SEARCH

SEARCH BY CITATION

Keywords:

  • fluorescent dyes;
  • Nile Red;
  • BODIPY;
  • DiO carbocyanine;
  • lipids;
  • green algae

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

When the fluorescence signal of a dye is being quantified, the staining protocol is an important factor in ensuring accuracy and reproducibility. Increasingly, lipophilic dyes are being used to quantify cellular lipids in microalgae. However, there is little discussion about the sensitivity of these dyes to staining conditions. To address this, microalgae were stained with either the lipophilic dyes often used for lipid quantification (Nile Red and BODIPY) or a lipophilic dye commonly used to stain neuronal cell membranes (DiO), and fluorescence was measured using flow cytometry. The concentration of the cells being stained was found not to affect the fluorescence. Conversely, the concentration of dye significantly affected the fluorescence intensity from either insufficient saturation of the cellular lipids or formation of dye precipitate. Precipitates of all three dyes were detected as events by flow cytometry and fluoresced at a similar intensity as the chlorophyll in the microalgae. Prevention of precipitate formation is, therefore, critical to ensure accurate fluorescence measurement with these dyes. It was also observed that the presence of organic solvents, such as acetone and dimethyl sulfoxide (DMSO), were not required to increase penetration of the dyes into cells and that the presence of these solvents resulted in increased cellular debris. Thus, staining conditions affected the fluorescence of all three lipophilic dyes, but Nile Red was found to have a stable fluorescence intensity that was unaffected by the broadest range of conditions and could be correlated to cellular lipid content. © 2012 International Society for Advancement of Cytometry


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

Traditional lipid extraction procedures allow for quantitative analysis of lipid content but can be complex, time consuming, labor-intensive and cause decomposition or oxidation of the lipids. As an alternative to extractions, there are several types of fluorescent lipophilic dyes available, including Nile Red (1–3), 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) (4), and 3,3′-dioctadecyloxacarbocyanine perchlorate (DiO) (5) (Fig. 1), for which their fluorescence intensities could be correlated to the lipid content within cells (1, 2). The rapidity and ease of use of a dye-based lipid assay makes its use to assess cellular lipid content in biofuel research highly desirable. However, staining protocols with these dyes, especially with microalgae, vary greatly (1–5). There is a lack of evidence showing the importance or insignificance of staining parameters such as dye concentration, cell concentration, or staining time. In order for lipophilic dye-dependent in vitro assays to be broadly applicable, the limitations of the dyes and specific staining procedures need to be identified.

thumbnail image

Figure 1. Structures of the lipophilic dyes.

Download figure to PowerPoint

Nile Red (Nile blue oxazone) is a benzophenoxazone dye synthesized from Nile Blue, a commonly used histological stain. The excitation and emission spectra of Nile Red shifts further toward the blue end of the spectrum as the polarity of the solvent decreases (6, 7). In an extremely nonpolar environment, such as hexane, the maximum Nile Red emission is in the green wavelengths between 525 and 560 nm. In neutral lipids of cholesterol esters and triacylglycerols, the maximum Nile Red emission is in the yellow wavelengths, ∼ 580 nm (3). As lipids increase in polarity with diacylglycerols and monoacylglycerols, the maximum Nile Red emission shifts to 610 and 640 nm accordingly (3). This is similar to the maximum emission of Nile Red in polar solvents, such as methanol. In water, Nile Red has very low solubility and may form an organized aggregate capable of fluorescence (7). Continual spectrum shifts means that Nile Red is not lipid specific, but a rather polarity specific dye and caution should be taken as it has been shown to bind proteins and other nonlipid cellular compartments (8).

BODIPY 505/515 is one of a group of boron-dipyrromethene dyes distinguished by their substituents. The vast majority of these dyes have extremely low water solubility and are insensitive to the pH or polarity of their environment (9). The nonpolar forms of BODIPY are considered lipophilic dyes that are useful for staining neutral lipids and when compared with Nile Red, BODIPY dyes have a narrower emission spectrum and higher sensitivity (10). Staining of microalgae for microscopy by Cooper et al. (4) has shown nonpolar BODIPY 505/515 distinctively labels lipid bodies in the green channel with less background cytoplasmic staining than Nile Red. Unlike Nile Red, the emission spectrum of BODIPY does not shift depending on the polarity of the environment the dye is in, but variation of the substituents of the dye varies the excitation and emission maxima of this family of dyes.

DiO-C18 is a member of the carbocyanine family of dyes that are amphiphilic probes containing a charged fluorophore attached to a pair of saturated alkyl chains. DiO has an excitation maximum of 484 nm and fluoresces green with an emission maximum of 501 nm. Carbocyanine dyes are popular for membrane staining and cell tracing applications (5, 11). Similar to the other lipophilic dyes, carbocyanine dyes have very low water solubility and weak fluorescence in water.

