The goal of this protocol is to form ensembles of molecular motors on DNA origami nanostructures and observe the ensemble motility using total internal reflection fluorescence microscopy.
Cytoskeletal motors are responsible for a wide variety of functions in eukaryotic cells, including mitosis, cargo transport, cellular motility, and others. Many of these functions require motors to operate in ensembles. Despite a wealth of knowledge about the mechanisms of individual cytoskeletal motors, comparatively less is known about the mechanisms and emergent behaviors of motor ensembles, examples of which include changes to ensemble processivity and velocity with changing motor number, location, and configuration. Structural DNA nanotechnology, and the specific technique of DNA origami, enables the molecular construction of well-defined architectures of motor ensembles. The shape of cargo structures as well as the type, number and placement of motors on the structure can all be controlled. Here, we provide detailed protocols for producing these ensembles and observing them using total internal reflection fluorescence microscopy. Although these techniques have been specifically applied for cytoskeletal motors, the methods are generalizable to other proteins that assemble in complexes to accomplish their tasks. Overall, the DNA origami method for creating well-defined ensembles of motor proteins provides a powerful tool for dissecting the mechanisms that lead to emergent motile behavior.
Dynein and kinesin are cytoskeletal motor proteins responsible for myriad functions in eukaryotic cells1. By converting the chemical energy of ATP hydrolysis into productive work, these motors translocate on microtubules to haul and distribute various intracellular cargos. They also coordinate in the massive intracellular rearrangements associated with mitosis, where they exhibit orchestrated forces that contribute to the positioning and separation of chromosomes. Structural, biochemical, and biophysical assays, including single molecule observations, have revealed the mechanisms of these motors at the individual level (well-reviewed in previous works2,3,4). However, many of the motors' tasks require them to work in small ensembles of both similar and mixed motor types. Comparatively less is understood about the mechanisms that coordinate the activity and ultimate emergent motility of these ensembles5,6. This knowledge gap is due, in part, to the difficulty in creating ensembles with controllable features, such as motor type and copy number. Over the past decade, the molecular construction techniques of DNA origami have been employed to solve this problem. For the microtubule based motors, some examples of these investigations include single molecule observations of ensembles of cytoplasmic dynein-17,8,9, intraflagellar dynein11, various kinesin motors12,13, and mixtures of both dyneins and kinesins7,14,15. Here, we provide details of the purification and oligonucleotide labeling of motors from yeast7,16,17,18,19,20, the folding and purification of segmented DNA origami with tunable compliance8, and the imaging of the yeast motors propelling the chassis structures7,18.
Constructing motor ensembles for in vitro single molecule observation requires three primary efforts. The first is the expression, purification and labeling of motor constructs suitable for attaching to DNA origami. The second is the production and purification of defined DNA origami structures (often termed "chassis"). And the third is the conjugation of the motors to the chassis structure followed by observation using total internal reflection fluorescence (TIRF) microscopy. Here, we provide established protocols for this process for recombinant microtubule-based motors purified from the yeast Saccharomyces cerevisiae7,16,17,18,19. DNA origami-based motor ensembles have been investigated using both recombinant kinesin15 and dynein7,8,18 constructs produced in this yeast expression system16,17,18,19. This protocol is valid for these constructs, given that they are controlled by the galactose induced promoter, and fused to the same protein tags for purification (ZZ and TEV protease linker) and for DNA oligo conjugation (SNAPtag).
Specific yeast strains produce specific motor constructs. For example, the dynein used to study the role of cargo compliance was purified from strain RPY10847,8. In general, strains containing motor constructs with the appropriate genetic modifications for expression and purification can be requested from the laboratories having published the use of those motors. Constructs with novel attributes such as mutations or tags can be made using recombinant genetic techniques, such as lithium acetate transformation21 and commercial kits. Detailed protocols for creating modified motor proteins in yeast for single molecule studies have been published19. In addition to the motors being fused to the SNAPtag, the oligos used to label the motors must be conjugated to the SNAP substrate, benzylguanine (BG); previously published protocols describe the formation and purification of BG-oligo conjugates18. The overall strategy described here has also been employed for actin-based motors (see previous works for examples22,23,24), and motors purified from other organisms and expression systems (see previous works for examples7,9,10,11,12,13,14).
