Limits...
Monitoring photosensitizer uptake using two photon fluorescence lifetime imaging microscopy.

Yeh SC, Diamond KR, Patterson MS, Nie Z, Hayward JE, Fang Q - Theranostics (2012)

Bottom Line: Fluorescence emission in PDT may be used to monitor the uptake process but fluorescence intensity is subject to variability due to scattering and absorption; the addition of fluorescence lifetime may be beneficial to probe site-specific drug-molecular interactions and cell damage.The fluorescence decays were analyzed using a bi-exponential model, followed by segmentation analysis of lifetime parameters.When Photofrin(®) was localized at the cell membrane, the slow lifetime component was found to be significantly shorter (4.3 ± 0.5 ns) compared to those at other locations (cytoplasm: 7.3 ± 0.3 ns; mitochondria: 7.0 ± 0.2 ns, p < 0.05).

View Article: PubMed Central - PubMed

Affiliation: 1. School of Biomedical Engineering, McMaster University, 1280 Main Street West, Hamilton, Ontario, L8S 4K1, Canada;

ABSTRACT
Photodynamic Therapy (PDT) provides an opportunity for treatment of various invasive tumors by the use of a cancer targeting photosensitizing agent and light of specific wavelengths. However, real-time monitoring of drug localization is desirable because the induction of the phototoxic effect relies on interplay between the dosage of localized drug and light. Fluorescence emission in PDT may be used to monitor the uptake process but fluorescence intensity is subject to variability due to scattering and absorption; the addition of fluorescence lifetime may be beneficial to probe site-specific drug-molecular interactions and cell damage. We investigated the fluorescence lifetime changes of Photofrin(®) at various intracellular components in the Mat-LyLu (MLL) cell line. The fluorescence decays were analyzed using a bi-exponential model, followed by segmentation analysis of lifetime parameters. When Photofrin(®) was localized at the cell membrane, the slow lifetime component was found to be significantly shorter (4.3 ± 0.5 ns) compared to those at other locations (cytoplasm: 7.3 ± 0.3 ns; mitochondria: 7.0 ± 0.2 ns, p < 0.05).

No MeSH data available.


Related in: MedlinePlus

(A) A sample confocal image showing the distribution of mitochondria. MLL cells were stained with 100 nM of MitoTracker® Green, as indicated by the green color. (B) Histogram (photon count per pixel) used for segmentation of FLIM images. Based on the morphological features shown in confocal images, the main peak corresponds to the lower pixel intensities from the cell nuclei, and the minor peak corresponds to the intensities from the cytoplasmic region, as indicated by the arrows and the range covered by the dashed lines. (C) Original intensity image of a single cell. The cell was stained with Photofrin® for 4.5 hours, when Photofrin® presented at both mitochondrial and cytoplasm regions. Note the image was from a small area of a field of view, therefore the image seems pixelated. The thresholds used to segment the cytoplasmic region were marked by the black arrows on the color bar (D) The segmentation was performed based on the morphological features and histogram, where signals with intensity > 60 counts/pixel (mitochondria) and < 15 counts/ pixel (cell nucleus and background) were eliminated.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3475212&req=5

Figure 2: (A) A sample confocal image showing the distribution of mitochondria. MLL cells were stained with 100 nM of MitoTracker® Green, as indicated by the green color. (B) Histogram (photon count per pixel) used for segmentation of FLIM images. Based on the morphological features shown in confocal images, the main peak corresponds to the lower pixel intensities from the cell nuclei, and the minor peak corresponds to the intensities from the cytoplasmic region, as indicated by the arrows and the range covered by the dashed lines. (C) Original intensity image of a single cell. The cell was stained with Photofrin® for 4.5 hours, when Photofrin® presented at both mitochondrial and cytoplasm regions. Note the image was from a small area of a field of view, therefore the image seems pixelated. The thresholds used to segment the cytoplasmic region were marked by the black arrows on the color bar (D) The segmentation was performed based on the morphological features and histogram, where signals with intensity > 60 counts/pixel (mitochondria) and < 15 counts/ pixel (cell nucleus and background) were eliminated.