This study examined the application of three lipophilic dyes (Nile Red, BODIPY 505/515, and DiO-C18) for the in vitro quantitative staining of lipids in microalgae, under a variety of staining conditions. Analysis via flow cytometry was used to evaluate the differences between the dyes and staining conditions using two strains of green microalgae often used in biofuel studies. The first strain, Scenedesmus dimorphus, are crescent shaped, thick cell-walled, 10–25 μm long, 2–7 μm wide, and grow in groups of four (12). The second strain, Chlorella vulgaris, are circular with diameters ranging from 0.5 to 5 μm. It was discovered during flow cytometry analysis that noncellular events developed under certain staining conditions and affected the mean fluorescence signal. To improve the correlation between fluorescence intensity and cellular lipid content, an optimal lipophilic dye and set of staining conditions were identified that reduced the noncellular events and fluorescence variability.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

Lipophilic Dyes

Vybrant DiO-C18 (Invitrogen, Burlington, Canada; cat no. V-22886) was purchased as a 1 mg/mL stock solution in ethanol, which was stored in the dark at room temperature. One mg/mL stock solution of BODIPY 505/515 (Invitrogen, cat no. D-3921) was made with acetone and another with DMSO, both were stored in the dark at −20°C. A total of 1 mg/mL stock solutions of Nile Red (Sigma, Oakville, Canada; cat no. N3013) dissolved in acetone or DMSO were made and stored in the dark at room temperature. These 1 mg/mL stocks were used for staining with dye concentrations of 5 and 25 μg/mL. Each dye solution was also diluted 1 in 10 to make a 0.1 mg/mL stock solution that was used for staining with dye concentrations of 0.5 to 2.5 μg/mL.

Cell Culture and Analysis

S. dimorphus #1237, purchased from the UTEX collection (University of Texas at Austin) and C. vulgaris #90, purchased from the CPCC collection (University of Waterloo, Canada), were maintained for 2 months by reinoculations in Bristol's medium (13), grown at 25°C, exposed to photosynthetic lights (2700 lux) in cycles of 12 h light and 12 h dark, and continuously agitated at 125 rpm in incubators (Infors, Montreal, Canada). Cells had relatively low lipid content and were used in all phases of growth. For correlation to lipid quantification, cells were in stationary phase and had higher lipid content.

Images were acquired with a Nikon (Mississauga, Canada) Eclipse 90i epi-fluorescence microscope and a Plan Apo 40×/0.95 objective. Cells were stained with each dye at a concentration of 10 μg/mL for 30 min before imaging. The green channel collected emissions between 510 and 560 nm, and the red channel collected emissions between 590 and 660 nm.

A FACSCanto II flow cytometer (Beckton Dickson Systems, San Jose, CA) equipped with a 488 nm argon laser was used to measure single cell fluorescence. Chlorophyll and precipitate fluorescence was measured using the red 780 ± 30 nm band pass filter (see emission spectra in Supporting Information Fig. 1). Nile Red fluorescence was measured using the yellow 585 ± 21 nm band pass filter, which corresponded to neutral lipid emissions. Fluorescence of BODIPY and DiO were measures using the green 530 ± 15 nm band pass filter. The voltages for forward scatter (FSC) and side scatter (SSC) were 400 and 500, respectively. Data was collected for 10,000 or 20,000 events using logarithmic amplification. Unstained cells were used as controls.

Standard conditions were 0.2 to 1 × 106 cells/mL in 40 mM pH 5.0 sodium phosphate, stained with 1 μg/mL dye and 1% solvent for at least 30 min at room temperature. These conditions were altered in accordance with the parameter being examined (see Supporting Information Methods). The parameters examined were dye concentration, cell concentration, duration of staining, acetone concentration, and DMSO concentration. The control condition for cell concentration, dye concentration, and staining time was unstained cells, and the control condition for the organic solvents was cells stained in 1% solvent with 1 μg/mL of dye.

Data Analysis

A similar gating strategy was used for all the data files to identify the total population and divide the population into cells, debris, and precipitate. The total population was gated in the FSC vs. SSC plot as the events between 0 and 250,000 of both axes. From the total population, the debris was gated in the infrared 780 ± 30 nm (PE-Cy7-A) channel based on an intensity < 1000. The remaining events were plotted in a 780 ± 30 nm (PE-Cy7-A) channel vs. 702 ± 33 nm (PerCP-Cy5.5-A) channel plot. A gate for precipitation events was setup by using unstained cell events as the boundary (see Supporting Information Fig. 2 for example of how events were gated). Cellular events were then the inverse of the precipitation gate. The relative fluorescence of a channel was calculated by dividing the mean fluorescence intensity of all the population for that condition by that channels mean fluorescence intensity of all the population of the control condition.

thumbnail image

Figure 2. The problem of gating flow cytometry events by chlorophyll signal alone when staining S. dimorphus with lipophilic dyes. These bivariate dot plots show representative data of unstained cells and cells stained with 10 μg/mL of Nile Red, BODIPY 505/515, or DiO-C18. The chlorophyll signal is in the infrared (780 ± 30 nm) channel and the lipid fluorescence is in the yellow (585 ± 21 nm) or green (530 ± 15 nm) channel. If chlorophyll fluorescence is used to gate for microalgae, care must be taken that dye precipitate is not present or it will be included in this gate. Blue dots represent debris, red dots represent precipitate, and green dots are cells. See Materials and methods section for how events were gated.

