An evaluation of DNA fluorochromes, staining techniques, and analysis for flow cytometry. I. Unperturbed cell populations

Several preparative techniques (detergent treatment, ethanol fixation, and hypotonic cell lysis), DNA fluorochromes, and methods of numerical analysis (planimetric or curve-fitting) were compared for the estimation of cell-cycle kinetic parameters (G1, S, G2 + M) by flow cytometry. In addition, coefficients of variation (CV), relative fluorescence, and G1/chicken erythrocyte (CRBC) ratios were measured and the effects of the proportion of cycling cells and cellular RNA content were examined. DNA fluorochromes were ranked by relative fluorescence: 4,6-diamidino-2-phenylindole > ethidium bromide/mithramycin > Hoechst 33342 > mithramycin > ethidium bromide > acridine orange approximately equal to propidium iodide. The first four (DNA-specific stains) gave lower CVs than the remainder (DNA intercalators). Detergent treatment also increased relative fluorescence and slightly lowered CVs. Comparable results were obtained for the kinetic parameters independently of stain or staining procedure; intercalating dyes with cells of a high RNA content not treated with RNAse and acridine orange being the exceptions. Of the two methods of numerical analysis, the planimetric technique was more consistant. Although highly consistant G1/CRBC ratios were obtained for any one stain, independently of staining procedures, variations between stains were noted. It is suggested that the detergent treatment in combination with DNA-specific stains provide optimal results.     

Flow cytometric estimation of DNA and RNA content in intact cells stained with Hoechst 33342 and pyronin Y

The addition of RNA content estimation to flow cytometric measurement of DNA content provides valuable information concerning cells' transitions between quiescent and proliferative states. Equilibrium staining methods employing acridine orange have been used for DNA/RNA content measurement but are difficult to apply to intact cells and impractical for use in conjunction with fluorescent antibodies or ligands for demonstration of cell surface structures. I have used a combination of Hoechst 33342 (HO342) and pyronin Y (PY) to stain intact cells for DNA/RNA content estimation with a dual source flow cytometer using UV and blue-green or green excitation, measuring HO342 fluorescence at 430--470 nm and PY fluorescence at 590--650 nm. Results obtained with cultured cells and stimulated lymphocytes are in good agreement with those obtained using acridine orange for DNA/RNA staining; about half of the PY fluorescence can be removed from ethanol-fixed cells stained with HO342 and PY by RNAse digestion. The HO342/PY method can be combined with fluorescein immunofluorescence for detection of cell surface markers. HO342 can be combined with other tricyclic heteroaromatic dyes for DNA/RNA estimation; the combination of HO342 and oxazine 1 can be excited in a dual source instrument using a mercury arc lamp and a helium-neon laser. The staining procedure is simple; cells in medium are incubated with 5 microM HO342 at 37 degrees C for 45 min, 5 microM PY (or oxazine 1) is then added and cells are analyzed without washing after an additional 45 min incubation. Suitability of these dye combinations for vital cell staining and sorting remains to be determined.     

Different sensitivity of chromatin to acid denaturation in quiescent and cycling cells as revealed by flow cytometry.

The properties of DNA in situ as reflected by its staining with acridine orange are different in quiescent nonstimulated lymphocytes as compared with interphase lymphocytes that have entered the cell cycle after stimulation by mitogens. The difference is seen after cell treatment with buffers at pH 1.5 (1.3-1.9 range) followed by staining with acridine orange at pH 2.6 (2.3-2.9). Under these conditions the red metachromatic fluorescence of the acridine orange-DNA complex is higher in quiescent cells than in the cycling lymphocytes while the orthochromatic green fluorescence is higher in the cycling, interphase cells. The results suggest that DNA in condensed chromatin of quiescent lymphocytes (as in metaphase chromosomes) is more sensitive to acid-denaturation than DNA in dispersed chromatin of the cycling interphase cells. The phenomenon is used for flow cytometric differentiation between G0 and G1 cells and between G2 and M cells. In contrast to normal lymphocytes the method applied to neoplastic cells indicates the presence of cell subpopulations with condensed chromatin but with DNA content characteristic not only of G1 but also of S and G2 cells. The possibility that these cells represent quiescent (resting) subpopulations, arrested in G1, S and/or G2, is discussed.     