Polymerized microtubules (MTs) are used in these experiments in two different procedures. MT affinity purification of functional motors requires MTs that are not labelled with other functional groups, while the motor-ensemble motility TIRF assay requires MTs labeled with biotin and fluorophores. In all cases, MTs are stabilized with taxol to prevent denaturation. The MT affinity purification step is used to remove any non-motile motors with a high MT affinity, as these motors can alter ensemble motility if conjugated to a chassis. During this process, active motors unbind the MTs and remain in solution, while tight-binding motors spin down in the MT pellet. This helps ensure all motors on the chassis are from an active population.
A variety of DNA origami structures have been used to study cytoskeletal motor ensembles. As the mechanistic understanding of ensemble transport has increased, the DNA origami structures employed in experiments have grown in complexity. In principle, any structure could be adapted for this purpose provided it is modified to include single-stranded DNA attachment sites for motors and fluorophores. Specific chassis designs and attributes may be useful for probing particular questions about the emergent behavior of motors ensembles. For example, rigid rods have been used to develop foundational knowledge of how copy number affects transport by teams of dyneins and kinesins7,15,18, and 2D platforms have been used to study myosin ensemble navigation of actin networks22. Structures with variable or tunable flexibility have been used to understand the roles of elastic coupling between motors and to probe how stepping synchronization affects motility8,24. More recently, spherical structures are being used to gain insight into how geometrical constraints to motor-track binding affect the dynamics of motility25.
In this protocol, we offer specific steps for ensemble experiments on segmented chassis with variable rigidity. Binding sites on the chassis are sometimes referred to as "handles", while complementary DNA sequences that bind these handles are termed "antihandles". The number of motors on these chassis is determined by which segments contain extended handle staples with complementarity to the antihandle oligo on the oligo-labeled motors. Using different handle sequences on different segments allows for binding of different types of motors to specific locations on the chassis. The chassis detailed here is composed of 7 sequential rigid segments, each comprised of 12 double-stranded DNA helices arranged in 2 concentric rings8. The rigid segments contain the motor handles and are connected through regions that can be either flexible single-stranded DNA or rigid double-stranded DNA, depending on the absence or presence, respectively, of "linker" staples. The compliance of the chassis structure is thus determined by the presence or absence of these "linker" staples. See previous reports for further details and specific DNA sequences8. In addition, multiple methods can be used to purify chassis26. The rate-zonal glycerol gradient centrifugation method27 is described here.
1. Growth, expression and harvesting of motor proteins controlled by a galactose induced promoter
2. Purification of motor proteins from yeast cells
3. Microtubule (MT) polymerization
4. Microtubule (MT) affinity purification
5. Production of segmented DNA origami chassis
6. Making slide assay chambers
7. Motor-ensemble motility TIRF assay
Successful purifications of motors and chassis structures were assayed by gel electrophoresis. SDS-PAGE analysis confirmed the successful extraction of dynein from yeast (Figure 2), as the final filtrate collected in step 2.3.7 showed a clear, sharp band at the position of ~350 kDa. As expected, this dynein band was absent from the flowthrough and wash that removed unwanted proteins, and the beads from which dynein was cleaved. The observation suggests that the IgG affinity purification and TEV protease cleavage were both highly efficient. Additionally, TEV protease was also present in the final filtrate and formed a clear band at ~50 kDa.
The successful MT affinity purification of dynein and kinesin proteins was also confirmed with SDS-PAGE analysis (Figure 3). While dynein showed up as a clear single band at ~350 kDa, kinesin showed up as slightly smearing multiple bands at ~120 kDa, possibly due to the presence of both phosphorylated and dephosphorylated forms of the protein14 and variable yields in oligo-labeling. A comparison between the dynein and kinesin bands before and after this MT affinity purification revealed a reduction in the motor concentration, as indicated by the decrease in the band intensity, which was likely due to the removal of non-functional motors. Despite the reduction, the concentrations of motors retained were sufficient for effective TIRF assays. In addition to the TEV protease, a noticeable amount of tubulin (~51 kDa) was present in the final supernatant, most likely due to the gradual decomposition of MTs during the experiment, or incomplete removal of the excess tubulin through centrifugation. However, the consistent motility of motor-chassis ensembles shown in TIRF assays suggests that tubulin and TEV protease did not interfere with motor functions (see Figure 6).
