM NaCl	4.1	Use cold dH2O to bring to volume.
	10 mM MgCl2
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	1 mM DTT
NM2b Storage Buffer	0.5 M NaCl	5.1	Keep on ice.
	10 mM MOPS, pH 7.2
	0.1 mM EGTA
	3 mM NaN3

Table 1: Buffers used in protein purification.

Buffer Name	Composition (M5a)	Composition (NM2b)	Step(s) Used (M5a/NM2b)	Comments
4X Motility Buffer (4X MB)	80 mM MOPS, pH 7.2	80 mM MOPS, pH 7.2		Vacuum filter and store in 4°C
	20 mM MgCl2	20 mM MgCl2
	0.4 mM EGTA	0.4 mM EGTA
	pH 7.4	pH 7.4
50 mM Salt Motility Buffer (50 mM MB)	25% v/v 4X MB	25% v/v 4X MB		Vacuum filter and store in 4°C
	50 mM KCl	50 mM NaCl
	Raise to volume with dH2O	Raise to volume with dH2O
500 mM Salt Motility Buffer (500 mM MB)	N/A	25% v/v 4X MB		Vacuum filter and store in 4°C
		500 mM NaCl
		Raise to volume with dH2O
Myosin	0.05-0.1 µM myosin	0.2 µM myosin	4.2/5.2	Keep on ice.
	1 mM DTT	1 mM DTT
	Dilute in 50 mM MB	Dilute in 500 mM MB
1 mg/mL Bovine Serum Albumin (BSA)	1 mg/mL BSA	1 mg/mL BSA	4.3/5.3	Keep on ice.
	Dilute in 50 mM MB	Dilute in 500 mM MB
	1 mM DTT	1 mM DTT
5 µM Unlabeled F-actin in 50 mM MB (black actin)	5 µM unlabeled F-actin	5 µM unlabeled F-actin	4.5/5.5	Keep on ice. Shear actin by pipetting up and down 5-10 times, or by using a syringe.
	1 μM calmodulin (CaM)	1 mM ATP
	1 mM ATP	0.2 mM CaCl2
	Dilute in 50 mM MB	1 μM CaM
		1–10 nM myosin light chain kinase (MLCK)
		Dilute in 50 mM MB
MB with 1 mM DTT and 1 mM ATP	1 mM DTT	1 mM DTT	4.6/5.6	Keep on ice.
	1 mM ATP	1 mM ATP
	Dilute in 50 mM MB	Dilute in 50 mM MB
MB with DTT	1 mM DTT	1 mM DTT	4.4, 4.7, 4.9/5.4, 5.7, 5.9	Keep on ice.
	Dilute in 50 mM MB	Dilute in 50 mM MB
20 nM Rhodamine-Phalloidin F-actin (Rh-Actin)	20 nM Rhodamine-phalloidin F-actin	20 nM Rhodamine-phalloidin F-actin	4.8/5.8	Keep on ice. Do not vortex.
	1 mM DTT	1 mM DTT
	Dilute in 50 mM MB	Dilute in 50 mM MB
Final Buffer	50 mM KCl	0.7% methylcellulose (optional)	4.10/5.10	Add in the glucose, glucose oxidase, and catalase immediately before performing the experiment. Keep on ice.
	20 mM MOPS, pH 7.2	50 mM NaCl
	5 mM MgCl2	20 mM MOPS, pH 7.2
	0.1 mM EGTA	5 mM MgCl2
	1 mM ATP	0.1 mM EGTA
	50 mM DTT	1 mM ATP
	1 μM calmodulin	50 mM DTT
	2.5 mg/mL glucose	1–10 nM MLCK
	100 μg/mL glucose oxidase	0.2 mM CaCl2
	40 μg/mL catalase	1 μM calmodulin
		2.5 mg/mL glucose
		100 μg/mL glucose oxidase
		40 μg/mL catalase

Table 2: Buffers used in gliding assay.

