Characterizing Individual Protein Aggregates by Infrared Nanospectroscopy and Atomic Force Microscopy

PREPARAÇÃO DO INSTRUTOR

CONCEITOS

PROTOCOLO DO ALUNO

JoVE Journal
Bioquímica

É necessária uma assinatura da JoVE para visualizar este conteúdo. Faça login ou comece sua avaliação gratuita.

JoVE Journal Bioquímica

Characterizing Individual Protein Aggregates by Infrared Nanospectroscopy and Atomic Force Microscopy

1. Aggregation assays on fluorescence plate readers

NOTE: The protocol described here is an example of how to study the aggregation of any protein or peptide by chemical kinetics. In particular, it describes an optimized protocol to study the aggregation of the Aβ42 peptide, which is involved in the onset and progression of Alzheimer’s disease⁵⁸^,⁵⁹. A similar protocol can be adjusted and adopted towards studying the aggregation of any protein or peptide.

Obtain a highly pure monomeric solution of Aβ42 through ion-exchange and size-exclusion chromatography techniques to distinguish protein fractions containing Aβ42, and to isolate the monomeric fraction from other aggregated forms of Aβ42 obtained⁵⁸^,⁵⁹.
Dilute the Aβ42 peptide in 20 mM sodium phosphate buffer, 200 µM EDTA at pH 8.0⁵⁸ to a desired final concentration ranging between 1−5 µM, in 1.5 mL low-bind tubes. Samples collected for AFM measurements should not contain the fluorescent dye used in the aggregation assay (thioflavin T, ThT), as it might introduce artefacts during sample deposition for AFM analysis. Thus, prepare triplicates of two identical solutions: i) the first containing monomeric Aβ42 to follow the process of aggregation by AFM; ii) the second containing the Aβ42 monomeric with addition of 20 µM ThT as a tracer to monitor the kinetics of aggregation.
NOTE: It is important to perform steps 1.1 and 1.2 on ice. This procedure is to ensure that the monomeric Aβ42 solution does not aggregate until initiated in the plate reader. When handling samples, careful pipetting is essential to avoid the introduction of any air bubbles that may affect the protein sample.
Pipette 80 µL of each sample in each well of a 96-well, half-area plate of black polystyrene with clear bottom and nonbinding surfaces.
Seal the plate with a foil to minimize the evaporation of the sample over the course of the aggregation.
Equilibrate the temperature of the plate reader to 37 °C.
Set up the aggregation protocol to read fluorescence measurements at fixed time-point intervals, at an excitation wavelength of 440 nm and an emission wavelength of 480 nm. The aggregation should be initiated in quiescent conditions.
Insert the plate into the plate reader.
NOTE: It is important to be careful while handling the plate, to ensure that the aggregation is not triggered prior to starting the plate reader. Use the gain adjustment settings to ensure the optimal readout of the fluorescence of each sample.
Start the measurement.
NOTE: The aggregation experiment is concluded when the ThT fluorescence sigmoidal curve reaches a plateau⁵⁸.

2. Sample preparation for AFM and nano-IR measurements

Glue highest quality grade V1 mica disc to high quality magnetic stainless-steel disc using adhesive tabs or double-sided tape.
Etch mica by placing a piece of adhesive tape onto the mica surface and pulling off the top layer of the mica. This procedure will produce clean, atomically flat surface, suitable for sample deposition.
NOTE: Etched surface must be uniform and smooth.
CAUTION: Do not touch or breathe directly above freshly etched mica, as this will introduce artefacts.
Pause/stop the plate reader measurement to collect the time point of interest during the aggregation process of the protein.
Remove the sealing foil and collect 10 µL aliquot of the samples without ThT from the well into a 1.5 mL low-bind tubes.
Deposit the 10 µL of the sample on the freshly etched mica. For AFM-IR measurements samples must be deposited on freshly stripped gold substrate.
Incubate the solution on the mica for 1 min to allow physisorbtion.
NOTE: Longer incubation time would allow better absorption on the surface but may induce artificial self-organization and self-assembly²⁵. To avoid these effects, spray deposition may be exploited²⁵.
Rinse three times with 1 mL of ultrapure water.
Dry under a gentle flow of nitrogen to measure the sample in air environment.

