We describe an 'open' design methodology for wide-field fluorescence, confocal and fluorescence lifetime imaging microscopy (FLIM), and how the resulting microscopes are being applied to radiation biology and protein activity studies in cells and human tissue biopsies. The design approach allows easy expansion as it moves away from the use of a monolithic microscope body to small, commercial off-the-shelf and custom made modular components. Details have been made available under an open license for non-commercial use at http://users.ox.ac.uk/~atdgroup. Two radiobiology 'end-stations' have been constructed which enable fast radiation targeting and imaging of biological material opening up completely novel studies, where the consequences of ionising radiation (signaling and protein recruitment) can be studied <i>in situ</i>, at short times following irradiation. One is located at Surrey University, UK, where radiation is a highly focused in beam (e.g. protons, helium or higher mass ions). The second is installed at the Gray Institute linear accelerator facility, Oxford University, which uses sub-microsecond pulses of 6 MeV electrons. FLIM capabilities have enhanced the study of protein-protein interactions in cells and tissues via Förster Resonance Energy Transfer (FRET). Extracting FRET signals from breast cancer tissue is challenging because of endogenous and fixation fluorescence; we are investigating novel techniques to measure this robustly. Information on specific protein interactions from large numbers of patient tumors will reveal prognostic and diagnostic information.
Photoswitchable and photoactivable proteins Dronpa and PhotoActivable mCherry (PA-mCherry) respectively, were used in order to perform FRET (Förster Resonance Energy Transfer) imaging at the single molecule level, using a FRET standard construct consisting of Dronpa and PA-mCherry separated by seven amino acids expressed in cells. Given Dronpa’s complex photophysical properties and the existence of a preswitched emissive state, irradiation conditions at 491 and 405 nm were optimised. We discuss strategies for observing FRET at the single molecule level with photoactivatable proteins by monitoring modifications in the donor and acceptors emissive states.
Fluorescence Lifetime Imaging (FLIM) is an intensity independent and sensitive optical technique for studying the cellular
environment but its accuracy is often compromised when low photon counts are available for analysis. We have developed
a photon-by-photon Bayesian analysis method targeted at the accurate analysis of low photon count time-domain FLIM
data collected using Time Correlated Single Photon Counting (TCSPC). Parameter estimates obtained with our mono-exponential
Bayesian analysis compare favorably with those using maximum likelihood, least squares, and phasor analysis,
offering robust estimation with greater precision at very low total photon counts, particularly in the presence of significant
background levels. Details of the Bayesian implementation are presented alongside results of mono-exponential analysis
of both real and synthetic data. We demonstrate that for low photon count data, obtained by imaging human epithelial
carcinoma cells expressing cdc42-GFP, Bayesian analysis estimates the green fluorescent protein (GFP) lifetime to a level
of accuracy not obtained using maximum likelihood estimation or other techniques. These results are echoed by the
analysis of synthetic decay data incorporating a 10% uniform background, with our Bayesian analysis routines yielding
lifetime estimates within an accuracy of 20% with about 50 counts. This level of precision is not achieved with maximum
likelihood nor phasor analysis techniques with fewer than 100 counts.
An automated high-content screening microscope has been developed which uses fluorescence anisotropy imaging and fluorescence lifetime microscopy to identify Förster resonant energy transfer between eGFP and mRPF1 in drug screening assays. A wide-field polarization resolved imager is used to simultaneously capture the parallel and perpendicular components of both eGFP and mRFP1 fluorescence emission to provide a high-speed measurement of acceptor depolarization. Donor excited state lifetime measurements performed using laser scanning microscopy is then used to determine the FRET efficiency in a particular assay. A proof-of-principle assay is performed using mutant Jurkat human T-cells to illustrate the process by which FRET is first identified and then quantified by our high-content screening system.
We have used an experimental arrangement comprising two photomultipliers and time-correlated single photon counting
(TCSPC) detection to measure time and polarization-resolved fluorescence decays and images simultaneously.
Polarization-resolved measurements can provide information which may be difficult to extract from lifetime
measurements alone. The combination of fluorescence lifetime and time-resolved anisotropy in an imaging modality
with two detectors minimizes the errors arising from bleaching of a sample between consecutive measurements.
Anisotropy measurements can provide evidence of fluorescence resonance energy transfer between chemically identical
fluorophores (homo-FRET). This phenomenon is not detectable in spectral or lifetime changes, yet a lowering of the
anisotropy and a faster anisotropy decay can provide evidence for close proximity (≤ 10 nm) of adjacent fluorophores
including dimerization and oligomerization of molecules. We have used FLIM and fluorescence anisotropy to measure
variations in fluorescence lifetimes and anisotropy of GFP-tagged proteins in cells in immunological synapse samples
and also acquire images of BODIPY-stained carcinoma cells.
