Ethical statement
We purified Myosin and actin from fast skeletal muscle from a euthanized rabbit, according to a procedure approved by the Regional Ethical Committee for Animal Experiments in Linköping, Sweden (ref. 17088 − 2020). Before euthanization, the rabbit was anesthetized by an intramuscular injection of 0.25 ml Zoletil (active substances: Zolazepam, 6 mg/kg; Tiletamin, 6 mg/kg och Medetomidin, 0.6 mg/kg). The rabbit was then euthanized by injection of 2ml of penthobarbital (100 mg/ml) in an ear vein. The Linnaeus University veterinary performed all procedures associated with the euthanization. Myosin and actin were from one given rabbit because our focus here was not on actin and myosin function per se but on the performance of a methodological approach. No in vivo experiments on live animals were performed. In view of the two last sentences the ARRIVE guidelines are not applicable.
Materials
The chemicals used for these experiments were purchased from Sigma Aldrich Stockholm, Sweden (now Merck), except for Rhodamine Phalloidin that was purchased from Thermo Fisher Scientific (Cat. No.: R415). The thin double-sided adhesive tape was purchased from 3M (3M-467MP and 3M-Scotch dispensered roll). Parafilm-M was purchased from Sigma Aldrich (Cat. No.: P7793). Type-F microscope immersion oil was purchased from Nikon Instruments (Cat. No.: MXA22192). Syringes with 1 ml volume were purchased from B. Braun (Omnifix-F Solo).
Protein isolation
Myosin and actin were isolated from rabbit skeletal muscle; myosin 59 from fast leg muscles and actin 60 from back muscles. Heavy meromyosin (HMM) was prepared from myosin 3 through limited proteolysis of myosin, using Nα-Tosyl-L-lysine chloromethyl ketone hydrochloride (TLCK) treated chymotrypsin. The protein concentration and purity were evaluated using absorbance spectrophotometry and sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).
Solutions for motility assays
A low-ionic strength solution (LISS), degassed as described below, was the main component in all solutions used. LISS has an ionic strength of 15 mM, pH of 7.4 and contains 10 mM 3-(N-morpholino) propanesulfonic acid (MOPS), 1 mM magnesium chloride (MgCl2) and 0.1 mM potassium ethylene glycol-bis(β-aminoethyl ether)-N,N,N´,N´-tetraacetic acid (K2EGTA). A wash buffer prepared in LISS included 50 mM KCl and 1 mM dithiothreitol (DTT). The assay solution prepared in LISS includes (final concentrations): 45 mM KCl, 10 mM DTT, 1mM magnesium adenosine triphosphate (MgATP), 2.5 mM creatine phosphate (CP) and 0.2 mg/ml creatine phosphokinase (CPK). This assay solution was denoted a60 solution due to its ionic strength of 60 mM. In some other experiments we instead used an assay solution (a45) with ionic strength of 45 mM, achieved by lowering the KCl concentration to 30 mM. Lastly, an oxygen scavenger mixture GOC: (3 mg ml-1 glucose, 0.1 mg/ml glucose oxidase, 0.02 mg/ml catalase, final concentrations) was included in the assay solutions for all experiments. HMM aliquots (100 µl) were retrieved from − 80 oC freezer and slowly thawed while buried in ice. Subsequently, the HMM was diluted in wash buffer to its incubation concentration of 120 µg/ml.
Solutions for experiments on microfluidic platform were similar to those described above for experiments in conventional flow cells. However, the assay solution a60 was modified in order to contain additional 2 nM of rhodamine-phalloidin labeled actin filaments.