Download figure to PowerPoint

Each condition was tested at least four times, and the means and standard deviations of each population were averaged. Percentage of events results are presented as means with standard deviations, and relative fluorescence results are presented as means with 95% confidence intervals.

Lipid Quantification

Total lipid concentration was determined by chemical titration (14). A cell concentration for each sample was determined before 100 mL of cells being spun down and dried at 60°C. The dried pellet of algae was resuspended in 20 mL of 95% ethanol containing 5% isopropanol. A total of 5 mL of 33% potassium hydroxide was then added before transferring to a 250 mL boiling flask. With the flask connected to a reflux condenser, the contents were boiled at 93°C on a hot plate for 30 min. After cooling, 8.5 mL of 25% hydrochloric acid was added and then 25 mL of pentane. The flask was topped and shaken manually for 1 min. After separation, 12.5 mL of the pentane-lipid layer was removed with a graduated pipette and placed in an 80 mL beaker. The beaker contents were evaporated to dryness on a hot plate. Ten milliliters of ethanol containing bromothymol blue indicator was added to the beaker, and the solution was titrated with 10 mM sodium hydroxide until the thymol blue displayed its color. The moles of NaOH added was equal to the moles of fatty acid extracted, so the concentration of fatty acids per cell was calculated by dividing the moles of NaOH added by the cell concentration of the sample and the original volume of cells spun down.

Quantifiability of Staining Methods

To compare the correlation between lipid quantity and fluorescence signal, microalgae from each of the cultures used for lipid quantification were also stained and measured by flow cytometry. Samples were vigorously agitated to break-up aggregates of cells before analysis. The different staining methods used for this experiment were chosen to vary all of the parameters that affect fluorescence intensity. Staining Method 1 was 5 μg/mL of Nile Red in 5% acetone for 30 min. Staining Method 2 was 1 μg/mL of BODIPY in 1% DMSO for 30 min. Staining Method 3 was 5 μg/mL DiO in 1% ethanol for 30 min. Staining Method 4 was 2 μg/mL of Nile Red in 1% acetone for 5 min. The optimized staining method was 2 μg/mL of Nile Red in 1% acetone for 30 min.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

Staining of microalgae lipids with fluorescent lipophilic dyes was investigated using both S. dimorphus and C. vulgaris. The effects of staining conditions were similar for both species, so to simplify presentation of the findings, only the results of S. dimorphus are shown in the majority of cases.

Presence of Noncellular Events

While optimizing a lipid staining protocol, flow cytometry revealed the presence of two types of noncellular events. Debris, including cell fragments and salts, were measured as events when their size was similar to the microalgae. This debris was gated based on its red fluorescence being lower than the strong chlorophyll fluorescence of intact cells.

However, gating based on chlorophyll signal alone was insufficient to isolate cellular events (Fig. 2). Another noncellular set of events appeared only when cells were in the presence of a lipophilic dye. These events were determined to be aggregated dye precipitate and were difficult to gate due to the fluorescence of the precipitates having similar intensities in the red channels as the chlorophyll of the microalgae. Especially for DiO, it was not possible using any combination of filters to isolate all the precipitate events from the cell events. The identification of the strongly red fluorescent events as precipitate of the dye was confirmed by microscopy (Fig. 3). Samples stained with Nile Red or BODIPY clearly had particulates fluorescing within the red channel (620 ± 30 nm), but the precipitate had a very weak signal in the green channel (535 ± 25nm). With DiO, the precipitate fluoresced less intensely than the cells in the red channel, but the precipitate had very intense fluorescence in the green channel. Green fluorescence in the cells stained by DiO was also inconsistent, as if there was DiO precipitate in the cells.

thumbnail image

Figure 3. Visualization of precipitates and microalgae cells when staining with lipophilic dyes. Fluorescence of the dyes in the presence of cellular lipids can be seen in the green channel. Fluorescence of DiO precipitate is also visible in the green channel. Microalgae have chlorophyll fluorescence in the red channel, and the precipitates of Nile Red and BODIPY also have strong red fluorescence. The merge images clearly show the absence of colocalization of the precipitates and cells. Cells were stained for 30 min with 10 μg/mL of dye before imaging.

