Hyperspectral imaging microscopy for identification and quantitative analysis of fluorescently‐labeled cells in highly autofluorescent tissue

Standard fluorescence microscopy approaches rely on measurements at single excitation and emission bands to identify specific fluorophores and the setting of thresholds to quantify fluorophore intensity. This is often insufficient to reliably resolve and quantify fluorescent labels in tissues due to high autofluorescence. Here we describe the use of hyperspectral analysis techniques to resolve and quantify fluorescently labeled cells in highly autofluorescent lung tissue. This approach allowed accurate detection of green fluorescent protein (GFP) emission spectra, even when GFP intensity was as little as 15% of the autofluorescence intensity. GFP‐expressing cells were readily quantified with zero false positives detected. In contrast, when the same images were analyzed using standard (single‐band) thresholding approaches, either few GFP cells (high thresholds) or substantial false positives (intermediate and low thresholds) were detected. These results demonstrate that hyperspectral analysis approaches uniquely offer accurate and precise detection and quantification of fluorescence signals in highly autofluorescent tissues. (© 2012 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

A dual fluorescence flow cytometric analysis of bacterial adherence to mammalian host cells

Flow cytometry has provided a powerful tool for analyzing bacteria-host cell associations. Established approaches have used bacteria, labeled either directly with fluorochromes or indirectly with fluorescently conjugated antibodies, to detect these associations. Although useful, these techniques are consistently unable to include all host cells in the analysis while excluding free, aggregated bacteria. This study describes a new flow cytometry method of assessing bacterial adherence to host cells based on direct fluorescent labeling of both bacteria and host cells. Eukaryotic host cells were labeled with PKH-26, a red fluorescent dye, and bacteria were labeled with fluorescein isothiocyanate, a green fluorescent dye. The red host cells were gated and the mean green fluorescence intensity (MFI) of these red cells was determined. We used MFI values obtained from control samples (unlabeled and labeled host cells with unlabeled bacteria) to eliminate contributions due to autofluorescence. The final MFI values represent fluorescence of host cells resulting from the adherent bacteria. Because all red fluorescent cells are analyzed, this method includes all the eukaryotic cells for analysis but excludes all free or aggregated bacteria that are not bound to target cells.     

Comparison of helium-neon and dye lasers for the excitation of allophycocyanin

Allophycocyanin (APC) has a broad absorption spectrum permitting several different lasers to be used to excite this dye in a flow cytometer. A comparison was made between a dye laser and a helium-neon (HeNe) laser for the excitation of APC as an immunofluorescent chromophore. The ratio of fluorescence of stained to unstained lymphocytes (signal to background) was used to assess differences in sensitivity. In determining the best wavelength for operating the dye laser, it was found that there was little difference in the ability to separate the positive-labelled cells from the unstained cells using 600 nm or 633 nm light for excitation of APC. A study of the effect of laser power on the signal to background identified a nonlinear relationship. It was found that the sensitivity obtained with 47 mW of 633 nm light from a HeNe laser was near the maximum attainable. This sensitivity was comparable to that obtained using phycoerythrin as an immunofluorescence chromophore. APC had the added advantage of being applicable to the study of highly autofluorescent cells. Exciting this chromophore using red light dramatically decreased the autofluorescence observed even on alveolar macrophages.     

Microspectroscopic imaging of nodulation factor-binding sites on living Vicia sativa roots using a novel bioactive fluorescent nodulation factor.

A novel bioactive fluorescent nodulation (Nod) factor, NodRlv-IV(BODIPY FL-C16), has been synthesized by attaching a BODIPY FL-C16 acyl chain to the primary amino group of chitotetraose deacetylated at the nonreducing terminus by recombinant NodB. The binding of the fluorescent Nod factor to root systems of Vicia sativa was investigated with fluorescence spectral imaging microscopy (FSPIM) and fluorescence ratio imaging microscopy (FRIM). Spatially resolved fluorescence spectra of living and labeled Vicia sativa root systems were measured by FSPIM. Strong autofluorescence, inherent to many plant systems when excited at 488 nm, was corrected for by utilizing the difference in fluorescence emission spectra of the autofluorescence and NodRlv-IV(BODIPY FL-C16). A methodology is presented to break down the in situ fluorescence emission spectra into spatially resolved autofluorescence and BODIPY FL fluorescence spectra. Furthermore, an FRIM method was developed for correcting autofluorescence in fluorescence micrographs for this system. After autofluorescence correction it was shown that NodRlv-IV(BODIPY FL-C16) was concentrated in the root hairs, but was also bound to other parts of the root surface.     