Buffer Name	Composition (M5a)	Composition (NM2b)	Step(s) Used (M5a/NM2b)	Comments
4X Motility Buffer (4X MB)	80 mM MOPS, pH 7.2	80 mM MOPS, pH 7.2		Vacuum filter and store in 4°C
	20 mM MgCl2	20 mM MgCl2
	0.4 mM EGTA	0.4 mM EGTA
	pH 7.4	pH 7.4
50 mM salt Motility Buffer (50 mM MB)	25% v/v 4X MB	25% v/v 4X MB		Vacuum filter and store in 4°C
	50 mM KCl	50 mM NaCl
	Raise to volume with dH2O	Raise to volume with dH2O
150 mM salt Motility Buffer (150 mM MB)		25% v/v 4X MB		Vacuum filter and store in 4°C
		150 mM NaCl
		Raise to volume with dH2O
Myosin		30 nM myosin	See "Final Buffer" Recipe/4.12	Keep on ice.
		1 mM DTT
		Dilute in 150 mM MB
2 mg/mL NeutrAvidin	2 mg/mL NeutrAvidin	2 mg/mL NeutrAvidin	3.5/4.8	Keep on ice.
	1 mM DTT	1 mM DTT
	Dilute in 50 mM MB	Dilute in 150 mM MB
1 mg/mL bovine serum albumin (BSA)	1 mg/mL BSA	1 mg/mL BSA	3.3/4.6	Keep on ice.
	1 mM DTT	1 mM DTT
	Dilute in 50 mM MB	Dilute in 150 mM MB
200 nM rhodamine-phalloidin biotinylated F-actin (bRh-Actin)	200 nM rhodamine-phalloidin biotinylated F-actin	200 nM rhodamine-phalloidin biotinylated F-actin	3.7/4.10	Avoid shearing by not vortexing or pipetting up and down. To mix, gently invert.
	1 mM DTT	1 mM DTT
	Dilute in 50 mM MB	Dilute in 150 mM MB
MB with DTT	50 mM DTT	50 mM DTT	3.2, 3.4, 3.6, 3.8/4.5, 4.7, 4.9, 4.11, 4.13	Keep on ice.
	Dilute in 50 mM MB	Dilute in 150 mM MB
Final Buffer	50 mM KCl	0.7% methylcellulose (optional)	3.9/4.14	Add in the glucose, glucose oxidase, and catalase immediately before performing the experiment. Keep on ice.
	20 mM MOPS, pH 7.2	50 mM NaCl
	5 mM MgCl2	20 mM MOPS, pH 7.2
	0.1 mM EGTA	5 mM MgCl2
	1 mM ATP	0.1 mM EGTA
	50 mM DTT	1 mM ATP
	1 μM calmodulin	50 mM DTT
	2.5 mg/mL glucose	1–10 nM MLCK
	100 μg/mL glucose oxidase	0.2 mM CaCl2
	40 μg/mL catalase	1 μM calmodulin
	10 nM myosin	2.5 mg/mL glucose
		100 μg/mL glucose oxidase
		40 μg/mL catalase

Table 3: Buffers used in TIRF assay.

Discussion

Presented here is a workflow for the purification and in vitro characterization of myosin 5a and nonmuscle myosin 2b. This set of experiments is useful for quantifying the mechanochemical properties of purified myosin constructs in a fast and reproducible manner. Although the two myosins shown here are just two specific examples out of the many possibilities, the conditions and techniques can be applied, with some tailoring, to most myosins and to many other motor proteins.

The protocols discussed here are subject to variations depending on the individual needs of the lab and experiments. For example, as discussed in the Expression and Molecular Biology section, the proteins used in this paper were generated from a co-infection of two or more viruses; however, successful protein expression can also be achieved with multi-expression vectors such as the p-FastBac dual expression vector or the BiGBac expression system⁵³.