Download figure to PowerPoint

Effect of Dye Concentration

To stain cells for quantification of their fluorescence intensity, it is important that the fluorescence remain independent of the concentration of the dye in the sample. For these measurements, the concentration of the dye solvent was kept constant, while varying only the concentration of dye. In addition to measuring the mean fluorescence intensity (Fig. 4, top panels), the composition of the fluorescence signal was determined to be comprised of cellular, debris, and precipitate events (Fig. 4, bottom panels). Although the majority of the fluorescent signal could be attributed to cellular events, staining the microalgae with increasing concentrations of any of the lipophilic dyes decreased the percentage of cell events and increased the percentage of precipitation events.

thumbnail image

Figure 4. Dye concentration affects the population fluorescence and composition. Top panels show dye fluorescence associated with lipid content was affected by increasing concentration of lipophilic dye. Bottom panels show the composition of the population that produced the fluorescence signal. Increasing the concentration of dye decreased the percentage of cell (□) and debris (•) events and increased the percentage of precipitate (×) events recorded by flow cytometry for all lipophilic dyes. After 30 min of staining, Nile Red fluorescence was measured in the yellow channel, while BODIPY and DiO fluorescence were measured in the green channel. Percentage of events data are shown as mean and standard deviation, and relative fluorescence data are shown as mean and 95% confidence interval relative to unstained cells.

Download figure to PowerPoint

For samples stained with lipophilic dye concentrations less than 1 μg/mL, the mean fluorescence increased with increasing concentration of dye, suggesting that at low concentrations cellular lipids were not saturated (Fig. 4, top panels). Based on the fluorescence signal, saturation of cellular lipids by the dyes was dependent on dye concentration and not cell concentration. Various concentrations of cells, from the lowest to the highest concentration measurable with the flow cytometer, were stained with 1 μg/mL of dye (Supporting Information Fig. 3). It was found that neither the percentage of precipitate events nor the fluorescence intensity changed significantly. Only when excess concentrations of dye were used did a high concentration of cells reduce the percentage of precipitate events. Therefore, the percentage of precipitation events was also a dye related effect more than a cell related effect.

With Nile Red, concentrations of 1 to 10 μg/mL resulted in mean fluorescent intensities in the yellow channel that were not significantly affected by the changing concentration of dye (Fig. 4). When the percentage of precipitation events increased with high concentrations of Nile Red, the mean yellow channel fluorescence decreased. BODIPY fluorescence increased to a maximum with a concentration of 2.5 μg/mL, and then, the presence of precipitate events decreased the fluorescence. These significant changes in BODIPY fluorescence with concentration indicated the absence of a broad range of usable concentrations and extreme care must be taken with the concentration of BODIPY. DiO also lacked a range of concentrations, where fluorescence was independent of dye concentration. With increased concentration of DiO, the relative fluorescence increased as well as the percentage of precipitation events. This absence of a concentration where the relative fluorescence of DiO reached a maximum precluded DiO from any further investigations.

Effect of Staining Duration

The length of time required for the stain to penetrate the cells and develop a strong signal was examined to determine when the cells should be measured after being stained. During the first 20 min of staining the microalgae with Nile Red or BODIPY, the mean fluorescence increased significantly (Fig. 5). From 30 min onward, both Nile Red and BODIPY fluorescence had no significant difference in mean fluorescence for the remainder of the 2 h test. The duration of staining had no effect on the presence of debris or precipitate.

thumbnail image

Figure 5. Staining duration affects fluorescence. Dye fluorescence increased significantly during the first 30 min of staining with Nile Red and BODIPY. After 30 min, no significant change in fluorescence occurred for either dye. Data are the mean and 95% confidence interval relative to unstained cells.

Download figure to PowerPoint

The samples stained with Nile Red for examination of the effect of staining duration (Fig. 5) had a significantly increased relative fluorescence compared with microalgae stained with Nile Red in Figure 4. This difference in relative fluorescence was due to the use of cells in varying stages of cell growth with different lipid contents.

Effect of an Organic Solvent

Being lipophilic dyes, Nile Red and BODIPY were dissolved in an organic solvent for use, in this case acetone. It was, therefore, important to examine the effects of acetone on Nile Red and BODIPY stained cells when the concentration of the dye and cells were kept constant. The effect of acetone concentration was significant for both dyes (Fig. 6). The only change in acetone concentration that did not affect the fluorescence was doubling the acetone concentration from 1% to 2% (v/v) with Nile Red. However, this small increase in acetone concentration from 1% to 2% caused a significant increase in BODIPY fluorescence. For both dyes, further increases in acetone concentration caused an increase in the percentage of events that were debris, until at 25% acetone when the mean relative fluorescence decreased and more than 50% of the total number of events was cell debris.

thumbnail image

Figure 6. Acetone is detrimental to cell staining. Top panels show dye fluorescence associated with lipid content was significantly affected by increasing concentrations of acetone. Bottom panels show the population composition that produced the fluorescence signal. Increasing the concentration of acetone decreased the percentage of cell (□), increased the percentage of debris (○) events, and did not change the percentage of precipitate (×) events. Percentage of events data are shown as mean and standard deviation, and relative fluorescence data are shown as mean and 95% confidence interval relative to cells stained in 1% acetone.