3. AFM imaging of the morphology of protein aggregates

NOTE: Morphology measurements can be performed both in contact and dynamic mode, in the following steps the latter is described since it reduces lateral forces to measure the 3D morphology of the sample with high resolution. AFM-IR measurements will be performed in contact mode to enhance AFM-IR signal-to-noise ratio.

Turn on the AFM system at least 30 min before the measurements in order to enable the system to reach thermal stability. Click on Setup, then on Probe. In Probe Change window click on Next to prepare Z, Focus Stage.
Turn off the AFM beam switch (if it is on). Unlock the dovetail locks, disconnect the head from the system and in Probe Change window click on Next.
Mount the AFM cantilever on the probe holder. Connect the head to the system and lock the dovetail locks.
NOTE: Optimal cantilevers to study biological samples in dynamic mode AFM have spring constant ranging between 2−40 N m^-1 and an apex radii ranging between 2−8 nm¹⁴.
Turn on the AFM laser beam switch and in Probe Change window click on Next.
In the pop-up window click on No if the cantilever type did not change. Adjust optical alignment knobs to find the cantilever. Adjust focus onto cantilever and click on Next.
In the pop-up window click on Yes if the focus is on the cantilever.
Position the laser beam at the end of the cantilever using the knobs controlling position of the detection laser. Maximize the total signal measured by the four-quadrant photodiode to at least >1 V using the knobs controlling the position of the deflected laser beam on the position sensitive photodiode (PSPD). In Probe Change window click on Next and then on Close.
Wait 15 min for the cantilever to reach thermal stability. Readjust the position of the deflected laser beam on the PSPD if necessary.
Click on NCM SWEEP, choose desired amplitude of oscillation, click on Use Phase, click on auto and tune the cantilever close to the maximum of its first free resonance of oscillation, which is of approximately 300 kHz for a cantilever with a spring constant of 40 N m^-1.
NOTE: Tuning the cantilever out of the maximum of its free resonance assures higher stability of the measurements¹⁴.
Place the sample on the sample holder. Choose suitable imaging parameters. Typical scanning rate is 0.3−1.0 Hz for a scan area of 1 x 1 µm² to 5 x 5 µm². Typical resolution needed is between 256 x 256 and 1024 x 1024 pixels. Click on Scan Area to choose the scan size and the pixel number.
Focus the optical view on the sample. Click on Approach to approach the sample surface. Once approach is completed click on Lift 100 µm to rise the AFM tip 100 µm above the surface of the sample. Click Expand on the optical image of the sample. Click on the Focus Stage bar to focus the view on the surface of the sample.
Use the arrows to move in the region of the sample of interest. Engage the surface by pressing Approach button. Click on Line Scan button, and check if the tip is following the surface well, if necessary, adjust the Set Point.
Start imaging the sample surface by pressing Scan button. During imaging, to avoid large imaging force and keep consistency of within independent samples, maintain a constant regime of phase change not exceeding Δ20°⁴².
Enter file name and choose base directory where the acquired image will be saved.

