Academia.eduAcademia.edu

Figure 24 – uploaded by Wolfgang Becker

See full PDFdownloadDownload figure
Figure 25 TCSPC lifetime images of the ocular fundus of a patient with wet AMD. (A) Amplitude-weighted lifetimes, tm. (B—D Relative intensity contributions of the fast, medium, and slow fluorescence components, A;, Az and A3. Autofluorescence wa: excited at 445 nm and collected in the wavelength interval from 490 nm to 560 nm (Data courtesy of Dietrich Schweitzer, Friedrict Schiller University Jena, Germany). Fig. 25 shows the ocular fundus of a patient with an age- related wet macula degeneration (AMD). The decay data in the pixels of the image were analysed by fitting a triple- exponential decay model. Fig. 25A shows a lifetime image of the amplitude-weighted average lifetime, tm. Fig. 25B—D shows the relative contributions, A;, Az, and A3 of the fast, medium, and slow fluorescence components to the total intensity. The images show massive changes in the In the last decade, speed and memory size of comput- ers has increased by more than an order of magnitude. As a result, the raw data (e.g. data for single photons) can be transferred directly to the computer, so that the FLIM histogram is built up in the computer memory. With the introduction of 64-bit operating systems, memory size is no longer a problem. In a 64-bit environment, FLIM data can be recorded at mega-pixel resolution (Studier et al., 2014).

Figure 25 TCSPC lifetime images of the ocular fundus of a patient with wet AMD. (A) Amplitude-weighted lifetimes, tm. (B—D Relative intensity contributions of the fast, medium, and slow fluorescence components, A;, Az and A3. Autofluorescence wa: excited at 445 nm and collected in the wavelength interval from 490 nm to 560 nm (Data courtesy of Dietrich Schweitzer, Friedrict Schiller University Jena, Germany). Fig. 25 shows the ocular fundus of a patient with an age- related wet macula degeneration (AMD). The decay data in the pixels of the image were analysed by fitting a triple- exponential decay model. Fig. 25A shows a lifetime image of the amplitude-weighted average lifetime, tm. Fig. 25B—D shows the relative contributions, A;, Az, and A3 of the fast, medium, and slow fluorescence components to the total intensity. The images show massive changes in the In the last decade, speed and memory size of comput- ers has increased by more than an order of magnitude. As a result, the raw data (e.g. data for single photons) can be transferred directly to the computer, so that the FLIM histogram is built up in the computer memory. With the introduction of 64-bit operating systems, memory size is no longer a problem. In a 64-bit environment, FLIM data can be recorded at mega-pixel resolution (Studier et al., 2014).

Related Figures (24)

Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting Figure 1 Principle of classic time-correlated single photon counting. The sample is excited by a pulsed laser source with a high repetition rate. Photons emitted by the sample are detected with a high-gain photomultiplier and the time with respect to the excitation pulse is measured. This information is used to build up a one-dimensional photon distribution over the time in the signal period (Figure reproduced from: Becker, 2014, p 69).
Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting Figure 1 Principle of classic time-correlated single photon counting. The sample is excited by a pulsed laser source with a high repetition rate. Photons emitted by the sample are detected with a high-gain photomultiplier and the time with respect to the excitation pulse is measured. This information is used to build up a one-dimensional photon distribution over the time in the signal period (Figure reproduced from: Becker, 2014, p 69).
Figure 2 Principle of multi-dimensional TCSPC. For every detected photon several parameters are determined, and a multi- dimensional photon distribution is built up (Figure reproduced from: Becker, 2014, p 69).
Figure 2 Principle of multi-dimensional TCSPC. For every detected photon several parameters are determined, and a multi- dimensional photon distribution is built up (Figure reproduced from: Becker, 2014, p 69).
Figure 3. TCSPC architecture for FLIM. Single photon events are accumulated in the memory of the TCSPC card with respect to their arrival time (t) and their spatial coordinates (x, y) which are calculated from the scan clock signals of the laser scanning microscope (Figure reproduced from: Becker, 2014, p 81).