Cytofluorometric and cytochemical comparisons of normal and abnormal human cells from the female genital tract.

Acridine orange staining of exfoliated cells from epithelial tissues facilitates discrimination between normal and abnormal cells: abnormal cells develop highly elevated nuclear fluorescence. Comparisons of acridine orange (AO) staining with propidium iodide (PI) or Feulgen staining have shown that: (a) PI staining also provides highly elevated nuclear fluorescence from abnormal cells; (b) the distributions of nuclear fluorescence following AO or PI staining were usually not significantly different as judged by the Kolmogorov-Smirnov test; (c) fluorescence emission spectra from AO and PI stained cells are consistent with the hypothesis that both fluorochromes bind to DNA within cell nuclei; (d) DNAse treatment of AO stained normal cells eliminates the nuclear fluorescence peak from slit-scan contours; RNAse treatment has no effect on nuclear fluorescence; (e) the distribution of abnormal cell nuclear fluorescence after AO staining is usually, but not always, significantly different from the distribution of abnormal cell nuclear absorbance after Feulgen staining, with relative nuclear fluorescence being greater than relative nuclear absorbance. The hypothesis currently most consistent with these results is that elevated Feulgen DNA content can account for only part of the discrimination provided by AO staining, and that the chromatin within abnormal cells is altered so as to increase accessibility of DNA to intercalating dyes.     

Discrimination of cycling and non-cycling lymphocytes by BUdR-suppressed acridine orange fluorescence in a flow cytometric system.

Stimulated lymphocytes which pass through the cell cycle may be distinguished from dormant G0 lymphocytes rapidly by flow cytometry. The method is based on cell incubation with 5-bromodeoxyuridine (BUdR) and their subsequent staining with acridine orange under conditions in which cellular DNA and RNA stain differentially. The DNA-specific green fluorescence of stimulated, cycling cells is suppressed while RNA-specific red fluorescence is affected only minimally. It is possible, therefore, to distinguish cycling vs non-cycling cells based on two entirely different parameters, i.e. BUdR incorporation and RNA content.     

Quantitative fluorescence spectrophotometry of acridine orange-stained unfixed cells. Potential for automated detection of human uterine cancer.

After staining with acridine orange (AO), the nuclei of unfixed cells from the human female genital tract exhibited the same fluorescence behavior previously observed for human and murine leukocytes and mouse ascites tumor cells. With staining conditions chosen to assure saturation of the green-fluorescing AO-nucleic acid complex in normal cells, corrected fluorescence emission spectra were recorded from the entire nucleus of 341 cells taken from 32 normal and 28 abnormal patients. Intensity of the recorded spectra was expressed in phosphor particle units, a fixed arbitrary unit of fluorescence intensity, to display intensity differences among the spectra from the various cell types. In all abnormal samples, one or more cells were found with 530-nm nuclear fluorescence intensity considerably greater than the maximum intensity recorded from normal cells. Determination of the adequacy of 530-nm nuclear fluorescence intensity as a criterion for cancer detection requires additional investigation. Additional criteria, if needed, may be supplied by the metachromasy of AO-stained unfixed cells.     

Mechanism of differential staining of BUdR-substituted Vicia faba chromosomes.

Chromosomes of the broad bean Vicia faba were isolated and air-dried on slides after incorporation of BUdR into DNA (BUdR substitution) for two rounds of replication. Then the preparations were embedded in a buffer solution containing trypsin as well as fluorescence dye (acridine orange or Hoechst 33258). We observed chromosomes with a fluorescence microscope at various times after embedding. After about 15 min one sister chromatid of some of the metaphase chromosomes showed enhanced darkening and disintegration within 1–4 min (melting effect) during observation. We suppose that fragmentation of BUdR-substituted DNA by the acridine orange-visible light system in acridine orange staining and by irradiation with wavelengths around the transition from UV to visible light in Hoechst 33258 staining is responsible for this phenomenon. The disintegration of one sister chromatid in BUdR-substituted chromosomes can also be produced by UV irradiation during trypsin treatment when fluorescence dyes are not present.     