Folding of DNA origami structures was assayed by agarose gel electrophoresis. Figure 4 depicts the results of a gel analyzing the folding of a flexible segmented chassis with 2 motor binding sites. The shift in mobility between the pure unfolded scaffold strand (Lane 1) and the folding reaction (Lane 2) indicates origami folding. Additionally, the folding reaction in Lane 2 indicates the presence of some multimerization of chassis structures. Multimerization typically occurs, and requires subsequent purification of the well-folded monomeric structures. The unincorporated excess staples were also visible, displaying a high degree of mobility through the gel.
Folded origami reactions were purified to remove excess unincorporated staples and multimers of the chassis structure. Figure 5 shows the results of a glycerol gradient purification of a flexible chassis with 7 motor binding sites. The early fractions correspond to the low glycerol density at the top of the tube. They contain the excess staples. The late fractions correspond to the high glycerol densities at the bottom of the tube and contain multimers and aggregates of folded structures. In this gel, fraction 7 indicates a suitable fraction containing well-folded monomeric chassis. Note that the well-folded structure is isolated from both excess staples and multimers. While this gel is representative, and fraction 7 often contains useable chassis, each purification experiment yields slightly different results and fractions should always be assayed to determine which fraction is best for motility assays.
The motility of motor-chassis ensembles is easily detectable and measurable on the kymographs generated from TIRF movies. For instance, the kymographs (Figure 6) of flexible chassis conjugated to seven dynein proteins ("7D" ensembles) show highly processive runs at relatively consistent velocities, demonstrating that the ensembles were active and motile in the reconstituted system, and that MT affinity purification successfully removed most of the non-functional, immobile dyneins that could slow or stall the ensembles. The same TIRF experiment has been successfully performed on other chassis types with different compliance and motor numbers to reveal the effects of these factors on dynein teamwork8,9.
Figure 1: Schematic of the slide assay chamber. The coverslip sits atop two strips of double-sided tape. Solutions are pipetted in one side, and extracted with filter paper on the other. Please click here to view a larger version of this figure.
Figure 2: SDS-PAGE analysis of the purification of dynein from yeast. Yeast cells were lysed and centrifuged to collect the lysate (lane 1) containing total soluble proteins. The affinity between IgG on column beads and the ZZ tag on the recombinant dynein construct was exploited for this purification. The flow-through (lane 2) was collected from the mixture of the lysate and beads after it passed through a chromatography column. The dynein-bound beads were washed with buffers, and the first wash with the wash buffer (lane 3) was collected. Dynein was then conjugated to a DNA oligo. After TEV protease cleavage, centrifugation in a spin column separated the filtrate containing dynein (lane 4) and the residual beads (lane 5). All samples collected from the purification were denatured with 1x LDS Sample Buffer and 1x Reducing Agent, and loaded onto a 4-12% Bis-Tris gel. The gel was run in 1x MOPS buffer at 200 V for ~1 h, and stained with SYPRO Red Protein Gel Stain for imaging under UV light. Notably, the clear, sharp band at ~350 kDa in lane 4 indicates the presence of concentrated purified dynein in the final filtrate, while the band at ~50 kDa in the same lane indicates the co-presence of TEV protease. Please click here to view a larger version of this figure.
Figure 3: SDS-PAGE analysis of the MT affinity purification of functional dynein and kinesin proteins. Kinesin and dynein purified from yeast (lanes 1 and 3) were mixed with polymerized MTs and ATP, and ultracentrifugation was performed to separate the functional motors in the supernatants (lanes 2 and 4) from the non-functional motors that co-pellet with MTs. The motor samples before and after purification were denatured with 1x LDS Sample Buffer and 1x Reducing Agent, and loaded onto a 4-12% Bis-Tris gel. The gel was run in 1x MOPS buffer at 200 V for ~1 h, and stained with SYPRO Red Protein Gel Stain for imaging under UV light. The intensity of the motor bands (~120 kDa for kinesin, and ~350 kDa for dynein) appeared to decrease after the MT affinity purification, indicating a reduction in the motor concentrations. Noticeable concentrations of tubulin and TEV protease were present in the post-purification motor samples. Please click here to view a larger version of this figure.