Figure 3. TCSPC architecture for FLIM. Single photon events are accumulated in the memory of the TCSPC card with respect to their arrival time (t) and their spatial coordinates (x, y) which are calculated from the scan clock signals of the laser scanning microscope (Figure reproduced from: Becker, 2014, p 81).
Figure 4 Time-domain analysis of TCSPC FLIM data. Left: Raw data, each pixel contains a decay function in a large number of time channels. Middle: Decay curve in a selected pixel. The model function is convoluted with the instrument response function (IRF, green curve), and the result (red curve) fits to the decay data (blue). Right: FLIM image. A selected decay parameter is used as colour, the photon number as intensity. Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 4 Time-domain analysis of TCSPC FLIM data. Left: Raw data, each pixel contains a decay function in a large number of time channels. Middle: Decay curve in a selected pixel. The model function is convoluted with the instrument response function (IRF, green curve), and the result (red curve) fits to the decay data (blue). Right: FLIM image. A selected decay parameter is used as colour, the photon number as intensity. Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 5 —Eight-channel parallel FLIM of a Drosophila eye. Autofluorescence is excited at 407nm and collected in a wavelength range from 480nm to 620nm. Images are obtained with the bh DCS-120 confocal scanning system. (A) Fluorescence lifetime image encoding the amplitude weighted mean lifetime recovered from a fit with a double-exponential decay model. (B) Fluorescence lifetime image encoding the relative amplitude of the fast decay component (Figure reproduced from Becker et al., 2014, p. 320).
Figure 5 —Eight-channel parallel FLIM of a Drosophila eye. Autofluorescence is excited at 407nm and collected in a wavelength range from 480nm to 620nm. Images are obtained with the bh DCS-120 confocal scanning system. (A) Fluorescence lifetime image encoding the amplitude weighted mean lifetime recovered from a fit with a double-exponential decay model. (B) Fluorescence lifetime image encoding the relative amplitude of the fast decay component (Figure reproduced from Becker et al., 2014, p. 320).
Figure 6 Principle of multi-wavelength TCSPC FLIM. The number of the channel, in which the photon was detected, is used to route the photon events into different data blocks of the memory (Figure adapted from Becker, 2014, p. 83).
Figure 6 Principle of multi-wavelength TCSPC FLIM. The number of the channel, in which the photon was detected, is used to route the photon events into different data blocks of the memory (Figure adapted from Becker, 2014, p. 83).
Figure 7 Multi-wavelength FLIM of plant tissue obtained with a Zeiss LSM 710 and a bh Simple-Tau 150 FLIM system in 16 wavelength intervals, 128 x 128 pixels, and 256 time channels, respectively. Fluorophores were exited by two-photon excitation at 850nm, fluorescence was detected in the indicated wavelength intervals. Amplitude weighted lifetimes of the double-exponential fit are encoded according to the colour scale at the bottom. As for single-wavelength FLIM, the result of the recording process is an array of pixels. The pixels of multi-wavelength FLIM data contain several decay curves for different wave- length. Each decay curve contains a large number of time channels; the time channels contain photon numbers for consecutive times after the excitation pulse. A result of multi-wavelength FLIM is shown in Fig. 7. (For other appli- cations please see Becker et al., 2007; Bird et al., 2004;
Figure 7 Multi-wavelength FLIM of plant tissue obtained with a Zeiss LSM 710 and a bh Simple-Tau 150 FLIM system in 16 wavelength intervals, 128 x 128 pixels, and 256 time channels, respectively. Fluorophores were exited by two-photon excitation at 850nm, fluorescence was detected in the indicated wavelength intervals. Amplitude weighted lifetimes of the double-exponential fit are encoded according to the colour scale at the bottom. As for single-wavelength FLIM, the result of the recording process is an array of pixels. The pixels of multi-wavelength FLIM data contain several decay curves for different wave- length. Each decay curve contains a large number of time channels; the time channels contain photon numbers for consecutive times after the excitation pulse. A result of multi-wavelength FLIM is shown in Fig. 7. (For other appli- cations please see Becker et al., 2007; Bird et al., 2004;
Figure 8 Excitation wavelength multiplexing and dual-wavelength detection. BPAE cell labelled with Alexa 488 phalloidin and Mito Tracker Red. By frame-by-frame multiplexing, fluorescence was exited at 488nm (A,B) and 579nm (C,D). Fluorescence was collected in the wavelength range of 500—550nm (A,C) and 590—650 nm (B,D). Fluorescence lifetime images consist of 256 x 256 pixels with 256 time channels each. Amplitude weighted lifetimes of a double-exponential decay model are encoded according to the colour scale at the right.