Changes in accessibility of DNA to various fluorochromes during spermatogenesis

Accessibility of mouse testicular and vas deferens (vas) sperm cell DNA to acridine orange, propidium iodide, ellipticine, Hoechst 33342, mithramycin, chromomycin A3, 4'6-diamidino-2-phenylindole (DAPI), and 7-amino-actinomycin D (7-amino-AMD) was determined by flow cytometry. Permeabilized cells were either stained directly or after pretreatment with 0.06 N HCl. For histone-containing tetraploid, diploid, and round spermatid cells, HCl extraction of nuclear proteins caused an approximately sixfold increase of 7-amino-AMD stainability but had no significant effect on DAPI stainability. For these same cell types, the stainability with other intercalating (acridine orange, propidium iodide, ellipticine) and externally binding (Hoechst 33342, mithramycin, chromomycin A3) dyes was increased by 1.6- to 4.0-fold after HCl treatment. In sharp contrast, HCl treatment of vas sperm did not increase the staining level of 7-amino-AMD, DAPI, or propidium iodide but did increase the staining level for the other intercalating dyes (1.3- to 1.5-fold) and external dyes (1.3- to 1.9-fold). Elongated spermatids that contain a mixture of protein types including histones, transition proteins, and protamines demonstrated the greatest variability of staining with respect to type of stain and effect of acid extraction of proteins. In general, for nearly all dyes, the round spermatids had an increased level and tetraploid cells had a decreased level of stainability relative to the same unit DNA content of diploid cells. The observed differential staining is discussed in the context of chromatin alterations related to the unique events of meiosis and protein displacement and replacement during sperm differentiation.     

Acridine Orange Staining and Fluorescence of Nucleic Acids in Plasmodia and Associated Host Erythrocytes

Acridine orange in daily doses of 1, 2 and 4 mg for 4 days was given to chicks averaging 50 gm in weight. Dosage was started 1, 2 and 3 days after infection with Plasmodium gallinaceum. Such doses were sufficient to stain the parasite in vivo, as shown by its bright fluorescence in UV light, but did not exhibit any antimalarial action. Staining of fresh blood samples from infected chicks with 0.01% acridine orange in Krebs-Ringer containing 0.1 M phosphate buffer (pH 6.0-6.2) resulted in differential fluorescence of the nucleic acids of the plasmodia, to show nuclear DNA bright green and cytoplasmic RNA orange-red. After optimum acid hydrolysis, as used for the Feulgen reaction, staining with 0.1% acridine orange produced intense red fluorescence of the nuclear DNA in the plasmodia. Nuclear DNA of the chick erythrocytes showed bright fluorescence both in vivo and in vitro.     

Identification of polymorphonuclear leukocytes in cytologic samples for flow cytometry.

Inflammatory cells are commonly present in cytologic specimens obtained for flow cytometry, and may interfere with the analysis of epithelial cells. We have found that detergent (Triton X-100) pretreatment in the two-step acridine orange staining procedure disrupts granulocyte cell membranes to yield bare nuclei; bladder epithelial and squamous cells on the other hand are quite resistant to the detergent treatment. Being deprived of their cytoplasmic RNA, the granulocytes lose red fluorescence. Moreover, the shearing forces in the cytometer extend the multisegmented granulocyte nuclei and align them in the direction of flow. Thus, they present as elongated objects in the measuring system, giving a large DNA fluorescence pulsewidth (nuclear size). These two phenomena make it possible to identify granulocytes in the recorded data, where they are discernible from the mononucleated leukocytes and from epithelial cells. By data selection the granulocytes can be excluded, rendering epithelial cell populations more amenable to analysis. This method may make it unnecessary to remove physically leukocytes from the specimen before flow cytometry; it may also provide a way to analyze the morphology of granulocyte nuclei and to assess methods to manipulate their membrane stability. Full protection from membrane disruption is accomplished by alcohol fixation, and partial protection by 20-30% serum.     

