|Home | About | Journals | Submit | Contact Us | Français|
Time-lapse atomic force microscopy (AFM) is widely used for direct visualization of the nanoscale dynamics of various biological systems. The advent of high-speed AFM instrumentation made it possible to image the dynamics of proteins and protein-DNA complexes within millisecond time range. This chapter describes protocols for studies of structure and dynamics of nucleosomes with time-lapse AFM including the high-speed AFM instrument. The necessary specifics for the preparation of chromatin samples for imaging with AFM including the protocols for the surface preparation are provided.
Chromatin dynamics is needed for executing all genetic function by the cell such as DNA replication and transcription. A fundamental unit of chromatin, the nucleosome core particle (NCP), is very compact and considered stable. A regular NCP includes 147 bp of DNA duplex that is tightly wrapped around an octameric core, comprising histones H2A, H2B, H3, and H4. The NCP structure is stabilized by electrostatic interactions, specific hydrogen bonds, and salt bridges, identified in the high-resolution crystal structure of NCPs . The high stability of the NCPs leads to the question of how the DNA within the nucleosome can be accessed by regulatory proteins and polymerases for transcription and DNA replication. Chromatin remodeling proteins are capable of the DNA dissociation from the histone core [2–5] suggesting that nucleosome dynamics should be involved into the nucleosome dissociation process. Indeed, studies performed in the past decade with the use of various techniques, including single-molecule approaches, showed that the NCP is not a static structure. Rather, DNA can spontaneously and transiently dissociate, and single-molecule fluorescence and time-resolved techniques revealed that nucleosomes undergo local dissociation of DNA in the absence of remodeling proteins [6–10] and this process occurs on the subsecond timescale . Atomic force microscopy (AFM) was instrumental in direct visualization of the nucleosome dynamics with the nanometer-range spatial resolution [11, 12]. Implementation of high-speed AFM (HS-AFM; (reviewed in refs. 13–15) capable of the nanometer resolution on the millisecond timescale made it possible to identify various pathways of the NCP dynamics [16, 17]. The AFM methodologies enabling direct visualization the chromatin dynamics using regular and high-speed time-lapse AFM modes are outlined in this chapter.
A schematic illustrating the principles of the AFM operation is shown in Fig. 1. A sharp stylus (AFM tip shown as a triangle) reads the sample topography (shown as a bumpy profile) while it moves over sample in a raster pattern termed scanning. The tip cantilever works as a spring pressing the tip against the sample during scanning. The vertical movement of the tip is detected by the optical lever principle in which the tip displacement is measured by the changing of the laser spot location on the position-sensitive photodetector (PSD). Note that no special contrasting sample is needed for AFM imaging. Additionally, scanning can be performed in any media at ambient conditions, including physiological conditions enabling direct visualization of the sample dynamics as described in this chapter. AFM instruments include a number of important features that enable the production of high-resolution images. First, the position of the sample relative to the tip is controlled by the scanner with an accuracy of less than 1 nm. Second, the tip can be atomically sharp. Third, the displacement of the tip relative to the surface is determined with subnanometer accuracy. All these features are critical for the use of AFM for biomedical studies including the chromatin dynamics.
Prepare all solutions using deionized water; Aquamax Water System (Aquamax Laboratory, Van Nuys, CA) produces low-conductivity water (18.2 MΩ) with a required quality. Use analytical grade reagents when preparing the solutions.
The 601 Widom sequence was used as a template for the nucleosome assembly which has the very high affinity for binding of the histone core compared to other sequences .
Histone octamers were assembled according to the protocol described in . Commercially available histones H2A, H2B, H3, and H4 (New England Biolab, Ipswich, MA) are suitable.
The sections below describe the procedure for the preparation of samples of nucleosomes for AFM imaging. This protocol utilizes the methodology of mica functionalization with 1-(3-aminopropyl) silatrane (APS-mica) that provides very reproducible and reliable results. The APS synthesis protocol is described in .
This section describes the AFM imaging procedures for imaging dried and wet samples.
For imaging in air, any type of tip with a spring constant of approximately 40 N/m and a resonant frequency between 300 and 340 kHz can be used. For example, Olympus silicon probes (Asylum Research, Santa Barbara, CA) with a spring constant of 40 N/m and a 300 kHz resonant frequency in air work reliably in the tapping/oscillating mode for imaging in air. Probes with similar characteristics are currently manufactured by a large number of other vendors.
