AFM for analysis of structure and dynamics of DNA and protein-DNA complexes

2.1. Materials

2.2. Functionalization of mica with 3-aminopropyltriethoxy silane (AP-mica preparation)

The protocol below describes the methodology for the preparation of positively charged mica surface provided by covalent binding of APTES to mica surface.

Place two plastic caps (cut them from regular 1.5 ml plastic tubes) on the bottom of 2 l desiccators and vacuum it with a regular vacuum pump and fill with argon. Cleave mica sheets (approx. 5×5 cm) to make them as thin as 0.1–0.05 mm and mount at the top of the desiccators. Place 30 μl of APTES into one plastic cap and 10 μl of N,N-diisopropylethylamine (Aldrich) into another cap and allow the reaction to proceed for 1–2 hours. After that remove the cap with APTES and purge argon for 2 min. Leave the sheets for 1–2 days in the desiccator to cure; AP-mica is ready for the sample deposition after that. This procedure allows one to obtain a weak cationic surface with rather uniform distribution of the charge. Dry argon atmosphere is crucial for obtaining the substrates for AFM studies and for storage of the substrate. Allow the gas to flow while you open the desiccator. With these precautions the AP-mica substrates retain their activity for several weeks.

2.3. Synthesis of 1-(3-aminopropyl)silatrane (APS)

APS is not commercially available, but can be synthesized with use of rather simple equipment. A catalytic amount of sodium metal (5 mg) is added to 15.0 mL (16.8 g, 0.11 mol) of triethanolamine (Aldrich) in a 250 mL round-bottom flask under argon or nitrogen atmosphere and allowed to form a solution. A rubber balloon is then attached to the flask via a rubber septum and a needle to allow hydrogen to escape without building up pressure. Moderate heat (up to 100°C) can be applied to accelerate the process, but the mixture should be allowed to cool to room temperature before the next step. An equivalent amount of (3-aminopropyl)triethoxysilane (26.4 mL or 25.0 g, 0.11 mol) is added to the mixture, then the flask is placed into a 60° C water bath and connected to the vacuum line to absorb ethanol released in the reaction. This reaction can be simply performed on a rotary evaporator. At the end of the reaction, the mixture loses approximately 17 g of ethanol and is turned into a solid. This process can take more than 24 hr, but mechanical or occasional manual stirring can reduce time to 1–2 hr. Evaporation at the end with 150 mL of xylenes at 60° can also accelerate the process and help crystallization. Vacuum-dried APS obtained by this method can be used directly for AFM, or for better results and higher stability, the product can be purified by crystallization. Minute amounts of sodium hydroxide in the product does not practically affect the performance of the reagent, or change the pH of stock solutions of APS used for AFM. Recrystallization from xylenes provides 20 g (80% yield) of colorless solid powder. The synthesized APS has the following characteristics: m.p. 91–94°C (open capillary tube); m.p. 87.2–87.9°C (sealed capillary tube). ¹H NMR (DMSO-d₆), ppm: 0.08–0.14 (2H, m, SiCH₂); 1.1 (2H, br. s, NH₂); 1.28–1.37 (2H, m, CH₂); 2.37 (2H, t J = 7.2 Hz, NCH₂); 2.77 (6H, t J = 5.9 Hz, NCH₂); 3.59 (6H, t J = 5.9 Hz, OCH₂).

2.4. Mica functionalization with APS

Figure 1 schematically illustrates the reaction of APS with mica. Prepare 50 mM APS stock solution in water and store it in refrigerator. The shelf life of the stock solution is not less than 6 months. Prepare working APS solution for mica modification dissolving the stock solution in 1:300 ratio in water; it can be stored at room temperature for several days. Cleave mica sheets of needed sizes (typically 1×3 cm) to make them as thin as 0.05–0.1 mm, place them in appropriate plastic tubes and pour working APS solution to cover the mica sheet and leave on the bench for 30 min. Remove the mica sheets wash with deionized water and dry them under Argon stream. The strips are ready for the sample preparation. Note, however, as prepared, the APS mica sheets can be stored under Argon for several days.

