CHAPTER
Reconstituting Functional Microtubule–Barrier Interactions
5
Nu´ria Taberner*, Georges Weber*, Changjiang You{, Roland Dries*, Jacob Piehler{, and Marileen Dogterom* *
FOM Institute AMOLF, Amsterdam, The Netherlands { Universita¨t Osnabru¨ck, Osnabru¨ck, Germany
CHAPTER OUTLINE Introduction .............................................................................................................. 70 5.1 Rationale........................................................................................................... 71 5.1.1 Assay 1: Interactions of Dynamic Microtubule Tips with Barrier-Attached Dynein............................................................. 72 5.1.2 Assay 2: Microtubule-Based Delivery and Anchoring of þTIPs to Barriers.......................................................................... 72 5.2 Materials........................................................................................................... 73 5.2.1 Microfabrication ............................................................................... 73 5.2.1.1 Special Equipment......................................................................73 5.2.1.2 Reagents....................................................................................73 5.2.2 Surface Functionalization.................................................................. 73 5.2.2.1 Special Equipment......................................................................73 5.2.2.2 Reagents....................................................................................74 5.2.3 Characterization of Surfaces and Microtubule Assays........................... 74 5.2.3.1 Special Equipment......................................................................74 5.2.3.2 Reagents....................................................................................75 5.3 Methods ............................................................................................................ 76 5.3.1 Microfabrication ............................................................................... 76 5.3.1.1 Wells with Gold Walls .................................................................76 5.3.1.2 Glass Wells with Overhang .........................................................78 5.3.2 Selective Barrier Functionalization..................................................... 79 5.3.2.1 Functionalization of Gold Walls with Biotinylated Dynein ............82 5.3.2.2 Selective Functionalization of Glass Walls with Ni(II)-NTA ..........82 5.3.3 FRAP Characterization of Ni(II)-NTA Surfaces Coated with His-Tagged Proteins........................................................................................... 83 5.3.4 Fluorescent Microscopy Assays of Dynamic Microtubules Interacting with Functionalized Barriers ............................................. 84
Methods in Cell Biology, Volume 120 Copyright © 2014 Elsevier Inc. All rights reserved.
ISSN 0091-679X http://dx.doi.org/10.1016/B978-0-12-417136-7.00005-7
69
70
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.3.4.1 Assay 1: Interactions of Dynamic Microtubule Ends with Barrier-Attached Dynein .........................................................................................84 5.3.4.2 Assay 2: Microtubule-Based Delivery and Anchoring of þTIPs to Barriers ...86 5.4 Discussion and Perspectives .............................................................................. 86 Acknowledgements ................................................................................................... 88 References ............................................................................................................... 88
Abstract Local interactions between the tips of microtubules and the cell cortex, or other cellular components such as kinetochores, play an important role in essential cellular processes like establishing cell polarity, distribution of organelles, and microtubule aster and chromosome positioning. Here we present two in vitro assays that specifically mimic microtubule–cortex interactions by employing selectively functionalized microfabricated barriers that allow for the immobilization of proteins with a range of affinities. We describe the microfabrication process to create gold or glass barriers and the subsequent functionalization of these barriers using self-assembled thiol monolayers or polylysine–poly(ethylene glycol), respectively. Near-permanent attachment of proteins is obtained using biotinylated surfaces combined with streptavidin and biotinylated proteins. Lower affinity interactions, further tunable with the addition of imidazole, are obtained using nickel-nitrilotriacetic acid (Ni(II)-NTA) functionalization combined with his-tagged proteins. Both mono-NTA and trisNTA compounds are used. We show an assay to reconstitute the “end-on” interaction between dynamic microtubule tips and barrier-attached dynein, mimicking the cellular situation at the cortex and at kinetochores. In a second assay, we reconstitute microtubule-based delivery of end-tracking proteins to functionalized barriers, mimicking the transport of cell-end markers to the cell poles in interphase fission yeast cells.
INTRODUCTION Microtubules are stiff long biopolymers with a fast growing plus end and a slow growing minus end that extend along the cytoplasm providing a physical connection between distant organelles in the cell. They serve as tracks for selective transport of cellular components by molecular motors (Goodman, Derr, & Reck-Peterson, 2012; Vale, 2003) and play an important role in cellular organization. Microtubule plus ends explore the cytoplasmic space by switching stochastically between states of polymerization and depolymerization. This intrinsic process is termed “dynamic instability”, where the sudden switch from polymerization to depolymerization is called “catastrophe” (Desai & Mitchison, 1997; Mitchison & Kirschner, 1984).
5.1 Rationale
In cells, dynamic instability is regulated by several microtubule-associated proteins (MAPs) as well as physical interactions with the cell boundaries. Interestingly, so-called microtubule plus-end tracking proteins (þTIPs), MAPs that interact preferentially with the growing plus ends of microtubules (Akhmanova & Steinmetz, 2008), can also mediate microtubule interactions with the cell cortex or other cell components such as kinetochores. In most cells, microtubule minus ends remain stably attached to microtubulenucleating organelles close to the nucleus while the plus ends grow toward the cell boundaries. At the cell boundary, microtubule polymerization can generate pushing forces up to several pN before undergoing catastrophe (Dogterom & Yurke, 1997; Fygenson, Braun, & Libchaber, 1994; Janson, de Dood, & Dogterom, 2003). When needed, cells can reduce the catastrophe rate by employing designated þTIPs such as cytoplasmic linker associated proteins (CLASPs) that stabilize microtubule ends while keeping them linked to the cell boundary. Also, during mitosis, kinetochore proteins link the ends of spindle microtubules to the chromatids, stabilizing them until chromosomal segregation. Shrinking microtubule ends can also produce pulling forces (Grishchuk, Molodtsov, Ataullakhanov, & McIntosh, 2005; Lombillo, Stewart, & McIntosh, 1995). For example during chromosomal segregation, the chromatids are pulled toward opposite sides of the cell by maintaining a dynamic link between kinetochore complexes and the ends of depolymerizing microtubules. Similarly, cortical dynein can capture growing microtubule plus ends, trigger catastrophes, and keep shrinking microtubule ends linked to the cortex (Laan, Pavin, et al., 2012). This generates pulling forces on anything linked to the other end of the microtubule, which in cells contributes to the positioning of microtubule organizing centers such as centrosomes (Laan, Pavin, et al., 2012). In addition to generating forces, microtubule ends may exchange molecules with the cortex when they reach cellular boundaries. Motor proteins such as kinesin and cytoplasmic dynein transport organelles and other proteins toward either the plus or the minus end. Proteins delivered to the cell boundary may associate with the cortex leading to cortical protein patterns. For example, in fission yeast, protein tea1, involved in polarized cell growth, is transported at polymerizing microtubule tips with the help of mal3 and the kinesin tea2 to be delivered at the cell poles where it accumulates (Dodgson et al., 2013; Sawin & Snaith, 2004).
5.1 RATIONALE In vitro reconstitution experiments allow for the study of biological processes using a bottom-up approach by adding only minimal components. Since we are interested in reconstituting functional interactions between dynamic microtubule ends and proteins that are either (semi-)permanently or transiently associated with cellular objects such as the cell boundary or chromosomes, we reconstitute microtubule systems in microwells with functionalized walls to which proteins can be specifically attached with tunable affinity.
71
72
CHAPTER 5 Functional Microtubule–Barrier Interactions
These experiments require the selective functionalization of the walls of the wells with desired proteins as well as passivation of the bottom surfaces. Here we describe two methods: the first one is adapted from Romet-Lemonne, VanDuijn, and Dogterom (2005) and consists of immobilizing proteins to gold walls via self-assembled monolayers (SAMs) of alkanethiols with a functional terminal group (Frasconi, Mazzei, & Ferri, 2010; Love, Estroff, Kriebel, Nuzzo, & Whitesides, 2005). This method was used before in our lab to produce “end-on” interactions between microtubule tips and cortical dynein (Laan & Dogterom, 2010; Laan, Pavin, et al., 2012). The second method consists of using both bottom surfaces and walls made of glass, where selectivity is achieved by irradiating the bottom surfaces with deep UV to remove the coating there. This technique has been successfully used on flat surfaces before with the help of an external mask (Azioune, Storch, Bornens, The´ry, & Piel, 2009). Here, we apply it to 3D structures using a microfabricated mask. To study the cellular functions described earlier, different affinities for protein binding are required. We distinguish two cases: proteins forming stable links with the organelle or cell cortex, such as kinetochore proteins, cortical dynein, or CLASPs; proteins that form transient links. For the first case, we immobilize biotinylated proteins at the walls using biotin–streptavidin sandwich linkages. For the second case, we use Ni(II)-NTA, which allows for a reversible immobilization of polyhistidine-tagged proteins, tunable by competition with imidazole (Lata & Piehler, 2005; Lata, Reichel, Brock, Tampe´, & Piehler, 2005). To achieve a wide range of affinities we use both mono-Ni(II)-NTA and tris-Ni(II)-NTA. We specifically describe the methodology to produce the two assays that are shown in Figs. 5.1, 5.6, and 5.7 and that are summarized below. Note however that different combinations of methods are also possible. Section 5.3.1 describes the steps for fabrication of gold and glass barriers; Section 5.3.2, the selective functionalization of barriers, Section 5.3.3, the characterization of the Ni(II)-NTA functionalized surfaces by fluorescence recovery after photobleaching (FRAP); and Section 5.3.4 explains the preparation and imaging of our two assays with dynamic microtubules.
5.1.1 Assay 1 (Fig. 5.1A): interactions of dynamic microtubule tips with barrier-attached dynein In this assay dynamic microtubules grow from freely floating guanylyl-(alpha, beta)methylene-diphosphonate (GMPCPP)-stabilized microtubule seeds (Hyman, Salser, Drechsel, Unwin, & Mitchison, 1992) inside microwells. The walls are 100 nm high made of gold. There is a 10 nm adhesion layer of chromium below the gold and a 25 nm one on top. The gold walls are coated with a SAM of biotin-terminated thiols followed by a multilayer of streptavidin tetramers and biotinylated BSA. Biotinylated dynein molecules are bound to a final streptavidin layer.
5.1.2 Assay 2 (Fig. 5.1B): microtubule-based delivery and anchoring of þTIPs to barriers In this assay dynamic microtubules grow from biotinylated GMPCPP-stabilized microtubule seeds bound to the bottom of microwells via biotin–streptavidin links. The wells are 300 nm deep, made of glass, with an overhang of 50 nm gold on top of
5.2 Materials
50 nm chromium. This overhang prevents the microtubules from growing over the walls and serves as a mask for functionalization steps. The walls are functionalized with PLL–PEG–tris-Ni(II)-NTA to which þTIPs with a his-tag can reversibly bind (Bhagawati, You, & Piehler, 2013). We use the end-tracking system from fission yeast consisting of the EB1 analog mal3, kinesin tea2, and protein tip1 (Bieling et al., 2007).
5.2 MATERIALS 5.2.1 Microfabrication 5.2.1.1 Special equipment All the microfabrication steps except evaporation were performed inside a clean room (class ISO 6). Coverslips No. 1 24 24 mm 170 mm (Menzel Gla¨sser, Germany). Delta 80 GYSET® Spin coater with a close lid (Su¨ss MicroTec, Germany). Hot plates. N2 gas (hand gun). Homemade turbo pumped vacuum system with base pressure 107 mbar equipped with a resistance heating evaporation system for tungsten boats loaded with chromium or gold. 2510 Ultrasonic Cleaner (Branson, USA). MJB3 mask aligner for UV exposure (Su¨ss MicroTec). Binary chromium/soda lime mask (Delta Mask, The Netherlands).
5.2.1.2 Reagents Hexamethyldisilazane (HMDS) primer (440191 Sigma-Aldrich, USA).
5.2.1.2.1 Photoresists
Shipley Microposit® S1805 G2 positive UV-resist (Microresist, Germany). Negative tone photoresist ma-N1410 (Microresist).
5.2.1.2.2 Developers
Microposit® MF®-319 developer (Microresist). Ma-D 533/S; alternatively ma-D332S (Microresist).
5.2.1.2.3 Etchants Standard gold etchant (651818 Sigma-Aldrich). Standard chromium etchant (651826 Sigma-Aldrich).
5.2.2 Surface functionalization 5.2.2.1 Special equipment Compact UV-Ozone cleaner (Uvotech, USA). Binary chromium/quartz mask (Delta Mask).
73
74
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.2.2.2 Reagents All reagents were dissolved in MRB80 buffer (80 mM KPipes, 4 mM MgCl2, 1 mM ethylene glycol tetraacetic acid (EGTA), pH 6.8) at the stated stock concentration and stored at 80 C unless stated otherwise. Albumin, biotin labeled bovine (A8549 Sigma-Aldrich); 20 mg/ml solution (biotinylated BSA). Streptavidin (43–4302 Invitrogen, USA); 1 mg/ml solution. Alexa Fluor® 488 streptavidin (S-11223 Invitrogen); 1 mg/ml solution. k-casein from bovine milk (C0406 Sigma-Aldrich); 5 mg/ml solution.