Total cellular RNA content: correlation between flow cytometry and ultraviolet spectroscopy

Total RNA content in Chinese hamster ovary and HeLa-S3 cells determined by ultraviolet spectroscopy is compared with the red fluorescence distribution of acridine orange-stained cells observed by flow cytometry. A correlation coefficient of 0.93 is obtained when these methods of estimating RNA content are compared after various RNAse treatments. These data suggest that acridine orange staining effectively quantitates total cellular RNA content when analyzed by flow cytometry, although DNA is also shown to contribute a low but significant background of red fluorescence.     

Acridine orange as a supravital fluorochrome indicating varying degrees of chromatin condensation

Summary When acridine orange is used at 10^–5 or 10^–6M concentrations in supravital preparations of mouse or grasshopper testicular cells, it is evident that condensed chromatin in spermatogenic cells displays considerably higher levels of fluorescence than extended chromatin in vesicular nuclei. This result differs from that obtained traditionally in appropriately fixed and stained cytological preparations in which fluorescence is greatest in extended chromatin and lowest in condensed chromatin (Rigler, 1966, and others). While an explanation for this reversed staining pattern in unfixed cells is not immediately available, it is clear that acridine orange may offer novel opportunities for the experimental manipulation and examination of the chromosomes (for deoxyribonucleoprotein complex) of living cells.     

Bertalanffy-like fluorescence staining with 3-dimethylamino-6-methoxyacridine

Three new acridine dyes, 3-dimethylamino-6-methoxyacridine 1, 3-amino-6-methoxyacridine 2 and 3-amino-7-methoxyacridine 3, have been prepared and tested as fluorochromes of LM- and HeLa-cells. The dyes are basic compounds (pKA: 1 8,76; 2 8,01; 3 7,65) and form cations in neutral or acidic aqueous solutions by addition of a proton to the aza-nitrogen atom of the heterocycle. The fluorochromes stain fixed LM- and HeLa-cells at pH = 6. The fluorescence shows metachromasy similar to the staining with acridine orange AO according to the technique of Bertalanffy. But there is less fading of the fluorescence. The dye 1 is the most suitable fluorochrome of the series. It was studied in detail. Using optimized staining conditions the fluorescence of the nucleus is yellow-green that of the cytoplasm and the nucleoli orange or brownish-red. Enzymatic digestion experiments show that the dye cations are bound to DNA in the nucleus and to RNA in the cytoplasm or nucleoli. The absorption and emission spectra of the stained cells have been studied by means of microspectrophotometry. The absorption spectra of the nucleus and the cytoplasm are very similar. The maximum of the long wave length absorption of both occurs at 21400 cm-1 (467 nm) with a shoulder at ca 20100 cm-1 (498 nm). The fluorescence spectra of nucleus and cytoplasm of metachromatically stained cells are different. The emission maximum of the cytoplasm and nucleoli, 16200 cm-1 (617 nm), is red-shifted relative to the maximum of the nucleus, 18200 cm-1 (549 nm). This shift causes the metachromatic fluorescence effect. In addition we studied the concentration dependence of the absorption and fluorescence spectra of the cation 1 in aqueous solution, pH = 6, in the concentration range 6 X 10(-6)-6 X 10(-4) M. Shape and maximum of the long wave length absorption and emission depend only slightly on the concentration: Mean value of absorption maximum ca 21500 cm-1 (465 nm), shoulder at ca 20300 cm-1 (493 nm), fluorescence maximum ca 18300 cm-1 (547 nm). With growing concentration diminishes the molar absorptivity. This decrease in absorptivity and isosbestic points in the absorption spectra indicate the formation of dimers with growing dye concentration. The absorption spectra of the metachromatically stained cells and of the dye in aqueous solution are very similar.(ABSTRACT TRUNCATED AT 400 WORDS)     