Download figure to PowerPoint

Occasionally (3, 4), DMSO has been used as the organic solvent to dissolve Nile Red or BODIPY, so the effect of DMSO concentration on cell staining was also examined. For both Nile Red and BODIPY, even the presence of 2.5% DMSO reduced the fluorescence signals significantly and increased the percentage of cellular debris (Fig. 7). Increasing concentrations of DMSO further, resulted in decreasing the fluorescence signals to negligible levels and the production of more than 90% of the events being debris at DMSO concentrations as low as 10%.

thumbnail image

Figure 7. DMSO is detrimental to cell staining. Top panels show dye fluorescence associated with lipid content was significantly reduced by increasing concentrations of DMSO. Bottom panels show increasing the concentration of DMSO decreased the percentage of cell (□), increased the percentage of debris (•) events, and did not change the percentage of precipitate (×) events. Percentage of events data are shown as mean and standard deviation, and relative fluorescence data are shown as mean and 95% confidence interval relative to cells stained in 1% DMSO.

Download figure to PowerPoint

Quantification of Cellular Lipids

The variation in fluorescence intensity caused by changes in staining conditions significantly affected the ability to quantify cellular lipids (Fig. 8). Cultures of S. dimorphus and C. vulgaris, having different cellular lipid contents, were stained using various conditions and dyes. The resulting fluorescence of each staining method was compared with the cellular fatty acid concentration of the culture determined by extraction and titration. Despite the broad range of cellular lipid contents, staining the cultures with DiO (Method 3) resulted in almost no difference in fluorescence intensity. A correlation was seen when the samples were stained with 1 μg/mL BODIPY in 1% DMSO (Method 2), except with one sample for which the fluorescence was higher than expected for its measured lipid content. Staining the cells with 5 μg/mL Nile Red in 5% acetone (Method 1) significantly reduced the fluorescence signal of the cells with high lipid content. The best correlation between measured lipid quantity and fluorescence intensity was obtained using the optimal dye concentration, solvent concentration, and a staining duration of 2 μg/mL Nile Red in 1% acetone for 30 min. However, reducing the staining time to 5 min with the optimal dye and solvent concentrations (Method 4) negated any correlation between lipid content and fluorescence intensity.

thumbnail image

Figure 8. Correlation between cellular lipid content and fluorescence intensity of microalgae was affected by staining conditions. The optimized staining method of 2 μg/mL Nile Red in 1% acetone for 30 min resulted in the highest correlation (solid line, R2 = 0.99) between fluorescence and lipid content. No correlation could be established with staining methods 3 (5 μg/mL DiO in 1% ethanol for 30 min) and 4 (2 μg/mL Nile Red in 1% acetone for 5 min). Staining Method 1 (5 μg/mL Nile Red in 5% acetone for 30 min) had some correlation (dotted line, R2 = 0.83). Staining Method 2 (1 μg/mL BODIPY in 1% DMSO for 30 min) had a good correlation (dashed line, R2 = 0.95). The concentration of fatty acid per cell was measured by titrating the extracted fatty acids with NaOH. The three samples with the highest fatty acid content per cell were stationary phase S. dimorphus cultures, and the three samples with the lowest fatty acid content per cell were stationary phase C. vulgaris cultures.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

Lipophilic dyes can be used to quantitatively measure cellular lipids. However, caution should be taken as staining conditions can dramatically affect the fluorescent intensity of these dyes, and therefore, the ability to accurately compare lipid content. Current publications quantifying lipids with these dyes, especially with microalgae, use a variety of staining conditions with few reports detailing optimization of the staining procedure (10, 15). Without an optimized staining procedure, it is difficult to reproduce results, adjust the assay for different species, or reliably use staining to quantify cellular lipids. Although flow cytometry and spectrofluorometric assays have both been used for fluorescence assays, for this study, flow cytometry was chosen for its ability to measure the fluorescence of individual events. This ensured measured fluorescence was independent of cell concentration and allowed distinction between cells, debris, or precipitate. Several variables of a staining procedure were examined, including concentration of the dye, concentration of cells, duration of staining, and the presence of organic solvents. Using three popular lipophilic dyes, Nile Red, BODIPY, and DiO, a staining procedure that can be used to quantify fluorescent intensity was then optimized.

Microalgae were chosen as the model microorganism due to the significant interest in the use of microalgae lipids for biofuels, the potential difficulty in stain penetrating the thick cell wall, and the variation among staining protocols of published lipid assays. Whilst the stage of cell growth varied the morphology and lipid content of the microalgae, these did not appear to influence how staining conditions affected fluorescence intensity with either of the microalgae species examined. Background noise was evident from both salts and cellular debris and could not be significantly reduced by washing the cells (results not shown), but gating out the debris based on the strong chlorophyll fluorescence in the red channel was possible and has been done previously (4, 16).