5.2.2.2.1 For gold surfaces Biotinylated PEG alkanethiol (CMT005-25 Nanoscience Instruments, USA); 0.1 mM solution in 100% ethanol stored at 20 C (HS–PEG–biotin).
5.2.2.2.2 For glass surfaces PLL(20)-g[3.5]–PEG(2) (SUSOS AG, Switzerland); 2 mg/ml solution (PLL–PEG). PLL(20)-g[3.1]–PEG(2)/PEG(3.4)–biotin(17.5%) (SUSOS AG); 2 mg/ml solution (PLL–PEG–biotin). PLL(20)-g[3.5]–PEG(3.5)–(3,4)NTA (SUSOS AG); 2 mg/ml solution (PLL–PEG–NTA). PLL-g3–PEG2k/PEG3.4K–tris-NTA (PLL–PEG–tris-NTA) made by a two-step coupling process, as described elsewhere (Bhagawati et al., 2013). Nickel(II) sulfate (656895-10G, Sigma-Aldrich), 10 mM NISO4 solution in MRB80 pH 7.5 stored at room temperature.
5.2.3 Characterization of surfaces and microtubule assays 5.2.3.1 Special equipment
Airfuge® Air-driven ultracentrifuge (Beackman Coulter, USA). Leica microscope equipped with a spinning disk confocal head (Yokogawa, Japan). 100 1.3 NA oil immersion objective and cooled EMCCD camera (C9100, Hamamatsu Photonics, Japan). Excitation lasers: Sapphire 488 nm (30 mW) (Coherent, Germany) and 561 nm class 3B (100 mW) (Melles Griot). Total internal reflection fluorescence microscope (TIRF) (Nikon Corporation, Japan) equipped with an Apo TIRF 100 1.49 N.A. oil objective, a motorized stage, Perfect Focus System, a motorized TIRF illuminator (Roper Scientific, France) and a QuantEM:512SC EMCCD camera (Photometrics, Roper Scientific). Excitation lasers: 561 nm (50 mW) Jive (Cobolt, Sweden) and a 488 nm (40 mW) Calypso (Cobolt) diode-pumped solid-state laser. For FRAP experiments the microscope is equipped with a MAG Biosystems FRAP-3D system (Photometrics, Roper Scientific).
5.2 Materials
Glass slides (Menzel Gla¨sser). Tesa® double-sided tape (15 mm wide). Diamond glass cutter. Valap (vaseline, lanolin, paraffin wax melted at equal concentrations).
5.2.3.2 Reagents All reagents were dissolved in MRB80 buffer at the stated stock concentration and stored at 80 C unless stated otherwise.
5.2.3.2.1 Microtubule polymerization and molecular motor assays Tubulin from porcine brain (T240 Cytoskeleton, Inc., USA); 100 mM solution. Rhodamine-labeled tubulin from porcine brain (TL590M Cytoskeleton, Inc.); 50 mM solution. Fluorescent HiLyte 488 tubulin from porcine brain (TL488M Cytoskeleton, Inc.); 50 mM solution. Biotinylated tubulin from porcine brain (T333P Cytoskeleton, Inc.); 50 mM solution. Guanosine 50 -triphosphate sodium salt hydrate (GTP) (G8877 Sigma-Aldrich); 50 mM solution. GMPCPP (NU-405 Jena BioScience); 10 mM solution. Glucose oxidase from Aspergillus niger (G2133 Sigma-Aldrich); 20 mg/ml dissolved in 200 mM DL-Dithiolthreitol (646563 Sigma-Aldrich) with 10 mg/ml catalase from bovine liver (C9322 Sigma-Aldrich) (glucose oxidase). D-(þ)-Glucose (G8270 Sigma-Aldrich). Adenosine 50 -triphosphate (ATP) (A7699 Sigma-Aldrich); 50 mM solution.
5.2.3.2.2 GMPCPP-stabilized microtubule seeds Microtubule seeds are prepared by a polymerization–depolymerization–polymerization cycle with GMPCPP in MRB80 (this cycling is done to get rid of residual GTP). The first polymerization step is performed by incubating a tubulin mix of biotinylated tubulin, fluorescently labeled and nonlabeled tubulin in a ratio 18:12:70 (total of 20 mM tubulin) with 1 mM GMPCPP at 37 C for 30 min. For nonbiotinylated seeds simply exchange the biotinylated tubulin by nonlabeled tubulin. The mix is then airfuged 5 min at 30 psi and the pellet is resuspended in 80% of the initial volume and left on ice for 20 min to depolymerize the microtubules. Then, 1 mM GMPCPP is added to the mix and left for 30 min at 37 C to polymerize again. The seeds are then airfuged again, resuspended in MRB80 with 10% glycerol and stored at 80 C. Thawed seeds can be kept at room temperature for a week.
5.2.3.2.3 Microtubule-associated proteins The dynein construct used is GFP-SNAP-GST-Dynein331-HaloTag (modified from Reck-Peterson et al. (2006)), where the GST allows it to dimerize. A biotin was added via the SNAP tag (S9110S New England BioLabs, UK) and a TMR (tetramethylrhodamine) label via HaloTag (G8251 Promega, France).
75
76
CHAPTER 5 Functional Microtubule–Barrier Interactions
Kinesin Tea2 was expressed in E. coli and purified as in Bieling et al. (2007). The following proteins were purified as in Maurer, Bieling, Cope, Hoenger, and Surrey (2011). 6his-TEV-Mal3-mCherry sequence was obtained by deleting the GFP sequence on 6his-Mal3-GFP (generous gift by Thomas Surrey, Maurer et al., 2011) and inserting the mCherry sequence. Unlabelled 6his-TEV-mal3 was obtained as in Bieling et al. (2007). The his-tag was removed by cleavage overnight with AcTEV™ protease (12575-015 Invitrogen) at 4 C according to the provider’s protocol. 6his-TEV-eGFP-Tip1 was obtained as in Bieling et al. (2007) and the oligo duplex 50 -tccgcgggtgagaatcttcagggcgcc-30 containing a TEV recognition site was added between the his-tag and the eGFP.
5.3 METHODS 5.3.1 Microfabrication Here we describe the production of two different designs of microfabricated wells: (1) wells with gold walls obtained by lift-off (used in assay 1, Figs. 5.1A and 5.2A), and (2) glass wells with an overhang obtained by wet etch (used in assay 2, Figs. 5.1B and 5.2B). An alternative method to produce gold walls using plasma etching can be found in Laan & Dogterom (2010).
5.3.1.1 Wells with gold walls (Fig. 5.2A and C) 5.3.1.1.1 Photolithography Start by cleaning the glass coverslips with base piranha (NH4OH:H2O2 in 3:1 at 75 C) for 15 min. Rinse them, first in ddH2O and then in isopropanol and blow dry with N2 flow. Bake the samples 5 min on a hot plate at 150 C, cool them down with N2 flow, and spin-coat a thin layer of HMDS primer (4000 rpm for 32 s). Bake the coverslips 5 min at 150 C, cool them down with N2 flow, and spin-coat a thin layer of negative tone photoresist ma-N1410 (3000 rpm for 30 s). Bake the slides 2 min on a hot plate at 120 C. This leads to a 500 nm layer of resist. Expose to a dose of 325 mJ/cm2 of 365 nm UV light using a Su¨ss MJB3 mask aligner with a binary chromium/soda lime mask on top. Immerse the sample in developer ma-D332S for 120 s while constantly moving the fluid (e.g., by shaking the container). The UV unexposed pattern will be developed (dissolved) while leaving the rest of the photoresist. Rinse in ddH2O and dry with N2 flow.
5.3.1.1.2 Evaporation of metals Evaporate a 10 nm layer of chromium, followed by 100 nm of gold, and finally 25 nm of chromium in an evaporation chamber under pressure below 106 mbar with a deposition rate of 0.06 nm/s. The chromium layer between the glass and the gold ensures the good deposition of the gold.
5.3 Methods
FIGURE 5.1 Schematic cartoons describing assays 1 (A) and 2 (B). (A) Assay 1: A dynamic microtubule grows against a gold wall functionalized with dynein. The bottom surface is passivated with PLL–PEG. (B) Assay 2: A dynamic microtubule grows from a stabilized microtubule seed against a glass wall selectively functionalized with tris-Ni(II)-NTA. The microtubule seed is attached to the bottom surface via biotin–streptavidin links. The microtubule delivers his-tagged tip1 to the wall.
5.3.1.1.3 Lift-off Lift off the photoresist by sonicating the coverslips in acetone for 1 h. This method often leaves deposited particles of chromium on the gold that will hinder the functionalization. If needed, they can be removed by rinsing the sample in chromium etchant for 10 s. Clean again the sample with base piranha to make it ready for functionalization steps. The sample will be very hydrophilic. Each coverslip can be cut with a glass cutter in four long pieces useful for four different flow cell experiments.
77
78
CHAPTER 5 Functional Microtubule–Barrier Interactions
FIGURE 5.2 Schematic drawings showing the steps of the microfabrication process for gold barriers (A) and glass barriers with overhang (B). (C and D) SEM images of microfabricated wells. A 10 nm layer of chromium was sputtered on top of the samples with a K575X Turbo Sputter Coater (EMITECH Ltd) allowing a good conduction of electrons. (D, right) The overhang was removed with wet etch so that the glass walls can be seen. Note that due to wet etching steps the walls can become irregular depending on the quality of the adhesion between different layers.
5.3.1.2 Glass wells with overhang (Fig. 5.2B and D) 5.3.1.2.1 Evaporation
Start by cleaning the coverslips with base piranha (NH4OH:H2O2 in 3:1 at 75 C) for 15 min and rinse them in ddH2O and isopropanol. Evaporate a 50 nm layer of chromium and 50 nm of gold (both rates 0.06 nm/s) in an evaporation chamber at pressure below 106 mbar. The gold layer protects the chromium from being attacked later by KOH and also prevents the chromium from peeling off due to mechanical strain.
5.3 Methods
5.3.1.2.2 Photolithography
Dry the samples 5 min at 150 C, cool them down with N2 flow, spin-coat a thin layer of HMDS primer (4000 rpm for 32 s), and bake 1 min at 150 C. Spin-coat a thin layer (around 300 nm) of S1805 positive photoresist (500 rpm for 8 s followed by 4000 rpm for 32 s), let dry 10 s, and soft bake at 115 C for 60 s. Expose to a dose of 25 mJ/cm2 of 365 nm UV light using a Su¨ss MJB3 mask aligner with a binary chromium/soda lime mask on top. For structures that do not require sharp corners (like straight lines), the sample can be postbaked 30 min at 130 C which melts a little photoresist providing smoother structures. Immerse the sample in developer MF-319 for 1 min. The UV-exposed pattern will be developed (dissolved) while leaving the rest of the photoresist.
5.3.1.2.3 Wet etching Immerse sequentially the sample in gold etchant (etch rate around 2.8 nm/s) and chromium etchant (etch rate around 4 nm/s). Each step has to be carefully checked with a microscope because the etching rates depend on the exact conditions. Remove the photoresist by sonicating the samples 20 min in acetone. If needed, further removal can be achieved by base piranha. Isotropically etch the glass by immersing it in a 40% KOH solution at 80 C (etch rate around 5 nm/min). We found that for Menzel Gla¨sser coverslips, KOH produced a smoother final surface than plasma etching or HF (not shown). A depth of 300 nm is convenient for later imaging of microtubules with confocal microscopy. Clean again the sample with base piranha to make it ready for functionalization steps and optionally cut each sample in four long pieces.