Highly Selective Acridine and Ethidium Staining of Bacterial DNA and RNA

The acridine dyes acridine orange (AO) and coriphosphine O (CPO) and ethidium bromide (EtBr) were used to stain bacterial digests after electrophoresis in native and denaturing (SDS) polyacrylamide gels and were shown to stain DNA and RNA preferentially over other subcellular components in the gels. Vegetative cell digests of Bacillus subtilis, Escherichia coli, Micrococcus luteus, and Staphylococcus aureus showed intense staining of DNA with AO and CPO near the top of the gel, but little or no staining of other cellular constituents. EtBr stained both DNA and RNA in the gels. Protein standards and non-nucleic acid cellular constituents stained faintly with high concentrations (> 100 μM) of AO, lower concentrations (13.9 μM) of CPO, and did not stain with 0.5 μg/ml EtBr in denaturing gels. The complete set of cellular biochemicals was visualized by silver staining, while the protein subset was detected by Coomassie blue staining. The highest concentrations of AO (120 μM) and CPO (13.9 μM) were shown to detect purified DNA in gels with a sensitivity in the range of 25–50 ng per band. This work demonstrates the specificity of acridine and ethidium dyes for nucleic acids, while illustrating the level of non-nucleic acid-specific interactions with other cellular components by staining of electrophoretically separated cellular components in a gel matrix.     

DNA stainability in aneuploid breast tumors: comparison of four DNA fluorochromes differing in binding properties.

The aim of this study was to evaluate whether or not the differences in chromatin structure between diploid stromal cells or lymphocytes, which are often used as DNA ploidy standard, and aneuploid breast tumor cells can significantly affect the estimates of the DNA index of these tumors. To this end, the DNA content estimates of 34 aneuploid breast tumors, differing in size, degree of differentiation, and presence or absence of estrogen and progesterone receptors and metastases, were compared using four common DNA fluorochromes: DAPI, Hoechst 33342, propidium iodide, and acridine orange. These dyes differ in their mode of interaction with DNA (binding to minor groove or intercalation) and for each of them binding to DNA is restricted to a different degree by nuclear proteins. It was expected, therefore, that if differences in chromatin structure play a role in DNA content estimates, the DNA index of the measured tumors may vary depending on the dye. The cell nuclei were isolated from the tumors using a detergent-based procedure and stained with each of the dyes and the DNA index was estimated using peripheral blood lymphocytes as a DNA content standard. For each of the tumors, the DNA index estimates with all four dyes correlated very well. When the results obtained with individual dyes were compared in pairs, the correlation coefficients (r) of DNA indices were all above 0.96 (correlation at p less than 0.001). The best concordance was seen between specimens stained with Hoechst 33342 and DAPI (r = 0.99), and the least between those stained with Hoechst 33342 and propidium iodide (r = 0.96). The data indicate that DNA content analysis of unfixed nuclei, utilizing the above fluorochromes, is not significantly biased by differences in chromatin structure of the measured cells.     

Flow cytometric analysis of RNA content in different cell populations using pyronin Y and methyl green

Pyronin Y (PY) was used, in flow cytometric (FCM) systems, to estimate the RNA content per cell in formalin fixed EL4 leukosis tumor cells, enzyme dispersed R3327-G rat prostatic adenocarcinoma cells, mouse spleen cells stimulated with concanavalin A, and human peripheral blood lymphocytes stimulated with phytohemagglutinin. Preincubation of the cells with methyl green (MG) blocked PY binding to DNA such that the intracellular fluorescence from MG-PY was due primarily to its binding to RNA. Treatment of the cells with ribonuclease resulted in a 3- to 5-fold reduction in the fluorescence intensity of intracellular MG-PY. Mitogen stimulation of either mouse or human lymphocytes resulted in an increase in DNA (propidium iodide fluorescence) and RNA (MG-PY fluorescence) content per cell over resting levels. Further, the changes in stimulated human lymphocyte DNA and RNA contents following 24, 48, and 72 hr of cell culture were monitored. The results showed that RNA levels were significantly increased prior to that of DNA. Also, the effects of different cell cycle phase specific blocking agents on lymphocyte cell cycle traverse were investigated. We found that: a) actinomycin D inhibited the increases in cellular RNA and DNA; b) hydroxyurea inhibited the increases in cellular RNA were only slightly reduced; c) tritiated thymidine caused an accumulation of cells having high DNA and RNA contents; and d) Colcemid promoted an accumulation of cells having high DNA contents while causing a reduction of cells having high RNA contents. These results were nearly identical to reports by other investigators using the metachromatic dye acridine orange to quantitate RNA per cell. Thus, the MG-PY technique described is indicated to provide a stable and accurate measure of RNA content per cell.     