Immunofluorescence analysis in flow cytometry: better selection of antibody-labeled cells after fluorescence overcompensation in the red channel

Selection of cells labeled with fluorescein isothiocyanate-conjugated (FITC) antibodies can be difficult if large autofluorescent cells are used and if the cells bind only a few molecules of antibody. We have developed a simple flow cytometric procedure that allows better selection of stained cells. When an argon ion laser emitting at 488 nm is used, the green fluorescence detected is the sum of cell autofluorescence and of the signal generated by the FITC antibody. Thus, when we subtract green signal from the red by fluorescence compensation, the signal of stained cells is on average reduced more than for the unstained counterpart. In this scenario, positive selection of cells with low red signal allows more efficient selection of stained cells. We tested the overcompensation procedure on mixtures of cells unstained and stained with a relevant FITC antibody. Cell mixtures were analyzed using normal vs increased levels of compensation in the red channel. Increased levels of compensation resulted in easier gating and higher recovery of stained cells. The efficiency of the overcompensation procedure was particularly high when using red filters with low cutoff (i.e., 560 or 570 nm), possibly because of the significant emission of fluorescein in the red channel, which caused separation between stained and unstained cells also in the red dimension. This method is useful for sorting cells expressing low levels of surface markers and facilitates selection of rare cells transfected with surface antigen genes. This technique is compatible with the use of propidium iodide for live/dead cell discrimination and with the subtraction of the cellular background of autofluorescence.     

Cell-substrate contacts illuminated by total internal reflection fluorescence

A technique for exciting fluorescence exclusively from regions of contact between cultured cells and the substrate is presented. The technique utilizes the evanescent wave of a totally internally reflecting laser beam to excite only those fluorescent molecules within one light wavelength or less of the substrate surface. Demonstrations of this technique are given for two types of cell cultures: rat primary myotubes with acetylcholine receptors labeled by fluorescent alpha- bungarotoxin and human skin fibroblasts labeled by a fluorescent lipid probe. Total internal reflection fluorescence examination of cells appears to have promising applications, including visualization of the membrane and underlying cytoplasmic structures at cell-substrate contacts, dramatic reduction of autofluorescence from debris and thick cells, mapping of membranes topography, and visualization of reversible bound fluorescent ligands at membrane receptors.     

MRI Detection of Nonproliferative Tumor Cells in Lymph Node Metastases Using Iron Oxide Particles in a Mouse Model of Breast Cancer

Cell tracking with magnetic resonance imaging (MRI) and iron nanoparticles is commonly used to monitor the fate of implanted cells in preclinical disease models. Few studies have employed these methods to study cancer cells because proliferative iron-labeled cancer cells will lose the label as they divide. In this study, we evaluate the potential for retention of the iron nanoparticle label, and resulting MRI signal, to serve as a marker for slowly dividing cancer cells. Green fluorescent protein-transfected MDA-MB-231 breast cancer cells were labeled with red fluorescent micron-sized superparamagnetic iron oxide (MPIO) nanoparticles. Cells were examined in vitro at multiple time points after labeling by staining for iron-labeled cells and by flow cytometric detection of the fluorescent MPIO. Severe combined immune deficiency (SCID) mice were implanted with 5 x 10⁵ MPIO-labeled or unlabeled cells in the mammary fat pad and MRI was performed weekly until 28 days after injection. Microscopy was performed to validate MRI. In vitro assays revealed a very small percentage of cells that retained MPIO at 14 days after labeling. Regions of signal loss were observed in MRI of primary tumors that developed from iron-labeled cancer cells. Small focal regions of signal loss were detected in images of the axillary and brachial nodes in six of eight mice, at day 14 or later, with microscopy confirming the presence of iron-labeled cancer cells. Our data suggest an interesting role for cell tracking with iron particles since label retention leads to persistent signal void, allowing proliferative status to be determined.     