4. Infrared nanospectroscopy measurements of protein aggregates

Turn on the AFM-IR system and the infrared (IR) laser⁶⁰ 30−60 min before measurements for thermal stabilization. Typical lasers for AFM-IR instruments are optical parameter oscillators (OPO) and quantum cascade lasers (QCL).
Open the built-in software to control the instrument. Click on file and then on new to open a new nanoIR file. Press the button initialise to start the AFM-IR system.
Open the instrument cover and mount on the AFM-IR system a silicon gold coated probe with a nominal radius of 30 nm and a spring constant of 0.2 N/m to measure the sample in contact mode.
Click on Load in the section AFM probe and then next. In focus on probe section click on the arrows to focus the camera on the cantilever. In the section sample XY movement, use the arrows to place the cross-air at the end of the cantilever.
Rotate the knobs controlling the position of the detection laser to position laser at the end of the cantilever. Rotate the laser knobs to detect and maximize the sum measured by the four-quadrant photodiode to a value >3 V. Rotate the deflection knob to adjust the cantilever deflection to -1 V and then click next. Close the cover of the instrument.
In the section focus on sample use the arrows to focus the camera on the sample. Then, in the section sample XY movement use the arrows to move in the region of interest of the sample and click on approach and engage.
In the microscope window select as inputs the channels of interest to map the biophysical properties of the sample. In particular, choose height for morphology, Amplitude 2 for IR absorption and PLL frequency to map tip-sample contact resonance.
In the AFM scan section of the controls window, use similar parameters as in section 3 (i.e., scan rate 0.1−1.0 Hz, between 256 x 256 and 1024 x 1024 pixels). Choose as value of the gains according to sample roughness: integral I = 1−10 and proportional P = 10−30. Then, click on scan to acquire a morphology map.
NOTE: The morphology can be similarly measured in the dynamic tapping mode, but AFM-IR measurements are performed here in the contact mode to have higher signal to noise ratios.
After the mapping of morphology is finished, in the microscope window, click on the height map to position the probe on the top of one aggregate. Then in the nanoIR section of the controls window, click on start IR to illuminate the sample with the IR laser.
To focus the infrared laser on the cantilever, click on optimisation. In this window, write a wavenumber where the sample is going to have high absorbance and click add. For protein, a typical value is 1655 cm^-1. Click on scan to find the IR laser position and click on update to align its position with the cantilever. Close the window.
In the section general of the nanoIR setting, write a wavenumber where high absorbance in the relative field is expected. Then, deactivate the Band Pass Filter option and look at the meter reading and at the fast Fourier transform (FFT) of the cantilever response. In the FFT window move the green cursor to read the resonance frequency of the cantilever. A typical value of the FFT of the resonance of the cantilevers is around 300 kHz. Write this resonance frequency value in the general section in the freq. centre field and use a freq. window of 50 kHz.
NOTE: Select a laser power that is low enough to not saturate the meter reading and to have distortion in the cantilever response and to not overheat the sample.
Click on laser pulse tune window to use the resonance enhanced mode. Choose a frequency centre of 300 kHz, a tune range of 50 kHz and a duty cycle of the laser of 5%. Click on acquire to sweep the pulse rate of the laser. By using the cursor in the graph, tune the laser pulse to the frequency of the mechanical response of the thermal expansion of the sample absorbing the IR light. Then, select the option PLL to monitor the contact resonance between the sample and the tip. Press the zero button in the PLL window and tick enable to track the sample-tip contact resonance. Choose an integral gain I = 0.5 and proportional gain P = 10. Close the window.
Open again the optimisation window and find the position of the IR laser for at least 3 wavenumbers corresponding to major absorbance bands of the sample (amide band I-II-III) and for at least one wavenumber for each chip of the laser.
Click on Tools | IR Background Calibration | novo. In the window select the spectroscopic region of interest (1800−1200 cm^-1 to study protein samples); choose the same pulse rate as determined in step 4.12 and a duty cycle of 5%; click on fast acquisition and select the laser speed, typical range for a quantum cascade laser is between 20−500 cm^-1. Click on acquire to measure the IR laser background. This background spectrum will be used for normalization of measured nanoscale localized spectra. Close the window.
NOTE: If a fast laser and resonance enhanced mode is not available, skip step 4.12 and select stepped spectra instead of fast in both the background and IR spectra acquisition windows. However, single aggregate sensitivity will not be reached.
In the IR spectra settings, choose an IR spectrum resolution between 1−4 cm^-1 and a number of co-averages of at least 64x. Click on acquire to measure a nanoscale localized IR spectrum in the protein range (1800−1200 cm^-1).
To acquire a nanoscale resolved chemical map, select IR imaging option, choose a wavenumber of interest (e.g., 1655 cm^-1 for amide band I) and click on scan in the AFM scan window.
Once the mapping is completed, go to file and save the measurements. To analyze the acquired maps of morphology, contact-resonance and chemistry, as well as the nanoscale localized spectra, use the built-in AFM image processing software. Spectra and Images can be saved as .csv or .axz files for further detailed analysis with commercial software.