Figure 4: Agarose gel analysis of folded DNA origami structure. Folded DNA origami structures are assessed by gel electrophoresis. Lane 1 is the pure unfolded scaffold strand while lane 2 is the product of the folding reaction. Gel was run at 70 V in an ice water bath for 90 min in 0.5x TBE buffer supplemented with 11 mM MgCl2. A shift in mobility indicates origami folding. Unincorporated staples and chassis multimers were also visible in Lane 2. Please click here to view a larger version of this figure.
Figure 5: Agarose gel analysis of glycerol gradient purification of DNA origami chassis. The quality of the glycerol gradient purification can be determined by gel electrophoresis of the fractions from the centrifuge gradient. Low number fractions are from the top of the tube and correspond to low density of glycerol. High numbered fractions are from the bottom of the tube and correspond to high densities of glycerol. The excess staples, monomeric chassis, and multimeric chassis are all visible. Fraction 7 contains chassis suitable for TIRF microscopy as they are monomeric and excess staples are absent. Gel was run at 70 V in an ice water bath for 120 min in 0.5x TBE buffer supplemented with 11 mM MgCl2. Please click here to view a larger version of this figure.
Figure 6: Kymographs showing the motility of dynein-chassis ensembles on single MTs. Dynein proteins extracted from yeast and purified with MTs were conjugated to flexible chassis structures. Each chassis had seven binding sites for dynein and formed a "7D" ensemble. The movement of these ensembles on MTs was recorded in a 10 min movie (200 ms exposure, 0.5 fps) during a TIRF assay. Kymographs from two MTs were generated from this movie in ImageJ by tracing along a single MT. The vertical and horizontal red bars in the top left corner of each image indicate 2 min and 20 µm, respectively. Each bright line records the movement of one ensemble, with the inverse of the slope indicating the velocity and horizontal displacement indicating the run length. With TIRF assays and kymography, the motility of motor-chassis ensembles becomes easily detectable and directly measurable. Please click here to view a larger version of this figure.
Buffer Name | Composition | Step(s) Used | Comment |
5x Lysis Buffer | 150 mM HEPES (pH 7.4) | – | Filter sterilize the buffer. It can be stored at RT in a properly sealed container for a year. |
250 mM KAcetate | |||
10 mM MgAcetate | |||
5 mM EGTA (pH 7.5) | |||
50% glycerol | |||
4x Lysis Buffer With Supplements | 4x Lysis Buffer | 2.1-2.2 | Make this 4x buffer from the 5x lysis buffer above. Prepare a buffer without PMSF first, and add the compound (dissolved in pure ethanol) to a small aliquot of the buffer right before each step that requires it. |
4 mM DTT | |||
0.4 mM Mg-ATP | |||
2 mM PMSF | |||
Wash Buffer | 1x Lysis Buffer With Supplements | 2.2 | To make the buffer, add KCl, Triton X-100, and ddH2O to the 4x lysis buffer with DTT and Mg-ATP, but do not add PMSF until right before use. |
250 mM KCl | |||
0.1% Triton X-100 | |||
5x TEV Buffer | 50 mM Tris-HCl (pH 8.0) | – | Filter sterilize the buffer. It can be stored at RT in a properly sealed container for a year. |
150 mM KCl | |||
10% Glycerol | |||
1x TEV Buffer | 1x TEV Buffer | 2.2-2.3 | To make the buffer, add DTT, Mg-ATP, Triton X-100, and ddH2O to the 5x stock TEV buffer, but do not add PMSF until right before use. |
1 mM DTT | |||
0.1 mM Mg-ATP | |||
0.1% Triton X-100 | |||
0.5 mM PMSF |
Table 1: Buffers for the IgG affinity purification of motor proteins from yeast cells (Protocol section 2).