Figure 8 Excitation wavelength multiplexing and dual-wavelength detection. BPAE cell labelled with Alexa 488 phalloidin and Mito Tracker Red. By frame-by-frame multiplexing, fluorescence was exited at 488nm (A,B) and 579nm (C,D). Fluorescence was collected in the wavelength range of 500—550nm (A,C) and 590—650 nm (B,D). Fluorescence lifetime images consist of 256 x 256 pixels with 256 time channels each. Amplitude weighted lifetimes of a double-exponential decay model are encoded according to the colour scale at the right.
Figure 9 FLIMz-stack (18 planes, 256 x 256 pixels, 256 time channels) recorded from a pig skin sample with a Zeiss LSM 710 NLO and bh Simple-Tau 152 FLIM system. Autofluorescence was excited by 2-photon excitation at 800 nm and detected in the wavelength range from 400 to 500 nm. The scan area was 212 x 212 um. Images were obtained from 5 1m to 90 um below the top of the sample. Acquisition time was 20s per plane.
Figure 9 FLIMz-stack (18 planes, 256 x 256 pixels, 256 time channels) recorded from a pig skin sample with a Zeiss LSM 710 NLO and bh Simple-Tau 152 FLIM system. Autofluorescence was excited by 2-photon excitation at 800 nm and detected in the wavelength range from 400 to 500 nm. The scan area was 212 x 212 um. Images were obtained from 5 1m to 90 um below the top of the sample. Acquisition time was 20s per plane.
Figure 10 Time series of the non-photochemical chlorophyll transient in the chloroplasts of a moss leaf. Recorded by a Becker & Hickl DCS—120 system, 2s per image.
Figure 10 Time series of the non-photochemical chlorophyll transient in the chloroplasts of a moss leaf. Recorded by a Becker & Hickl DCS—120 system, 2s per image.
Figure 11 Time series recorded by mosaic imaging. 64 mosaic elements (image size of each element 128 x 128 pixels, 256 time channels) are obtained for consecutive times after turning-on the excitation light. Acquisition time per element was 1 second, total time for the sequence was 64s. Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 11 Time series recorded by mosaic imaging. 64 mosaic elements (image size of each element 128 x 128 pixels, 256 time channels) are obtained for consecutive times after turning-on the excitation light. Acquisition time per element was 1 second, total time for the sequence was 64s. Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 12 Principle of Microsecond FLIM. To record phospho- rescence lifetimes, the laser is turned off for the rest of the pixel time. Within the on-time the laser excites fluorescence, and builds up phosphorescence. Within the laser-off time the phosphorescence signal is observed (Figure reproduced from Becker, 2014, p.332).
Figure 12 Principle of Microsecond FLIM. To record phospho- rescence lifetimes, the laser is turned off for the rest of the pixel time. Within the on-time the laser excites fluorescence, and builds up phosphorescence. Within the laser-off time the phosphorescence signal is observed (Figure reproduced from Becker, 2014, p.332).