Several factors can hamper the successful production of the recombinant protein. To prevent protein degradation, it is imperative that every step of the protein purification is completed at the appropriate temperature, with all centrifugation steps occurring at 4 °C and all other steps being performed on ice. Excess bands may be apparent in the gel of the dialyzed protein product. This could be indicative of degradation or contamination. Subsequent purification via size-exclusion or ionic-exchange chromatography can enhance the purity of the samples⁵⁴. To that effect, it is recommended to save aliquots at each step of the protein purification process for troubleshooting, should there be a low yield of myosin or suspected protein contamination. Sometimes, there can be inadequate binding of the lysate to the resin in step 1.8, which can result in the loss of the myosin to the supernatant in subsequent centrifugations. This can be resolved by varying the duration of resin binding during this step, even leaving it to bind overnight, if necessary. Longer incubation times introduce a greater risk of protein degradation if contaminant proteases are not sufficiently inhibited, and myosins with proteolytically sensitive regions will be adversely affected. Additionally, improper washing of the resin both before and after use may result in the elution of an undesired protein product, so it is imperative that the proper protocols are followed before and after using the FLAG-affinity resin. If the resin is washed immediately after use and stored appropriately, it can be reused up to 20 times.

It is important to note that protein degradation also occurs during the expression stage and shortening expression times may be advantageous in terms of degradation, although this can be at the expense of the total yield. Following the procedures outlined here to monitor the protein sample at different stages of the protocol (i.e., before, during, and after purification) will help to determine the stages necessary for optimization. For myosins that repeatedly resist successful expression and purification, common problems can be co-expression with insufficient or inappropriate light chains as well as improper folding during overexpression. Appropriate light chains must be selected based on known interactions when possible and heavy chain to light chain baculovirus ratios must be tested in small-scale experiments to determine the optimum. For myosins that aggregate or yield little or no soluble active product, co-expression with chaperones can aid in successfully obtaining active protein^54,55.

Purified myosin products inevitably contain a small population of damaged myosin, referred to as "dead heads," which can be addressed in two ways. One method, outlined in this protocol, involves flowing unlabeled, or "black", actin through the chamber in the gliding actin filament assay. Subsequently washing with ATP causes functional myosins to dissociate from the black actin while dead heads will remain bound to this unlabeled actin due to their inability to hydrolyze ATP and due to their high affinity. While performing the black actin wash, a syringe can be used to shear the actin effectively. Additional shearing can be accomplished by vortexing, provided that any resultant bubbles are removed by centrifugation. An alternative method is to selectively pellet the dead heads from the myosin sample by mixing myosin with F-actin and Mg-ATP at high salt (0.5 M) concentrations and sedimenting in a table-top ultracentrifuge. The myosins capable of hydrolyzing ATP under these conditions do not stay bound to the actin due to their low affinity for actin under these high salt conditions and are found in the supernatant, whereas the myosin dead heads remain bound to the actin in the pellet⁵⁶. Similarly, a sedimentation with actin and resuspension of the pellet can also be used to remove myosins which are incapable of binding to actin in the absence of ATP. Note that a small proportion of this type of dead heads will have less of an impact on these types of assays. By doing a nucleotide-free sedimentation and resuspension followed by an ATP-bound centrifugation and resuspension, myosins that are competent for both actin-binding and ATP-dependent release from actin can be isolated.