For imaging in liquid a regular AFM, Si3 N4, 100 μm long probes (SNL, Bruker-Nano/Veeco, Santa Barbara, CA) with a spring constant of approximately 0.06 N/m and a resonance frequency around 7–10 kHz are recommended. AFM probes with similar characteristics from other vendors are available. The protocols described below assume the use of Multi Mode AFM (Bruker-Nano/Veeco), but they can be adapted to any type of AFM. Protocol for imaging with HS AFM is given in a separate chapter.
This section describes the protocol for the preparation of nucleosome samples for imaging with high-speed AFM (HS AFM) instrument designed by T. Ando . This paper also provides a protocol for the instrument operation. The instrument is designed for mica disks as small as 1 mm and 1.5 mm available from RIBM (Tsukuba, Japan).
Schematics in Fig. 7 explain how the structural parameters of nucleosomes can be obtained from the AFM images.
The DNA template (353 bp; panel A) contains the sequence with a high specificity for nucleosome binding (147 bp, Widom 601 motif; blue color) located at different distances from the ends. Such an asymmetry was beneficial for structural analysis of nucleosome allowing us to distinguish the dynamics of the left and right flanks during the time-lapse imaging. DNA wraps around the histone core with 1 and ¾ turns to make a crystallographic structure . However the dynamics of this structure can lead to nucleosomes with variable number of DNA turns around the histone octamer and schematically this dynamics is shown in Fig. 7b. The starting position in this set is the nucleosome with one turn around the core. Rotation of the arm by a quarter of turn leads to the nucleosome with 90° angle between the flanks. Further wrapping produces the structure with the parallel orientation of the flanks and additional one-quarter wrapping step leads to the crystallographic structure with 1.75 DNA turns with 147 bp around the core. One more rotation leads to the structure that is geometrically similar to the starting one, but the lengths of flanks are twice shorter. Table c summarizes the structural parameters of the structures described above.
Two major parameters measured in assigning the number of turns on the AFM images of nucleosomes are the lengths of the free DNA flanks and angles between the flanks. DNA contributes substantially to the nucleosome volume; therefore this parameter can also be used for additional validation of the DNA wrapping values.
The analysis is essentially similar to the one described above and is performed over considerably larger dataset. The length and angle measurements are typically sufficient for the analysis.
The authors thank the members of the Lyubchenko lab and specifically N. Filenko and M. Atsushi for their contribution to the protocols on handling of nucleosomes for AFM imaging. The work is supported by grants from NIH (P01 GM091743, 1R01 GM096039) and NSF (EPS—1004094).
1Depending on the size of the mica strip, the plastic disposable 3 mL cuvettes or plastic 15 mL tubes are suitable for these purposes.
2As prepared, the APS mica sheets can be stored dry (plastic tubes or cuvettes) in the argon atmosphere for at least a week.
3Nitrogen gas can be used but it is recommended to use Ar gas as it is heavy and does evaporate from the tube.
4The length measurements can be performed in nanometers. These values can be converted in the number of base pairs. The conversion coefficient can be obtained from the length measurements of free DNA. Generate the histogram from multiple measurements (100 molecules is sufficient) and divide 353 bp by the mean length number. For APS mica procedure this value is very close to 0.34 nm for B-DNA base pair spacing.
5The volume measurements can be used, but these are not direct measurements of the number of turns due to the tip convolution effect and the model used for the volume measurements. We recommend to make a calibration plot using the volume measurements for a set of nucleosomes with the number of turns unambiguously determined by the angle and length measurements. Make a graph as a dependence of the volume on the number of the DNA turns. Use the same model of nucleosome for all measurements. The effect of the tip convolution effect can be incorporated by the measurements of the DNA heights and width. Similar parameters justify the measurements.
6The estimates of the DNA turns obtained by different methods should not exactly be the same, but typically they are close. Each type of measurements has limitation. The length measurements are limited by the identification of the DNA detaching point on the nucleosome image. The nucleosome size on the AFM images is enlarged due to the tip convolution effect and therefore the measured DNA length is shorter as it should be. Therefore, the length of wrapped DNA is increased leading to the elevated values for the DNA turns as it is seen in Fig. 8c. However, the dependence of the values over time should be similar and this is illustrated in Fig. 8c.