Silanization of surfaces for HMM adsorption
Silanization of glass coverslips (Menzel-Gläzer, No. 0; 24x 60 mm2) was performed using trimethylchlorosilane (TMCS), following approaches developed previously 42, 53, 61. Molecular sieves (pore size-4Å, diameter − 3.2 mm, Honeywell Fluka, Cat. 334294) were transferred to 3 different glass bottles (1000 ml; 500 ml; 500 ml) in ~ 1/10 (v/v) ratio. Bottles with sieves were placed in an oven (120 oC, 3h) for evaporation of remaining water. Subsequently, after removal from the oven, the bottles were kept under a fume hood to cool down for at least 1 hour. They were then filled with chloroform (1000 ml and 500 ml) and acetone (500 ml) and kept sealed overnight. On the day of silanization, the solutions were split into several dry glass jars: chloroform (4 jars), acetone (1 jar), methanol (1 jar), approx. 300 ml/jar). Silanization of the glass slides was performed for 10 slides at a time, using a custom-made Teflon holder that was manipulated by modified Pasteur pipettes with hooks made by exposing the pipette ends over a flame. The silanization procedure encompassed the following steps: i. Incubation in sufficient volume of Piranha solution (95–97% H2SO4 and 30% H2O2 in a 7:3 ratio) to cover all the glass slides at 80 oC for 5 minutes. Caution: Piranha solution is a highly corrosive and acidic solution that reacts violently with organic materials. ii. Incubation in dH2O (approx. 800 ml per glass jar) for 2 minutes, 3 times. iii. Incubation in methanol for 2 minutes. iv. Incubation in acetone for 2 minutes v. Incubation in chloroform for 2 minutes. vi. Drying under a nitrogen gas stream for 1–2 minutes (until sufficiently dry). vii. Incubation in chloroform with TMCS (5% v/v). Opening and closing the lid to transfer the glass slides in and out of a particular jar was performed under a nitrogen gas stream. viii. Incubation in chloroform for 2 minutes, 2 times. ix. Drying under a nitrogen gas stream for 1–2 minutes. The prepared glass slides were stored at room temperature in 100 mm cell culture dishes (2 glass slides per dish) that were sealed with parafilm.
Degassing and storage of solutions
LISS was the main constituent of all solutions and was therefore carefully degassed. The LISS buffer was first transferred to a flask with a side outlet that connects to an air suction line. The top of the flask was covered with a rubber stopper allowing creation of low air pressure during suction. Using this approach, 20 ml of the LISS solution was degassed for a minimum of 40 minutes at 0.7–0.8 bar vacuum under slight mixing using magnetic stirrer. The prepared buffer solutions were stored on ice in 1.5 ml centrifuge tubes and, as extra precaution against air entrance, the tubes were wrapped in parafilm. The assay buffer solution was instead transferred to a 1 ml syringe fitted with a hypodermic needle (0.40 x 20 mm, KD-Fine Safety, Germany)) (cf. 38, 45). The syringe was then covered in aluminum foil and the needle-syringe connection was wrapped with parafilm. The syringe was buried in ice until use.
Assembly of Multi-Channel Flow Cells of Conventional Type
A glass coverslip functionalized with TMCS, as described above, was used as the bottom surface for assembling a flow cell of the type conventionally used for in vitro motility assays. These are the flow cells used here unless otherwise stated. On top of the TMCS-derivatized coverslip, 3 strips of double-sided adhesive tape were placed parallel to each other with a 5 mm gap between neighboring strips. A square coverslip (18 x 18 mm2) was then applied to the top of the tape strips, creating 2 flow cells of 10 µl volume. The flow cell openings can be sealed using silicone vacuum grease to hinder oxygen from entering. The latter approach was used only in some experiments in this study. Complete flow cell assembly is illustrated in Fig. 1A-C.
Flow cells for use with microfluidic pump were assembled as described previously38. Briefly, a silanized glass coverslip prepared as above (No. 0; 24 × 60 mm2), was used as floor. Above it, an adhesive multi-channel microfluidic slide (sticky-Slide VI 0.4, Ibidi, Cat. No. 80608) was properly secured. After final addition of assay solution, entry points in one of the channels were sealed by vacuum grease. To the entry points of the channel used for continuous flow of the assay solution, a tubing primed with the same assay solution was attached.
Microfluidic platform
The microfluidic platform setup is depicted in Fig. S7A. The continuous flow of assay solution was achieved by a Pump 33 Dual Drive System (Harvard Apparatus) to which a 3 ml luer-lock tip syringe (Plastipak, BD) filled with assay solution was secured. The syringe was connected to the microfluidic channel inlet (prepared as described above) with silicone tubing (inner diameter 0.8 mm, Ibidi, Cat. No. 10841) aided with the male elbow luer connector (Ibidi, Cat. No. 10802) on a microfluidic channel side and a needle on the syringe side. To the outlet of the microfluidic channel, similar tubing was attached leading to the waste tube. When connecting the tubing, special attention was dedicated to have it properly primed with the solution so that no air bubble was introduced to the system. A bag of ice was put on the syringe to cool the assay solution during pump action. The flow rate was set to 0.2 µl/ml unless otherwise specified.
