The Optimized Quantum Dots Mediated Thermometry Reveals the Eciency of Myosin Extracted from Muscle Mini Bundles

Background: The quantum dots (QD) has been investigated as thermometrical sensor in biological microenvironment and applied to measure the muscle eciency with underlying mechanisms, i.e., a reduction in uorescent intensity of QD reects an increase in temperature caused by heat release during ATP hydrolysis, denoting the eciency of the motor protein myosin. The aim of this study is to optimize the QD mediated thermometry for measuring the eciency of freshly extracted myosin from muscle mini bundles rather than pre-puried myosin and test this approach in preparations with different myosin isoform. Methods: The protocol of myosin extraction used in the single muscle ber in vitro motility assay was modied slightly for extracting myosin from the muscle mini bundles. Moreover, the quantitation of extracted myosin was calculated from the total extracted protein since the ratio of myosin to total protein was constant, performing through spectrophotometric measurement of UV absorbance at 280 nm. The change in uorescence intensity of QD thermometry measurement of myosin ATPase enzymatic reaction was plotted over time, and the slope of the linear negative regression between time course and relatively decreased uorescence intensity was used to reect the eciency of extracted myosin. Results: The optimized QD mediated thermometry is established for evaluating the eciency of myosin extracted from muscle mini bundles. Moreover, myosin isoform specic differences in the myosin eciency were observed in comparison between slow myosin and fast myosin, i.e., the low myosin has lower eciency than fast myosin, evidenced by a higher heat release. Conclusions: The optimized QD mediated thermometric measure of myosin eciency in muscle mini bundles provides a nanoscale approach to evaluate myosin function based on a minimal amount of muscle, which is essentially required in the muscle research.


Background
The motor protein myosin is a mechanoenzyme converting chemical energy from ATP hydrolysis to mechanical work during muscle contraction and is expressed in different isoforms determining the contractile properties of the muscle. To date, nine distinct myosin heavy chains (MyHC) isoforms have been identi ed in mammalian striated muscle, and four of them (type I, IIa, IIx and IIb MyHCs) are expressed in normal adult limb and trunk mammalian skeletal muscle [1]. There is a close relationship between the maximum velocity of unloading shortening (Vo), the actin-activated ATPase activity of myosin, and the myosin isoform expression of single muscle bres, which means different isoforms of myosin confer distinct contractile properties to muscle bres [2][3][4][5][6][7][8][9]. Correspondingly, the changes in isoform expression, speci c mutations or post-translational modi cations of myosin will in uence muscle function [10][11][12][13][14]. So far, several miniature approaches, such as single bre or myo bril contractile measurement or single bre in vitro motility assays, have been applied to evaluate muscle function at cellular, subcellular or molecular levels [9,15,16]. However, the biochemical properties of myosin from these preparations remain incomplete and the introduction of cadmium telluride quantum dots (CdTe QD) mediated thermometry to measure myosin e ciency at the protein level has complemented the understanding of the myosin function in vitro [12,17]. The underlying mechanism of the approach is that the heat loss during ATP hydrolysis by myosin ATPase enzymatic reaction will increase the temperature in the microenvironment, and QD uorescence intensity will accordingly decrease with thermal resolution of ~ 1 mK. In addition, small amounts of muscle tissue are routinely collected in investigation of pathophysiological mechanisms or diagnosis of neuromuscular disorders and there is a compelling need to use a minimum muscle tissue. In this study, we have veri ed and quanti ed the extracted myosin content from muscle mini bundles, optimized the procedures of QD mediated thermometry for measuring the myosin e ciency and observed isoform speci c differences in myosin e ciency.