Visualization of thrombin receptors on mouse embryo fibroblasts using fluorescein-amine conjugated human α-thrombin

The localization of thrombin receptors on mouse embryo (ME) cells has been examined by direct fluorescence microscopy using a fluorescein aminelabeled thrombin. Two fluorescein amines, 4-(N-6-aminoethyl thioureal)-fluorescein and 4-(N-6-aminohexyl thioureal)-fluorescein, were synthesized and attached to the carbohydrate moiety of highly purified human α-thrombin by periodate oxidation of the carbohydrate and selective reduction of the Schiff's base using sodium cyanoborohydride. Preparations of fluorescent thrombin with from 1 to 4 fluoresceins per molecule of thrombin retained their ability to proteolytically cleave fibrinogin to form fibrin clots, to bind to thrombin receptors on ME cells, and to initiate cell division. After incubating mitogenic concentrations of the fluorescein amine labeled thrombin with ME cells at 4°C, a diffuse fluorescent pattern was observed over the surface of the ME cells. This diffuse pattern was specific: it was not observed on cells from parallel cultures incubated with fluorescent thrombin plus a 20-fold excess of unlabeled thrombin. Thus, thrombin receptors appear to be distributed randomly over the surface of ME cells prior to interaction with thrombin. Increasing the temperature to 37°C following binding at 4° C resulted in a rapid dissociation of the fluorescent pattern from the cells leaving only the autofluorescent vesicles. This result may reflect the unique ability of thrombin to proteolytically cleave its own receptor.     

A simple fluorescent labeling technique to study virus adsorption in Newcastle disease virus infected cells

The present study demonstrates that the fluorescent general membrane dyes PKH67 and PKH26 are suitable to label Newcastle disease virus, an enveloped virus belonging to the family of paramyxoviridae. Adsorption of the labeled virus particles was tracked, visualized and quantitated using confocal laser scanning microscopy. The specificity of PKH-labeling was determined by colocalization analysis of the PKH signal with NDV-specific immunolabeling, and by using mock-infected controls and infection with detergent-pretreated labeled virus particles. The infectivity of the NDV particles was not affected by the labeling procedure as indicated by the results of a cytotoxicity ATP assay, an apoptosis assay and detection of virus-specific RNA and protein by qPCR and Western blotting, respectively, in cells infected with PKH-labeled and unlabeled virus particles. This technique can be used as an inexpensive, sensitive and rapid alternative method in the analysis of adsorption and internalization of enveloped viruses by the infected cells.     

Long-term tracking of lymphocytes in vivo: the migration of PKH-labeled lymphocytes

Studies of in vivo cell migration using cell markers such as 51Cr, 111In, FITC, or XRITC have been limited to short time periods due to the elution, toxicity, or rapid loss of label detectability. We have labeled sheep lymphocytes in vitro with PKH-2, a new fluorescent cell membrane label, and, after their intravenous injection back into donor sheep, have been able to detect them in efferent lymph, using flow cytometry, for longer than 38 days. The PKH-2-labeled lymphocytes migrated with similar kinetics, efficiency, and tissue specificity as lymphocytes labeled with cell markers used previously. PKH-2-labeled cells mediated graft versus host reactions indistinguishable from those mediated by unlabeled cells, and cell surface antigens were equally detectable on the surface of labeled and unlabeled lymphocytes. According to the slow, consistent loss of fluorescence intensity of the labeled cells in vivo, we predict that labeled lymphocytes could remain detectable by flow cytometry for greater than 7 weeks with the labeling protocol used in these experiments.     