In vitro motility assay procedure using conventional flow cells
The in vitro motility assays were performed following a standard procedure developed from previous work 3, 38, 62. First, 20 µl of heavy meromyosin (HMM) at 120 µg/ml (343 nM) was added to each chamber of the flow cell. After incubation for 5 minutes, 20 µl Bovine Serum Albumin (BSA) solution (1 mg/ml BSA in wash buffer) was added to the chamber and incubated for 2 minutes. Before the addition of actin filaments, each chamber was washed with 20 µl wash buffer. Then, 20 µl of Rhodamine phalloidin labeled actin filaments (10 nM prepared in wash buffer) were added and incubated for 2 minutes. Finally, another wash step took place, followed by addition of 2–3 drops of assay solution (from a syringe; see above). A few extra drops of assay solution were added to both sides of the flow cell openings to reduce the risk of drying (Fig. 1C). The latter procedure was not used in the limited number of experiments where the flow cell openings were sealed (Fig. 1B) with silicon vacuum grease (Dow Corning® high-vacuum silicone grease, ref. number Z273554, now Molykote, Dupont). Importantly, the HMM used in the experiments was of sufficient quality, to not require an affinity purification step (ultracentrifugation with actin filaments in the presence of millimolar MgATP) for removal of rigor like (“dead”) heads before application to the flow cell 62. Neither was a step with blocking actin (non-fluorescent actin filaments added to flow cell at near micromolar concentration) used to block dead heads 62. Despite the lack of these refinements, the fraction of motile filaments ranged between 0.7 and 1 for all conditions tested.
Unless otherwise stated below, the assay solution was exchanged every 30 minutes during experiments lasting for up to 8 h. Each exchange of assay solution was associated with addition of new actin filaments as follows: After rinsing with 20 µl of wash buffer, once or twice (as specified below), actin filaments were added, followed by 2 min incubation. Then, another wash was performed using wash buffer, and fresh assay solution was added. In all cases (unless otherwise stated), a drop of assay solution was placed outside each of the flow cell opening to prevent drying (Fig. 1C).
In between observations, the flow cells were kept in one of two ways: i) on wet paper to reduce the risk of drying and covered with lid (a small box for tips wrapped in aluminum foil) at 20-22oC (room temperature) (Fig. S1). ii) on the microscope stage in contact with the temperature-controlled objective under a lid (for further details, see under Results).
In vitro motility assays using the microfluidics platform
The procedure was modified from that for conventional flow cells as follows. The volume of solutions added/aspirated to the flow cells during the different incubation steps to prepare for a motility assay (with HMM etc.) was 40 µl. The pipetting procedure to ensures complete solution complete exchange in the microfluidic channel follows the manufacturer (Ibidi) manuals particularly the video protocols for sticky-Slide VI 0.4 (obtained at https://ibidi.com/sticky-slides/65-sticky-slide-vi-04.html). Briefly, the solution additions were done by pipetting the solution against the inner inlet wall (closer to the chamber) while the solution removal was done by placing the tip against the outer wall of the inlet (away from the chamber). For sealed conditions, after the final addition of assay solution to the microfluidic channel, the inlets were filled with an extra 40 µl of assay solution on both sides followed by their sealing with vacuum grease. For continuous flow conditions, the inlets were filled with assay solution following careful attachment to the primed silicone tubing. The latter tubing was further connected inlet with the syringe on the pump for the inlet and with the waste container for the outlet. The first recordings (0 h) were performed 2–3 min after the addition of assay solution and then after approx. every hour till 8 hours. In between observations, the microfluidic slide was kept on the microscope stage in contact with the temperature-controlled objective in the empty channel between sealed and continuous flow conditions channels (see also Fig. S7A). Qualitative analysis of motility was performed for each time point (see Fig. S7B, C) and for time points 0, 4, 7, 8 h we quantified the quality of motility by determining filament velocities and fraction of motile filaments as decried below.