5.3.2 Selective barrier functionalization Both assays shown in Fig. 5.1 require selective functionalization of the walls of the microwells. In the first assay, this is achieved by using different substrates. Poly(L-lysine)–poly(ethylene glycol) (PLL–PEG) chains spontaneously bind (mostly by electrostatic interactions) to negatively charged surfaces like glass (Kenausis et al., 2000) (Fig. 5.3B–D), whereas thiols (SH) form SAMs specifically on gold (Fig. 5.3A) (Frasconi et al., 2010). In the second assay where both surfaces consist of glass, selectivity can be achieved by first coating all surfaces with the compound desired for the walls, irradiating the bottom surfaces with deep UV to photocleave the molecules there, and then coating the irradiated surfaces with the second compound. Selective irradiation of the bottom surfaces is automatically achieved due to the microfabricated overhang on the walls (Fig. 5.4B). For quasi-permanent immobilization of proteins a high binding affinity is required so that proteins do not unbind on experimental timescales. We use biotin–streptavidin links, whose equilibrium dissociation constant Kd is reported to be of the order of 1014 M in solution (Green, 1990) and 109 M on beads (Buranda, Lopez, Keij, Harris, & Sklar, 1999). In our assays, while a single biotin–streptavidin layer (Fig. 5.3B) is enough to immobilize biotinylated microtubule seeds (with several
79
FIGURE 5.3 Cartoons showing the surface functionalization used for protein immobilization on gold (A) and glass (B–D). (A) Biotinylated PEG-alkanethiols forming SAMs on gold substrates in a mixture with nonfunctionalized 11-Mercapto-1-undecanol molecules. The biotin is covered with a multilayer of streptavidin tetramers and biotinylated BSA. Biotinylated proteins can bind to the streptavidin tetramers. (B–D) Poly(L-lysine)-based coating of glass surfaces. (B) Immobilization of biotinylated proteins via PLL–PEG–biotin and one layer of streptavidin. (C) PLL–PEG–tris-Ni(II)-NTA for immobilization of histidine-tagged proteins. Each of the three NTAs chelates one nickel ion to which two histadines can bind. The binding affinity can be tuned by competition with imidazole. (D) PLL–PEG–Ni(II)-NTA bound to a glass surface. In this case four of the histidines remain unbound.
5.3 Methods
FIGURE 5.4 Cartoons illustrating the functionalization steps of gold (A) and glass (B) barriers. (C and D) Confocal images of fluorescent proteins bound to the surfaces: (C) dynein before (top) and after (bottom) buffer wash, (D) Alexa Fluor® 488 streptavidin at the bottom surface and histagged mal3-mCherry at the walls. When adding 1 M imidazole only mal3 unbinds. (C and D(ii)) Z-stacks along the dotted line in (i). Scale bars 10 mm.
binding sites), multiple layers of streptavidin and biotinylated BSA produce denser layers of functional proteins at the barriers (Fig. 5.3A) (Romet-Lemonne et al., 2005), suggesting an effective enhancement of the density of available binding sites. For reversible binding of þTIPs the affinity should be comparable to that for microtubule tips. Reversible binding of proteins can be achieved using the affinity of histidines to nickel ions chelated with nitrilotrilotriacetic acid (Ni(II)-NTA)
81
82
CHAPTER 5 Functional Microtubule–Barrier Interactions
(Hochuli, Do¨beli, & Schacher, 1987). Proteins with a hexahistidine tag (his-tag) fused at one of the termini will bind to Ni(II)-NTA with affinity tunable by competition with imidazole. We use both mono-Ni(II)-NTA (Fig. 5.3D, Kd 10 mM) (Zhen et al., 2006) and tris-Ni(II)-NTA (Fig. 5.3D, Kd 10 nM) depending on the affinity required. The latter one contains three chelator heads (NTAs) and binds his-tagged proteins stoichiometrically with higher affinity than mono-Ni(II)-NTA whose effective affinity depends strongly on the density of Ni(II)-NTAs at the surface (Lata & Piehler, 2005).
5.3.2.1 Functionalization of gold walls with biotinylated dynein (Fig. 5.4A) Drop a 20 ml solution of 0.2 mg/ml PLL–PEG on top of a microfabricated coverslip from Section 5.3.1.1. Insert the coverslip in a sealable plastic bag and keep the surface wet by close contact between the fluid and the inside of the bag. Sonicate the bag 5 seconds to remove possible air bubbles from the microwells. Further incubate for 30 min, later wash it with MRB80, and dry it with N2 flow. Now drop 50 ml solution of 1 mM HS–PEG–biotin on the sample and place it between two parafilm sheets. In order to form a good SAM, incubate between 10 and 14 h (overnight). Rinse the sample with ethanol and blow dry. Prepare a flow cell on a hellmanex® (HellmaAnalytics, UK) cleaned glass slide by attaching two parallel strips of double-sided TESA® tape separated by 5 mm. Place the coverslip on top with the microfabricated wells facing the glass slide (10–20 ml flow cell volume). Now you can flow in the subsequent solutions by putting absorbent paper on the other side of the cell. Avoid any dry out of the flow cell. Clean thoroughly the flow cell with MRB80 buffer and incubate 15 min with a 0.2 mg/ml PLL–PEG solution to passivate the slide surface. Wash with MRB80 and perform alternated 10 min incubations of 1 mg/ml streptavidin (three times) and 1.5 mg/ml biotinylated BSA (two times). Wash thoroughly with MRB80 in between incubations. Further passivate the surfaces with 1.2 mg/ml k-casein for 10 min. Incubate with a mix containing 10 nM dynein, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, and 50 mM glucose in MRB80 for 10 min. Wash with MRB80. Figure 5.4C shows examples of confocal microscopy of dynein-coated walls before and after the washing step. For a control experiment without dynein we simply omit this final incubation step.
5.3.2.2 Selective functionalization of glass walls with Ni(II)-NTA (Fig. 5.4B) Drop 20 ml solution of 0.1 mM PLL–PEG–NTA or PLL–PEG–tris-NTA on top of a microfabricated coverslip from Section 5.3.1.2. Put the sample in a sealable plastic bag and immerse it in an ultrasonication bath for 5 seconds. Incubate for 30 min, wash with MRB80, and dry with N2 flow. Expose to deep UV (185 and 254 nm) with an Ozone Cleaner for 10 min (the time depends on the power of the lamp and the distance to the sample). Prepare a flow cell on a glass slide with the coverslip and double-sided TESA® tape as described in the previous section. Load the NTA with nickel ions by incubating a 10 mM NiSO4 solution for 30 min.
5.3 Methods
Clean thoroughly the flow cell with MRB80 buffer and incubate 15 min with 0.2 mg/ml PLL–PEG–biotin, followed by 1 mg/ml streptavidin, and passivate with 1.2 mg/ml k-casein (10 min incubation each, rinsed in between with MRB80). For a control experiment to visualize the selective coatings, omit the incubation of nonlabeled streptavidin and incubate a mix containing 500 nM his-tagged mal3mCherry, 0.1 mg/ml 488 alexa streptavidin, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, and 50 mM glucose in MRB80. This should lead to a bottom surface coated with streptavidin and only the walls coated with mal3 (Fig. 5.4D). When washing with 1 M imidazole, mal3 detaches from the walls while fluorescent streptavidin remains at the bottom surface.
5.3.3 FRAP characterization of Ni(II)-NTA surfaces coated with his-tagged proteins We perform FRAP measurements to determine whether there is a turnover of proteins at the functionalized surfaces. These measurements are done on flat surfaces with micropatterned regions of PLL–PEG–Ni(II)-NTA around islands passivated with PLL–PEG. Since the density of protein immobilized by Ni(II)-NTA increases with protein concentration (Lata & Piehler, 2005), fluorescently labeled his-tagged þTIPs are incubated at the same concentration as in tip tracking experiments. Once the density of immobilized protein has reached equilibrium, we bleach a region and measure the recovery of fluorescence signal due to turnover of proteins between the bulk and the surface. Drop 20 ml solution of 0.1 mM PLL–PEG–NTA or PLL–PEG–tris-NTA on top of a base piranha cleaned glass coverslip and place it between two parafilm sheets for 30 min. Wash the sample with MRB80 and dry in N2 flow. Place on top a binary chromium/quartz mask with micropatterns (with the chromium side in good contact with the functionalized surface of the sample; use an extra weight if needed). Expose through the mask to deep UV with an Ozone Cleaner for 20 min. Prepare a flow cell on a glass slide with double-sided TESA® tape and the exposed coverslip as described earlier. Incubate with a 10 mM NiSO4 solution for 30 min. Clean thoroughly with MRB80 and incubate 15 min with 0.2 mg/ml PLL–PEG to passivate the regions exposed to UV. Further passivate with 1.2 mg/ml k-casein for 10 min and rinse with MRB80. Prepare a protein mix containing 100 nM his-tagged eGFP-tip1 or 200 nM histagged mal3-mCherry, 50 mM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose in MRB80. Airfuge the mix at 30 psi for 9 min and incubate in the flow cell. We image the surface of the sample with TIRF at 500 ms frame rate and photobleach a circle of 4.7 mm diameter for 1 s (Fig. 5.5A). The recovery of the fluorescence intensity of the bleached area, I, is followed over time and normalized to lie between the fluorescent signal at a nonbleached region that was passivated (I set to 0) and a nonbleached region that was functionalized with Ni(II)-NTA
83
84
CHAPTER 5 Functional Microtubule–Barrier Interactions
FIGURE 5.5 FRAP characterization of Ni(II)-NTA surfaces. (A) TIRF snapshots of a patterned surface with Ni(II)-NTA with 200 nM his-tagged mal3-mCherry in solution. A circular area in the center of the images is bleached and the recovery is followed over time. Scale bar 2 mm. (B) Normalized intensity recovery of the bleached area for an incubation with 100 nM his-tagged eGFP-tip1. Error bars correspond to the standard deviation between five different recoveries.
(I set to 1) (I ¼ (1/IbeforeFRAP)((IFRAP Ipassivated)/(InoFRAP Ipassivated))afterFRAP with IbeforeFRAP ¼ ((IFRAP Ipassivated)/(InoFRAP Ipassivated))beforeFRAP). Figure 5.5B shows examples of recovery curves for mono-Ni(II)-NTA and trisNi(II)-NTA surfaces for samples in equilibrium (bleached more than 10 min after incubation). Tip1 has a fast turnover on mono-Ni(II)-NTA surfaces (65% of the protein exchanges with a half time of 50 s) while it recovers, as expected, much slower on tris-Ni(II)-NTA surfaces.
5.3.4 Fluorescent microscopy assays of dynamic microtubules interacting with functionalized barriers 5.3.4.1 Assay 1: interactions of dynamic microtubule ends with barrier-attached dynein Prepare on ice a tubulin mix in MRB80 containing 14.25 mM tubulin, 0.75 mM HiLyte 488 labeled tubulin (5% of the total amount of tubulin), 50 mM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose, 0.1% methyl cellulose, 1 mM GTP, and 2 mM ATP. Airfuge the mix 9 min at 30 psi, warm up to room temperature, add rhodamine-labeled GMPCPP microtubule seeds (to a final 10 times dilution), and incubate in the flow cell from Section 5.3.2.1, that was coated before with dynein. Seal the flow cell with melted valap so that it does not dry out.
5.3 Methods
We image the sample at 27 C using spinning disk confocal microscopy with 488 and 561 nm lasers, taking images every 5 s (exposure 300 ms). For 100 nm barriers, both the bottom surface and the walls can be imaged in the same focal plane. However, some microtubules may grow over the barriers. We search for microtubules that grow with their plus ends towards the barrier. In control experiments without dynein, microtubules continue to polymerize at their plus end when contacting the wall. Since in this assay the microtubule seed is not attached to the surface, the addition of tubulin dimers at the wall pushes the microtubule away from the wall (Fig. 5.6A and B). For barriers functionalized with dynein we often observe the same behavior. However, in some cases the microtubule is now observed to be pulled toward the wall (Fig. 5.6C and D). In these cases dynein keeps the plus end tip associated with the wall while the microtubule depolymerizes (Fig. 5.6B and C). In Fig. 5.6C and D, the depolymerizing microtubule eventually detaches from the wall.
FIGURE 5.6 (A and C) Confocal images (time average over three snapshots) at different times of dynamic microtubules polymerizing against gold walls (barriers are indicated by dashed areas) without dynein (A) and with dynein (C). (B and D) Respective kymographs for the microtubules with the arrows. Since the microtubule fluctuates to the sides of its growth axis, the kymograph shows the maximum intensity along six pixels (1 mm) perpendicular to the microtubule growth axis. (B) When the microtubule polymerizes against the wall, the microtubule seed is pushed away from the barrier. (D) The plus end stays attached to the dynein-coated barrier while depolymerizing. As a result, the seed is being pulled towards the wall until the plus end detaches and the microtubule diffuses away.