Pattern of lipofuscin pigmentation in nitrergic and non-nitrergic,neurofilament immunoreactive myenteric neuron types of human small intestine

Lipofuscin, an autofluorescent age pigment, occurs in enteric neurons. Due to its broad excitation and emission spectra, it overlaps with commonly used fluorophores in immunohistochemistry. We investigated the pattern of lipofuscin pigmentation in neurofilament (NF)-reactive nitrergic and non-nitrergic human myenteric neuron types. Subsequently, we tested two methods for reduction of lipofuscin-like autofluorescence. Myenteric plexus/longitudinal muscle wholemounts of small intestines of five patients undergoing surgery for carcinoma (aged between 18 and 69 years) were double stained for NF and neuronal nitric oxide synthase (nNOS). Lipofuscin pigmentation patterns were semiquantitatively evaluated by using confocal laser scanning microscopy with three different excitation wave lengths (one for undisturbed lipofuscin autofluorescence and two for specific labellings). Two pigmentation patterns could be detected in the five NF-reactive neuron types investigated. In nitrergic/spiny as well as in non-nitrergic/stubby neurons, coarse, intensely autofluorescent pigment granules were prominent. In non-nitrergic type II, III and V neurons, a fine granular, diffusely distributed and less intensely autofluorescent pigment was obvious. After incubation of wholemounts in either CuSO₄ or Sudan black B solutions, unspecific autofluorescence could be substantially reduced whereas specific NF and nNOS fluorescence remained largely unaffected. We conclude that NF immunohistochemistry is useful for morphological representation of subpopulations of human myenteric neurons. The lipofuscin pigmentation in human myenteric neurons reveals at least two different patterns which can be related to distinct neuron types. Incubations of multiply stained whole mounts in both CuSO₄ or Sudan black B are suitable methods for reducing autofluorescence thus facilitating discrimination between specific (immunohistochemical) and non-specific (lipofuscin) fluorescence.     

Spectral decomposition of intracellular complex fluorescence using multiple-wavelength phase modulation lifetime determination: technical approach and preliminary applications

Lifetime-based spectral decomposition using a frequency-domain phase/modulation technique is developed on a microspectrofluorimeter prototype. In a fluorescent mixture with strongly overlapping components, such measurements enable us to not only obtain excited state lifetimes of each fluorescent component but also determine the specific spectral contribution of each species without the use of any model spectra. Examples of such applications are first given for complex mixtures of highly overlapping fluorescent components in solution. Preliminary results concerning cellular applications are also reported. This allows us to follow the cellular uptake and intracellular stability of fluorescent labeled modified oligonucleotides in the context of antisense strategy studies. Indeed, the intracellular signal from the fluorescent label bound to oligonucleotides can be distinguished from those of the free label by its specific excited state lifetime.     

Isotopic labeling of DNA in rat adipose tissue: evidence for proliferating cells associated with mature adipocytes.

The intraperitoneal administration of [3H]thymidine to adult rats resulted in the rapid appearance of label in the adipocyte fraction of collagenase digests of adipose tissue. Low-speed centrifugation followed by freezing and slicing showed the label to be uniformly distributed in the adipocyte fraction. The presence of label in DNA was confirmed by hydrolysis with deoxyribonuclease and by inhibition of incorporation with hydroxyurea. Organelle fractionation revealed that the label was predominantly in nuclei, and radioautography showed that only a few adipocyte nuclei were labeled. The label in the adipocyte fraction could not be reduced by increased collagenase digestion or by trypsin treatment. Mixing of labeled adipocytes with unlabeled stroma did not result in decrease of label and addition of labeled stroma to unlabeled adipocytes did not cause significant transfer of radioactivity. Addition of [3H]thymidine to the collagenase digestion medium of unlabeled adipose tissue resulted in more incorporation by adipocytes than by stroma, suggesting the presence of a very rapidly proliferating cell type associated more with adipocytes than with stroma. In vivo turnover studies of labeled DNA indicated that there are two components in both adipocytes and stroma, a rapidly labeled component with a half-life of only several days and another with a half-life of several months. These experiments suggest that there is a rapidly proliferating cell type in adipose tissue, closely associated with mature adipocytes, that may be an adipocyte progenitor or may have some other unknown function.     