Differences in the red fluorescence of acridine orange bound to single-stranded RNA and DNA.

Fluorescence of acridine orange bound to RNA or DNA in the single-stranded form including single-stranded synthetic polyribo- or polydeoxyribonucleotides was measured in the expectation that some distinct structural characteristic between single-stranded RNA and DNA might be reflected by a specific fluorescent behaviour of bound dyes. It was found that the complex of the dye with single-stranded RNA emits a weaker red fluorescence around 650 nm than the complex with single-stranded DNA at low phosphate-to-dye ratios. The fact could be explained neither by a direct interaction of bound dyes with the 2′-hydroxyl group of ribose in RNA nor by the difference in the G-C content of the nucleic acids. On the basis of the character of dye molecules emitting the red fluorescence, it was suggested that the bases in single-stranded RNA might be buried in some hydrophobic environment that would make the dyes less likely to interact with them, compared with the bases in single-stranded DNA. It was further inferred that some conformational rigidity of single-stranded RNA may partially be responsible for the weaker red fluorescence.     

A novel technique for viable cell determinations

We have developed a simple method to determine cell viability using two fluorescent dyes, Hoechst 33258 and acridine orange. When these dyes are used in combination, dead cells fluoresce brilliant blue and live cells fluoresce green. This method works over a range of dye concentrations (Hoechst 33258, 0.25-2 micrograms/ml; acridine orange, 1-5.0 micrograms/ml) and the fluorescence spectra of the two dyes are such that only one set of filters is required to visualize the effects of both dyes simultaneously. It is insensitive to a wide range of exogenous serum concentrations and is read with greater uniformity by different observers.     

The Changes of the Vitality of Armillaria mellea and the Histochemicam Localization of Some Enzymes in the Hyphae Digested Period of Gastrodia elara

The changes of the vitality of Armillaria mellea of infecting corm of Gastrodia elata were observed by appling the live body staining method of acridine orange and by means of fluorescence microscopy. The green fluorescence of vitality was emitted by the first infected hyphae, the yellow one by the decrepit hyphae; but the orange to red fluorescence with the lost vitality were emitted by the fragmentary hyphae and clump form bodies. The large cells containing rich, RNA and protein had been confirmed by the method of the induced fluorescence which the acridine orange and by the method at pH 2.2 which the fast-green staining. The acid-phosphatase was mainly distributed within the cortical cells filled with the infected hyphae. There were few such deposits in the socalled large cells except their walls. The activity of the esterase was shown in the cortical cells filled with the infected hyphae. It were also shown in the clump form bodies and the collapsed nuclei of the large cells. The activities of adenosine triphosphatase (ATPase), peroxidase and polyphenol oxidase was notably shown in the cortical cells filled with the infected hyphae and the large cells.     

Bacterial Growth on Surfaces: Automated Image Analysis for Quantification of Growth Rate-Related Parameters

A fast routine method for estimating bacterial cell growth rates by using the metachromatic dye acridine orange is described. The method allows simultaneous estimates of cellular RNA and DNA contents of single cells. Acridine orange staining can be used as a nonspecific supplement to quantitative species-specific hybridizations with fluorescence-labelled ribosomal probes to estimate the single-cell concentration of RNA. By automated analysis of digitized images of stained cells, we determined four independent growth rate-related parameters: cellular RNA and DNA contents, cell volume, and the frequency of dividing cells in a cell population. These parameters were used to compare physiological states of liquid-suspended and surface-growing Pseudomonas putida KT2442 in chemostat cultures. The major finding is that the correlation between substrate availability and cellular growth rate found for the free-living cells was not observed for the surface-bound cells; in contrast, the data indicate an almost constant growth rate for attached cells which was independent of the dilution rate in the chemostat.