Additional gating was required for each dye because another population appeared under certain conditions. This population was identified as precipitation of the dyes for three reasons. First, this additional population only appeared when dye was present with or without cells, and its prevalence increased with increasing concentration of the dye. Second, microscopy showed an accumulation of large red fluorescent particles with increased concentrations of dyes. Third, in water, both Nile Red and BODIPY have quenched red fluorescence (17), but aggregation of dye molecules can boost this red emission (18). The fluorescence of these events occurred in the same red channels as chlorophyll and made the precipitate difficult to gate out, especially with DiO. DiO precipitate also fluoresced in the green channel but Nile Red and BODIPY precipitates did not. Therefore, mistakenly including precipitation with cells will increase the mean green fluorescence for DiO and decrease the mean yellow fluorescence for Nile Red and green fluorescence for BODIPY. Staining of cells with these lipophilic dyes must, therefore, ensure that concentrations are sufficient to saturate the cells but be minimal to prevent precipitate formation.

The desired staining protocol would be one with negligible formation of precipitation or cell debris, whilst ensuring fluorescence is stable and reproducible. The conditions for staining also needed to have as large a margin of error as possible, such that if a parameter was increased or decreased slightly it would not affect the fluorescence. Cell concentration during staining was the most tolerant parameter, as fluorescence and population composition were unaffected by cell concentration changes when the optimal dye concentrations were used. For the other conditions, if slight fluctuations of the parameter caused significant changes in fluorescence, the dye was not considered useful for the quantification of cellular lipids.

It was apparent that DiO was not suitable for lipid quantification because changes in fluorescence appeared to be due to dye concentration rather than lipid content of the algae. A significant green fluorescence signal was only detected once precipitation events were already present. There was no range of DiO concentration, where the fluorescence of the cells was independent of the dye concentration. This was potentially due to DiO precipitating in the cells or the saturation of DiO in liposomes being greater than the concentrations tested. DiI, another member of the carbocyanine family of lipophilic dyes, was also tested and showed similar precipitation and concentration dependence (results not shown), indicating a general problem with this family of dyes. Due to the difficulty of gating DiO precipitate events and the absence of a range of concentrations where fluorescence was independent of DiO concentration, no other parameters could be examined using DiO or other carbocyanine dyes.

Maximum BODIPY fluorescence occurred at concentrations that were slightly lower than the concentration at which precipitate formed. This saturation effect was because excess dye precipitated in the buffer allowing an equilibrium to be maintained between the dye inside the cell and the dye in the buffer. However, by reaching a maximum fluorescence BODIPY is usable for quantification, but there was no range of dye concentrations, where the fluorescence was stable. It could be possible to gate out the BODIPY precipitate and increase its use with flow cytometry quantification studies, but use in spectrofluorometric assays should be avoided.

Relative fluorescence intensity for Nile Red remained independent of dye concentration within the range of 1–10 μg/mL. This concentration range is consistent with other reports that suggest Nile Red concentrations below 0.5 μg/mL were insufficient to stain all the cells (2, 15), and concentrations above 5 μg/mL may have developed precipitation effects, although they were not reported as such (2, 18). A 1 μg/mL concentration of Nile Red was used in previous studies (3, 10, 19), and for this study, 1 μg/mL was just within the reproducibly concentration independent range for Nile Red, but it would be more robust for assays to use a final concentration of 2 μg/mL.

Staining duration has varied greatly in publications, from measuring fluorescence immediately upon staining (2), incubating for 10 min (1), to monitoring for 40 min (20). The significant increase in Nile Red and BODIPY fluorescence during the first 20 min of staining indicates the dyes require time to penetrate the cells, but once in the cells the fluorescence signal was stable for 1.5 h, as was observed previously (15). Therefore, placing a set of samples on the cytometer 30 min after staining and reading them within 1 h is suggested.

The presence of organic solvent during staining cannot be avoided due to the hydrophobic nature of the lipophilic dyes and their insolubility in water. However, with the increased solubility in organic solvents and the fluorescence shift of Nile Red depending on the polarity of the environment, the variation in reported organic solvent concentration from < 0.1% to 100% (v/v) (1–3, 8, 15) must have some effect on cell staining. Instead of preventing dye precipitation, increasing concentrations of both acetone and DMSO caused cell debris to be formed. The lack of green, yellow, or red fluorescence of these debris events suggests that the organic solvents could be extracting the lipids and chlorophyll from the cells (21–23). Although this extraction might not affect a spectrofluorometric assay, it was not usable with a cytometry assay. Concentrations of acetone below 5% (v/v) did not result in debris formation but did affect the fluorescence intensity of BODIPY significantly, which would be problematic for fluorescence quantification. Only Nile Red had a range below 2% (v/v), where fluorescence was independent of acetone concentration.