Figure 13 Fluorescence lifetime image (A) and phosphorescence lifetime image (B) of yeast cells stained with tris(2,2’- bipyridyl)dichlororuthenium(Il)hexahydrate. Amplitude weighted lifetimes of a double-exponential decay model are encoded according to the colour scale at the bottom of the images. An example is shown in Fig. 13. It shows yeast cells stained with tris (2,2’-bipyridyl) dichlororuthenium (II) hexa- hydrate. On the ps time scale, the yeast cells emit autofluorescence from NADH and FAD. On the ys time scale, phosphorescence of the ruthenium dye is emitted. The fluorescence lifetime image is shown on the left, the phos- phorescence lifetime image on the right. Fig. 14 shows decay curves for a selected spot in the images. The dots are the photon numbers in the subsequent time channels, the solid curve is a fit with a double-exponential decay model, and the grey curve is the effective instrument response function IRF). Please note that the IRF of the fluorescence decay is the laser pulse convoluted with the detector response, The principle is shown in Fig. 12. A high-frequency-pulsed laser is used for excitation. However, the laser is not puls- ing continuously as for normal FLIM. Instead, the laser is turned on for a short period of time, T,,, at the beginning of each pixel, and then turned off for the rest of the pixel time. Within the on-time the laser excites fluorescence, and builds up phosphorescence. Within the laser-off time the phospho- rescence signal is observed. Photon times are determined both with respect to the laser pulses and with respect to the modulation pulse. The times from the laser pulse, t, are used
Figure 13 Fluorescence lifetime image (A) and phosphorescence lifetime image (B) of yeast cells stained with tris(2,2’- bipyridyl)dichlororuthenium(Il)hexahydrate. Amplitude weighted lifetimes of a double-exponential decay model are encoded according to the colour scale at the bottom of the images. An example is shown in Fig. 13. It shows yeast cells stained with tris (2,2’-bipyridyl) dichlororuthenium (II) hexa- hydrate. On the ps time scale, the yeast cells emit autofluorescence from NADH and FAD. On the ys time scale, phosphorescence of the ruthenium dye is emitted. The fluorescence lifetime image is shown on the left, the phos- phorescence lifetime image on the right. Fig. 14 shows decay curves for a selected spot in the images. The dots are the photon numbers in the subsequent time channels, the solid curve is a fit with a double-exponential decay model, and the grey curve is the effective instrument response function IRF). Please note that the IRF of the fluorescence decay is the laser pulse convoluted with the detector response, The principle is shown in Fig. 12. A high-frequency-pulsed laser is used for excitation. However, the laser is not puls- ing continuously as for normal FLIM. Instead, the laser is turned on for a short period of time, T,,, at the beginning of each pixel, and then turned off for the rest of the pixel time. Within the on-time the laser excites fluorescence, and builds up phosphorescence. Within the laser-off time the phospho- rescence signal is observed. Photon times are determined both with respect to the laser pulses and with respect to the modulation pulse. The times from the laser pulse, t, are used
Figure 15 Pig skin sample stained with DTTCC. Measurements were done with a Zeiss LSM 710 NLO and a Becker & Hickl Simple- Tau 152 FLIM system with a HPM—100-50 hybrid detector. A Ti:Sa laser was used for one-photon excitation at 780nm; fluores- cence was collected in the wavelength range from 800nm to 900 nm. The lateral size of the image is 212 x 212 wm, the depth from surface is about 30 um. Near-infrared FLIM systems. There is an increasing inter- est in performing FLIM at near-infrared wavelengths. In the NIR, the emission of exogenous fluorophores can be detected without contamination from autofluorescence. Moreover, the fluorescence decay signature of NIR fluo- rophores is important to diffuse optical imaging techniques. For these techniques, the fluorescence decay parameters and their dependence on local environment parameters must be known to determine biologically relevant tissue properties. NIR FLIM can be performed by using the Ti:Sa laser of a multiphoton microscope as a one-photon excita- tion source. Of course, the scanning microscope must have a confocal detection path, and a confocal output to a FLIM detector (see Becker and Shcheslavskiy, 2013 for details). An example of NIR FLIM is shown in Fig. 15.