Buffer Name | Composition | Step(s) Used | Comment |
5x BRB80 | 400 mM PIPES | – | Filter sterilize the buffer. It can be stored at RT in a properly sealed container for a year. |
10 mM MgCl2 | |||
5 mM EGTA | |||
Adjust pH to 6.8 with KOH | |||
1x BRB80 With Supplements | 1x BRB80 | 3.3 & 4.1-4.2 | Must be freshly made from the 5x BRB80 stock for every experiment. |
20 µM Taxol (dissolved in DMSO) | |||
1 mM DTT | |||
2x Polymerization Mix | 2x BRB80 (without supplements) | 3.3 | Flash freeze the mix in small aliquots and store at -80 oC. |
2 mM DTT | |||
2 mM Mg-GTP | |||
20% DMSO | |||
Reconstitution Buffer | 1x BRB80 (without supplements) | 3.1 | Must be freshly made for every experiment. |
1 mM DTT | |||
1 mM Mg-GTP |
Table 2: Buffers for the polymerization of microtubules (Protocol section 3).
Buffer Name | Composition | Step(s) Used | Comment |
Taxol-Supplemented Lysis Buffer | 1x Lysis Buffer | 4.1 | Must be freshly made for every purification. |
20 µM Taxol (dissolved in DMSO) | |||
1 mM DTT | |||
5x ATP/Taxol Mix | 1x Lysis Buffer | 4.2 | Must be freshly made for every purification. |
25 mM Mg-ATP | |||
50 µM Taxol (dissolved in DMSO) | |||
5 mM DTT |
Table 3: Buffers for the microtubule affinity purification of functional motor proteins (Protocol section 4).
Buffer Name | Composition | Step(s) Used | Comment |
20x Origami Folding Buffer | 100 mM Tris pH 8.0 | – | Can be stored at RT in a properly sealed container for up to a year. |
20 mM EDTA | |||
200 mM MgCl2 | |||
1x Origami Folding Buffer | 5 mM Tris pH 8.0 | 5.1-5.2 | Make the buffer fresh by diluting the 20x stock with ddH2O before every experiment. |
1 mM EDTA | |||
10 mM MgCl2 | |||
0.5x TBE | 45 mM TrispH 8.0 | 5.1-5.2 | Can be stored at RT in a properly sealed container for up to a year. |
45 mM Boric Acid | |||
1 mM EDTA |
Table 4: Buffers for the production of segmented DNA origami chassis (Protocol section 5).
Buffer Name | Composition | Step(s) Used | Comment |
DTT-Supplemented BRB80 | 1x BRB80 | 7.3 | Must be freshly made before every TIRF experiment. |
1 mM DTT | |||
Taxol-Supplemented Lysis Buffer | 1x Lysis Buffer | 7.1 & 7.3 | Must be freshly made before every TIRF experiment. |
20 µM Taxol (dissolved in DMSO) | |||
1 mM DTT | |||
Casein-Taxol-Supplemented Lysis Buffer | 1x Lysis Buffer | 7.1-7.3 | Must be freshly made before every TIRF experiment. |
20 µM Taxol (dissolved in DMSO) | |||
1 mM DTT | |||
~2.5 mg/ml Casein (dissolved in Tris-HCl at pH 8.0) | |||
1x Lysis Buffer | 30 mM HEPES (pH 7.4) | 7.2 | Make this buffer fresh by diluting the 5x stock (recipe detailed in Table 1) with ddH2O. |
50 mM KAcetate | |||
2 mM MgAcetate | |||
1 mM EGTA (pH 7.5) | |||
10% glycerol | |||
4x Energy Mix | 22.5 µL 1x Casein-Taxol-Supplemenfted Lysis Buffer | 7.3 | Must be freshly made before every TIRF experiment; volumes indicated are for a final volume of 25 µL. |
1 µL 0.1 M Mg-ATP | |||
1 µL 40% Glucose | |||
0.5 µL β-Mercaptoethanol | |||
4x Scavenger Mix | 24 µL 1x Casein-Taxol-Supplemented Lysis Buffer | 7.3 | Must be freshly made before every TIRF experiment; volumes indicated are for a final volume of 25 µL. |
1 µL 1x Oxygen Scavenger System |
Table 5: Buffers and mixes for the motor-ensemble motility TIRF assay (Protocol section 7). Details on the Oxygen Scavenger System used to make the Scavenger Mix can be found elsewhere17.