85
86
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.3.4.2 Assay 2: microtubule-based delivery and anchoring of þTIPs to barriers Take the flow cell from Section 5.3.2.2 prepared with tris-Ni(II)-NTA. Incubate for 5 min with the biotinylated microtubule seeds 10 times diluted in MRB80 with 0.1% methyl cellulose so that they bind to the surface coated with PLL–PEG–biotin– streptavidin. Prepare on ice a tubulin mix with end-binding proteins in MRB80 containing 14.25 mM tubulin, 0.75 mM rhodamine tubulin, 100 nM his-tagged eGFP-tip1, 8 nM tea2, 200 nM mal3, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose, 0.1% methyl cellulose, 1 mM GTP, and 2 mM ATP. For assays of delivery of mal3, use his-tagged mal3-mCherry and HiLyte 488 labeled tubulin without tea2 and tip1. Airfuge 9 min at 30 psi, incubate in the flow cell, and seal it with valap. Since these wells are 300 nm deep, we take confocal z-stacks every 5 s (300 ms exposure, 300 nm step size). Each image of the z-stack is later background corrected and projected by maximum intensity. When using walls functionalized with tris-Ni(II)-NTA, his-tagged proteins in solution have a relatively high affinity for the walls, leading to spontaneous coverage of the walls (Fig. 5.7). Additionally, his-tagged end-binding proteins have a high affinity for growing microtubule ends. His-tagged mal3 binds to microtubule ends with a dwell time of about 0.3 s, leading to the appearance of mal3 comets at growing microtubule ends (Bieling et al., 2007). When these comets reach a barrier, they lead to a locally enhanced concentration of mal3 at the barrier while the microtubule is in contact with the wall, but the “extra” mal3 vanishes when the microtubule depolymerizes (Fig. 5.7A and B). This shows that microtubules cannot “deliver” mal3 to barriers that are precoated with mal3. By contrast, Fig. 5.7C and D shows a microtubule in the presence of his-tagged tip1 reaching a wall functionalized with tris-Ni(II)-NTA. When the microtubule reaches the wall an apparent cluster of tip1 proteins remains attached to the wall, even after the microtubule depolymerizes. This preliminary observation suggests that effective clustering of proteins such as observed for tip1, but not for mal3, may be important for cortical tethering of proteins.
5.4 DISCUSSION AND PERSPECTIVES In this chapter, we presented two methods to immobilize proteins selectively at barriers of gold or glass, using functional groups that allow for a broad range of affinities. Selective immobilization on gold walls had already been used to mimic cortical interactions (Laan & Dogterom, 2010; Laan, Pavin, et al., 2012). Here we additionally introduced a method for functionalization of glass barriers, using only PLL– PEGs combined with deep UV irradiation (Azioune et al., 2009). PLL–PEGs assemble faster than SAMs and suffer less from unspecific binding (Zhen et al., 2006), making them more suitable for immobilizations at high Kd. Selective irradiation
5.4 Discussion and Perspectives
FIGURE 5.7 (A and C) Confocal images (time average over five snapshots) at different times, before, during, and after contact of the microtubule end with a glass wall functionalized with trisNi(II)-NTA, in the presence of 200 nM his-tagged mCherry-mal3 (A) or in the presence of mal3, tea2, and his-tagged tip1-eGFP (C). (B and D) Respective kymographs for the microtubules with the arrows. His-tagged mal3 delivered to the wall by a microtubule does not remain anchored at the wall upon microtubule catastrophe (B), while his-tagged tip1 does (D). Spatial scale bar 2 mm.
of the bottom surface is easily achieved with the overhang of chromium produced by wet etch during the microfabrication procedure. Note, however, that an overhang of chromium amplifies the detected fluorescence intensity of proteins by up to a factor of 2 (data not shown). This is likely due to reflections of the excitation and emission light at the chromium layer. To avoid this effect, we are currently testing the use of an 80 nm overhang of TiO2, which is opaque to deep UV but 70–80% transparent to the wavelengths of interest (480–620 nm) (data not shown). An alternative way to selectively expose regions of the sample with deep UV is by employing an external binary chromium/quartz mask complementary to the 3D structures. In this case a proper alignment and close contact between the mask and the sample needs to be achieved. This method offers the additional possibility to selectively irradiate a subset of walls or wells in the sample. This would allow for control experiments with nonfunctionalized barriers in the same sample as experiments with functionalized barriers.
87
88
CHAPTER 5 Functional Microtubule–Barrier Interactions
We presented a method based on semipermanent biotin–streptavidin bonds to reconstitute dynamic microtubule “end-on” interactions for the specific case of cortical dynein (assay 1). However, the same technique could be used to study microtubule interactions with other cortical factors like CLASPs, or microtubule interactions with kinetochore complexes. Alternatively, “end-on” interactions have recently been achieved in emulsion droplets (Laan, Roth, & Dogterom, 2012) and using beads with the addition of a long tether (Volkov et al., 2013). Potentially, functionalized microwells could furthermore provide a framework to study controlled lateral interactions; for example, by placing cells inside wells where the walls mimic lateral cellular adhesion conditions. Ni(II)-NTA based immobilization does not require protein biotinylation and suffers little from protein inactivation (Bieling, Telley, Hentrich, Piehler, & Surrey, 2010). Moreover, since it is reversible and tunable with the addition of imidazole, it is better suited to study cellular processes that involve transient bindings. We showed in assay 2 that microtubule-based protein delivery and docking of his-tagged þtips can be achieved with tris-Ni(II)-NTA at walls. To also allow for diffusion of tethered proteins along the barrier, one could alternatively use emulsion droplets with Ni(II)-NTA lipids using a method analogous to Laan, Roth, et al. (2012).
Acknowledgements We thank Gijs Vollenbroek, Dimmitry Lamers, Andries Lof, and Hans Zeijlemaker for advice and help with microfabrication. Liedewij Laan for initiating this project and advice. Magdalena Preciado-Lopez, Sophie Roth, Cristina Manatschal, and Michel Steinmetz for teaching us the purification of proteins and for discussions. Vanda Sunderlı´kova´ for help with protein purification. We thank Pierre Recouvreux for teaching us cloning techniques and for discussions, and David Minde for careful reading of the manuscript. This work is part of the research program of the “Stichting voor Fundamenteel Onderzoek de Materie (FOM)” which is financially supported by the “Nederlandse organisatie voor Wetenschappelijk Onderzoek (NWO).” This work is further supported by NanoNextNL, a micro and nanotechnology consortium of the Government of the Netherlands and 130 partners.
References Akhmanova, A., & Steinmetz, M. O. (2008). Tracking the ends: A dynamic protein network controls the fate of microtubule tips. Nature Reviews. Molecular Cell Biology, 9, 309–322. Azioune, A., Storch, M., Bornens, M., The´ry, M., & Piel, M. (2009). Simple and rapid process for single cell micro-patterning. Lab on a Chip, 9, 1640–1642. Bhagawati, M., You, C., & Piehler, J. (2013). Quantitative real-time imaging of proteinprotein interactions by LSPR detection with micropatterned gold nanoparticles. Analytical Chemistry 85, 9564-9571. Bieling, P., Laan, L., Schek, H., Munteanu, E. L., Sandblad, L., Dogterom, M., et al. (2007). Reconstitution of a microtubule plus-end tracking system in vitro. Nature, 450, 1100–1105.
References
Bieling, P., Telley, I. A., Hentrich, C., Piehler, J., & Surrey, T. (2010). Fluorescence microscopy assays on chemically functionalized surfaces for quantitative imaging of microtubule, motor, and þTIP dynamics. Methods in Cell Biology, 95, 555–580. Buranda, T., Lopez, G. P., Keij, J., Harris, R., & Sklar, L. A. (1999). Peptides, antibodies, and FRET on beads in flow cytometry: A model system using fluoresceinated and biotinylated b-endorphin. Cytometry, 37, 21–31. Desai, A., & Mitchison, T. J. (1997). Microtubule polymerization dynamics. Annual Review of Cell and Developmental Biology, 13, 83–117. Dodgson, J., Chessel, A., Yamamoto, M., Vaggi, F., Cox, S., Rosten, E., et al. (2013). Spatial segregation of polarity factors into distinct cortical clusters is required for cell polarity control. Nature Communications, 4, 1834. Dogterom, M., & Yurke, B. (1997). Measurement of the force–velocity relation for growing microtubules. Science, 278, 856–860. Frasconi, M., Mazzei, F., & Ferri, T. (2010). Protein immobilization at gold–thiol surfaces and potential for biosensing. Analytical and Bioanalytical Chemistry, 398, 1545–1564. Fygenson, D. K., Braun, E., & Libchaber, A. (1994). Phase-diagram of microtubules. Physical Review E, 50, 1579–1588. Goodman, B. S., Derr, N. D., & Reck-Peterson, S. L. (2012). Engineered, harnessed, and hijacked: Synthetic uses for cytoskeletal systems. Trends in Cell Biology, 22, 644–652. Green, N. (1990). Avidin and streptavidin. Methods in Enzymology, 184, 51–67. Grishchuk, E. L., Molodtsov, M. I., Ataullakhanov, F. I., & McIntosh, J. R. (2005). Force production by disassembling microtubules. Nature, 438, 384–388. Hochuli, E., Do¨beli, H., & Schacher, A. (1987). New metal chelate adsorbent selective for proteins and peptides containing neighbouring histidine residues. Journal of Chromatography, 411, 177–184. Hyman, A. A., Salser, S., Drechsel, D. N., Unwin, N., & Mitchison, T. J. (1992). Role of GTP hydrolysis in microtubule dynamics: Information from a slowly hydrolyzable analogue, GMPCPP. Molecular Biology of the Cell, 3, 1155–1167. Janson, M. E., de Dood, M. E., & Dogterom, M. (2003). Dynamic instability of microtubules is regulated by force. Journal of Cell Biology, 161, 1029–1034. Kenausis, G. L., Vo, J., Elbert, D. L., Huang, N., Hofer, R., Ruiz-Taylor, L., et al. (2000). Poly(L-lysine)-g-poly(ethylene glycol) layers on metal oxide surfaces: Attachment mechanism and effects of polymer architecture on resistance to protein adsorption. Journal of Physical Chemistry, 1(04), 3298–3309. Laan, L., & Dogterom, M. (2010). In vitro assays to study force generation at dynamic microtubule ends. Methods in Cell Biology, 95, 617–639. Laan, L., Pavin, N., Husson, J., Romet-Lemonne, G., van Duijn, M., Preciado Lo´pez, M., et al. (2012). Cortical dynein controls microtubule dynamics to generate pulling forces that position microtubule asters. Cell, 148(3), 502–514. Laan, L., Roth, S., & Dogterom, M. (2012). End-on microtubule-dynein interactions and pulling-based positioning of microtubule organizing centers. Cell Cycle, 11, 3750–3757. Lata, S., & Piehler, J. (2005). Stable and functional immobilization of histidine-tagged proteins via multivalent chelator headgroups on a molecular poly(ethylene glycol) brush. Analytical Chemistry, 77, 1096–1105. Lata, S., Reichel, A., Brock, R., Tampe´, R., & Piehler, J. (2005). High-affinity adaptors for switchable recognition of histidine-tagged proteins. Journal of the American Chemical Society, 127, 10205–10215.
89
90
CHAPTER 5 Functional Microtubule–Barrier Interactions
Lombillo, V. A., Stewart, R. S., & McIntosh, J. R. (1995). Minus-end-directed motion of kinesin-coated microspheres driven by microtubule depolymerization. Nature, 373, 161–164. Love, J. C., Estroff, L. A., Kriebel, J. K., Nuzzo, G. R., & Whitesides, G. M. (2005). Selfassembled monolayers of thiolates on metals as a form of nanotechnology. Chemical Reviews, 105, 1103–1609. Maurer, S. P., Bieling, P., Cope, J., Hoenger, A., & Surrey, T. (2011). GTPgS microtubules mimic the growing microtubule end structure recognized by end-binding proteins (EBs). Proceedings of the National Academy of Sciences of the United States of America, 108(10), 3988–3993. Mitchison, J., & Kirschner, M. (1984). Dynamic instability of microtubule growth. Nature, 312, 237–242. Reck-Peterson, S. L., Yildiz, A., Carter, A. P., Gennerich, A., Zhang, N., & Vale, R. D. (2006). Single-molecule analysis of dynein processivity and step-ping behavior. Cell, 126, 335–348. Romet-Lemonne, G., VanDuijn, M., & Dogterom, M. (2005). Three-dimensional control of protein patterning in microfabricated devices. Nano Letters, 12, 1350–2354. Sawin, K. E., & Snaith, H. A. (2004). Role of microtubules and tea1p in establishment and maintenance of fission yeast cell polarity. Journal of Cell Science, 117, 689–700. Vale, R. D. (2003). The molecular motor toolbox for intracellular transport. Cell, 112, 467–480. Volkov, V. A., Zaytsev, A. V., Gudimchuk, N., Grissom, P. M., Gintsburg, A. L., Ataullakhanov, F. I., et al. (2013). Long tethers provide high-force coupling of the Dam1 ring to shortening microtubules. Proceedings of the National Academy of Sciences of the United States of America, 110, 7708–7713. Zhen, G., Falconnet, D., Kuennemann, E., Vo¨ro¨s, J., Spencer, N. D., Textor, M., et al. (2006). Nitrilotriacetic acid functionalized graft copolymers: A polymeric interface for selective and reversible binding of histidine-tagged proteins. Advanced Functional Materials, 16, 243–251.