Phosphorylation of the chromatin protein H2AX (forming γH2AX) is implicated in the repair of DNA double strand
breaks (DSB's); a large number of H2AX molecules become phosphorylated at the sites of DSB's. Fluorescent staining
of the cell nuclei for γH2AX, via an antibody, visualises the formation of these foci, allowing the quantification of DNA
DSB's and forming the basis for a sensitive biological dosimeter of ionising radiation.
We describe an automated fluorescence microscopy system, including automated image processing, to count γH2AX
foci. The image processing is performed by a Hough transform based algorithm, CHARM, which has wide applicability
for the detection and analysis of cells and cell colonies. This algorithm and its applications for cell nucleus and foci
detection will be described. The system also relies heavily on robust control software, written using multi-threaded cbased
modules in LabWindows/CVI that adapt to the timing requirements of a particular experiment for optimised
slide/plate scanning and mosaicing, making use of modern multi-core processors. The system forms the basis of a
general purpose high-content screening platform with wide ranging applications in live and fixed cell imaging and tissue
micro arrays, that in future, can incorporate spectrally and time-resolved information.
We present details of the development of a optical biochip, with integrated on-chip laser excitation, for fluorescence
intensity cell based assays. The biochip incorporates an "active surface" for the control and manipulation of fluorescent
species placed directly on the device. The active elements of the biochip are one-dimensional periodic sub-wavelength
corrugations fabricated on a thin gold film. We have made fluorescence intensity measurements of both an organic dye
(Cy5), and immobilized and fluorescently labeled (with 705 nm emitting quantum dots), mammalian tumor cells in
contact with the active surface. Here we show that the presence of the periodic grating can be used to control both the
excitation and fluorescence generation process itself. We demonstrate that the gratings convert evanescent surface optical
modes into well-defined beams of radiation in the far-field and at the surface of the device this produces highly
contrasting regions of fluorescence excitation providing regions of high spatial selectivity.
Group velocity dispersion (GVD) and pulse front distortion of ultrashort pulses are of critical importance in
efficient multiphoton excitation microscopy. Since measurement of the pulse front distortion due to a lens is not trivial we have developed an imaging interferometric cross-correlator which allows us to measure temporal delays and pulse-widths across the spatial profile of the beam. The instrument consists of a modified Michelson interferometer with a reference arm containing a voice-coil delay stage and an arm which contains the optics under test. The pulse replicas are recombined and incident on a 22×22 lenslet array. The beamlets are focused in a 0.5 mm thick BBO crystal (cut for Type I second harmonic generation), filtered to remove the IR component of the beam and imaged using a 500 fps camera. The GVD and pulse front distortion are extracted from the temporal stack of beamlet images to produce a low resolution spatio-temporal map.
FLIM/FRET is an extremely powerful technique that can microscopically locate nanometre-scale protein-protein interactions within live or fixed cells, both <i>in vitro</i> and <i>in vivo</i>. The key to performing sensitive FRET, via FLIM, besides the use of appropriate fluorophores, is the analysis of the time-resolved data present at each image pixel. The fluorescent transient will, in general, exhibit multi-exponential kinetics: at least two exponential components arise from both the interacting and non-interacting protein. We shall describe a novel method and computer program for the global analysis of time resolved data, either at the single level or through global analysis of grouped pixel data. Kinetic models are fitted using the Marquardt algorithm and iterative convolution of the excitation signal, in a computationally-efficient manner. The fitting accuracy and sensitivity of the algorithm has been tested using modelled data, including the addition of simulated Poisson noise and repetitive excitation pulses which are typical of a TCSPC system. We found that the increased signal to noise ratio offered by both global and invariance fitting is highly desirable. When fitting mono-exponential data, the effects of a <i>ca</i>. 12.5 ns (<i>ca</i>. 80 MHz) repetitive excitation do not preclude the accurate extraction of populations with lifetimes in the range 0.1 to 10 ns, even when these effects are not represented in the fitting algorithm. Indeed, with global or invariance fitting of a 32x32 pixel area, the error in extracted lifetime can be lower than 0.4% for signals with a peak of 500 photon counts or more. In FRET simulations, modelling GFP with a non-interacting lifetime of 2.15 ns, it was possible to accurately detect a 10% interacting population with a lifetime of 0.8 ns.
The spatio-temporal localization of molecular interactions within cells in situ is of great importance in elucidating the key mechanisms in regulation of fundamental process within the cell. Measurements of such near-field localization of protein complexes may be achieved by the detection of fluorescence (or Forster) resonance energy transfer (FRET) between protein-conjugated fluorophores. We demonstrate the applicability of time-correlated single photon counting multiphoton microscopy to the spatio-temporal localization of protein-protein interactions in live and fixed cell populations. Intramolecular interactions between protein hetero-dimers are investigated using green fluorescent protein variants. We present an improved monomeric form of the red fluorescent protein, mRFP1, as the acceptor in biological fluorescence resonance energy transfer (FRET) experiments using the enhanced green fluorescent protein as donor. We find particular advantage in using this fluorophore pair for quantitative measurements of FRET. The technique was exploited to demonstrate a novel receptor-kinase interaction between the chemokine receptor (CXCR4) and protein kinase C (PKC) α in carcinoma cells for both live and fixed cell experiments.