Figure 15 Pig skin sample stained with DTTCC. Measurements were done with a Zeiss LSM 710 NLO and a Becker & Hickl Simple- Tau 152 FLIM system with a HPM—100-50 hybrid detector. A Ti:Sa laser was used for one-photon excitation at 780nm; fluores- cence was collected in the wavelength range from 800nm to 900 nm. The lateral size of the image is 212 x 212 wm, the depth from surface is about 30 um. Near-infrared FLIM systems. There is an increasing inter- est in performing FLIM at near-infrared wavelengths. In the NIR, the emission of exogenous fluorophores can be detected without contamination from autofluorescence. Moreover, the fluorescence decay signature of NIR fluo- rophores is important to diffuse optical imaging techniques. For these techniques, the fluorescence decay parameters and their dependence on local environment parameters must be known to determine biologically relevant tissue properties. NIR FLIM can be performed by using the Ti:Sa laser of a multiphoton microscope as a one-photon excita- tion source. Of course, the scanning microscope must have a confocal detection path, and a confocal output to a FLIM detector (see Becker and Shcheslavskiy, 2013 for details). An example of NIR FLIM is shown in Fig. 15.
Figure 14 Decay curves in a selected spot of Fig. 13. (A) Fluorescence decay. (B) Phosphorescence decay. The dots correspond t the photon numbers recorded in the respective time channels, the solid curve is a fit with double-exponential decay model, and th grey curve is the effective instrument response function. Amplitude-weighted lifetime and double-exponential decay parameter are indicated in ps for fluorescence and in ns for phosphorescence, respectively.
Figure 14 Decay curves in a selected spot of Fig. 13. (A) Fluorescence decay. (B) Phosphorescence decay. The dots correspond t the photon numbers recorded in the respective time channels, the solid curve is a fit with double-exponential decay model, and th grey curve is the effective instrument response function. Amplitude-weighted lifetime and double-exponential decay parameter are indicated in ps for fluorescence and in ns for phosphorescence, respectively.
Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting Figure 16 Cl~ measurement in the spinal ganglion of a mouse by quenching of MQAE. The data were analysed with a double- exponential incomplete decay model. Decay curves in selected spots are shown on the right (Data courtesy of Thomas Gensch, Julich Research Centre, Germany).
Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting Figure 16 Cl~ measurement in the spinal ganglion of a mouse by quenching of MQAE. The data were analysed with a double- exponential incomplete decay model. Decay curves in selected spots are shown on the right (Data courtesy of Thomas Gensch, Julich Research Centre, Germany).
Figure 17 Lifetime image of skin tissue stained with BCECF. The lifetime corresponds to the pH. Images were acquired by a Zei: LSM 510 NLO with a bh SPC—830 TCSPC FLIM module (Courtesy by Theodora Mauro, University of San Francisco).
Figure 17 Lifetime image of skin tissue stained with BCECF. The lifetime corresponds to the pH. Images were acquired by a Zei: LSM 510 NLO with a bh SPC—830 TCSPC FLIM module (Courtesy by Theodora Mauro, University of San Francisco).
Figure 18 Barley root tip loaded with Oregon green. Data were acquired with a Leica SP5 MP with a bh SPC-150 FLIM module (Courtesy of Feifei Wang, Zhonghua Chen and Anya Salih, Confocal Bioimaging Facility, University of Western Sydney, Australia).
Figure 18 Barley root tip loaded with Oregon green. Data were acquired with a Leica SP5 MP with a bh SPC-150 FLIM module (Courtesy of Feifei Wang, Zhonghua Chen and Anya Salih, Confocal Bioimaging Facility, University of Western Sydney, Australia).
Figure 19 Cell containing GFP fusion proteins (donor) and Cy3-labelled antibodies (acceptor). Donor fluorescence lifetimes have been recorded with a Zeiss LSM 710 microscope and a Becker & Hickl Simple-Tau 152 FLIM system. Donor fluorescence decays have been analysed with a double-exponential decay mode, and amplitude-weighted mean lifetimes have been encoded according to the colour bar.
Figure 19 Cell containing GFP fusion proteins (donor) and Cy3-labelled antibodies (acceptor). Donor fluorescence lifetimes have been recorded with a Zeiss LSM 710 microscope and a Becker & Hickl Simple-Tau 152 FLIM system. Donor fluorescence decays have been analysed with a double-exponential decay mode, and amplitude-weighted mean lifetimes have been encoded according to the colour bar.