Reconstituting Functional Microtubule–Barrier Interactions
5
Nu´ria Taberner*, Georges Weber*, Changjiang You{, Roland Dries*, Jacob Piehler{, and Marileen Dogterom* *
FOM Institute AMOLF, Amsterdam, The Netherlands { Universita¨t Osnabru¨ck, Osnabru¨ck, Germany
CHAPTER OUTLINE Introduction .............................................................................................................. 70 5.1 Rationale........................................................................................................... 71 5.1.1 Assay 1: Interactions of Dynamic Microtubule Tips with Barrier-Attached Dynein............................................................. 72 5.1.2 Assay 2: Microtubule-Based Delivery and Anchoring of þTIPs to Barriers.......................................................................... 72 5.2 Materials........................................................................................................... 73 5.2.1 Microfabrication ............................................................................... 73 5.2.1.1 Special Equipment......................................................................73 5.2.1.2 Reagents....................................................................................73 5.2.2 Surface Functionalization.................................................................. 73 5.2.2.1 Special Equipment......................................................................73 5.2.2.2 Reagents....................................................................................74 5.2.3 Characterization of Surfaces and Microtubule Assays........................... 74 5.2.3.1 Special Equipment......................................................................74 5.2.3.2 Reagents....................................................................................75 5.3 Methods ............................................................................................................ 76 5.3.1 Microfabrication ............................................................................... 76 5.3.1.1 Wells with Gold Walls .................................................................76 5.3.1.2 Glass Wells with Overhang .........................................................78 5.3.2 Selective Barrier Functionalization..................................................... 79 5.3.2.1 Functionalization of Gold Walls with Biotinylated Dynein ............82 5.3.2.2 Selective Functionalization of Glass Walls with Ni(II)-NTA ..........82 5.3.3 FRAP Characterization of Ni(II)-NTA Surfaces Coated with His-Tagged Proteins........................................................................................... 83 5.3.4 Fluorescent Microscopy Assays of Dynamic Microtubules Interacting with Functionalized Barriers ............................................. 84
Methods in Cell Biology, Volume 120 Copyright © 2014 Elsevier Inc. All rights reserved.
ISSN 0091-679X http://dx.doi.org/10.1016/B978-0-12-417136-7.00005-7
69
70
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.3.4.1 Assay 1: Interactions of Dynamic Microtubule Ends with Barrier-Attached Dynein .........................................................................................84 5.3.4.2 Assay 2: Microtubule-Based Delivery and Anchoring of þTIPs to Barriers ...86 5.4 Discussion and Perspectives .............................................................................. 86 Acknowledgements ................................................................................................... 88 References ............................................................................................................... 88
Abstract Local interactions between the tips of microtubules and the cell cortex, or other cellular components such as kinetochores, play an important role in essential cellular processes like establishing cell polarity, distribution of organelles, and microtubule aster and chromosome positioning. Here we present two in vitro assays that specifically mimic microtubule–cortex interactions by employing selectively functionalized microfabricated barriers that allow for the immobilization of proteins with a range of affinities. We describe the microfabrication process to create gold or glass barriers and the subsequent functionalization of these barriers using self-assembled thiol monolayers or polylysine–poly(ethylene glycol), respectively. Near-permanent attachment of proteins is obtained using biotinylated surfaces combined with streptavidin and biotinylated proteins. Lower affinity interactions, further tunable with the addition of imidazole, are obtained using nickel-nitrilotriacetic acid (Ni(II)-NTA) functionalization combined with his-tagged proteins. Both mono-NTA and trisNTA compounds are used. We show an assay to reconstitute the “end-on” interaction between dynamic microtubule tips and barrier-attached dynein, mimicking the cellular situation at the cortex and at kinetochores. In a second assay, we reconstitute microtubule-based delivery of end-tracking proteins to functionalized barriers, mimicking the transport of cell-end markers to the cell poles in interphase fission yeast cells.
INTRODUCTION Microtubules are stiff long biopolymers with a fast growing plus end and a slow growing minus end that extend along the cytoplasm providing a physical connection between distant organelles in the cell. They serve as tracks for selective transport of cellular components by molecular motors (Goodman, Derr, & Reck-Peterson, 2012; Vale, 2003) and play an important role in cellular organization. Microtubule plus ends explore the cytoplasmic space by switching stochastically between states of polymerization and depolymerization. This intrinsic process is termed “dynamic instability”, where the sudden switch from polymerization to depolymerization is called “catastrophe” (Desai & Mitchison, 1997; Mitchison & Kirschner, 1984).
5.1 Rationale
In cells, dynamic instability is regulated by several microtubule-associated proteins (MAPs) as well as physical interactions with the cell boundaries. Interestingly, so-called microtubule plus-end tracking proteins (þTIPs), MAPs that interact preferentially with the growing plus ends of microtubules (Akhmanova & Steinmetz, 2008), can also mediate microtubule interactions with the cell cortex or other cell components such as kinetochores. In most cells, microtubule minus ends remain stably attached to microtubulenucleating organelles close to the nucleus while the plus ends grow toward the cell boundaries. At the cell boundary, microtubule polymerization can generate pushing forces up to several pN before undergoing catastrophe (Dogterom & Yurke, 1997; Fygenson, Braun, & Libchaber, 1994; Janson, de Dood, & Dogterom, 2003). When needed, cells can reduce the catastrophe rate by employing designated þTIPs such as cytoplasmic linker associated proteins (CLASPs) that stabilize microtubule ends while keeping them linked to the cell boundary. Also, during mitosis, kinetochore proteins link the ends of spindle microtubules to the chromatids, stabilizing them until chromosomal segregation. Shrinking microtubule ends can also produce pulling forces (Grishchuk, Molodtsov, Ataullakhanov, & McIntosh, 2005; Lombillo, Stewart, & McIntosh, 1995). For example during chromosomal segregation, the chromatids are pulled toward opposite sides of the cell by maintaining a dynamic link between kinetochore complexes and the ends of depolymerizing microtubules. Similarly, cortical dynein can capture growing microtubule plus ends, trigger catastrophes, and keep shrinking microtubule ends linked to the cortex (Laan, Pavin, et al., 2012). This generates pulling forces on anything linked to the other end of the microtubule, which in cells contributes to the positioning of microtubule organizing centers such as centrosomes (Laan, Pavin, et al., 2012). In addition to generating forces, microtubule ends may exchange molecules with the cortex when they reach cellular boundaries. Motor proteins such as kinesin and cytoplasmic dynein transport organelles and other proteins toward either the plus or the minus end. Proteins delivered to the cell boundary may associate with the cortex leading to cortical protein patterns. For example, in fission yeast, protein tea1, involved in polarized cell growth, is transported at polymerizing microtubule tips with the help of mal3 and the kinesin tea2 to be delivered at the cell poles where it accumulates (Dodgson et al., 2013; Sawin & Snaith, 2004).
5.1 RATIONALE In vitro reconstitution experiments allow for the study of biological processes using a bottom-up approach by adding only minimal components. Since we are interested in reconstituting functional interactions between dynamic microtubule ends and proteins that are either (semi-)permanently or transiently associated with cellular objects such as the cell boundary or chromosomes, we reconstitute microtubule systems in microwells with functionalized walls to which proteins can be specifically attached with tunable affinity.
71
72
CHAPTER 5 Functional Microtubule–Barrier Interactions
These experiments require the selective functionalization of the walls of the wells with desired proteins as well as passivation of the bottom surfaces. Here we describe two methods: the first one is adapted from Romet-Lemonne, VanDuijn, and Dogterom (2005) and consists of immobilizing proteins to gold walls via self-assembled monolayers (SAMs) of alkanethiols with a functional terminal group (Frasconi, Mazzei, & Ferri, 2010; Love, Estroff, Kriebel, Nuzzo, & Whitesides, 2005). This method was used before in our lab to produce “end-on” interactions between microtubule tips and cortical dynein (Laan & Dogterom, 2010; Laan, Pavin, et al., 2012). The second method consists of using both bottom surfaces and walls made of glass, where selectivity is achieved by irradiating the bottom surfaces with deep UV to remove the coating there. This technique has been successfully used on flat surfaces before with the help of an external mask (Azioune, Storch, Bornens, The´ry, & Piel, 2009). Here, we apply it to 3D structures using a microfabricated mask. To study the cellular functions described earlier, different affinities for protein binding are required. We distinguish two cases: proteins forming stable links with the organelle or cell cortex, such as kinetochore proteins, cortical dynein, or CLASPs; proteins that form transient links. For the first case, we immobilize biotinylated proteins at the walls using biotin–streptavidin sandwich linkages. For the second case, we use Ni(II)-NTA, which allows for a reversible immobilization of polyhistidine-tagged proteins, tunable by competition with imidazole (Lata & Piehler, 2005; Lata, Reichel, Brock, Tampe´, & Piehler, 2005). To achieve a wide range of affinities we use both mono-Ni(II)-NTA and tris-Ni(II)-NTA. We specifically describe the methodology to produce the two assays that are shown in Figs. 5.1, 5.6, and 5.7 and that are summarized below. Note however that different combinations of methods are also possible. Section 5.3.1 describes the steps for fabrication of gold and glass barriers; Section 5.3.2, the selective functionalization of barriers, Section 5.3.3, the characterization of the Ni(II)-NTA functionalized surfaces by fluorescence recovery after photobleaching (FRAP); and Section 5.3.4 explains the preparation and imaging of our two assays with dynamic microtubules.
5.1.1 Assay 1 (Fig. 5.1A): interactions of dynamic microtubule tips with barrier-attached dynein In this assay dynamic microtubules grow from freely floating guanylyl-(alpha, beta)methylene-diphosphonate (GMPCPP)-stabilized microtubule seeds (Hyman, Salser, Drechsel, Unwin, & Mitchison, 1992) inside microwells. The walls are 100 nm high made of gold. There is a 10 nm adhesion layer of chromium below the gold and a 25 nm one on top. The gold walls are coated with a SAM of biotin-terminated thiols followed by a multilayer of streptavidin tetramers and biotinylated BSA. Biotinylated dynein molecules are bound to a final streptavidin layer.
5.1.2 Assay 2 (Fig. 5.1B): microtubule-based delivery and anchoring of þTIPs to barriers In this assay dynamic microtubules grow from biotinylated GMPCPP-stabilized microtubule seeds bound to the bottom of microwells via biotin–streptavidin links. The wells are 300 nm deep, made of glass, with an overhang of 50 nm gold on top of
5.2 Materials
50 nm chromium. This overhang prevents the microtubules from growing over the walls and serves as a mask for functionalization steps. The walls are functionalized with PLL–PEG–tris-Ni(II)-NTA to which þTIPs with a his-tag can reversibly bind (Bhagawati, You, & Piehler, 2013). We use the end-tracking system from fission yeast consisting of the EB1 analog mal3, kinesin tea2, and protein tip1 (Bieling et al., 2007).
5.2 MATERIALS 5.2.1 Microfabrication 5.2.1.1 Special equipment All the microfabrication steps except evaporation were performed inside a clean room (class ISO 6). Coverslips No. 1 24 24 mm 170 mm (Menzel Gla¨sser, Germany). Delta 80 GYSET® Spin coater with a close lid (Su¨ss MicroTec, Germany). Hot plates. N2 gas (hand gun). Homemade turbo pumped vacuum system with base pressure 107 mbar equipped with a resistance heating evaporation system for tungsten boats loaded with chromium or gold. 2510 Ultrasonic Cleaner (Branson, USA). MJB3 mask aligner for UV exposure (Su¨ss MicroTec). Binary chromium/soda lime mask (Delta Mask, The Netherlands).
5.2.1.2 Reagents Hexamethyldisilazane (HMDS) primer (440191 Sigma-Aldrich, USA).
5.2.1.2.1 Photoresists
Shipley Microposit® S1805 G2 positive UV-resist (Microresist, Germany). Negative tone photoresist ma-N1410 (Microresist).
5.2.1.2.2 Developers
Microposit® MF®-319 developer (Microresist). Ma-D 533/S; alternatively ma-D332S (Microresist).
5.2.1.2.3 Etchants Standard gold etchant (651818 Sigma-Aldrich). Standard chromium etchant (651826 Sigma-Aldrich).
5.2.2 Surface functionalization 5.2.2.1 Special equipment Compact UV-Ozone cleaner (Uvotech, USA). Binary chromium/quartz mask (Delta Mask).
73
74
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.2.2.2 Reagents All reagents were dissolved in MRB80 buffer (80 mM KPipes, 4 mM MgCl2, 1 mM ethylene glycol tetraacetic acid (EGTA), pH 6.8) at the stated stock concentration and stored at 80 C unless stated otherwise. Albumin, biotin labeled bovine (A8549 Sigma-Aldrich); 20 mg/ml solution (biotinylated BSA). Streptavidin (43–4302 Invitrogen, USA); 1 mg/ml solution. Alexa Fluor® 488 streptavidin (S-11223 Invitrogen); 1 mg/ml solution. k-casein from bovine milk (C0406 Sigma-Aldrich); 5 mg/ml solution.