Imaging, filament tracking and filament analysis
The myosin propelled actin filaments were observed using an inverted fluorescence microscope (Axio Observer.D1 from Zeiss) with a 63 x (Apochromat, NA = 1.4, WD = 0.19 mm) oil immersion objective. The rhodamine phalloidin was excited by a short-arc mercury lamp (103 W/2, from OSRAM) using the fluorescence filter set denoted as Cy3 (Exc: 545/25nm, Di: 565nm, Em: 605/70nm). Later the mercury lamp was permanently replaced with a LED light source (Colibri 5, Zeiss; used in all experiments with the microfluidics platform). Imaging was performed using an electron multiplying charged-coupled device (EMCCD) camera (C9100-12PHX1, from Hamamatsu photonics). Videos were acquired at 4.95 frames per second (FPS), at a pixel size of 0.24 x 0.24 µm2 for a 63x objective. Three videos were acquired, starting 1.5-2 minutes after completed exchange of assay solution. Using a ring-shaped temperature objective heater (Objective Heater 2000, Pecon), connected to a temperature controller (TempController 2000-2, Pecon), a temperature of 24–26 oC was maintained (unless otherwise stated) on the flow cells when motility was observed in the in vitro motility assay. Temperature was measured from a water droplet placed on top of the flow cell, using a 51 II Handheld Digital Probe Thermometer (Fluke). The fraction of motile filaments was estimated from data based on manually counting the total number of filaments and the total number of non-motile filaments in each given video. If needed, the contrast in images/videos was adjusted using the ImageJ function (Image/Adjust/Brightness/Contrast). The sliding velocity was estimated by manually following the leading or trailing end of each filament using a Matlab (Mathworks, Natick, Ma) routine developed previously 16, 48.
Estimates of fluid evaporation from conventional flow cells
Fluid evaporation from flow cells over time, was estimated from the change in weight of flow cell assemblies with the flow cells filled in the standard way (Fig. 1C) including droplets outside the openings (but using MilliQ water instead of assay solution). The weighing of the flow cell assemblies at different time points was performed using an analytic balance (ME205, Mettler Toledo).
Estimates of changes in pH in open flow cells
Changes in pH of standard assay solution with time was estimated using a dedicated set of 5 flow cell assemblies (0, 30, 60, 90 and 120 min), filled in the standard way with assay solution (Fig. 1C). Each of the assemblies/flow cells was used to estimate pH after a given storage time. The pH was estimated using pH indicator paper (pH-Box, ref. number 1095650001, from Merck, with three paper rolls: blue roll is corresponding to pH range 0.5-5.0 (ACILIT™), yellow to pH range 5.5-9.0 (NEUTRALIT™), and orange to pH range 9.5–13.0 (ALKALIT™), see also Fig. S2 – pictures of pH paper). Fluid from the flow cells were withdrawn by bringing the pH paper into contact with the droplet outside the flow cell. A similar procedure was used for sealed flow cells, expect that the vacuum grease needed to be carefully removed and/or the glass coverslip roof needed to be broken in order to reach the small amount of liquid inside the chamber.
Oxygen concentration measurements
The measurements were performed by a multi-parameter meter (Multi 3620 IDS, WTW) attached to the dissolved oxygen probe (FDO 925, WTW) in a 50 ml tube filled with 20 ml of MilliQ water. The volume of MilliQ water for degassing (20 ml) mirrors the typical volume of major assay buffer (LISS) degassed on a typical experimental day using a conical flask with a side outlet connected to the air suction line, placed on a magnetic stirrer plate for slow stirring of the solution during degassing38.
Statistical analysis and reproducibility
No sample size calculations were performed prior to the experiments as the main purpose was not to demonstrate differences in observed variables between protein population groups or from a given population value. Instead, the aim was to test the effect of long-time incubation and solution exchange including information of the variability.
Both velocity data and the fraction of motile filaments were assumed to be normally distributed, consistent with results of the Shapiro-Wilks test, showing no significant difference from the normal distribution (p > 0.05). Moreover, each filament and each observed region of interest on a flow cell was assumed to represent an independent random sample for the velocity and fraction of motile filaments, respectively (cf. 11, 54, 62, 63 for further motivation). The data are presented as mean ± 95% confidence interval (CI) and non-overlapping such confidence intervals are assumed to indicate statistically significant differences between groups (p < ∼0.05). All statistical analyses and data visualization were performed using Graph Pad Prism v. 9.2.0 (Graph Pad Software, San Diego, Ca).