5. Image processing and analysis of cross-sectional dimensions

Flatten raw images¹⁴^,¹⁷^,¹⁹^,⁴¹^,⁴²^,⁵⁹ using built-in AFM image processing software or commercial software.
NOTE: The aggregates should be masked from the calculation to avoid artefacts of analysis and underestimation of their measured height.
Flatten the image by 0 order plane fit subtraction.
Flatten the image by plane and then line by line at a 1^st regression order, the second step is repeated until the flat baseline in line profile of the image is reached.
If the sample is very crowded, contains exceptionally high aggregates or scanner bow artefact is present, flatten image using 2^nd regression order fit.
Measure aggregate height and width from a line profile taken perpendicular to the fibril axis¹⁴^,¹⁷^,¹⁹^,⁴¹^,⁴²^,⁵⁹.
Measure fibril length parallel to the fibril axis¹⁴^,¹⁷^,¹⁹^,⁴¹^,⁴²^,⁵⁹.

Characterizing Individual Protein Aggregates by Infrared Nanospectroscopy and Atomic Force Microscopy

Learning Objectives

A representative time course of Aβ42 aggregation, as measured by the ThT fluorescence assay, is shown in Figure 1. The aggregation process is commonly characterized by a sigmoidal curve, where a lag phase is initially observed, and is followed by a steep growth phase, before the curve reaches a plateau when an equilibrium steady state is reached⁶^,⁷^,⁵⁸. It is essential to ensure that an optimized aggregation protocol is used to generate high-quality data to study the molecular details pertaining to aggregation processes⁵⁸.

High resolution of AFM enables to investigate the morphology and heterogeneity of the aggregated species at different time points of the process. During the aggregation, monitored by the ThT assay, aliquots of the sample in the plates are prepared for single aggregate investigation by AFM and nano-IR (Figure 1). The typical process flow of manual sample preparation is shown in Figure 2. At the completion of the measurement of the 3D morphology of the sample by high resolution and phase-controlled AFM, the maps are flattened to remove non-linearity of the piezoelectric scanner and reduce sources of error in post-processing analysis of sample morphology (Figure 3). Subsequently, an accurate and sensitive single molecule statistical analysis can be performed, as shown in Figure 4. From a 3D morphology map, it is possible to extract aggregate cross-sectional height, width and length (or diameter in case of spheroidal particles), which allows to distinguish and characterize distinct species of aggregates present during the aggregation time course²³^,³⁸. Typical time points of interest to investigate the process of aggregation are the lag phase, the growth phase and the plateau phase (Figure 1). During the lag phase, monomeric and oligomeric species of Aβ42 are primarily present. When visualized by AFM, monomeric and oligomeric species of Aβ42 typically appear as spheroidal particles 1−15 nm in diameter and 0.3−2 nm in height (Figure 1, bottom left)³⁸^,³⁹. Formation of elongated protofilaments, protofibrils and fibrils is visible during the growth phase of Aβ42 aggregation time course (Figure 1, bottom middle)³⁸^,³⁹. Typically, protofilaments appear as elongated features hundreds of nanometers in length and 0.5−2 nm in height, while protofibrils appear as elongated linear or curvilinear aggregates 1−5 nm in height and hundreds of nanometers in length³⁸^,³⁹. During the plateau phase, fibrils are the dominant species of Aβ42 aggregates. Aβ42 fibrils typically appear as unbranched, thread-like structures, with a cross-sectional diameter of 6−10 nm and length in the order if micrometers (Figure 3, bottom right)³⁸^,³⁹. Remarkably, this schematic representation of the morphological properties of the aggregates is a general feature of most aggregating proteins and peptides¹³^,¹⁴.