2.5. Sample preparation for AFM imaging in air

Two procedures are described. AP-mica is mentioned in the protocols below, in the droplet and immersion procedures, however the same protocol is also applied for APS-mica. The first procedure is attractive due to the use of small amount of the sample, whereas the immersion procedure is recommended for procedures requiring the control of environmental conditions such as temperature, pH and salt concentration.

2.6. Procedure for AFM imaging in air

The procedure for imaging in air is straightforward: mount the sample and start approaching. Although both contact and intermittent (tapping) modes can be used, the latter is preferable and allows one to get images of DNA and DNA-protein complexes routinely. Any types of probes designed for non-contact imaging can be used. NanoProbe TESP tips (Digital Instruments, Inc.) or Olympus and conical sharp silicon probes of K-TEK International (Portland, OR) work well. Typical tapping frequency of 240–380 KHz; scanning rate of 1.5–2.5 Hz allows one to obtain stable images.

2.7. Procedures for imaging in solution

The capability of AFM to perform scanning in liquid is its most attractive feature for numerous biological applications allowing imaging at conditions close to physiological ones. In addition, this mode of imaging permits one to eliminate undesirable resolution-limiting capillary effect typical for imaging in air (e.g., (8, 40, 41)). As a result, images of DNA filaments as thin as ~3 nm were obtained in water solutions (42) and helical periodicity was observed when dried DNA samples were imaged in propanol (11). In addition to APS-mica, our previously developed procedure with the use of AP-mica can be used as a substrate for imaging in liquid (10); note that the first images of DNA in fully hydrated state were obtained by the use of AP-mica (11). This type of imaging is recommended in cases when dynamics is studied. The following procedure is described for the use of MultiMode AFM (Veeco, Santa Barbara, CA) and can be easily adapted for other systems. Install an appropriate tip in the holder designed for imaging in liquid (fluid cell). Use Si₃N₄ big triangle with thick legs or small triangle with thin legs cantilevers (42). Mount the AP –mica or APS mica sheet on the stage of the microscope. Mica pieces of 1cm × 1cm are sufficient for MultiMode AFM design of fluid cell. Attach the head of the microscope with installed fluid cell and make appropriate adjustments of the microscope. Approach the sample to the tip manually, leaving ~50–100 μm gap between the tip and the surface. This gap is sufficient to inject ~50 μl of the sample. Allow the microscope to approach the sample and to engage the tip. Adjust the values for the set point voltage and drive amplitude to improve the quality of images and start the data acquisition using the continuous mode scanning. The best resonance frequencies are usually around 6–9 kHz. APS modified mica also can be used in MFP 3D instrument (Asylum Research, Santa Barbara, CA). For that, the piece of mica should be glued to the glass slide first and then modified as described above with APS solution.

2.8. AFM force spectroscopy

This section outlines specifics for the immobilization of molecules utilizing maleimide silatrane (MAS) methodology (43), the suitability of which has been proven in a number of single molecule experiments including AFM force spectroscopy (22, 43–45).

The immobilization chemistry is shown in Fig. 2. A freshly cleaved mica surface (cartoon 1) is treated with an aqueous solution of maleimide silatrane (MAS, 2) for several minutes and then rinsed, leading to a surface containing immobilized maleimide units as depicted in cartoon 3. Surface 3 is then treated with a buffered solution containing the solution of thiol-modified DNA. The terminal thiol group covalently adds to the maleimide units on the surface leading to the immobilized DNA as shown in cartoon 4.