Motility assays can also be modified in several ways. For example, in the gliding actin filament assay for the NM2b, the NM2b is phosphorylated in the chamber via the addition of MLCK, calmodulin, calcium, and ATP in the black actin step, as well as in the Final Buffer. However, the NM2b can also be phosphorylated in a tube, before performing the assay. By doing so, the percent of phosphorylated NM2b can be quantified by running a native gel with a urea-containing sample buffer or performing mass spectrometry^57,58. The effect of temperature on myosin activity can also be investigated. This can be accomplished by employing an objective heating system on the microscope or an environmental enclosure, so that the flow cell is maintained at constant temperature. Ionic strength is another important consideration. For many myosins, actin affinity and enzymatic activity will be increased at lower ionic strength; for others, higher ionic strength is necessary⁵⁹. In addition to providing valuable information about the myosin mechanism, lowering the ionic strength can enhance motility and make myosin more accessible to investigation with many assays. In contrast, some motors will exhibit electrostatic tethering effects, which will slow motility at lower ionic strengths. Finally, when assaying the movement of NM2b filaments, it is crucial to maintain ionic strength within a narrow range (150-200 mM ionic strength), approximating those found in most cell types. The use of lower ionic strengths results in aggregation of the myosin filaments, while the filaments depolymerize at higher ionic strengths.

With many myosins, particularly those with low duty ratios, the conditions of the final buffer given for M5a would result in the fluorescently labeled actin filaments being only loosely bound to the surface or dissociating altogether. This results in erratic movements that complicate quantification. Better quality movement can often be obtained using methylcellulose (0.7%) in the Final Buffer. Methylcellulose is a viscous crowding agent and forces actin filaments to remain close to the surface even when the density of attached myosin motors is sparse⁶⁰. Similarly, it has been observed that the inclusion of methylcellulose in the final buffer of the single filament motility assay is necessary to observe movement with NM2a, and the same phenomenon was reported for smooth muscle myosin filaments^29,61. This also increases the processivity of NM2b filaments. One potentially unwanted side effect of using methylcellulose in this assay is that the crowding agent properties can promote the lateral association of myosin filaments into bundles. Alternatives to methylcellulose when troubleshooting a lack of movement or loosely bound actin filaments in the gliding actin filament assay are to lower the salt concentration in the motility buffers or to increase the myosin surface density. As stated above, a high ionic strength in the motility buffers has been shown to lower the ability of some myosins to bind to actin^29,34,62.

Another variation of the gliding actin filament assay is the use of antibodies to anchor the myosin onto the glass coverslip. For example, if a GFP is present at the C-terminal end of the myosin construct, an anti-GFP antibody can be used to fix the GFP-myosin to the coverslip^36,63,64. This can aid with obtaining successful motility in situations where the geometry of the system may otherwise hamper actin translocation, such as in the case of testing artificial or short lever arms^54,64. Additionally, the effect of load on translocation velocities can be investigated in the gliding actin filament assay by employing actin-binding proteins such as α-actinin or utrophin^39,50,65. Such a measurement can be useful to compare the effect of load on an ensemble of myosins versus the load-dependent kinetics of a single myosin that can be measured using an optical trapping assay^66,67. This can be accomplished by adding increasing amounts of an actin-binding protein along with the myosin in the initial step. The actin-binding protein binds to the surface and exerts a frictional load on the actin filaments that are being moved by myosins, which results in a graded velocity as the concentration of actin-binding protein on the surface is increased³⁹.

The single molecule/ensemble motility assay can also be adapted to investigate the effect of various actin structures on myosin movement. For example, rather than observing myosin movement on top of single actin filaments, fascin- or α-actinin-mediated actin bundles can be studied as an in vitro reconstitution of the actin filament network found in cells^68,69. The effect of actin-binding proteins such as tropomyosin can also be studied^65,70,71,72.

Of note is the versatility in choosing a label for the single molecule/ensemble motility assay. In this report, a GFP label was used on both the M5a-HMM and NM2b; however, many other labels can be used. Examples include HaloTag or SNAP-tag, which can be genetically fused to the myosin and covalently bind a synthetic dye. The benefit of HaloTag technology lies in its versatility for several experimental adaptations, such as labeling with different colors or adding a biotin affinity tag^29,73. Additionally, the use of quantum dot technology can be employed to improve the resolution of single molecule fluorescence tracking, which also addresses the limitation of GFP's low brightness and the tendency for photobleaching⁷⁴. Tags can be successfully attached to light chains as well as the heavy chain^11,75,76.