Figure 20 HEK cell expressing a CFP and a YFP fusion protein. Double-exponential FRET analysis. (A) Amplitude weighted mea lifetimes are encoded by colour. (B) Ratio of amplitudes, A;/A2, of the fast and slow decay component. (C) Ratio of the fast (z, and the slow (r2) decay component. Data have been acquired by a Zeiss LSM 510 NLO laser scanning microscope with a bh SPC-83( FLIM module as described in Becker et al., 2004b. In this figure, the data have been reanalysed with an incomplete decay model.
Figure 20 HEK cell expressing a CFP and a YFP fusion protein. Double-exponential FRET analysis. (A) Amplitude weighted mea lifetimes are encoded by colour. (B) Ratio of amplitudes, A;/A2, of the fast and slow decay component. (C) Ratio of the fast (z, and the slow (r2) decay component. Data have been acquired by a Zeiss LSM 510 NLO laser scanning microscope with a bh SPC-83( FLIM module as described in Becker et al., 2004b. In this figure, the data have been reanalysed with an incomplete decay model.
Figure 21 Fluorescence lifetime imaging of human salivary gland stem cells. (A) The amplitude weighted mean lifetime tm is encoded by colour according to the colour bar. (B) The ratio A1/A2 of the relative amplitudes is used to calculate the colour. Differentiated cells have a significantly longer mean lifetime and a lower A;/A 2 ratio (Data courtesy of Aisada Uchugonova and Karsten Konig, Saarland University, Saarbriicken). Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 21 Fluorescence lifetime imaging of human salivary gland stem cells. (A) The amplitude weighted mean lifetime tm is encoded by colour according to the colour bar. (B) The ratio A1/A2 of the relative amplitudes is used to calculate the colour. Differentiated cells have a significantly longer mean lifetime and a lower A;/A 2 ratio (Data courtesy of Aisada Uchugonova and Karsten Konig, Saarland University, Saarbriicken). Fluorescence lifetime imaging by multi-dimensional time correlated single photon counting
Figure 22 Two-photon FLIM of pig skin. Images have been recorded with a Zeiss LSM 710 NLO and a bh Simple-Tau 152 FLIM system. The sample was excited by two-photon excitation, fluorescence was collected by non-descanned detectors. (A) Channel for wavelengths <480 nm, colour shows percentage of SHG in the recorded signal. (B) Channel for wavelength > 480nm, colour shows amplitude-weighted mean lifetime.
Figure 22 Two-photon FLIM of pig skin. Images have been recorded with a Zeiss LSM 710 NLO and a bh Simple-Tau 152 FLIM system. The sample was excited by two-photon excitation, fluorescence was collected by non-descanned detectors. (A) Channel for wavelengths <480 nm, colour shows percentage of SHG in the recorded signal. (B) Channel for wavelength > 480nm, colour shows amplitude-weighted mean lifetime.
Figure 23 FLIM image of human skin recorded by a Jenlab Dermainspect Multiphoton Tomograph and a Becker & Hickl SPC- 150 TCSPC FLIM system. Two-photon excitation at 750nm (Data courtesy of Michael Roberts and Washington Sanchez, Brisbane, Australia).
Figure 23 FLIM image of human skin recorded by a Jenlab Dermainspect Multiphoton Tomograph and a Becker & Hickl SPC- 150 TCSPC FLIM system. Two-photon excitation at 750nm (Data courtesy of Michael Roberts and Washington Sanchez, Brisbane, Australia).
Figure 24 LIM of the fundus of the author’s left eye. (A) Fluorescence intensity image. (B) FLIM image. Fluorescence decay curves have been approximated by an bi-exponential model. Amplitude weighted mean lifetimes are shown.
Figure 24 LIM of the fundus of the author’s left eye. (A) Fluorescence intensity image. (B) FLIM image. Fluorescence decay curves have been approximated by an bi-exponential model. Amplitude weighted mean lifetimes are shown.
Related topics:
PhysicsFRET- FLIMFlim StudiesFluorescence life time imaging (FLIM)
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved