We instrumented a combined fluorescence spectroscopy and imaging system to characterize the single- and two-photon excited autofluorescence in epithelial tissue. Single-photon fluorescence (SPF) are compared with two-photon fluorescence (TPF) measured at the same location in epithelial tissue. It was found that the SPF and TPF signals excited at corresponding wavelengths are similar in nonkeratinized epithelium, but the SPF and TPF spectra in the keratinized epithelium and the stromal layer are significant different. Specifically, the comparison of SPF signals with TPF signals in keratinized epithelial and stromal layers shows that TPF spectral peaks always have about 15-nm redshift with respect to SPF signals, and the TPF spectra are broader than SPF spectra. The results were generally consistent with the SPF and TPF measurements of pure nicotinamide adenine dinucleotide, flavin adenine dinucleotide, keratin and collagen, the major fluorophores in epithelium and stroma, respectively. The double peak structure of TPF spectra measured from keratinized layer suggests that there may be an unknown fluorophore responsible for the spectral peak in the long wavelength region. Furthermore, the TPF signals excited in a wide range of wavelengths provide accurate information on epithelial structure, which is an important advantage of TPF over SPF spectroscopy in the application for the diagnosis of tissue pathology.
Autofluorescence spectroscopy has been a widely explored noninvasive technique to detect the precancerous
development in epithelial tissue, where NADH and FAD fluorescence are metabolism related. In this study, we
investigated the methods to monitor cellular metabolism based on the ratio of NADH over FAD fluorescence and the
ratio of free NADH and protein-bound NADH fluorescence, respectively. The signals of free NADH, protein-bound
NADH and FAD were isolated from the intracellular autofluorescence using wavelength- and time-resolved fluorescence
spectroscopy. We demonstrated that the wavelength- and time-resolved intracellular autofluorescence can be used to
monitor the cellular metabolic pathways and differentiate the normal cells from the cancer cells.
A time-resolved confocal fluorescence spectroscopy system was instrumented utilizing the multi-channel time-correlated
single photon counting (TCSPC) technique. The system provided a unique approach to investigate the relationship
between wavelength- and time-resolved cell autofluorescence and cellular metabolic status. The experiments were
carried out on monolayered cell cultures including normal and cancer ectocervical cells. With UV excitation at 365 nm,
the decay of cellular fluorescence can be well described by a dual-exponential function, consisting of a short lifetime
component ((tau)<sub>1</sub>~ 0.40 - 0.47 ns) and a long lifetime component (((tau)<sub>2</sub> ~ 3.3 - 4.0 ns). By analyzing the decay-associated
spectra of the short and long lifetime components, we found that the long lifetime component carried the information of
protein-bound NADH and short lifetime component was mainly determined by free NADH with certain interference
from bound NADH. Moreover, it was found that the ratio of the amplitudes of two lifetime components, dominated by
free/bound NADH, was sensitive to cell metabolism. Overall, this study demonstrated that wavelength- and timeresolved
autofluorescence can be potentially used as an important contrast mechanism to detect epithelial pre-cancer.
Autofluorescence spectroscopy has been a widely explored technique for <i>in vivo</i> and noninvasive diagnosis of pre-cancer
lesions in epithelium where 90% cancers originate. For extracting more accurate fluorescence information for cancer
diagnosis, depth-resolved fluorescence measurements are crucial to assess NADH and FAD in non-keratinized epithelial
layer and collagen in stromal layer, respectively. In this study, we achieved the depth-resolved fluorescence spectral
measurements of squamous epithelial tissue based on confocal technique. We found that in non-keratinized epithelial
layer the fluorescence signals excited at 405 nm were the combination of NADH and FAD fluorescence and could be
used for evaluating the redox ratio. Moreover, we found that confocal time-resolved autofluorescence measurements of
epithelial tissue with 405 nm excitations could provide the information on the layered tissue structure. All depth-resolved
autofluorescence decays were accurately fitted with a dual-exponential function consisting of a short lifetime (0.4 ~ 0.6
ns) and a long lifetime (3 ~ 4 ns) components. The short lifetime component dominated the decay of non-keratinzied
epithelial fluorescence while the decay of the signals from keratinized epithelium and stroma were mainly determined by
the long lifetime component. The ratio of the amplitudes of two components could be used to differentiate the layered
structure of epithelial tissue. In general, the results in this study demonstrated that the combined depth- and timeresolved
fluorescence measurements can produce the information on the layered structure and localized biochemistry of
epithelial tissue for the diagnosis of tissue pathology.
Autofluorescence of rabbit and human epithelial tissues were studied by using a depth-resolved fluorescence spectroscopy system with multiple excitations. Keratinization was found to be common in the squamous epithelium. Strong keratin fluorescence with excitation and emission characteristics similar to collagen were observed in the topmost layer of the keratinized squamous epithelium. The keratin signal created interference in the assessment of the endogenous fluorescence signals (NADH/FAD fluorescence in epithelium and collagen fluorescence in stroma) associated with the development of epithelial precancer. Furthermore, the keratinized epithelial layer attenuated the excitation light and reduced the fluorescence signals from underlying tissue layers. The autofluorescence of columnar epithelium was found to be dominated by NADH and FAD signals, identical to the autofluorescence measured from nonkeratinized squamous epithelium. The study also demonstrated that a fluorescence signal excited at 355 nm produced sufficient contrast to resolve the layered structure of epithelial tissue, while the signal excited at 405 nm provided the information for a good estimation of epithelial redox ratios that are directly related to tissue metabolism. Overall, the depth-resolved measurements are crucial to isolate the fluorescence signals from different sublayers of the epithelial tissue and provide more accurate information for the tissue diagnosis.