For achieving success in the single molecule TIRF motility assay, a key factor is using a well blocked and functionalized surface. A simple method to achieve moderate blocking is to use biotinylated-BSA bound to a nitrocellulose surface. Although this will work well enough to characterize many motors, including M5a, the level of nonspecific binding on such a surface is prohibitive for reproducing clean movement with samples such as NM2b. A key breakthrough in this regard was the transition to PEGylated surfaces doped with biotin-PEG for functionalization⁷⁷. The PEG surfaces provide a far superior level of surface blocking and a defect-free PEGylated surface can remain free of nonspecific binding for very long periods of time. The specific protocol detailed here allows the production of biotinylated PEG surfaces in a matter of hours and if immediately stored as described, the surfaces can be used for several weeks with only a marginal decline in quality.

A key consideration before collecting data for tracking is the acquisition frame rate. Movement between subsequent frames must be large enough to avoid oversampling errors. High sampling rates will yield overestimated velocities due to the division of localization errors by a small time interval and increase the apparent error of the measurement. In cases where the raw data is too finely sampled, the data can be down-sampled by taking every Nth frame to create a new stack and considering the change in frame rate that results. Subpixel movements between frames must be avoided and movements of several hundred nanometers are required to obtain accurate values. In all cases where a new sample is being characterized, the results generated by automated analysis must be compared to a small dataset of manually tracked filaments for consistency.

When analyzing data from single molecule motility experiments, care must be taken when choosing which parameters to measure, how to filter data, and how to fit data. As stated above, the sampling rate can be an important factor when analyzing velocity data. For many myosins, processive runs will be short and well approximated by a straight line. In such cases, the start to endpoint distance of the track may provide a good measure of the run length and this can be divided by the duration of the track to yield a good estimate of velocity. In cases where the tracks are very long and follow curved paths around bent filaments, this type of analysis will yield inaccurate results and a total distance traveled must be used, using an acquisition rate that allows for successive localization points to be sufficiently well spaced to avoid oversampling errors as described above, while being close enough together that the straight line distance between them remains a good approximation of the curve between those points. In addition, for motor proteins with long run lengths in relation to the length of the track, additional statistics such as the Kaplan-Meier estimator must be made when calculating run lengths⁷⁸. The same is true for situations in which photobleaching is sufficiently likely to occur before the end of a processive run. Another phenomenon that can be observed in single molecule fluorescence studies is photoblinking, in which fluorophores switch between the on and off state rapidly and appear to blink. This typically does not occur in these motility experiments; however, if this does occur, the laser intensity and exposure times can be decreased which should minimize the effect. Several chemicals, including β-mercaptoethanol, Trolox, cyclooctateraene, n-propyl gallate, 4-nitrobenzyl alcohol, and 1,4-diazabicyclo[2.2.2]octane can be utilized to mitigate this as well⁷⁹.

In summary, this article presents detailed protocols that are robust in their ability to quantify mechanochemical properties such as actin translocation velocity, myosin translocation velocity, and myosin run length. These assays are reproducible and can be used to determine the quality of the purified myosin even in situations where the motile characteristics are not the specific end goal of the study. In addition, changes such as pH, temperature, and chemical regulators can be introduced to these assays to examine how the mechanochemistry of the studied myosin is affected. Taken together, the actin gliding and inverted motility assays can allow for a better understanding of myosin ensemble behavior and intermolecular variations in molecular motor mechanics and kinetics. The fluorescence microscopy-based assays described here support a reductionist's approach to cytoskeletal research and can be a powerful tool to understand protein-protein dynamics in vitro. Together, data collected from these highly controlled experiments can be used to advise mechanobiologists of key actomyosin behaviors that may be relevant at the cell biological level, and beyond.