5.2.2.2.1 For gold surfaces Biotinylated PEG alkanethiol (CMT005-25 Nanoscience Instruments, USA); 0.1 mM solution in 100% ethanol stored at 20 C (HS–PEG–biotin).
5.2.2.2.2 For glass surfaces PLL(20)-g[3.5]–PEG(2) (SUSOS AG, Switzerland); 2 mg/ml solution (PLL–PEG). PLL(20)-g[3.1]–PEG(2)/PEG(3.4)–biotin(17.5%) (SUSOS AG); 2 mg/ml solution (PLL–PEG–biotin). PLL(20)-g[3.5]–PEG(3.5)–(3,4)NTA (SUSOS AG); 2 mg/ml solution (PLL–PEG–NTA). PLL-g3–PEG2k/PEG3.4K–tris-NTA (PLL–PEG–tris-NTA) made by a two-step coupling process, as described elsewhere (Bhagawati et al., 2013). Nickel(II) sulfate (656895-10G, Sigma-Aldrich), 10 mM NISO4 solution in MRB80 pH 7.5 stored at room temperature.
5.2.3 Characterization of surfaces and microtubule assays 5.2.3.1 Special equipment
Airfuge® Air-driven ultracentrifuge (Beackman Coulter, USA). Leica microscope equipped with a spinning disk confocal head (Yokogawa, Japan). 100 1.3 NA oil immersion objective and cooled EMCCD camera (C9100, Hamamatsu Photonics, Japan). Excitation lasers: Sapphire 488 nm (30 mW) (Coherent, Germany) and 561 nm class 3B (100 mW) (Melles Griot). Total internal reflection fluorescence microscope (TIRF) (Nikon Corporation, Japan) equipped with an Apo TIRF 100 1.49 N.A. oil objective, a motorized stage, Perfect Focus System, a motorized TIRF illuminator (Roper Scientific, France) and a QuantEM:512SC EMCCD camera (Photometrics, Roper Scientific). Excitation lasers: 561 nm (50 mW) Jive (Cobolt, Sweden) and a 488 nm (40 mW) Calypso (Cobolt) diode-pumped solid-state laser. For FRAP experiments the microscope is equipped with a MAG Biosystems FRAP-3D system (Photometrics, Roper Scientific).
5.2 Materials
Glass slides (Menzel Gla¨sser). Tesa® double-sided tape (15 mm wide). Diamond glass cutter. Valap (vaseline, lanolin, paraffin wax melted at equal concentrations).
5.2.3.2 Reagents All reagents were dissolved in MRB80 buffer at the stated stock concentration and stored at 80 C unless stated otherwise.
5.2.3.2.1 Microtubule polymerization and molecular motor assays Tubulin from porcine brain (T240 Cytoskeleton, Inc., USA); 100 mM solution. Rhodamine-labeled tubulin from porcine brain (TL590M Cytoskeleton, Inc.); 50 mM solution. Fluorescent HiLyte 488 tubulin from porcine brain (TL488M Cytoskeleton, Inc.); 50 mM solution. Biotinylated tubulin from porcine brain (T333P Cytoskeleton, Inc.); 50 mM solution. Guanosine 50 -triphosphate sodium salt hydrate (GTP) (G8877 Sigma-Aldrich); 50 mM solution. GMPCPP (NU-405 Jena BioScience); 10 mM solution. Glucose oxidase from Aspergillus niger (G2133 Sigma-Aldrich); 20 mg/ml dissolved in 200 mM DL-Dithiolthreitol (646563 Sigma-Aldrich) with 10 mg/ml catalase from bovine liver (C9322 Sigma-Aldrich) (glucose oxidase). D-(þ)-Glucose (G8270 Sigma-Aldrich). Adenosine 50 -triphosphate (ATP) (A7699 Sigma-Aldrich); 50 mM solution.
5.2.3.2.2 GMPCPP-stabilized microtubule seeds Microtubule seeds are prepared by a polymerization–depolymerization–polymerization cycle with GMPCPP in MRB80 (this cycling is done to get rid of residual GTP). The first polymerization step is performed by incubating a tubulin mix of biotinylated tubulin, fluorescently labeled and nonlabeled tubulin in a ratio 18:12:70 (total of 20 mM tubulin) with 1 mM GMPCPP at 37 C for 30 min. For nonbiotinylated seeds simply exchange the biotinylated tubulin by nonlabeled tubulin. The mix is then airfuged 5 min at 30 psi and the pellet is resuspended in 80% of the initial volume and left on ice for 20 min to depolymerize the microtubules. Then, 1 mM GMPCPP is added to the mix and left for 30 min at 37 C to polymerize again. The seeds are then airfuged again, resuspended in MRB80 with 10% glycerol and stored at 80 C. Thawed seeds can be kept at room temperature for a week.
5.2.3.2.3 Microtubule-associated proteins The dynein construct used is GFP-SNAP-GST-Dynein331-HaloTag (modified from Reck-Peterson et al. (2006)), where the GST allows it to dimerize. A biotin was added via the SNAP tag (S9110S New England BioLabs, UK) and a TMR (tetramethylrhodamine) label via HaloTag (G8251 Promega, France).
75
76
CHAPTER 5 Functional Microtubule–Barrier Interactions
Kinesin Tea2 was expressed in E. coli and purified as in Bieling et al. (2007). The following proteins were purified as in Maurer, Bieling, Cope, Hoenger, and Surrey (2011). 6his-TEV-Mal3-mCherry sequence was obtained by deleting the GFP sequence on 6his-Mal3-GFP (generous gift by Thomas Surrey, Maurer et al., 2011) and inserting the mCherry sequence. Unlabelled 6his-TEV-mal3 was obtained as in Bieling et al. (2007). The his-tag was removed by cleavage overnight with AcTEV™ protease (12575-015 Invitrogen) at 4 C according to the provider’s protocol. 6his-TEV-eGFP-Tip1 was obtained as in Bieling et al. (2007) and the oligo duplex 50 -tccgcgggtgagaatcttcagggcgcc-30 containing a TEV recognition site was added between the his-tag and the eGFP.
5.3 METHODS 5.3.1 Microfabrication Here we describe the production of two different designs of microfabricated wells: (1) wells with gold walls obtained by lift-off (used in assay 1, Figs. 5.1A and 5.2A), and (2) glass wells with an overhang obtained by wet etch (used in assay 2, Figs. 5.1B and 5.2B). An alternative method to produce gold walls using plasma etching can be found in Laan & Dogterom (2010).
5.3.1.1 Wells with gold walls (Fig. 5.2A and C) 5.3.1.1.1 Photolithography Start by cleaning the glass coverslips with base piranha (NH4OH:H2O2 in 3:1 at 75 C) for 15 min. Rinse them, first in ddH2O and then in isopropanol and blow dry with N2 flow. Bake the samples 5 min on a hot plate at 150 C, cool them down with N2 flow, and spin-coat a thin layer of HMDS primer (4000 rpm for 32 s). Bake the coverslips 5 min at 150 C, cool them down with N2 flow, and spin-coat a thin layer of negative tone photoresist ma-N1410 (3000 rpm for 30 s). Bake the slides 2 min on a hot plate at 120 C. This leads to a 500 nm layer of resist. Expose to a dose of 325 mJ/cm2 of 365 nm UV light using a Su¨ss MJB3 mask aligner with a binary chromium/soda lime mask on top. Immerse the sample in developer ma-D332S for 120 s while constantly moving the fluid (e.g., by shaking the container). The UV unexposed pattern will be developed (dissolved) while leaving the rest of the photoresist. Rinse in ddH2O and dry with N2 flow.
5.3.1.1.2 Evaporation of metals Evaporate a 10 nm layer of chromium, followed by 100 nm of gold, and finally 25 nm of chromium in an evaporation chamber under pressure below 106 mbar with a deposition rate of 0.06 nm/s. The chromium layer between the glass and the gold ensures the good deposition of the gold.
5.3 Methods
FIGURE 5.1 Schematic cartoons describing assays 1 (A) and 2 (B). (A) Assay 1: A dynamic microtubule grows against a gold wall functionalized with dynein. The bottom surface is passivated with PLL–PEG. (B) Assay 2: A dynamic microtubule grows from a stabilized microtubule seed against a glass wall selectively functionalized with tris-Ni(II)-NTA. The microtubule seed is attached to the bottom surface via biotin–streptavidin links. The microtubule delivers his-tagged tip1 to the wall.
5.3.1.1.3 Lift-off Lift off the photoresist by sonicating the coverslips in acetone for 1 h. This method often leaves deposited particles of chromium on the gold that will hinder the functionalization. If needed, they can be removed by rinsing the sample in chromium etchant for 10 s. Clean again the sample with base piranha to make it ready for functionalization steps. The sample will be very hydrophilic. Each coverslip can be cut with a glass cutter in four long pieces useful for four different flow cell experiments.
77
78
CHAPTER 5 Functional Microtubule–Barrier Interactions
FIGURE 5.2 Schematic drawings showing the steps of the microfabrication process for gold barriers (A) and glass barriers with overhang (B). (C and D) SEM images of microfabricated wells. A 10 nm layer of chromium was sputtered on top of the samples with a K575X Turbo Sputter Coater (EMITECH Ltd) allowing a good conduction of electrons. (D, right) The overhang was removed with wet etch so that the glass walls can be seen. Note that due to wet etching steps the walls can become irregular depending on the quality of the adhesion between different layers.
5.3.1.2 Glass wells with overhang (Fig. 5.2B and D) 5.3.1.2.1 Evaporation
Start by cleaning the coverslips with base piranha (NH4OH:H2O2 in 3:1 at 75 C) for 15 min and rinse them in ddH2O and isopropanol. Evaporate a 50 nm layer of chromium and 50 nm of gold (both rates 0.06 nm/s) in an evaporation chamber at pressure below 106 mbar. The gold layer protects the chromium from being attacked later by KOH and also prevents the chromium from peeling off due to mechanical strain.
5.3 Methods
5.3.1.2.2 Photolithography
Dry the samples 5 min at 150 C, cool them down with N2 flow, spin-coat a thin layer of HMDS primer (4000 rpm for 32 s), and bake 1 min at 150 C. Spin-coat a thin layer (around 300 nm) of S1805 positive photoresist (500 rpm for 8 s followed by 4000 rpm for 32 s), let dry 10 s, and soft bake at 115 C for 60 s. Expose to a dose of 25 mJ/cm2 of 365 nm UV light using a Su¨ss MJB3 mask aligner with a binary chromium/soda lime mask on top. For structures that do not require sharp corners (like straight lines), the sample can be postbaked 30 min at 130 C which melts a little photoresist providing smoother structures. Immerse the sample in developer MF-319 for 1 min. The UV-exposed pattern will be developed (dissolved) while leaving the rest of the photoresist.
5.3.1.2.3 Wet etching Immerse sequentially the sample in gold etchant (etch rate around 2.8 nm/s) and chromium etchant (etch rate around 4 nm/s). Each step has to be carefully checked with a microscope because the etching rates depend on the exact conditions. Remove the photoresist by sonicating the samples 20 min in acetone. If needed, further removal can be achieved by base piranha. Isotropically etch the glass by immersing it in a 40% KOH solution at 80 C (etch rate around 5 nm/min). We found that for Menzel Gla¨sser coverslips, KOH produced a smoother final surface than plasma etching or HF (not shown). A depth of 300 nm is convenient for later imaging of microtubules with confocal microscopy. Clean again the sample with base piranha to make it ready for functionalization steps and optionally cut each sample in four long pieces.