After optimizing the conditions to stain microalgae with a lipophilic dye, the impact of staining conditions on the correlation of lipid content to fluorescence intensity was examined. The measured lipid content of six cultures was compared with the various fluorescence intensities of the cultures resulting from staining using different conditions. Staining with DiO showed no relationship between fluorescence intensity and lipid quantity. This reinforced the earlier decision that DiO is not appropriate for lipid quantification in microalgae. BODIPY fluorescence did have a correlation to lipid content, but the reproducibility of this correlation is questionable based on the fluorescence variability with both dye and DMSO concentration. Staining the microalgae with Nile Red under nonoptimal conditions resulted in poor correlations between fluorescence and lipid content. For Staining Method 1, the 5% acetone concentration increased the cellular debris, and the 5 μg/mL dye concentration increased the amount of precipitate. These both contributed to the fluorescence intensity plateauing for cells with high lipid contents. Even at the determined optimal Nile Red and acetone concentrations of 2 μg/mL and 1%, respectively, staining time remained critical. Measuring the fluorescence too quickly negated any correlation between fluorescence and lipid content (Method 4). When the optimal Nile Red and acetone concentration were used and the fluorescence was measured 30 min after staining, the correlation (R2 = 0.9912) between fluorescence intensity and lipid quantity was the best of any of the staining methods. The six cultures had different cell concentrations, and the strength of this correlation also shows the robustness of staining to variations in cell concentration. Therefore, optimization of staining conditions is critical to develop a correlation between fluorescence intensity and cellular lipid content, and the optimal conditions identified for Nile Red produce the best correlation.