Materials

Name	Company	Catalog Number	Comments
1 mL Syringe	BD	309628
2 M CaCl2 Solution	VWR	10128-558
2 M MgCl2 Solution	VWR	10128-298
27 Gauge Needle	Becton Dickinson	309623
5 M NaCl Solution	KD Medical	RGE-3270
Acetic Acid	ThermoFisher Scientific	984303
Amyl Acetate	Ladd Research Industries	10825
Anti-FLAG M2 Affinity Gel	Millipore Sigma	A2220	https://www.sigmaaldrich.com/content/dam/sigma-aldrich/docs/Sigma/Bulletin/a2220bul.pdf
ATP	Millipore Sigma	A7699
Biotinylated G-Actin	Cytoskeleton, Inc.	AB07
Bovine Serum Albumin	Millipore Sigma	5470
bPEG-silane	Laysan Bio, Inc	Biotin-PEG-SIL-3400-1g
Bradford Reagent Concentrate	Bio-Rad	5000006
Calmodulin		PMID: 2985564
Catalase	Millipore Sigma	C40
Cell Line (Sf9) in SF-900 II SFM	ThermoFisher Scientific	11496015	http://tools.thermofisher.com/content/sfs/manuals/bevtest.pdf https://tools.thermofisher.com/content/sfs/manuals/bactobac_man.pdf
Circular Filter Paper - Gliding Assay	Millipore Sigma	WHA1001125
Circular Filter Paper - Inverted Assay	Millipore Sigma	WHA1001090
cOmplete, EDTA-Free Protease Inhibitor Tablets	Millipore Sigma	5056489001	This should be stored at 4 °C. The tablets can be used directly or can be reconstituted as a 25x stock solution by dissolving 1 tablet in 2 mL of distilled water. The resulting solution can be stored at 4 °C for 1-2 weeks or at least 12 weeks at -20 °C.
Concentrating Tubes (100,000 MWCO)	EMD Millipore Corporation	UFC910024	The MWCO of the tube is not necessarily "one size fits all," as long as the MWCO is less than the total molecular weight of the protein being purified. The NM2b herein was concentrated with a 100,000 MWCO tube and the M5a was concentrated with a 30,000 MWCO tube.
Coomassie Brilliant Blue R-250 Dye	ThermoFisher Scientific	20278
Coverslip Rack	Millipore Sigma	Z688568-1EA
Coverslips: Gliding Acting Filament Assay	VWR International	48366-227
Coverslips: Inverted Motility Assay	Azer Scientific	ES0107052
Dialysis Tubing (3500 Dalton MCWO)	Fischer Scientific	08-670-5A	The diameter of the dialysis tube can vary, but the MWCO should be the same. The NM2b used herein was dialyzed in an 18 mm dialysis tube. The tubes can be stored in 20% alcohol solution at 4 °C.
DL-Dithiothreitol	Millipore Sigma	D0632
Double-Sided Tape	Office Depot	909955
DYKDDDDK Peptide	GenScript	RP10586	This can be dissolved in a buffer containing 0.1 M NaCl, 0.1 mM EGTA, 3 mM NaN3, and 10 mM MOPS (pH 7.2) to a final concentration of 50 mg/mL. This can be stored at -20 °C as 300 µL aliquots.
EGTA	Millipore Sigma	E4378
Elution Column	Bio-Rad	761-1550	These can be reused. To clean, rinse the column with 2-3 column volumes of PBS and distilled water. Chill the column at 4° C before use.
Ethanol	Fischer Scientific	A4094
G-actin		PMID: 4254541	G-actin stock can be stored at 200 μM in liquid N2.
Glucose	Millipore Sigma	G8270
Glucose Oxidase	Millipore Sigma	G2133
Glycine Buffer Solution, 100 mM, pH 2-2.5, 1 L	Santa Cruz Biotechnology	sc-295018
HaloTag	Promega	G100A
HCl	Millipore Sigma	320331
KCl	Fischer Scientific	P217-500