5.3.2 Selective barrier functionalization Both assays shown in Fig. 5.1 require selective functionalization of the walls of the microwells. In the first assay, this is achieved by using different substrates. Poly(L-lysine)–poly(ethylene glycol) (PLL–PEG) chains spontaneously bind (mostly by electrostatic interactions) to negatively charged surfaces like glass (Kenausis et al., 2000) (Fig. 5.3B–D), whereas thiols (SH) form SAMs specifically on gold (Fig. 5.3A) (Frasconi et al., 2010). In the second assay where both surfaces consist of glass, selectivity can be achieved by first coating all surfaces with the compound desired for the walls, irradiating the bottom surfaces with deep UV to photocleave the molecules there, and then coating the irradiated surfaces with the second compound. Selective irradiation of the bottom surfaces is automatically achieved due to the microfabricated overhang on the walls (Fig. 5.4B). For quasi-permanent immobilization of proteins a high binding affinity is required so that proteins do not unbind on experimental timescales. We use biotin–streptavidin links, whose equilibrium dissociation constant Kd is reported to be of the order of 1014 M in solution (Green, 1990) and 109 M on beads (Buranda, Lopez, Keij, Harris, & Sklar, 1999). In our assays, while a single biotin–streptavidin layer (Fig. 5.3B) is enough to immobilize biotinylated microtubule seeds (with several
79
FIGURE 5.3 Cartoons showing the surface functionalization used for protein immobilization on gold (A) and glass (B–D). (A) Biotinylated PEG-alkanethiols forming SAMs on gold substrates in a mixture with nonfunctionalized 11-Mercapto-1-undecanol molecules. The biotin is covered with a multilayer of streptavidin tetramers and biotinylated BSA. Biotinylated proteins can bind to the streptavidin tetramers. (B–D) Poly(L-lysine)-based coating of glass surfaces. (B) Immobilization of biotinylated proteins via PLL–PEG–biotin and one layer of streptavidin. (C) PLL–PEG–tris-Ni(II)-NTA for immobilization of histidine-tagged proteins. Each of the three NTAs chelates one nickel ion to which two histadines can bind. The binding affinity can be tuned by competition with imidazole. (D) PLL–PEG–Ni(II)-NTA bound to a glass surface. In this case four of the histidines remain unbound.
5.3 Methods
FIGURE 5.4 Cartoons illustrating the functionalization steps of gold (A) and glass (B) barriers. (C and D) Confocal images of fluorescent proteins bound to the surfaces: (C) dynein before (top) and after (bottom) buffer wash, (D) Alexa Fluor® 488 streptavidin at the bottom surface and histagged mal3-mCherry at the walls. When adding 1 M imidazole only mal3 unbinds. (C and D(ii)) Z-stacks along the dotted line in (i). Scale bars 10 mm.
binding sites), multiple layers of streptavidin and biotinylated BSA produce denser layers of functional proteins at the barriers (Fig. 5.3A) (Romet-Lemonne et al., 2005), suggesting an effective enhancement of the density of available binding sites. For reversible binding of þTIPs the affinity should be comparable to that for microtubule tips. Reversible binding of proteins can be achieved using the affinity of histidines to nickel ions chelated with nitrilotrilotriacetic acid (Ni(II)-NTA)
81
82
CHAPTER 5 Functional Microtubule–Barrier Interactions
(Hochuli, Do¨beli, & Schacher, 1987). Proteins with a hexahistidine tag (his-tag) fused at one of the termini will bind to Ni(II)-NTA with affinity tunable by competition with imidazole. We use both mono-Ni(II)-NTA (Fig. 5.3D, Kd 10 mM) (Zhen et al., 2006) and tris-Ni(II)-NTA (Fig. 5.3D, Kd 10 nM) depending on the affinity required. The latter one contains three chelator heads (NTAs) and binds his-tagged proteins stoichiometrically with higher affinity than mono-Ni(II)-NTA whose effective affinity depends strongly on the density of Ni(II)-NTAs at the surface (Lata & Piehler, 2005).
5.3.2.1 Functionalization of gold walls with biotinylated dynein (Fig. 5.4A) Drop a 20 ml solution of 0.2 mg/ml PLL–PEG on top of a microfabricated coverslip from Section 5.3.1.1. Insert the coverslip in a sealable plastic bag and keep the surface wet by close contact between the fluid and the inside of the bag. Sonicate the bag 5 seconds to remove possible air bubbles from the microwells. Further incubate for 30 min, later wash it with MRB80, and dry it with N2 flow. Now drop 50 ml solution of 1 mM HS–PEG–biotin on the sample and place it between two parafilm sheets. In order to form a good SAM, incubate between 10 and 14 h (overnight). Rinse the sample with ethanol and blow dry. Prepare a flow cell on a hellmanex® (HellmaAnalytics, UK) cleaned glass slide by attaching two parallel strips of double-sided TESA® tape separated by 5 mm. Place the coverslip on top with the microfabricated wells facing the glass slide (10–20 ml flow cell volume). Now you can flow in the subsequent solutions by putting absorbent paper on the other side of the cell. Avoid any dry out of the flow cell. Clean thoroughly the flow cell with MRB80 buffer and incubate 15 min with a 0.2 mg/ml PLL–PEG solution to passivate the slide surface. Wash with MRB80 and perform alternated 10 min incubations of 1 mg/ml streptavidin (three times) and 1.5 mg/ml biotinylated BSA (two times). Wash thoroughly with MRB80 in between incubations. Further passivate the surfaces with 1.2 mg/ml k-casein for 10 min. Incubate with a mix containing 10 nM dynein, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, and 50 mM glucose in MRB80 for 10 min. Wash with MRB80. Figure 5.4C shows examples of confocal microscopy of dynein-coated walls before and after the washing step. For a control experiment without dynein we simply omit this final incubation step.
5.3.2.2 Selective functionalization of glass walls with Ni(II)-NTA (Fig. 5.4B) Drop 20 ml solution of 0.1 mM PLL–PEG–NTA or PLL–PEG–tris-NTA on top of a microfabricated coverslip from Section 5.3.1.2. Put the sample in a sealable plastic bag and immerse it in an ultrasonication bath for 5 seconds. Incubate for 30 min, wash with MRB80, and dry with N2 flow. Expose to deep UV (185 and 254 nm) with an Ozone Cleaner for 10 min (the time depends on the power of the lamp and the distance to the sample). Prepare a flow cell on a glass slide with the coverslip and double-sided TESA® tape as described in the previous section. Load the NTA with nickel ions by incubating a 10 mM NiSO4 solution for 30 min.
5.3 Methods
Clean thoroughly the flow cell with MRB80 buffer and incubate 15 min with 0.2 mg/ml PLL–PEG–biotin, followed by 1 mg/ml streptavidin, and passivate with 1.2 mg/ml k-casein (10 min incubation each, rinsed in between with MRB80). For a control experiment to visualize the selective coatings, omit the incubation of nonlabeled streptavidin and incubate a mix containing 500 nM his-tagged mal3mCherry, 0.1 mg/ml 488 alexa streptavidin, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, and 50 mM glucose in MRB80. This should lead to a bottom surface coated with streptavidin and only the walls coated with mal3 (Fig. 5.4D). When washing with 1 M imidazole, mal3 detaches from the walls while fluorescent streptavidin remains at the bottom surface.
5.3.3 FRAP characterization of Ni(II)-NTA surfaces coated with his-tagged proteins We perform FRAP measurements to determine whether there is a turnover of proteins at the functionalized surfaces. These measurements are done on flat surfaces with micropatterned regions of PLL–PEG–Ni(II)-NTA around islands passivated with PLL–PEG. Since the density of protein immobilized by Ni(II)-NTA increases with protein concentration (Lata & Piehler, 2005), fluorescently labeled his-tagged þTIPs are incubated at the same concentration as in tip tracking experiments. Once the density of immobilized protein has reached equilibrium, we bleach a region and measure the recovery of fluorescence signal due to turnover of proteins between the bulk and the surface. Drop 20 ml solution of 0.1 mM PLL–PEG–NTA or PLL–PEG–tris-NTA on top of a base piranha cleaned glass coverslip and place it between two parafilm sheets for 30 min. Wash the sample with MRB80 and dry in N2 flow. Place on top a binary chromium/quartz mask with micropatterns (with the chromium side in good contact with the functionalized surface of the sample; use an extra weight if needed). Expose through the mask to deep UV with an Ozone Cleaner for 20 min. Prepare a flow cell on a glass slide with double-sided TESA® tape and the exposed coverslip as described earlier. Incubate with a 10 mM NiSO4 solution for 30 min. Clean thoroughly with MRB80 and incubate 15 min with 0.2 mg/ml PLL–PEG to passivate the regions exposed to UV. Further passivate with 1.2 mg/ml k-casein for 10 min and rinse with MRB80. Prepare a protein mix containing 100 nM his-tagged eGFP-tip1 or 200 nM histagged mal3-mCherry, 50 mM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose in MRB80. Airfuge the mix at 30 psi for 9 min and incubate in the flow cell. We image the surface of the sample with TIRF at 500 ms frame rate and photobleach a circle of 4.7 mm diameter for 1 s (Fig. 5.5A). The recovery of the fluorescence intensity of the bleached area, I, is followed over time and normalized to lie between the fluorescent signal at a nonbleached region that was passivated (I set to 0) and a nonbleached region that was functionalized with Ni(II)-NTA
83
84
CHAPTER 5 Functional Microtubule–Barrier Interactions
FIGURE 5.5 FRAP characterization of Ni(II)-NTA surfaces. (A) TIRF snapshots of a patterned surface with Ni(II)-NTA with 200 nM his-tagged mal3-mCherry in solution. A circular area in the center of the images is bleached and the recovery is followed over time. Scale bar 2 mm. (B) Normalized intensity recovery of the bleached area for an incubation with 100 nM his-tagged eGFP-tip1. Error bars correspond to the standard deviation between five different recoveries.
(I set to 1) (I ¼ (1/IbeforeFRAP)((IFRAP Ipassivated)/(InoFRAP Ipassivated))afterFRAP with IbeforeFRAP ¼ ((IFRAP Ipassivated)/(InoFRAP Ipassivated))beforeFRAP). Figure 5.5B shows examples of recovery curves for mono-Ni(II)-NTA and trisNi(II)-NTA surfaces for samples in equilibrium (bleached more than 10 min after incubation). Tip1 has a fast turnover on mono-Ni(II)-NTA surfaces (65% of the protein exchanges with a half time of 50 s) while it recovers, as expected, much slower on tris-Ni(II)-NTA surfaces.
5.3.4 Fluorescent microscopy assays of dynamic microtubules interacting with functionalized barriers 5.3.4.1 Assay 1: interactions of dynamic microtubule ends with barrier-attached dynein Prepare on ice a tubulin mix in MRB80 containing 14.25 mM tubulin, 0.75 mM HiLyte 488 labeled tubulin (5% of the total amount of tubulin), 50 mM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose, 0.1% methyl cellulose, 1 mM GTP, and 2 mM ATP. Airfuge the mix 9 min at 30 psi, warm up to room temperature, add rhodamine-labeled GMPCPP microtubule seeds (to a final 10 times dilution), and incubate in the flow cell from Section 5.3.2.1, that was coated before with dynein. Seal the flow cell with melted valap so that it does not dry out.
5.3 Methods
We image the sample at 27 C using spinning disk confocal microscopy with 488 and 561 nm lasers, taking images every 5 s (exposure 300 ms). For 100 nm barriers, both the bottom surface and the walls can be imaged in the same focal plane. However, some microtubules may grow over the barriers. We search for microtubules that grow with their plus ends towards the barrier. In control experiments without dynein, microtubules continue to polymerize at their plus end when contacting the wall. Since in this assay the microtubule seed is not attached to the surface, the addition of tubulin dimers at the wall pushes the microtubule away from the wall (Fig. 5.6A and B). For barriers functionalized with dynein we often observe the same behavior. However, in some cases the microtubule is now observed to be pulled toward the wall (Fig. 5.6C and D). In these cases dynein keeps the plus end tip associated with the wall while the microtubule depolymerizes (Fig. 5.6B and C). In Fig. 5.6C and D, the depolymerizing microtubule eventually detaches from the wall.
FIGURE 5.6 (A and C) Confocal images (time average over three snapshots) at different times of dynamic microtubules polymerizing against gold walls (barriers are indicated by dashed areas) without dynein (A) and with dynein (C). (B and D) Respective kymographs for the microtubules with the arrows. Since the microtubule fluctuates to the sides of its growth axis, the kymograph shows the maximum intensity along six pixels (1 mm) perpendicular to the microtubule growth axis. (B) When the microtubule polymerizes against the wall, the microtubule seed is pushed away from the barrier. (D) The plus end stays attached to the dynein-coated barrier while depolymerizing. As a result, the seed is being pulled towards the wall until the plus end detaches and the microtubule diffuses away.