The molecular construction techniques of DNA origami provide a unique way to construct motor ensembles with defined architectures, motor numbers, and types, enabling studies of how emergent behavior arises from specific motor configurations31. As structural and cellular studies continue to elucidate examples of cytoskeletal motors working in teams, techniques for isolating and investigating the biophysical and biochemical mechanisms of motors in ensembles are growing in utility. For example, cryo-EM has shown that dynactin can bind 2 individual dynein motors, and that such pairings have different motility than the individual motors10. In addition, DNA-based construction was used to determine if mammalian dynein, when activated by dynactin and bicD2, could match the force of kinesin-1 in a tug of war scenario14. In another mixed-motor study, DNA origami was used to decouple the effects of opposite polarity motors and their regulatory binding proteins by spatially separating them on an origami structure15. As more structural and regulatory determinants of motility are found, DNA-origami-based techniques should prove useful in determining the specific biochemical and biophysical contributors to the emergent motility of motor ensembles. The molecular construction techniques enabled by DNA origami are particularly useful because the emergent phenomenological outcomes of ensembles are difficult to predict. This is due in part to the myriad factors that contribute to the motility of the individual motors within the ensemble5,6,31.
Examples of our previous structures include monolithic rods and segmented rods with variable rigidity. Current efforts explore spherical structures as well25. Others have employed morphologies such as planar structures and rods22,24. Likewise, by using motor handles with orthogonal DNA sequences, different types of motors can be bound to the same chassis structure7,14,15,18. This approach enables studies of the opposing actions of dyneins and kinesins, minus- and plus-end directed kinesins, and minus- and plus-end directed myosins. It also enables the introduction of a mutant among otherwise wild-type ensembles, allowing the specific biochemical contributors to the emergent motility to be deciphered7. Because of the ability to bind multiple fluorophores to each individual structure, imaging in TIRF and subsequent analysis by kymography or particle tracking is possible. Previous reports show analysis of kymography data and statistical evaluation of chassis structures with variable compliance8. While cytoskeletal motors have proven to be an exciting early application of using DNA origami as a molecular breadboard32, other proteins and protein systems will also benefit from these methods.
The authors have nothing to disclose.
We thank K. Chau, J. Morgan, and A. Driller-Colangelo for contributing to the techniques of the segmented DNA origami chassis. We also thank former members of the Reck-Peterson and Shih laboratories for helpful discussions and contributions to the original development of these techniques. We thank J. Wopereis and the Smith College Center for Microscopy and Imaging and L. Bierwert and the Smith College Center for Molecular Biology. We gratefully acknowledge the NSF MRI program for the acquisition of a TIRF microscope.
2 mL Round Bottom Tube | USA Scientific | 1620-2700 | |
Biotin labeled tubulin protein: porcine brain, >99% pure | Cytoskeleton.com | T333P-A | |
Biotin-BSA | Sigma | A8549-10MG | |
Bottle Assembly, Polycarbonate, 250 mL, 62 x 120 mm | Beckman Coulter | 356013 | |
Bottle, with Cap Assembly, Polycarbonate, 10.4 mL, 16 x 76 mm | Beckman Coulter | 355603 | |
Centrifugal Filter Unit | Millipore Sigma | UFC30VV00 | |
IgG Sepharose 6 Fast Flow, 10 mL | GE Healthcare | 17096901 | |
Micro Bio-Spin Chromatography Columns, empty | Bio-Rad | 7326204EDU | |
P8064 Scaffold | Tilibit | 2 mL at 400nM | |
Poly-Prep Chromatography Columns | Bio-Rad | 731-1550 | |
ProTev Protease | Promega | V6101 | |
Scotch Double Sided Tape with Dispenser | amazon.com | N/A | |
Sephacryl S-500 HR | GE Healthcare | 17061310 | |
Streptavidin | Thermo Fisher | 434302 | |
SYBR Safe DNA stain | Invitrogen | ||
Tubulin protein (>99% pure): porcine brain | Cytoskeleton.com | T240-B | |
Tubulin, HiLyte 647 | Cytoskeleton.com | TL670M-A | |
Ultra-Clear Centrifuge Tubes | Beckman Coulter | 344090 |