In many clinical studies, including those of cancer, it is highly desirable to acquire images of whole tumour sections whilst retaining a microscopic resolution. A usual approach to this is to create a composite image by appropriately overlapping individual images acquired at high magnification under a microscope. A mosaic of these images can be accurately formed by applying image registration, overlap removal and blending techniques. We describe an optimised, automated, fast and reliable method for both image joining and blending. These algorithms can be applied to most types of light microscopy imaging. Examples from histology, from in vivo vascular imaging and from fluorescence applications are shown, both in 2D and 3D. The algorithms are robust to the varying image overlap of a manually moved stage, though examples of composite images acquired both with manually-driven and computer-controlled stages are presented. The overlap-removal algorithm is based on the cross-correlation method; this is used to determine and select the best correlation point between any new image and the previous composite image. A complementary image blending algorithm, based on a gradient method, is used to eliminate sharp intensity changes at the image joins, thus gradually blending one image onto the adjacent 'composite'. The details of the algorithm to overcome both intensity discrepancies and geometric misalignments between the stitched images will be presented and illustrated with several examples.
We describe how spectral imaging, linear un-mixing and cluster computing have been combined to aid biomedical researchers and allow the spatial segmentation and quantitative analysis of immunohistochemically stained tissue section images. A novel cost-effective spectral imager, with a bandwidth of 15 nm between 400 and 700 nm, allows us to record both spatial and spectral data from absorptive and fluorescent chemical probes. The linear un-mixing of this data separates the stain distributions revealing areas of co-localisation and extracts quantitative values of optical density. This has been achieved at the single-pixel level of an image by non-negative least squares fitting. This process can be computationally expensive but great processing speed increases have been achieved through the use of cluster computing. We describe how several personal computers, running Microsoft WindowsXP, can be used in parallel, linked by the MPI (Message Passing Interface) standard. We describe how the free MPICH libraries have been incorporated into our spectral imaging application under the C language and how this has been extended to support features of MPI2 via the commercial WMPI II libraries. A cluster of 8 processors, in 4 dual-Athlon-2600+ computers, offered a speed up of a factor of 5 compared to a singleton. This includes the time required to transfer the data throughout the cluster and reflects a processing efficiency of 0.62 (a Cluster Efficacy of 3.0). The cluster was based on a 1000Base-T Ethernet network and appears to be scalable efficiently beyond 8 processors.
We demonstrate the applicability of time-correlated single photon counting multiphoton microscopy to the spatio-temporal localisation of protein-protein interactions in situ. Examples of new fluorescent protein variants with enhanced properties are given and the development of FRET biosensors for simultaneous measurement of multiple intra- and inter-molecular interactions is illustrated by experimental evidence of an energy transfer cascade via multiple acceptors. The juxtaposition of interacting population and FRET efficiency is elucidated, with <i>a priori</i> knowledge, by multi-exponential analysis.
The understanding of tumour angiogenesis and response to vascular-targeted drugs are of increasing interest in cancer research. We present 3D images of the <i>in vivo</i> tumour vasculature captured utilising multi-photon microscopy together with the results of manual and semi-automated delineation of the vascular network using novel in-house-developed software and algorithms. The software presented is aimed at aiding in these investigations and other problems where linear or dendritic structures are to be delineated from 3D data sets. A new algorithm, CHARM, based on a compact Hough transform and the formation of a radial map, has been used to automatically locate vessel centres and measure diameters. The robustness of this algorithm to image smoothing and noise has been investigated. Statistical information characterising the network in terms of vascular parameters as well as more complex analyses, such as fractal dimension, are now possible and examples are presented.
Recent interest in vascular targeting and anti-angiogenic drug treatments for cancer has stimulated fundamental research regarding the modes of action of these drugs as well as studies of the development and re-modeling of the vascular network following treatment. Multiphoton fluorescence microscopy is employed for in vivo mapping of three-dimensional blood vessel distribution in tumors grown in rodent dorsal skin-flap window chamber preparations. Accurate visualization of the vasculature in three-dimensions allows us to perform dynamic experiments in thick biological specimens in vivo. Examples of in vivo imaging of tumor vasculature are given and compared to normal tissue vasculature. The dynamic responses of blood vessels to treatment with the vascular targeting drug combretastatin A4-P are presented and discussed. The implementation of time-domain imaging by reversed stop-start time-correlated single photon counting (RSS-TCSPC) is discussed as a method for feature extraction in the presence of exogenous and endogenous fluorophores. In particular, the segmentation of the vascular network is demonstrated. Additional contrast, indicative of probe environmental factors, may also be realized. We present examples of in vivo lifetime imaging as a method to elucidate the physiological processes of the tumor microenvironment.