Cell-by-cell autofluorescence correction for low signal-to-noise systems: application to epidermal growth factor endocytosis by 3T3 fibroblasts

Autofluorescence of cells can be a major portion of the fluorescence signal in many systems, especially when fluorescent conjugates are used to study receptor-ligand systems for which there are less than 70,000 receptors per cell. We have devised a method for the cell-by-cell correction of autofluorescence for flow cytometric data by using an additional parameter to measure and correct for autofluorescence in the fluorescence channel. The principle has been extended to allow simultaneous correction for autofluorescence and dual-fluorescence spillover compensation in samples labeled with two different fluorochromes; all corrections were done in software, making them applicable to any flow cytometer. The autofluorescence correction method was used to analyze the acidification of epidermal growth factor (EGF) by Swiss 3T3 cells. EGF is acidified to pH 6.2 starting two min after labeling, with a half-time for acidification of 45 s.     

Flow cytometric, phase-resolved fluorescence measurement of propidium iodide uptake in macrophages containing phagocytized fluorescent microspheres

BACKGROUND: Spectral interference (overlap) from phagocytosed green-yellow (GY) microspheres in the flow cytometric, red fluorescence emission measurement channel causes errors in quantifying damaged/dead alveolar macrophages by uptake of propidium iodide. METHODS: Particle burdens of uniform GY fluorescent microspheres phagocytosed by rat alveolar macrophages and the discrimination of damaged/dead cells as indexed by propidium iodide uptake were assessed with conventional and phase-sensitive flow cytometry. RESULTS: The fluorescence spectral emission from phagocytosed microspheres partly overlapped the propidium iodide red fluorescence emission and interfered with the measurement of damaged/dead cells when using conventional flow cytometry without subtractive compensation. This caused errors when estimating the percentage of nonviable, propidium iodide-positive, phagocytic macrophages. The interference was eliminated by employing phase-sensitive detection in the red fluorescence measurement channel based on differences in fluorescence lifetimes between the fluorescent microspheres and propidium iodide. Intrinsic cellular autofluorescence, whose fluorescence lifetime is approximately the same as that of the phagocytosed microspheres, also was eliminated in the phase-sensitive detection process. Because there was no detectable spectral interference of propidium iodide in the green fluorescence (phagocytosis) measurement channel, conventional fluorescence detection was employed. CONCLUSIONS: Phase-resolved, red fluorescence emission measurement eliminates spectral overlap errors caused by autofluorescent phagocytes that contain fluorescent microspheres in the analyses of propidium iodide uptake.Cytometry 39:45-55, 2000. Published 2000 Wiley-Liss, Inc.     

Multicolour flow cytometry analyses and autofluorescence in chlorophytes: lessons from programmed cell death studies in Chlamydomonas reinhardtii

Flow cytometry is a valuable tool in phycological studies. However, endogenous cellular compounds like nicotinamide adenine dinucleotide and chlorophyll a and b autofluoresce, potentially interfering with fluorescent markers. Furthermore, autofluorescent properties are not uniform across algae, nor are their effects consistent in different cytometers. The choice of instrument and fluorescent marker, therefore, requires careful consideration. We investigated the suitability of fluorescent markers by using standard four-colour and advanced multicolour flow cytometers in relation to the effects of autofluorescence over ranges of parameters including fluorophore excitation and emission spectra, band-pass filter configurations, voltage gains and the effects of growth in the light and dark. The unicellular chlorophyte and model organism, Chlamydomonas reinhardtii, was used and findings were correlated with investigations of programmed cell death. As previously found C. reinhardtii autofluoresces in the red, far-red and infrared spectra. This is independent of laser excitation wavelength, and autofluorescence emits and spills over into detection channels of both four-colour and multicolour instruments. Band-pass filter configurations capturing longer wavelength emissions or fluorophores excited or emitted in these longer wavelengths are generally unsuitable. Furthermore, neither dark nor light incubation impacted the autofluorescent signals. Consideration of these algal autofluorescent properties and their spillover effects is required to avoid erroneous results. Recommendations for the use of a range of fluorophores in programmed cell death and other studies in C. reinhardtii using four-colour and multicolour instruments are made.     

A new technique to identify hybrid myotubes in vitro without culture fixation.