A fluorescence spectroscopy system combining depth- and time-resolved measurements is developed to investigate the layered fluorescence temporal characteristics of epithelial tissue. It is found that esophageal tissue structure can be resolved well by means of the autofluorescence time-resolved decay process with 375-, 405- and 435- nm excitation. The decay of the autofluorescence signals can be accurately fitted with a dual-exponential function consisting of a short lifetime (0.4 ~ 0.6 ns) and a long lifetime (3 ~ 4 ns) components. The short lifetime component dominates the decay of normal epithelial fluorescence while the decay of the signals from keratinized epithelium and stroma are mainly determined by the long lifetime component. The ratio of the amplitudes of two components provides the information of fine structure of epithelial tissue. This study demonstrates that the combined depth- and time-resolved measurements can potentially provide accurate information for the diagnosis of tissue pathology.
The depth-resolved autofluorescence ofrabbit oral tissue, normal and dysplastic human ectocervical tissue within l20μm depth were investigated utilizing a confocal fluorescence spectroscopy with the excitations at 355nm and 457nm. From the topmost keratinizing layer of oral and ectocervical tissue, strong keratin fluorescence with the spectral characteristics similar to collagen was observed. The fluorescence signal from epithelial tissue between the keratinizing layer and stroma can be well resolved. Furthermore, NADH and FADfluorescence measured from the underlying non-keratinizing epithelial layer were strongly correlated to the tissue pathology. This study demonstrates that the depth-resolved fluorescence spectroscopy can reveal fine structural information on epithelial tissue and potentially provide more accurate diagnostic information for determining tissue pathology.
The depth-resolved autofluorescence of normal and dysplastic human ectocervical tissue within 120um depth were investigated utilizing a portable confocal fluorescence spectroscopy with the excitations at 355nm and 457nm. From the topmost keratinizing layer of all ectocervical tissue samples, strong keratin fluorescence with the spectral characteristics similar to collagen was observed, which created serious interference in seeking the correlation between tissue fluorescence and tissue pathology. While from the underlying non-keratinizing epithelial layer, the measured NADH fluorescence induced by 355nm excitation and FAD fluorescence induced by 457nm excitation were strongly correlated to the tissue pathology. The ratios between NADH over FAD fluorescence increased statistically in the CIN epithelial relative to the normal and HPV epithelia, which indicated increased metabolic activity in precancerous tissue. This study demonstrates that the depth-resolved fluorescence spectroscopy can reveal fine structural information on epithelial tissue and potentially provide more accurate diagnostic information for determining tissue pathology.
Diagnostic techniques based on optical spectroscopy have the potential to link the biochemical and morphological properties of tissues. Light-induced fluorescence (LIF) spectroscopy as a noninvasive “optical biopsy” method has been widely used to detect small lesions <i>in vivo</i>. Confocal fluorescence spectroscopy provides a tool for optical sectioning of tissue and provides an approach for identifying small shifts in the emission spectra that are caused by intracellular microenvironment factors. The ability to demarcate abnormal and normal tissue with a confocal spectroscopy system depends on the ability of interpreting the source of fluorescence within the samples. It is realized by spatially dispersing the fluorescence collected with a fiber that serves as the pinhole aperture of the confocal system. In order to tracing the autofluorescence spectral signals of tissue layer by layer, a confocal fluorescent spectroscopy system has been set-up. Experiments have been carried out with fluorescent phantom and animal models. With an axial resolution of 10um in animal tissues, this confocal spectral system observed the spectral differences in spectral shape and spectral peak position among different layers of tissue illuminated with 349nm laser. It was also found that the fluorescence intensity is depth-dependent. In conclusion, confocal fluorescence spectroscopy can provide more diagnostic information due to its ability of optical sectioning. It’s hopeful that a confocal spectral system can detect cancer at much earlier stage.
Endogenous fluorophores, such as NAD(P)H/FAD and collagen/elastin, have been regarded as in vivo quantitative fluorescence biomarkers for precancerous changes of epithelial tissue. However, the fluorescence signal measured by conventional spectroscopy is a mixture of autofluorescence from the epithelium and deep structures. The dominant fluorescence of collagen/elastin from connective tissue in deep layers creates serious challenge for extracting the epithelial fluorescence of NAD(P)H/FAD that is weak, but important for the characterization of tissue pathology. In this work, we instrumented a confocal fluorescence spectroscopy system and a two-photon excited fluorescence spectroscopy system to measure the depth-resolved single- and two-photon fluorescence spectra from the rabbit esophageal tissues. The excitation wavelengths were 349 nm and 735 nm, respectively. Both systems provided good optical sectioning. The information obtained from depth-resolved fluorescence was generally consistent with the histology of the examined tissue sample. The NAD(P)H signals from epithelial layers were clearly separated from the collagen signal from deep layers. In addition, strong second harmonic generations given by collagen fibers were observed. This work demonstrates that depth-resolved fluorescence spectroscopy may produce more accurate information on the diagnosis of tissue pathology.