Large-Orifice Pipet Tips	Fischer Scientific	02-707-134
Leupeptin Protease Inhibitor	ThermoFisher Scientific	78435
Mark12 Unstained Standard Ladder	ThermoFisher Scientific	LC5677
Methanol	Millipore Sigma	MX0482
Methylcellulose	Millipore Sigma	M0512
Microscope Slides	Fischer Scientific	12-553-10
MOPS	Fischer Scientific	BP308-100
mPEG-silane	Laysan Bio, Inc	MPEG-SIL-2000-1g
Myosin Light Chain Kinase		PMID: 23148220	FLAG-tagged MLCK can be purified the same way that the FLAG-tagged myosin was purified herein.
NaN3	Millipore Sigma	S8032
NeutrAvidin	ThermoFisher Scientific	31050
Nitrocellulose	Ladd Research Industries	10800
NuPAGE 4 to 12% Bis-Tris Mini Protein Gel, 15-well	ThermoFisher Scientific	NP0323PK2
NuPAGE LDS Sample Buffer (4X)	ThermoFisher Scientific	NP0007
Phosphate-Buffered Saline, pH 7.4	ThermoFisher Scientific	10010023
PMSF	Millipore Sigma	78830	PMSF can be made as a 0.1 M stock solution in isopropanol and stored in 4 °C. Isopropanol addition results in crystal precipitation, which can be dissolved by stirring at room temperature. Immediately before use, PMSF can be added dropwise to a rapidly stirring solution to a final concentration of 0.1 mM.
Razor Blades	Office Depot	397492
Rhodamine-Phalloidin	ThermoFisher Scientific	R415	Stock can be diluted in 100% methanol to a final concentration of 200 μM.
Sf9 Media	ThermoFisher Scientific	12658-027	This should be stored at 4° C. Its shelf life is 18 months from the date of manufacture.
Tissue Culture Dish - Gliding Assay	Corning	353025	Each tissue culture dish can hold approximately nine coverslips.
Tissue Culture Dish - Inverted Assay	Corning	353003	Each tissue culture dish can hold approximately four coverslips.
Smooth-sided 200 µL Pipette Tips	Thomas Scientific	1158U38
EQUIPMENT
Centrifuge	ThermoFisher Scientific	75006590
Microscope	Nikon	Model: Eclipse Ti with H-TIRF system with 100x TIRF Objective (N.A. 1.49)
Microscope Camera	Andor	Model: iXon DU888 EMCCD camera (1024 x 1024 sensor format)
Microscope Environmental Control Box	Tokai HIT	Custom Thermobox
Microscope Laser Unit	Nikon	LU-n4 four laser unit with solid state lasers for 405nm, 488nm, 561nm,and 640nm
Mid Bench Centrifuge	ThermoFisher Scientific	Model: CR3i
Misonix Sonicator	Misonix	XL2020
Optima Max-Xp Tabletop Ultracentrifuge	Beckman Coulter	393315
Plasma-Cleaner	Diener electronic GmbH + Co. KG	System Type: Zepto
Sonicator Probe (3.2 mm)	Qsonica	4418
Standard Incubator	Binder	Model: 56
Waverly Tube Mixer	Waverly	TR6E
SOFTWARE
ImageJ FIJI	https://imagej.net/Fiji/Downloads
FAST (Version 1.01)	http://spudlab.stanford.edu/fast-for-automatic-motility-measurements	FAST is available for Mac OSX and Linux based systems.
Image Stabilizer Plugin	https://imagej.net/Image_Stabilizer
ImageJ TrackMate	https://imagej.net/TrackMate
Imaging Software	NIS Elements (AR package)
	http://www.cs.cmu.edu/~kangli/code/Image_Stabilizer.html
	File:TrackMate-manual.pdf
	https://github.com/turalaksel/FASTrack
	https://github.com/turalaksel/FASTrack/blob/master/README.md