85
86
CHAPTER 5 Functional Microtubule–Barrier Interactions
5.3.4.2 Assay 2: microtubule-based delivery and anchoring of þTIPs to barriers Take the flow cell from Section 5.3.2.2 prepared with tris-Ni(II)-NTA. Incubate for 5 min with the biotinylated microtubule seeds 10 times diluted in MRB80 with 0.1% methyl cellulose so that they bind to the surface coated with PLL–PEG–biotin– streptavidin. Prepare on ice a tubulin mix with end-binding proteins in MRB80 containing 14.25 mM tubulin, 0.75 mM rhodamine tubulin, 100 nM his-tagged eGFP-tip1, 8 nM tea2, 200 nM mal3, 50 nM KCl, 0.6 mg/ml k-casein, 0.4 mg/ml glucose oxidase, 50 mM glucose, 0.1% methyl cellulose, 1 mM GTP, and 2 mM ATP. For assays of delivery of mal3, use his-tagged mal3-mCherry and HiLyte 488 labeled tubulin without tea2 and tip1. Airfuge 9 min at 30 psi, incubate in the flow cell, and seal it with valap. Since these wells are 300 nm deep, we take confocal z-stacks every 5 s (300 ms exposure, 300 nm step size). Each image of the z-stack is later background corrected and projected by maximum intensity. When using walls functionalized with tris-Ni(II)-NTA, his-tagged proteins in solution have a relatively high affinity for the walls, leading to spontaneous coverage of the walls (Fig. 5.7). Additionally, his-tagged end-binding proteins have a high affinity for growing microtubule ends. His-tagged mal3 binds to microtubule ends with a dwell time of about 0.3 s, leading to the appearance of mal3 comets at growing microtubule ends (Bieling et al., 2007). When these comets reach a barrier, they lead to a locally enhanced concentration of mal3 at the barrier while the microtubule is in contact with the wall, but the “extra” mal3 vanishes when the microtubule depolymerizes (Fig. 5.7A and B). This shows that microtubules cannot “deliver” mal3 to barriers that are precoated with mal3. By contrast, Fig. 5.7C and D shows a microtubule in the presence of his-tagged tip1 reaching a wall functionalized with tris-Ni(II)-NTA. When the microtubule reaches the wall an apparent cluster of tip1 proteins remains attached to the wall, even after the microtubule depolymerizes. This preliminary observation suggests that effective clustering of proteins such as observed for tip1, but not for mal3, may be important for cortical tethering of proteins.
5.4 DISCUSSION AND PERSPECTIVES In this chapter, we presented two methods to immobilize proteins selectively at barriers of gold or glass, using functional groups that allow for a broad range of affinities. Selective immobilization on gold walls had already been used to mimic cortical interactions (Laan & Dogterom, 2010; Laan, Pavin, et al., 2012). Here we additionally introduced a method for functionalization of glass barriers, using only PLL– PEGs combined with deep UV irradiation (Azioune et al., 2009). PLL–PEGs assemble faster than SAMs and suffer less from unspecific binding (Zhen et al., 2006), making them more suitable for immobilizations at high Kd. Selective irradiation
5.4 Discussion and Perspectives
FIGURE 5.7 (A and C) Confocal images (time average over five snapshots) at different times, before, during, and after contact of the microtubule end with a glass wall functionalized with trisNi(II)-NTA, in the presence of 200 nM his-tagged mCherry-mal3 (A) or in the presence of mal3, tea2, and his-tagged tip1-eGFP (C). (B and D) Respective kymographs for the microtubules with the arrows. His-tagged mal3 delivered to the wall by a microtubule does not remain anchored at the wall upon microtubule catastrophe (B), while his-tagged tip1 does (D). Spatial scale bar 2 mm.
of the bottom surface is easily achieved with the overhang of chromium produced by wet etch during the microfabrication procedure. Note, however, that an overhang of chromium amplifies the detected fluorescence intensity of proteins by up to a factor of 2 (data not shown). This is likely due to reflections of the excitation and emission light at the chromium layer. To avoid this effect, we are currently testing the use of an 80 nm overhang of TiO2, which is opaque to deep UV but 70–80% transparent to the wavelengths of interest (480–620 nm) (data not shown). An alternative way to selectively expose regions of the sample with deep UV is by employing an external binary chromium/quartz mask complementary to the 3D structures. In this case a proper alignment and close contact between the mask and the sample needs to be achieved. This method offers the additional possibility to selectively irradiate a subset of walls or wells in the sample. This would allow for control experiments with nonfunctionalized barriers in the same sample as experiments with functionalized barriers.
87
88
CHAPTER 5 Functional Microtubule–Barrier Interactions
We presented a method based on semipermanent biotin–streptavidin bonds to reconstitute dynamic microtubule “end-on” interactions for the specific case of cortical dynein (assay 1). However, the same technique could be used to study microtubule interactions with other cortical factors like CLASPs, or microtubule interactions with kinetochore complexes. Alternatively, “end-on” interactions have recently been achieved in emulsion droplets (Laan, Roth, & Dogterom, 2012) and using beads with the addition of a long tether (Volkov et al., 2013). Potentially, functionalized microwells could furthermore provide a framework to study controlled lateral interactions; for example, by placing cells inside wells where the walls mimic lateral cellular adhesion conditions. Ni(II)-NTA based immobilization does not require protein biotinylation and suffers little from protein inactivation (Bieling, Telley, Hentrich, Piehler, & Surrey, 2010). Moreover, since it is reversible and tunable with the addition of imidazole, it is better suited to study cellular processes that involve transient bindings. We showed in assay 2 that microtubule-based protein delivery and docking of his-tagged þtips can be achieved with tris-Ni(II)-NTA at walls. To also allow for diffusion of tethered proteins along the barrier, one could alternatively use emulsion droplets with Ni(II)-NTA lipids using a method analogous to Laan, Roth, et al. (2012).
Acknowledgements We thank Gijs Vollenbroek, Dimmitry Lamers, Andries Lof, and Hans Zeijlemaker for advice and help with microfabrication. Liedewij Laan for initiating this project and advice. Magdalena Preciado-Lopez, Sophie Roth, Cristina Manatschal, and Michel Steinmetz for teaching us the purification of proteins and for discussions. Vanda Sunderlı´kova´ for help with protein purification. We thank Pierre Recouvreux for teaching us cloning techniques and for discussions, and David Minde for careful reading of the manuscript. This work is part of the research program of the “Stichting voor Fundamenteel Onderzoek de Materie (FOM)” which is financially supported by the “Nederlandse organisatie voor Wetenschappelijk Onderzoek (NWO).” This work is further supported by NanoNextNL, a micro and nanotechnology consortium of the Government of the Netherlands and 130 partners.
References Akhmanova, A., & Steinmetz, M. O. (2008). Tracking the ends: A dynamic protein network controls the fate of microtubule tips. Nature Reviews. Molecular Cell Biology, 9, 309–322. Azioune, A., Storch, M., Bornens, M., The´ry, M., & Piel, M. (2009). Simple and rapid process for single cell micro-patterning. Lab on a Chip, 9, 1640–1642. Bhagawati, M., You, C., & Piehler, J. (2013). Quantitative real-time imaging of proteinprotein interactions by LSPR detection with micropatterned gold nanoparticles. Analytical Chemistry 85, 9564-9571. Bieling, P., Laan, L., Schek, H., Munteanu, E. L., Sandblad, L., Dogterom, M., et al. (2007). Reconstitution of a microtubule plus-end tracking system in vitro. Nature, 450, 1100–1105.
References
Bieling, P., Telley, I. A., Hentrich, C., Piehler, J., & Surrey, T. (2010). Fluorescence microscopy assays on chemically functionalized surfaces for quantitative imaging of microtubule, motor, and þTIP dynamics. Methods in Cell Biology, 95, 555–580. Buranda, T., Lopez, G. P., Keij, J., Harris, R., & Sklar, L. A. (1999). Peptides, antibodies, and FRET on beads in flow cytometry: A model system using fluoresceinated and biotinylated b-endorphin. Cytometry, 37, 21–31. Desai, A., & Mitchison, T. J. (1997). Microtubule polymerization dynamics. Annual Review of Cell and Developmental Biology, 13, 83–117. Dodgson, J., Chessel, A., Yamamoto, M., Vaggi, F., Cox, S., Rosten, E., et al. (2013). Spatial segregation of polarity factors into distinct cortical clusters is required for cell polarity control. Nature Communications, 4, 1834. Dogterom, M., & Yurke, B. (1997). Measurement of the force–velocity relation for growing microtubules. Science, 278, 856–860. Frasconi, M., Mazzei, F., & Ferri, T. (2010). Protein immobilization at gold–thiol surfaces and potential for biosensing. Analytical and Bioanalytical Chemistry, 398, 1545–1564. Fygenson, D. K., Braun, E., & Libchaber, A. (1994). Phase-diagram of microtubules. Physical Review E, 50, 1579–1588. Goodman, B. S., Derr, N. D., & Reck-Peterson, S. L. (2012). Engineered, harnessed, and hijacked: Synthetic uses for cytoskeletal systems. Trends in Cell Biology, 22, 644–652. Green, N. (1990). Avidin and streptavidin. Methods in Enzymology, 184, 51–67. Grishchuk, E. L., Molodtsov, M. I., Ataullakhanov, F. I., & McIntosh, J. R. (2005). Force production by disassembling microtubules. Nature, 438, 384–388. Hochuli, E., Do¨beli, H., & Schacher, A. (1987). New metal chelate adsorbent selective for proteins and peptides containing neighbouring histidine residues. Journal of Chromatography, 411, 177–184. Hyman, A. A., Salser, S., Drechsel, D. N., Unwin, N., & Mitchison, T. J. (1992). Role of GTP hydrolysis in microtubule dynamics: Information from a slowly hydrolyzable analogue, GMPCPP. Molecular Biology of the Cell, 3, 1155–1167. Janson, M. E., de Dood, M. E., & Dogterom, M. (2003). Dynamic instability of microtubules is regulated by force. Journal of Cell Biology, 161, 1029–1034. Kenausis, G. L., Vo, J., Elbert, D. L., Huang, N., Hofer, R., Ruiz-Taylor, L., et al. (2000). Poly(L-lysine)-g-poly(ethylene glycol) layers on metal oxide surfaces: Attachment mechanism and effects of polymer architecture on resistance to protein adsorption. Journal of Physical Chemistry, 1(04), 3298–3309. Laan, L., & Dogterom, M. (2010). In vitro assays to study force generation at dynamic microtubule ends. Methods in Cell Biology, 95, 617–639. Laan, L., Pavin, N., Husson, J., Romet-Lemonne, G., van Duijn, M., Preciado Lo´pez, M., et al. (2012). Cortical dynein controls microtubule dynamics to generate pulling forces that position microtubule asters. Cell, 148(3), 502–514. Laan, L., Roth, S., & Dogterom, M. (2012). End-on microtubule-dynein interactions and pulling-based positioning of microtubule organizing centers. Cell Cycle, 11, 3750–3757. Lata, S., & Piehler, J. (2005). Stable and functional immobilization of histidine-tagged proteins via multivalent chelator headgroups on a molecular poly(ethylene glycol) brush. Analytical Chemistry, 77, 1096–1105. Lata, S., Reichel, A., Brock, R., Tampe´, R., & Piehler, J. (2005). High-affinity adaptors for switchable recognition of histidine-tagged proteins. Journal of the American Chemical Society, 127, 10205–10215.
89
90
CHAPTER 5 Functional Microtubule–Barrier Interactions
Lombillo, V. A., Stewart, R. S., & McIntosh, J. R. (1995). Minus-end-directed motion of kinesin-coated microspheres driven by microtubule depolymerization. Nature, 373, 161–164. Love, J. C., Estroff, L. A., Kriebel, J. K., Nuzzo, G. R., & Whitesides, G. M. (2005). Selfassembled monolayers of thiolates on metals as a form of nanotechnology. Chemical Reviews, 105, 1103–1609. Maurer, S. P., Bieling, P., Cope, J., Hoenger, A., & Surrey, T. (2011). GTPgS microtubules mimic the growing microtubule end structure recognized by end-binding proteins (EBs). Proceedings of the National Academy of Sciences of the United States of America, 108(10), 3988–3993. Mitchison, J., & Kirschner, M. (1984). Dynamic instability of microtubule growth. Nature, 312, 237–242. Reck-Peterson, S. L., Yildiz, A., Carter, A. P., Gennerich, A., Zhang, N., & Vale, R. D. (2006). Single-molecule analysis of dynein processivity and step-ping behavior. Cell, 126, 335–348. Romet-Lemonne, G., VanDuijn, M., & Dogterom, M. (2005). Three-dimensional control of protein patterning in microfabricated devices. Nano Letters, 12, 1350–2354. Sawin, K. E., & Snaith, H. A. (2004). Role of microtubules and tea1p in establishment and maintenance of fission yeast cell polarity. Journal of Cell Science, 117, 689–700. Vale, R. D. (2003). The molecular motor toolbox for intracellular transport. Cell, 112, 467–480. Volkov, V. A., Zaytsev, A. V., Gudimchuk, N., Grissom, P. M., Gintsburg, A. L., Ataullakhanov, F. I., et al. (2013). Long tethers provide high-force coupling of the Dam1 ring to shortening microtubules. Proceedings of the National Academy of Sciences of the United States of America, 110, 7708–7713. Zhen, G., Falconnet, D., Kuennemann, E., Vo¨ro¨s, J., Spencer, N. D., Textor, M., et al. (2006). Nitrilotriacetic acid functionalized graft copolymers: A polymeric interface for selective and reversible binding of histidine-tagged proteins. Advanced Functional Materials, 16, 243–251.