Thiol group was incorporated into DNA during a standard chemical synthesis. Specifics for the preparation and purification of DNA can be found in (45). To reduce protected thiol groups of the oligonucleotides, the duplexes were treated with 10 mM buffered solution (10 mM HEPES, 50 mM NaCl, pH 7.0) of TCEP-hydrochloride (Tris(2-carboxyethyl)phosphine) at 25°C for 10 min. Silicon nitride (Si₃N₄) or MLCT AFM tips were initially washed in ethanol by immersion for 30 min and then activated by UV treatment for 30 min t. Activated tips and freshly cleaved mica surface were treated with 167 μM solution of maleimide-silatrane for 3 hours followed by rinsing with deionized water. Maleimide functionalized AFM tips and mica were incubated for 1 hour respectively in 10 μM and 80 μM of double stranded DNA. After washing with HEPES buffer solution (10 mM HEPES, 50 mM NaCl, pH 7.0), unreacted maleimide moieties were quenched with buffered 10 mM β-mercaptoethanol solution by 10 min treatment at room temperature. The modified tips and mica were washed with 10 mM HEPES, 50 mM NaCl, pH 7.0 and stored in the same buffer until use.

3.1. AFM imaging with the use of functionalized surfaces

The procedure for the preparation of samples with the use of functionalized AP-mica or APS-mica procedures is straightforward. The sample is placed onto the surface to allow the sample to diffuse to the surface and bind. After that, the sample is rinsed with water, dried with a clean gas and placed on the microscope stage for imaging. Figure 3 shows typical images obtained with the use of described protocols. We were able to image as well such an alternative DNA structure as supercoil stabilized cruciforms (46–48). The following features of AP- and APS-mica protocol were critical in reliable and reproducible detection of these dynamic structures.

1. The samples can be deposited in a wide range of ionic conditions, temperatures and pH.

2. No divalent cations are needed for the sample binding to the surface.

3. The substrate is stable and retains the binding activity during several days after the preparation.

4. The samples deposited on the substrates are stable and do not accumulate impurities during several months with minimal precautions for the sample storage.

The following issue related to the effect of the surface on the DNA conformation should be taken into a consideration. Deposition of DNA molecules on a surface certainly leads to distortion of its structure, but the extent of such distortion depends on the physical and chemical characteristics of the surface and the DNA-surface interactions. Quite popular methods utilizing the use of divalent cations work very well for imaging of linear DNA (1, 49, 50). At the same time, the use of a modified Mg-assisted technique the images of supercoiled DNA appear with large loops between the very tightly twisted segments of the plectonemic superhelix was a typical feature of such images (51–55). Such large loops should not present in DNA in solution and they are not appear on the computer simulated images (56); therefore the appearances of large loops in supercoiled DNA are probably induced by the immobilization procedure. The use of functionalized mica allowed us to eliminate or at least to minimize above mentioned problems and to obtain the images of supercoiled DNA (42) fully consistent with the data obtained in solution (57) and the predictions of theoretical analysis for the conformations of supercoiled DNA (56). In addition, supercoiled DNA in low salt buffer, but in the presence of Mg²⁺ cations are observed as tightly interwound plectonemic molecules (19, 42). This observation is in perfect agreement with data in solution (58), suggesting that the charge neutralization effect of Mg²⁺ cations is much stronger than that of monovalent cations.

Altogether, these data further warrant the application of the functionalized mica techniques to studies of global DNA conformations and the structure and dynamics of negative DNA supercoiling, which is a natural state of DNA at physiological conditions and within cells in particular. Local DNA structures are very labile and dynamic, so such conditions as ionic strength and pH are critical for their visualization. Furthermore, our AFM study revealed the interplay between the conformational transition of a cruciform and the global conformation of supercoiled DNA can be a molecular trigger for the slithering of DNA chains, the process that facilitates communication between distant sites in the molecule (24, 48). Note that both AP- and APS-mica have very similar characteristics in terms of the surface smoothness, holding the sample on the surface provided and the high reproducibility of the results. Both surfaces were very useful for studies of protein-DNA complexes and specifics for these experiments can be found in a number of papers (25–28, 59, 60).