Mentions: Time-lapse FLIM data were fitted with a bi-exponential model using the vendor supplied software (SPCImage, v3.0.8.0, Becker and Hickl). The sample measured decay trace and its fitting curve from a single pixel were demonstrated in Figure 1. The retrieved individual lifetime and amplitude parameters of every pixel were then imported to Matlab® (Mathworks, Natick, MA) for futher analysis using a program developed in-house. Three repeated trials for each incubation period were acquired, with each trial containing multiple regions of interest. From each lifetime image, the mean value and standard deviation of individual parameters such as the bi-exponential lifetime components (τ2, τ1), and corresponding coefficients (A2, A1) of every pixel in the image were plotted. Using these values, the weighted mean and standard deviation of three repetitive measurements were calculated. The regions of interest, for example, mitochondria or cytoplasm, were segmented based on intensity. The segmentation was only performed on images collected at 4 hours and 4.5 hours of incubation, where the Photofrin® localization was undergoing a “transition” stage, thus the FLIM images showed lifetime values from different locations. The classification of the intracellular components was first visually verified based on preliminary confocal images as shown in Figure 2A, where the cells were stained with MitoTracker® Green (FM 7514, Invitrogen) to reveal the mitochondrial distribution. The segmentation was then done based on thresholding the histogram of pixel intensity (photon counts/ pixel) in the intensity images. Intensity values ranging from 15 - 60 counts per pixel were classified as “cytoplasm”, and greater than 60 counts per pixel were classified as mitochondria. Sample confocal and segmentation images are shown in Figure 2C and 2D. As the signal of Photofrin® is extremely low and comparable with the autofluorescence at the beginning of incubation (less than 1 hour), the cell membrane was selected using the Matlab® built-in periphery detection function (bwperim) instead of segmentation.


Monitoring photosensitizer uptake using two photon fluorescence lifetime imaging microscopy.

Yeh SC, Diamond KR, Patterson MS, Nie Z, Hayward JE, Fang Q - Theranostics (2012)

(A) A sample confocal image showing the distribution of mitochondria. MLL cells were stained with 100 nM of MitoTracker® Green, as indicated by the green color. (B) Histogram (photon count per pixel) used for segmentation of FLIM images. Based on the morphological features shown in confocal images, the main peak corresponds to the lower pixel intensities from the cell nuclei, and the minor peak corresponds to the intensities from the cytoplasmic region, as indicated by the arrows and the range covered by the dashed lines. (C) Original intensity image of a single cell. The cell was stained with Photofrin® for 4.5 hours, when Photofrin® presented at both mitochondrial and cytoplasm regions. Note the image was from a small area of a field of view, therefore the image seems pixelated. The thresholds used to segment the cytoplasmic region were marked by the black arrows on the color bar (D) The segmentation was performed based on the morphological features and histogram, where signals with intensity > 60 counts/pixel (mitochondria) and < 15 counts/ pixel (cell nucleus and background) were eliminated.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3475212&req=5