Animals, muscle tissues and bundle preparations
Fast-and slow-twitch skeletal muscles were collected from the hindlimbs of young Sprague Dawley rats, i.e., the fast-twitch extensor digitorum longus (EDL) muscle and the slow-twitch soleus (SOL) muscle. The animal experiments were carried out according to the guidelines of the Swedish Board of Agriculture and approved by the ethical committees at Karolinska Institutet. The EDL and SOL muscle tissues were dissected into bundles of approximately 50 bers (-5 mm long) in relaxing solution containing 50% (vol/vol) glycerol at 4 °C and tied to glass capillaries, stretched to about 110% of their resting slack length. Afterwards the bundles were chemically skinned by treatment for 24 hours at 4 °C in a relaxing solution, and then stored at -20 °C. Within one week after above skinning treatment, the bundles were cryo-protected by transferring them to relaxing solutions containing increasing concentrations of sucrose (0, 0.5, 1.0, 1.5, and 2.0 M) at 30-minute intervals, and then frozen in liquid propane chilled by liquid nitrogen. The frozen bundles were stored at -140 °C. Before the experiment, the bundle was incubated in sucrose solutions with decreasing concentrations (2.0, 1.5, 1.0 and 0.5M) sequentially at 30-minute intervals and then kept in the skinning solution at -20 °C for 2 weeks or shorter prior to usage. [9,18] The methods are presented in a schematic diagram ( Figure 1) The extraction of myosin and the spectrophotometric quantitation A muscle mini bundle consisting of 10-20 bers (that may be adjusted according to the measured concentration of extracted protein) was separated gently and incubated in the microtube with 20 μl a high-salt buffer (0.5 m KCI, 25 mm Hepes, 4 mm MgCl2, 4 mm EGTA, pH adjusted to 7.6 before adding 2 mm ATP and 1% β-mercaptoethanol) at 4 °C for 30 minutes. The solution containing extracted myosin and other minor proteins was kept on ice for the application to the different sections of the experiment, i.e., the spectrophotometric quanti cation of the concentration, QD-mediated thermometry and 12% SDS-PAGE to determine the relative content of myosin in the extracted proteins. The concentration of extracted protein was quanti ed by conventional spectrophotometry using absorbance at 280 nm (Nanodrop, Thermo Scienti c), which does not require any standard curve, but the blank control (contains high-salt buffer only) and internal control (different concentration of BSA, such as 1, 2, 3 and 4 mg/ml) were applied for quality control. Every sample and control for A280 measurement have been brie y and su ciently vortexed and the quanti cation was repeated at least 3 times for consistency. The extracted myosin was proportional to the extracted total protein (see the section of results). The A280 value of the extracted total protein was therefore used to represent the concentration of the extracted myosin in most situations unless speci ed otherwise.
The optimized QD-mediated thermometry of myosin ATPase enzymatic reaction The assay for determining the saturated ATP concentration was performed rstly. Commercial control myosin protein (1 mg, Cytoskeleton, Inc., Denver, CO) was dissolved in 100 μL of myosin resuspension buffer (15 mM Tris HCl of pH 7.5, 0.2 M KCl and 1 mM MgCl2) then diluted to the concentration of 0.2 μM. For each measurement, 1 µL of the control myosin was pipetted to a well containing 30 µL low-salt buffers (25mM KCI, 25 mm Hepes, 4 mm MgCl2, 1 mm EGTA, pH adjusted to 7.6 before adding 1% βmercaptoethanol) on a black 384-well microtiter plate, followed by the addition of 1 µL Cadmium telluride core-type QD (1 mg/mL) (Sigma-Aldrich) and 30 µL of the low-salt buffers containing different ATP concentrations (0, 3.5, 4.5, 5, and 6 mM ATP in low-salt buffers) respectively. After the quantitation, the extracted myosin was optimally diluted for actual measurement, then 1 µL of extracted myosin was pipetted to a well containing 30 µL low-salt buffer, followed by the addition of 1 µL QD (1 mg/mL) and 30 µL of the blank control (0mM ATP) or ATP solutions (5mM ATP), respectively. The addition of blank control/ATP solutions and the detection of uorescence signal were performed at the same time to avoid any inconsistency with time course due to the instantaneous enzymatic reaction. The QD uorescence signal was recorded every 15 seconds for 5 minutes by a uorescence spectrophotometer (TECAN, In nite M200, Switzerland), while the excitation and emission wavelengths were xed at 310 and 530 nm, respectively. The measurements for each preparation were performed in sextuplicate with both the blank control and ATP solution at 25°C. The low salt buffer and QD were kept at room temperature (22°C) and the remaining solutions were kept on ice.

