There are numerous types of single molecule imaging techniques where the trick is to reduce the density of fluorophores being imaged at the same time, allowing for the single molecules to be localised post-imaging for each frame and a molecular map to be created. Depending on your sample set up you will either use STORM (Stochastic Optical Reconstruction Microscopy (STORM) or PALM (Photoactivtable Localisation Microscopy). Both are performed in fixed samples with PALM requiring exogenous molecules tagged to photoactivatable fluorophores and STORM performed on endogenous molecules tagged with fluorescent probes. SPT (single particle tracking) PALM is a technique possible in live cells to track single particles over time.
An array of single molecules imaged all at the same time are indistinguishable as their fluorescent signals will overlap If these same molecules are imaged sequentially the centroid position of the signal from each individual molecule can be established and a molecular map can be produced. For this to occur the activation of single fluorophores needs to be temporally controlled. Photoactivatable fluorophores (ie. photoactivatable mCherry) were developed to possess the ability to be switched on and off in an irreversible manner. Unlike conventional fluorophores which simply fluorescent after excitation by a specific wavelength of light; photoactivatable fluorophores require an additional photoactivation step to engage the fluorescence mechanism. This photoactivation step, normally illumination with UV light, turns on the fluorophore so they can undergo normal fluorescence. Once activated these fluorophores are subsequently bleached off, an irreversible process that prevents the same molecule being re-imaged within the sample. When combined with controlled activation, see below, this process allows only one molecule to only be localised once within a cell. The quantal activation and bleaching of these fluorophores (it occurs in one step) is important in identifying when a single molecule is activated and makes PALM a highly accurate super-resolution technique.
The crucial condition for PALM is that the number of molecules activated in a single step must be controlled to prevent overlapping point spread functions. PALM activation is therefore a cyclic process and by applying a short burst of controlled intensity UV within one cycle only a small subset of molecules are activated, these are then imaged under their normal excitation wavelength and their positions recorded before being bleached off. The x-y co-ordinates are then stored, and a second activation can then be applied to repeat the process on a new subset of molecules. This process is repeated many times to enable the localisation of tens of thousands of molecules within a cell, enabling quantification of the spatial organisation within a cellular structure such as the plasma membrane of a neuroendocrine cell.
Stochastic Optical resolution microscopy, like PALM, relies on stochastic switching on and off of fluorescence emitted from probes, both require collection of multiple images for structural assignment. Data from either PALM or STORM can be analysed using statistical algorithms relying on localisation of the centroid of intensity generation a probability map of the fluorescent signals, this is usually described by localisation accuracy. STORM differs from PALM because it makes use of chemical fluorescent dyes. STORM works on the principle that fluorescent chemical dyes can be induced to randomly reversibly photoswitch between states where fluorescence is emitted ‘on state’ and not emitted ‘off state’.
By imaging dyes such as Alexafluor 647 at a specific laser intensity the dye will be pushed into the off state. Mark Bates and Xiowei Zhuang observed that by placing the AlexaFluor 647 next to a FRET donor dye such as AlexaFluor532 caused the Alexafluor647 dye to switch back nn from the off state. This reversible photoswitching of dye molecules is facilitated by the chemical environment of the dye. Later work by Mike Heilemann, Stephan Van Linde and Marcus Sauer showed that placing the fluorescent samples in a reducing environment using Thiol based buffers facilitated the switch from off state to the on state without a donor dye needing to be present, this technique is called dSTORM. Very recent work shows mounting samples in Vectashield, and correcting the refractive index with glycerol can also induce this effect.
Reactive oxygen species break the chemical bonds which make up the dye chromophore causing it to be irreversibly damaged and unable to emit fluorescent light. By making use of an Oxygen scavenging buffer, or chemical antifades such as DABCO and NPG the likelihood of Bleaching / Photodamaging the sample is reduced. For STORM single molecule localisation microscopy (SMLM) optimisation of the buffer for imaging is critical and also can be complex. Certain fluorophores are better suited to STORM and a full list of dyes and recommended buffers for photoswitching is available following this link: http://www.nature.com/nprot/journal/v6/n7/full/nprot.2011.336.html
Both techniques can be performed at Heriot-Watt University (TIRF only) and Edinburgh University (TIRF and 3D), find out more at Our Facilities for full specifications.
Techniques such as Photoactivated localisation microscopy (PALM) and Ground state depletion imaging microscopy (GSDIM) provide invaluable information about the organisation of single molecules within a fixed cellular structure. Cells are of course alive and molecules within cells move around; often very quickly. Molecular imaging approaches can be applied to visualise the movement of many 1000s of these molecules using photoactivatable fluorophores, typically at the membrane of living cells. This is referred to as single particle tracking PALM (sptPALM).
To track single molecules within a live sample the image needs to be acquired very quickly. Only around 2000 photons (not a lot of light) can be collected from a single molecule before it permanently photobleaches and so low background noise is ideally required to identify and track these molecules. Unfortunately imaging at a higher frame rate decreases the number of collected photons per frame, increasing the noise on the image and makes the segregation between signal (the light from the molecules) and the noise (from the detectors and background) very difficult. Fortunately these issues can be overcome using de-noising algorithms, which can identify each molecule and deliver quantitative data describing each molecule's trajectory.
Above is a figure from a recent paper by Rhodri Wilson which shows single SNARE molecules tracked throughout a neuronal cell and analysed to show molecule trajectory over time. This work was developed at ESRIC and published in 2016.
SPT-PALM can be performed at Heriot-Watt University (TIRF only), find out more at Our Facilities for full specifications.