In conclusion, all three of the lipophilic dyes tested formed a precipitate at high concentrations that fluoresced in the same channels as the chlorophyll of green microalgae. An optimized staining protocol for these dyes is required or the presence of precipitate causes errors in quantifying cellular fluorescence intensity. Rather than eliminating dye precipitation, organic solvents caused extraction of lipids and chlorophyll from the cells with increasing concentration. This extraction is detrimental to lipid quantification with cytometry, and therefore, organic solvents should be kept to a minimum when staining with lipophilic dyes. It was found that cellular DiO fluorescence was strongly dependent on staining DiO concentration, and BODIPY fluorescence was dependent on small changes in low concentrations of acetone and BODIPY. Therefore, neither DiO nor BODIPY were considered suitable for flow cytometry-based fluorescence quantification assays. The suggested protocol for green microalgae is to use a stock solution of Nile Red with a concentration greater than 200 μg/mL to stain cells at a working concentration of 2 μg/mL and ensure < 1% acetone in the sample. It is also recommended that the wait time is at least 30 min before measuring the fluorescence. With this optimized staining procedure for Nile Red, it is possible to develop a reliable correlation between fluorescence intensity and cellular lipid content.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information
  • 1
    de la Jara A,Mendoza H,Martel A,Molina C,Nordstron L,de la Rosa V,Diaz R. Flow cytometric determination of lipid content in a marine dinoflagellate, Crypthecodinium cohnii. J Appl Phycol 2003; 15: 433438.
  • 2
    Cooksey KE,Guckert JB,Williams SA,Callis PR. Fluorometric-determination of the neutral lipid-content of microalgal cells using nile red. J Microbiol Methods 1987; 6: 333345.
  • 3
    Chen W,Sommerfeld M,Hu Q. Microwave-assisted Nile red method for in vivo quantification of neutral lipids in microalgae. Bioresour Technol 2011; 102: 135141.
  • 4
    Cooper MS,Hardin WR,Petersen TW,Cattolico RA. Visualizing “green oil” in live algal cells. J Biosci Bioeng 2010; 109: 198201.
  • 5
    Sengupta P,Holowka D,Baird B. Fluorescence resonance energy transfer between lipid probes detects nanoscopic heterogeneity in the plasma membrane of live cells. Biophys J 2007; 92: 35643574.
  • 6
    Greenspan P,Fowler SD. Spectrofluorometric studies of the lipid probe, Nile red. J Lipid Res 1985; 26: 781789.
  • 7
    Dutta AK,Kamada K,Ohta K. Spectroscopic studies of Nile red in organic solvents and polymers. J Photochem Photobiol Chem 1996; 93: 5764.
  • 8
    O'Rourke EJ,Soukas AA,Carr CE,Ruvkun G. C. elegans major fats are stored in vesicles distinct from lysosome-related organelles. Cell Metab 2009; 10: 430435.
  • 9
    Loudet A,Burgess K. BODIPY dyes and their derivatives: Syntheses and spectroscopic properties. Chem Rev 2007; 107: 48914932.
  • 10
    Kacmar J,Carlson R,Balogh SJ,Srienc F. Staining and quantification of poly-3-hydroxybutyrate in Saccharomyces cerevisiae and Cupriavidus necator cell populations using automated flow cytometry. Cytometry Part A 2006; 69A: 2735.
  • 11
    Bradley J,Luo R,Otis TS,DiGregorio DA. Submillisecond optical reporting of membrane potential in situ using a neuronal tracer dye. J Neurosci 2009; 29: 91979209.
  • 12
    Waqar-ul-Haq,Zarina A,Masud-ul-Hasan,Shameel M. Taxonomic study of the family Scenedesmaceae (volvocophyta shameel) in certain north-eastern areas of Pakistan. Proc Pakistan Acad Sci 2008; 45: 2328.
  • 13
    Bold H. The morphology of Chlamydomonas chlamydogama sp. nov. Bull Torrey Botan Club 1949; 76: 101107.
  • 14
    Fine, KD,Ogunji, F. A new method of quantitative fecal fat microscopy and its correlation with chemically measured fecal fat output. Am J Clin Pathol 2000: 113: 528534.
  • 15
    Chen W,Zhang C,Song L,Sommerfeld M,Hu Q. A high throughput Nile red method for quantitative measurement of neutral lipids in microalgae. J Microbiol Methods 2009; 77: 4147.
  • 16
    Guzman HM,de la Jara Valido A,Duarte LC,Presmanes KF. Analysis of interspecific variation in relative fatty acid composition: Use of flow cytometry to estimate unsaturation index and relative polyunsaturated fatty acid content in microalgae. J Appl Phycol 2011; 23: 715.
  • 17
    Greenspan P,Mayer EP,Fowler SD. Nile red—A selective fluorescent stain for intracellular lipid droplets. J Cell Biol 1985; 100: 965973.
  • 18
    Hu R,Lager E,Aguilar-Aguilar A,Liu J,Lam JWY,Sung HHY,Williams ID,Zhong Y,Wong KS,Pena-Cabrera E, et al. Twisted intramolecular charge transfer and aggregation-induced emission of BODIPY derivatives. J Phys Chem C 2009; 113: 1584515853.
  • 19
    Wang ZT,Ullrich N,Joo S,Waffenschmidt S,Goodenough U. Algal lipid bodies: Stress induction, purification, and biochemical characterization in wild-type and starchless Chlamydomonas reinhardtii. Eukaryot Cell 2009; 8: 18561868.
  • 20
    Elsey D,Jameson D,Raleigh B,Cooney MJ. Fluorescent measurement of microalgal neutral lipids. J Microbiol Methods 2007; 68: 639642.
  • 21
    Barnes J,Balaguer L,Manrique E,Elvira S,Davison A. A reappraisal of the use of DMSO for the extraction and determination of chlorophylls-A and chlorophylls-B in lichens and higher-plants. Environ Exp Bot 1992; 32: 85100.
  • 22
    Gurtovenko AA,Anwar J. Modulating the structure and properties of cell membranes: The molecular mechanism of action of dimethyl sulfoxide. J Phys Chem B 2007; 111: 1045310460.
  • 23
    Radin NS. Lipid extraction. In: Boulton AA,Baker GB,Horrocks LA, editors. Lipids and Related Compounds. Totowa, NJ: The Humana Press Inc.; 1989. pp 161.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. LITERATURE CITED
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
CYTO_22066_sm_SuppFig1.tif1309KSupplemental Figure 1. Emission spectra of lipophilic dyes and cells. The fluorescence of 0.5 x 106 S. dimorphus cells/mL was measured with and without 1 µg/mL of each lipophilic dye. The fluorescence of each lipophilic dye was measured at 10 µg/mL when dye precipitation was occurring. Excitation was at 488 nm. Arrows indicate the wavelengths of the channels used during flow cytometry.
CYTO_22066_sm_SuppFig2.tif14876KSupplemental Figure 2. Dot plots showing the gating strategy for cells, debris, and precipitate. These bivariate dot plots show representative S. dimorphous data of unstained cells and cells stained with 10 µg/mL of Nile Red, BODIPY 505/515, or DiO-C18. The top panels of far red (670 – 735nm) versus infrared (750-810nm) fluorescence show the gating strategy used to isolate the precipitate and debris from the cellular events. There is clear separation of some of the precipitate from the cells, but it is not complete. The use of FSC, in the bottom panels, does not show any clear separation between the precipitate and cellular events. Blue dots represent debris, red dots represent precipitate, and green dots are cells.
CYTO_22066_sm_SuppFig3.tif525KSupplemental Figure 3. Top panels show dye fluorescence was not significantly affected by increasing the concentration of cells during staining with 1 µg/mL of lipophilic dye. There was some variation due to the need to change flow rates with different concentrations of cells. Bottom panels show the composition of the population that produced the fluorescence signal. Only at the lowest and highest cell concentrations was there a slight decrease in the percentage of cell (□) events and increase in the percentage of debris (•) precipitate (×) events. Nile Red fluorescence was measured in the yellow channel, while BODIPY fluorescence was measured in the green channel. Percentage of events data are shown as mean and standard deviation and relative fluorescence data are shown as mean and 95% confidence interval relative to unstained cells.
CYTO_22066_sm_SuppInfo.doc55KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.