3.2. AFM imaging in aqueous solutions

Time-lapse imaging mode is one of the attractive modes of topographic AFM applications. In the pioneering work (61), AFM was applied to the imaging the fibrin self assembly. In the series of works from the group of Bustamante, AFM was applied to the observation of RNA polymerase movement along DNA (8, 9, 62–64). We used time lapse imaging and AP-mica methodology for studies dynamics of local and global dynamics of supercoiled DNA with the focus on structural dynamics of cruciforms (42, 47). However, for the experiments in aqueous solutions APS-mica, there appears to be a better substrate than AP-mica. We observed that the APS mica surface remains smooth during experiments in liquid, rendering APS-mica superior to AP-mica (18). Controlling the surface density of amines is another important advantage of the APS mica method for the time-lapse AFM imaging experiments. These two features were critical, allowing single molecule AFM detection of the extensive dynamics of three-way DNA junctions (16), observations consistent with our earlier DNA cyclization studies in solution (65, 66). The observation of the H-to-B- form transition (18, 24) is another example. The results on direct visualization of recombination DNA dynamics enabling us to test the models for branch migration (20, 67) along with visualization of dynamics of nucleosomes are briefly described in subsections below.

3.3. AFM force spectroscopy: Stability and dynamics of synaptic DNA-protein complexes

AFM force spectroscopy allowing measurements of intermolecular interaction as well as intramolecular stability is the area of AFM application with fast growing demand. In this mode, the AFM measures the mechanical stability of the system positioned between the surface and AFM tip. Covalent attachment of the system at a well defined anchoring point is required for the force pulling experiments. We have shown that functionalized surfaces open prospects for immobilizing the system in fast, efficient and reproducible fashion (22, 23, 38, 44, 45, 75–79). The most promising is the approach with the use of silatranes with specific reactive groups allowing anchoring the molecule via formation of covalent bonds. The example below illustrates the immobilization approach utilizing maleimide silatrane capable of binding thiol-containing molecules.

Our AFM topographic studies of the complex formed by SfiI restriction enzyme and DNA revealed a number of important structural characteristics of this system, suggesting interesting dynamic properties of this system. We have developed single molecule AFM force spectroscopy approach enabling one to measure the stability of site-specific protein-DNA complexes involving the interaction of two DNA sites performed by a specific complex (45). The experimental setup is shown schematically in Fig. 7. The identical DNA duplexes containing SfiI recognition sites (shown as two parallel red thick lines) are covalently attached to the AFM substrate and tip via flexible linker (zigzag green lines, section A). The synaptic complex between the two duplexes is formed by SfiI when the duplexes are in the proximity to each other and SfiI (blue tetrameric oval) binds to them (section B). The complex was formed by the protein present in the solution. Upon retraction of the tip, the flexible linkers are stretched (section C) and the complex is ruptured when the stretching limit is achieved (section D).

The stability of the synaptic complex formed by the enzyme and two DNA duplexes was probed in a series of approach-retraction cycles during which the force distance curved are captured. A typical force-distance curve obtained using the experimental setup for studying the synaptic complex schematically represented by Figure 7A–D is shown in Figure 8. An adhesive peak at the beginning of the force curve (section B of the force curve) is accompanied by a force extension curve (section C) corresponding to the stretching of polymer linkers, followed by a rupture of the synaptic complexes event (section D). A series of control experiments allowed us to justify this assignment (45). The maleimide silatrane immobilization strategy allowed us to determine the dissociation rate (k_off) with a high accuracy due to the possibility to acquire a large set of the data. For example, the DFS experiments performed with two different sequences within the middle part of the DNA recognition region revealed the sequence-dependent difference in the stability of synaptic complex that correlated with the effect of the sequence on the SfiI enzymatic activity (80). Note again that the key in these experiments was the high yield of rupture events rupture events ca.10–20% vs. ca. 1% for multistep procedures, e.g., (81).