Figure 2: (A) A sample confocal image showing the distribution of mitochondria. MLL cells were stained with 100 nM of MitoTracker® Green, as indicated by the green color. (B) Histogram (photon count per pixel) used for segmentation of FLIM images. Based on the morphological features shown in confocal images, the main peak corresponds to the lower pixel intensities from the cell nuclei, and the minor peak corresponds to the intensities from the cytoplasmic region, as indicated by the arrows and the range covered by the dashed lines. (C) Original intensity image of a single cell. The cell was stained with Photofrin® for 4.5 hours, when Photofrin® presented at both mitochondrial and cytoplasm regions. Note the image was from a small area of a field of view, therefore the image seems pixelated. The thresholds used to segment the cytoplasmic region were marked by the black arrows on the color bar (D) The segmentation was performed based on the morphological features and histogram, where signals with intensity > 60 counts/pixel (mitochondria) and < 15 counts/ pixel (cell nucleus and background) were eliminated.
Mentions: Time-lapse FLIM data were fitted with a bi-exponential model using the vendor supplied software (SPCImage, v3.0.8.0, Becker and Hickl). The sample measured decay trace and its fitting curve from a single pixel were demonstrated in Figure 1. The retrieved individual lifetime and amplitude parameters of every pixel were then imported to Matlab® (Mathworks, Natick, MA) for futher analysis using a program developed in-house. Three repeated trials for each incubation period were acquired, with each trial containing multiple regions of interest. From each lifetime image, the mean value and standard deviation of individual parameters such as the bi-exponential lifetime components (τ2, τ1), and corresponding coefficients (A2, A1) of every pixel in the image were plotted. Using these values, the weighted mean and standard deviation of three repetitive measurements were calculated. The regions of interest, for example, mitochondria or cytoplasm, were segmented based on intensity. The segmentation was only performed on images collected at 4 hours and 4.5 hours of incubation, where the Photofrin® localization was undergoing a “transition” stage, thus the FLIM images showed lifetime values from different locations. The classification of the intracellular components was first visually verified based on preliminary confocal images as shown in Figure 2A, where the cells were stained with MitoTracker® Green (FM 7514, Invitrogen) to reveal the mitochondrial distribution. The segmentation was then done based on thresholding the histogram of pixel intensity (photon counts/ pixel) in the intensity images. Intensity values ranging from 15 - 60 counts per pixel were classified as “cytoplasm”, and greater than 60 counts per pixel were classified as mitochondria. Sample confocal and segmentation images are shown in Figure 2C and 2D. As the signal of Photofrin® is extremely low and comparable with the autofluorescence at the beginning of incubation (less than 1 hour), the cell membrane was selected using the Matlab® built-in periphery detection function (bwperim) instead of segmentation.

Bottom Line: Fluorescence emission in PDT may be used to monitor the uptake process but fluorescence intensity is subject to variability due to scattering and absorption; the addition of fluorescence lifetime may be beneficial to probe site-specific drug-molecular interactions and cell damage.The fluorescence decays were analyzed using a bi-exponential model, followed by segmentation analysis of lifetime parameters.When Photofrin(®) was localized at the cell membrane, the slow lifetime component was found to be significantly shorter (4.3 ± 0.5 ns) compared to those at other locations (cytoplasm: 7.3 ± 0.3 ns; mitochondria: 7.0 ± 0.2 ns, p < 0.05).

View Article: PubMed Central - PubMed

Affiliation: 1. School of Biomedical Engineering, McMaster University, 1280 Main Street West, Hamilton, Ontario, L8S 4K1, Canada;

ABSTRACT
Photodynamic Therapy (PDT) provides an opportunity for treatment of various invasive tumors by the use of a cancer targeting photosensitizing agent and light of specific wavelengths. However, real-time monitoring of drug localization is desirable because the induction of the phototoxic effect relies on interplay between the dosage of localized drug and light. Fluorescence emission in PDT may be used to monitor the uptake process but fluorescence intensity is subject to variability due to scattering and absorption; the addition of fluorescence lifetime may be beneficial to probe site-specific drug-molecular interactions and cell damage. We investigated the fluorescence lifetime changes of Photofrin(®) at various intracellular components in the Mat-LyLu (MLL) cell line. The fluorescence decays were analyzed using a bi-exponential model, followed by segmentation analysis of lifetime parameters. When Photofrin(®) was localized at the cell membrane, the slow lifetime component was found to be significantly shorter (4.3 ± 0.5 ns) compared to those at other locations (cytoplasm: 7.3 ± 0.3 ns; mitochondria: 7.0 ± 0.2 ns, p < 0.05).

No MeSH data available.


Related in: MedlinePlus