Myosin isoform expression and relative quantitation
After myosin extraction, the mini bundle was placed in SDS sample buffer in a microfuge tube and stored at −80 °C. The composition of MyHC isoforms was determined by 6% SDS-PAGE. The acrylamide concentration was 4% (wt/vol) in the stacking gel and 6% in the running gel, and the gel matrix included 30% glycerol. Sample loads were kept small to improve the resolution of the MyHC bands (type I, IIa, IIx and IIb). Electrophoresis was performed at 120 V for 22 h with a Tris-glycine electrode buffer (pH 8.3) at 10 °C (SE 600 vertical slab gel unit; Hoefer Scienti c Instruments, Holliston, MA, USA). The gels were silver-stained and subsequently scanned in a GS-900 Calibrated Densitometer (Bio-Rad). The volume integration function (Image Lab software 6.0, Bio-Rad) was used to quantify the relative amount of each MyHC isoform when more than one isoform was expressed. After the QD-mediated thermometry assay, the remaining myosin preparations were kept in urea buffer in a microfuge tube and stored at −80 °C. The relative quantitation of MyHC contents in total extracted protein was determined by 12% SDS-PAGE. After centrifugation and heating (90°C for 2 minutes) a volume of 4 µl was loaded on 12% SDS-PAGE. The total acrylamide and Bis concentrations were 4% (wt/vol) in the stacking gel and 12% in the running gel. The gel matrix included 10% glycerol. Electrophoresis was performed for 5 h with a Tris-glycine electrode buffer (pH 8.3) at 15°C (SE 600 vertical slab gel unit, Hoefer Scienti c Instruments). The gels were stained with Coomassie blue (SimplyBlue SafeStain, Invitrogen), as this staining shows high reproducibility and the ability to penetrate the gel and stain all proteins present, i.e., allowing accurate quantitative protein analyses. The gels were subsequently scanned to determine the relative contents of myosin heavy chain in total extracted protein [9,19,20].

Data analysis and statistic
QD uorescence signals were detected and normalized to the starting uorescent value and the corresponding relative uorescence intensity formed a negative hyperbolic regression plotted over time.
The linear part of the curve corresponding to the initial 60 seconds, was used to calculate the slope of the relative uorescence intensity over time. The slope of the blank control subtracted from the slope of the "real" reaction then normalized to the concentration of the extracted myosin, indicating myosin e ciency.
For the sextuplicate measurement of each preparation, the subtracted and normalized slope values were evaluated individually according to the criterion, i.e., calculated slopes which fell outside one standard deviation were excluded and the remaining quali ed slopes were included, and the slopes in negative values (indicating null reactions) were excluded ( Figure 2). Statistical analyses were performed by SigmaPlot software version14. The data were presented as mean ± standard deviation and analyzed by the Student's unpaired t test.

Quantitation of extracted myosin and QD-mediated thermometry
The ratio of myosin to total extracted protein was determined by 12% SDS-PAGE. The relative contents of total myosin in both the commercially puri ed myosin preparation and the extracted myosin from mini bundles were 87±3 and 86±3% (Fig. 3a), respectively, i.e., the ratio of myosin to total extracted proteins was stable and similar to the commercially puri ed control myosin, indicating that the A280 measure of the total extracted proteins can be used as an estimate of the extracted myosin concentration. The blank control (the high salt buffer) and the internal control (BSA in different concentrations in the high salt buffer) were measured by spectrophotometric A280 method prior to quantifying the total protein extracted from mini bundles (Fig. 3b).
The uorescence signal of QD itself is stable in the time course, which was veri ed by applying H 2 O as substrate in the reaction. However, QD uorescence intensity decayed signi cantly when exposed to low salt buffer and control myosin (Fig. 3c). Therefore, the interference from self-decaying QD uorescence was corrected in every measurement by the introduction of blank control measurements. To determine the concentration of ATP corresponding to the saturated reaction, different concentrations of ATP were applied, and nally 5 mM was chosen with reference to the previous study (Fig. 3,c) [12].

QDs thermometric measurement of the myosin ATPase reaction
According to the selection criterion, 77% of the measurements had at least three out of sextuplicate measurements in both blank and 5mM ATP reactions and were enrolled for statistical analysis. The other 23% had only two out of sextuplicate measurement and were omitted (Table 1 and Fig. 4). The correlation coe cients of the linear regression during 60 and 90seconds time courses were signi cantly different, and the plots in the 60 seconds course had a stronger linear relationship and were chosen in the following analysis (Table 2). Table 1 The number of preparations with sextuplicate measurements that meet the selection criterion, and only  Table 2 The correlation coe cients of the linear regression in 60 and 90seconds courses The calculated slopes, i.e., the subtracted and normalized slopes for EDL (12.9±3.9) and SOL (19.2±2.7) were signi cantly different (p=0.006) while the SOL presented a steeper decay of the QD uorescence intensity than EDL (Fig. 4a-d). The average MyHC isoform composition in the EDL was 55 ± 17% IIb, 37 ± 10 % IIx, 6 ± 9 % IIa, and 1 ± 3 % I, in the SOL was 89 ± 9 % I and 11 ± 9 % IIa (Fig. 4e).
In all measurements of myosin ATPase reactions, the experimental temperature was set at 25℃.
Considering the temperature of the blank control measurement as the baseline, a large majority of corresponding temperatures of the ATPase reaction uctuated in the range of 24.9 to 25.1℃. However, a small number of preparations exceeded this range but were within 25±0.2℃, such as 11% (6/57) and 3% (2/70) of individual measurements in EDL and SOL groups, respectively (Fig. 4f).

Discussion
Myosin, the dominant motor protein in skeletal muscle, is a mechanoenzyme converting chemical energy from ATP hydrolysis into mechanical work. Myosin is expressed in different isoforms that determines the contractile properties of skeletal muscle. At the skeletal muscle ber level, unloaded shortening velocity is directly related to, and dependent on, the speci c activity of myosin ATPase that is predominantly determined by the myosin isoform [21,22]. However, our understanding of myosin ATPase properties at the muscle ber level remains incomplete. As a semiconductor nanocrystal, quantum dots were selected as a nanoscale thermometer [12,17] because of their superior uorophores and thermal sensitivity compared to organic dyes or other uorescent proteins. Furthermore, QD have been investigated to measure altered temperature of the ATPase reaction by puri ed myosin to de ne myosin e ciency [12,17]. The decreased uorescence intensity of QD re ects higher amount of heat production in the process of ATP hydrolysis, signifying lower work e ciency of the myosin motor protein [12,17,23]. In this study, we optimized this approach for QD-mediated thermometry measurement [12] to evaluate the e ciency of myosin freshly extracted from muscle mini bundles rather than pre-puri ed myosin and observed the myosin isoform effect on the myosin e ciency.
It must be noted that the concept of e ciency in the present and the founding study [12] was de ned from the perspective of thermometry rather than enzymology where the Michaelis constant (Km) and the catalytic constant for the conversion of substrate to product (kcat) are used to evaluate the enzymatic reaction and even sometimes the ratio kcat/Km has been used to represent the catalytic e ciency of the enzyme [24,25]. Theoretically, the total amount of energy released during ATP hydrolysis is the same, but the fraction that is converted into chemical or mechanical energy or heat seems to be modulated by the ATPase [26]. From this point of view, the present study reveals that the SOL myosin ATPase causes more heat production but less mechanical energy than EDL myosin ATPase. As a crucial factor, the ambient temperature affects both QD uorescence and enzyme activity, therefore, the reaction temperature was carefully controlled and kept stable to minimize the error between measurements. The reaction rate is expected to vary between 4 and 8% per degree centigrade [24]. Thus, the 0.1℃ temperature uctuations will result in a negligible ± 0.6% error.
In the actin-myosin interaction, ATP binding rstly induces a conformational change in myosin and causes myosin to detach from actin, followed by ATP being hydrolyzed to ADP and inorganic phosphate (Pi), then myosin rebinds to actin and the force generating power-stroke accompanies subsequent Pi release, consequently, ADP is released, and the cycle repeats upon ATP binding [27]. Real time acquisition of reaction with millisecond time resolution is needed to measure the myosin ATPase cycle rate constants. Accordingly, the stopped-ow apparatus provides satisfactory and reproducible results and the related studies showed that myosin isoforms possess different enzymatic activity and different e ciency for the conversion of chemical into mechanical energy [21]. For example, muscle bers expressing the IIb MyHC isoform have the highest rate of ATPase activity, followed by IIx, IIa and I bers [22]. In the absence of actin, most myosin binds ATP rapidly and irreversibly, and release phosphate very slowly [21], the present QD-mediated thermometry is sensitive to detect the changing temperature in this process.
Like the single ber in vitro motility assay, myosin can also be extracted from muscle mini bundles successfully, which has been veri ed by 12% SDS-PAGE. Moreover, there was no signi cant difference in the ratio of myosin to total protein among the extracted myosin preparations themselves or when comparing with commercially puri ed myosin preparations, i.e., the quantitation of total extracted proteins is considered a reliable measure of myosin quantity. The extraction of myosin from mini bundles needs the addition of ATP, which means the myosin ATPase reaction has been initiated prior to the QD measurement, which indicates the necessity to assess blank control in all measurements. The advantage of using A280 measurement is that it can be accomplished faster, avoiding any delay that may cause myosin inactivation prior to the QD-mediated thermometry. The spectra of high salt buffer (blank control), BSA and extracted myosin in the same buffer exhibit signi cant, compatible, and reproducible absorbances, supporting the reliability of this quantitation method.
Even though QD has super uorescence, its photoluminescence signal in a highly crowded bioenvironment incline to be more sensitive to the presence of a variety of ions and molecules [28]. This feature may explain the phenomenon that the uorescent intensity of QD decayed signi cantly when incubated in salt solution. Since both the high or low salt solution are critical and necessary for the extraction and function of myosin in vitro, it is imperative to have the blank control introduced during the QD-mediated thermometry measurement to correct the self-decaying interference. All preparations of extracted myosin were assayed, paired with controls, in sextuplicate replicates to minimize error, and over 75% of all preparations have at least three quali ed measurements, being su cient to produce reliable results. Figure 1 The work ow chart of the QD-mediated thermometry experiments.

Figure 2
An example of data analysis of the sextuplicate assays for an extracted myosin preparation. In this case, 4 values of 6 subtracted and normalized slope were evaluated to be quali ed and rest 2 values (marked with star) were unquali ed because they fell outside of the range of mean ± SD according to the criterion.  The individual plot of QDs relative uorescence intensity over time in EDL a and SOL b groups, respectively, demonstrate higher e ciency of EDL. c The averaged plot of relative uorescence intensity over the time. The values are presented as means ± SEM. d The distribution of the slopes (-60seconds) in EDL and SOL groups, respectively. The values are presented as means ± SD and scatter plot. The open circles dots represent the preparations contain at least 3 quali ed measurement, and the black circles represent those that contain only 2 quali ed measurements (not included in statistical analyses), and the red cross represent mean values per individual rat. (*P<0.05). e The individually relative contents of MyHC isoforms in EDL (green lines) and SOL (yellow lines), respectively. f The uctuations of temperature in QD-mediated thermometry of EDL (green lines) and SOL (yellow lines) experiments (25℃ is considered as zero in y axis).