Fluorescent latex microspheres (FLMs) were used to label myoblasts and to permit the observation of hybrid myotubes before culture fixation. This type of labeling did not affect survival, development, or fusion of these cells. The FLMs were retained for several weeks. Labeled mouse myoblasts were co-cultured with unlabeled rat myoblasts to verify whether the marker was released and spread from labeled to unlabeled cells. The nuclear stain Hoechst 33258 was used to distinguish the myoblasts from both species and permitted the demonstration that there was virtually no re-uptake. Hybrid myotubes were also obtained by co-culturing mouse myoblasts containing rhodamine FLMs and rat myoblasts containing green FLMs. These mixed cultures were observed repeatedly with a fluorescent microscope without any cytotoxic effect. Several myotubes were observed before fixation of the cultures to contain both types of fluorescent labels. Subsequent fixation and staining with Hoechst dye confirmed that these myotubes were hybrids.     

Optical lock-in detection of FRET using synthetic and genetically encoded optical switches

The Förster resonance energy transfer (FRET) technique is widely used for studying protein interactions within live cells. The effectiveness and sensitivity of determining FRET, however, can be reduced by photobleaching, cross talk, autofluorescence, and unlabeled, endogenous proteins. We present a FRET imaging method using an optical switch probe, Nitrobenzospiropyran (NitroBIPS), which substantially improves the sensitivity of detection to <1% FRET efficiency. Through orthogonal optical control of the colorful merocyanine and colorless spiro states of the NitroBIPS acceptor, donor fluorescence can be measured both in the absence and presence of FRET in the same FRET pair in the same cell. A SNAP-tag approach is used to generate a green fluorescent protein-alkylguaninetransferase fusion protein (GFP-AGT) that is labeled with benzylguanine-NitroBIPS. In vivo imaging studies on this green fluorescent protein-alkylguaninetransferase (GFP-AGT) (NitroBIPS) complex, employing optical lock-in detection of FRET, allow unambiguous resolution of FRET efficiencies below 1%, equivalent to a few percent of donor-tagged proteins in complexes with acceptor-tagged proteins.     

Improved in vivo whole-animal detection limits of green fluorescent protein-expressing tumor lines by spectral fluorescence imaging

Green fluorescent protein (GFP) has been used for cell tracking and imaging gene expression in superficial or surgically exposed structures. However, in vivo murine imaging is often limited by several factors, including scatter and attenuation with depth and overlapping autofluorescence. The autofluorescence signals have spectral profiles that are markedly different from the GFP emission spectral profile. The use of spectral imaging allows separation and quantitation of these contributions to the total fluorescence signal seen in vivo by weighting known pure component profiles. Separation of relative GFP and autofluorescence signals is not readily possible using epifluorescent continuous-wave single excitation and emission bandpass imaging (EFI). To evaluate detection thresholds using these two methods, nude mice were subcutaneously injected with a series of GFP-expressing cells. For EFI, optimized excitation and emission bandpass filters were used. Owing to the ability to separate autofluorescence contributions from the emission signal using spectral imaging compared with the mixed contributions of GFP and autofluorescence in the emission signal recorded by the EFI system, we achieved a 300-fold improvement in the cellular detection limit. The detection limit was 3 x 10(3) cells for spectral imaging versus 1 x 10(6) cells for EFI. Despite contributions to image stacks from autofluorescence, a 100-fold dynamic range of cell number in the same image was readily visualized. Finally, spectral imaging was able to separate signal interference of red fluorescent protein from GFP images and vice versa. These findings demonstrate the utility of the approach in detecting low levels of multiple fluorescent markers for whole-animal in vivo applications.     

一种标记cDNA芯片探针的新方法

探讨mRNA长片段反转录PCR技术(RT-LDPCR)在cDNA芯片微量探针标记和信号放大中的应用.首先提取BEP2D细胞的总RNA,然后用两种不同的方法进行标记,一种为RT-LDPCR,用荧光素Cy3-dCTP进行标记;另一种为传统的RNA反转录,用荧光素Cy5-dCTP进行标记.将两种方法标记好的探针等量混合后与含有440个点(44个基因)的cDNA芯片同时杂交,发现二者具有很高的一致性(0.5＜Cy3/Cy5＞2.0).由于RNA反转录法为cDNA芯片探针标记的传统方法,从而验证了RT-LDPCR用于cDNA芯片探针标记的可行性.RT-LDPCR具有对样品总RNA的需要量少和可对样品中信号进行放大的优点,特别适合于对材料来源受到限制的RNA进行标记.