After the investigation of the sample morphology, nano-IR can be used to investigate the chemical properties of the individual protein aggregate species present during the process of aggregation, by acquiring nanoscale-resolved IR maps and spectra both in air and native liquid environment²⁴^,²⁶^,³⁸^,⁴²⁴⁹^,⁵⁰^,⁵¹^,⁵²^,. Figure 5 shows a schematic illustration of the AFM-IR setup. An IR source is used to illuminate the sample from the bottom in total internal reflection or directly from the top as in Figure 5a. If the IR light is absorbed by the sample, it will excite the corresponding molecular vibrational energy transition levels of the species present. The vibrational energy is dissipated inside the sample in the form of thermal heating, which causes the thermal expansion of the sample. This expansion is measured at the nanoscale by the AFM tip in contact with the sample with a resolution in the order of 10 nm. At each pulse, the cantilever detects the thermal expansion. In particular, the fast expansion of the sample kicks out the cantilever from contact with the sample, and after the kick out the cantilever rings down at its natural frequencies. In order to enhance the sensitivity of AFM-IR, it is possible to tune the laser pulse frequency at the same frequency of the oscillation of the cantilever⁵⁴. In order to work in this resonance-enhanced mode, it is necessary to have IR sources that can be pulsed in a wide range of frequencies, such as quantum cascade lasers that operate typically in a range between 1−1000 kHz. The peak-to-peak amplitude and the fast Fourier transform of the ringdown signal, termed IR amplitude, of the raw cantilever deflection are proportional to the IR light absorbed. These signals are detected in real time by measuring the deflection of a red laser focused on the top of the cantilever. In order to acquire chemical maps, the laser wavenumber is fixed at a certain wavenumber and the IR amplitude signal is collected at each point of the map, while, to acquire IR spectra, the position of the cantilever is fixed in a position of interest and the laser wavelength is swept along the spectroscopic range of interest. The ultimate resolution of AFM-IR enables the measurement of the chemical properties of protein aggregates with a cross-sectional height of approximately 5 nm, as represented in Figure 6.

Figure 1: Monitoring of the aggregation time course in vitro via ThT fluorescence and AFM.
Samples taken during the lag phase, growth phase, and plateau phase of the aggregation process as detected by ThT fluorescence were imaged via high-resolution AFM. This figure has been adapted from Ruggeri et al.³⁸. Please click here to view a larger version of this figure.

Figure 2: Schematic representation of the substrate preparation and sample deposition for AFM measurements.
(A, B) Mica etching using adhesive tape. (C) Sample deposition. (D) Sample incubation. (E) Sample rinsing with ultrapure water. (F) Sample drying under gentle flow of nitrogen. (G) Sample imaging using microfabricated AFM cantilever with sharp tip on its end. (H) Processed image of amyloid fibrils. Please click here to view a larger version of this figure.

Figure 3: AFM image processing procedure⁴².
The top of each panel shows a profile line of the sample surface (red line) illustrated by the corresponding AFM image, while the lower part shows a histogram of the height of all pixels in the image. (a) Raw AFM image before image flattening procedure. (b) AFM image after the processing procedure using the whole plane flattening. Fibrillar structures (pink color) were masked from the flattening procedure. (c) Image processed using line-by-line flattening procedure. (d) Final image after the image processing procedure. This figure has been adapted from Ruggeri et al.⁴².

Figure 4: Single aggregate statistical analysis of AFM images.
(a) Example of the tracing of the heights and lengths of fibrillar structures, indicated with 1 and 2. (b) Graph with sections of the traced fibrils and their average height. (c) Example of a histogram showing the average height of fibrillar structures. (d) Graph with normalized profile of the traced fibrils 1 and 2. (e) Histogram distribution of the normalized profile height points. This figure has been adapted from Ruggeri et al.⁴². Please click here to view a larger version of this figure.

Figure 5: Principle of function of the AFM-IR method.
(a) Absorbed IR light causes the thermal expansion of the sample, exciting the mechanical resonances of the AFM cantilever in contact with the sample. The amplitude of the cantilever oscillations is proportional to the IR absorption. (b) IR absorption maps are obtained scanning the cantilever on the sample while fixing the laser wavelength. (c) Tip and sample in contact behave as a system of coupled springs whose resonant frequency increases monotonically with the intrinsic stiffness of the sample⁵⁰. (d) Localized spectra are obtained by sweeping the laser wavelength while fixing the position of the AFM cantilever. (e) IR spectrum of protein. (f) In summary, AFM-IR enables the simultaneous study of morphological, mechanical and chemical properties at the nanoscale²⁶. Please click here to view a larger version of this figure.

Figure 6: Infrared nanospectroscopy (nano-IR) of a single amyloid aggregate.
(a) AFM morphology map. (b) IR absorption map at the laser resonance peak at 1658 cm^-1. (c) Cross sectional dimensions of the fibril height. (d) IR spectra on different positions of the fibrillar structure (marked in panel b with blue circles) and the substrate (marked in panel b with green circles). The average net signal deriving from the aggregate structure (solid black line) was obtained by subtracting the averaged background signal (solid green line) from the averaged fibril signal (solid blue line). This figure has been adapted from Ruggeri et al.⁴².

List of Materials

AFM-IR system	Anasys Instruments	nanoIR 2 or 3	Systems to measure thermal expansion in contact and resonance mode
Corning 96-well Half Area Black/Clear Bottom Polystyrene NBS Microplate	Corning	3881
Corning Microplate Aluminium Sealing Tape	Corning	6570
Double Sided Adhesive Discs	AGAR Scientific	AGG3347N
FLUOstar Omega	BMG Labtech	415-101	Platereader
Mica Disc 10mm V1	AGAR Scientific	AGF7013
Park NX10 AFM system	Park Systems	N/A	Atomic Force Microscope
Platypus Ultra-Flat Gold Chips	Platypus Technologies	AU.1000.SWTSG
PPP-NCHR-10 cantilevers	Park Systems	PPP-NCHR-10
Protein LowBind Tubes, 2.0mL	Eppendorf	30108132
Silicon gold coated cantilevers	Anasys Instruments	PR-EX-nIR2
SPM Specimen Discs 12mm	AGAR Scientific	AGF7001

Preparação do Laboratório

The phenomenon of protein misfolding and aggregation results in the formation of highly heterogeneous protein aggregates, which are associated with neurodegenerative conditions such as Alzheimer’s and Parkinson’s diseases. In particular low molecular weight aggregates, amyloid oligomers, have been shown to possess generic cytotoxic properties and are implicated as neurotoxins in many forms of dementia. We illustrate the use of methods based on atomic force microscopy (AFM) to address the challenging task of characterizing the morphological, structural and chemical properties of these aggregates, which are difficult to study using conventional structural methods or bulk biophysical methods because of their heterogeneity and transient nature. Scanning probe microscopy approaches are now capable of investigating the morphology of amyloid aggregates with sub-nanometer resolution. We show here that infrared (IR) nanospectroscopy (AFM-IR), which simultaneously exploits the high resolution of AFM and the chemical recognition power of IR spectroscopy, can go further and enable the characterization of the structural properties of individual protein aggregates, and thus offer insights into the aggregation mechanisms. Since the approach that we describe can be applied also to the investigations of the interactions of protein assemblies with small molecules and antibodies, it can deliver fundamental information to develop new therapeutic compounds to diagnose or treat neurodegenerative disorders.

Characterizing Individual Protein Aggregates by Infrared Nanospectroscopy and Atomic Force Microscopy

PREPARAÇÃO DO INSTRUTOR

CONCEITOS

PROTOCOLO DO ALUNO

Characterizing Individual Protein Aggregates by Infrared Nanospectroscopy and Atomic Force Microscopy

Learning Objectives

List of Materials

Preparação do Laboratório

Procedimento

Tags