A comparison of TO-PRO-1 iodide and 5-CFDA-AM staining methods for assessing viability of planktonic algae with epifluorescence microscopy

https://doi.org/10.1016/j.mimet.2012.03.005Get rights and content

Abstract

Two fluorescent dyes, TO-PRO-1 iodide and 5-CFDA-AM, were evaluated for LIVE/DEAD assessment of unicellular marine algae Brachiomonas submarina and Tetraselmis suecica. Epifluorescence microscopy was used to estimate cell viability in predetermined mixtures of viable and non-viable algal cells and validated using microplate growth assay as reference measurements. On average, 5-CFDA-AM underestimated live cell abundance by ~ 25% compared with viability estimated by the growth assay, whereas TO-PRO-1 iodide provided accurate viability estimates. Furthermore, viability estimates based on staining with TO-PRO-1 iodide were not affected by a storage period of up to one month in − 80 °C, making the assay a good candidate for routine assessment of phytoplankton populations in field and laboratory studies.

Highlights

► We compared TO-PRO-1 iodide and 5-CFDA-AM for LIVE/DEAD assessment of algae. ► Using radiation-killed cells, test mixtures with 0–100% dead cells were prepared. ► Viability was estimated with epifluorescence microscopy and microplate growth assay. ► Whereas TO-PRO-1 provided accurate estimate, CFDA underestimated viability by ~ 25%. ► Staining with TO-PRO-1 was not affected by a storage for ≤ 1 month in − 80 °C.

Introduction

Determining the viability of algae has applications in many areas of aquatic ecology, ecotoxicology and applied microbiology, including public health, water industry and biotechnology. Various environmental factors (e. g., UV radiation: Sinha et al., 2001; light and nutrient availability: Berges and Falkowski, 1998, Brookes et al., 2000), infections (reviewed by Brussaard, 2004), and physiological processes involved in natural, internally driven cell death (Franklin and Berges, 2004), drive phytoplankton mortality in nature, including collapsing blooms and other mass-death events. Therefore, development of various, mostly flow cytometry-based techniques to distinguish viable and dead algal cells in natural populations has attracted considerable attention (Dorsey et al., 1989, Adler et al., 2007). Moreover, ecotoxicological studies and applied issues, such as testing efficacy of biological decontamination of ship ballast water to prevent transport of unwanted species (i.e., biological invasions; Binet and Stauber, 2006, Steinberg et al., 2011) and evaluating physiological state of algal cultures in laboratory and industrial cultivations (Franklin et al., 2004), emphasize a great need to develop fast and reliable detection methods for algal viability.

Recent advances in staining technology provide great tools for assessing functional capacity of cells. Today, numerous fluorescent viability probes that detect changes in the metabolism of both eukaryotic and prokaryotic cells using flow cytometry or epifluorescence microscopy are available and continue to be developed (Molecular Probes Handbook; Molecular Probes Inc., Eugene, OR). The existing approaches for cell viability evaluation are largely based on two principles: membrane integrity and metabolic activity measurements (Veldhuis et al., 1997, Rijstenbil et al., 2000, Brussaard et al., 2001, Agusti and Sanchez, 2002, Binet and Stauber, 2006). The most common stains for identifying dead cells are membrane impermeant nucleic acid stains, because the integrity of cytoplasmic membrane is important in determining which molecules enter or leave the cytoplasm. Probes used in the determination of cytoplasmic membrane integrity are fluorescent only when bound to nucleic acids, and thus, cell membrane permeability characterizes necrotic and advanced apoptotic cells (Berges and Falkowski, 1998, Binet and Stauber, 2006). A possible problem with this technique is the detection of injured cells which have temporarily compromised membranes (Duffy et al., 2000). Non-vital or mortal stains that have been found useful for algal viability analysis are, for example, SYTOX-Green (Franklin and Berges, 2004, Timmermans et al., 2007) and TUNEL (terminal-deoxynucleotidyl-transferase-mediated dUTP nick end labeling) assay (Franklin and Berges, 2004). Another fluorochrome that has been used for detection of non-viable algal cells is TO-PRO-1 iodide (Okochi et al., 1999), a monomeric cyanine dye with a single cationic side chain and low base selectivity, which enters cells with compromised membrane and binds to nucleic acids resulting in a bright green fluorescence. Although generally used as a DNA electrophoresis stain, it possesses the chemical characteristics necessary for a viability probe (Molecular Probes Handbook; Molecular Probes Inc., Eugene, Oregon). Moreover, it performed exceptionally well for diatoms (Okochi et al., 1999) that have been reported as challenging for staining (Garvey et al., 2007). The utility of this and analogues cyanine dyes has been demonstrated in microscopic and flow cytometric LIVE/DEAD assays with various algae (Li et al., 1995, Marie et al., 1996).

For staining live cells, non-polar substrate molecules that can be passively loaded into cells are recommended, such as the most commonly used esterified fluorogenic substrates fluorescein diacetate (FDA) and carboxyfluorescein diacetate (CFDA), and their derivatives. As compared with FDA, CFDA contains extra negative charges and is therefore better retained in cells. One of the commonly used derivatives is carboxyfluorescein diacetate acetoxymethyl ester of CFDA (5-CFDA-AM), which is electronically neutral and can be loaded into cells at lower concentrations than CFDA. As other CFDA derivatives, this is a lipophilic substrate moderately permeant to most cell membranes (Haugland, 2006). Once inside the cell, diacetate is cleaved from the molecule by intracellular non-specific esterases, producing fluorescent carboxyfluorescein (CF) which is efficiently retained by live cells with intact plasma membrane. Hence, the conversion to CF by the cells indicates the integrity of the plasma membrane, since only an intact membrane can maintain the cytoplasmic milieu which is needed to support esterase activity. The FDA- and CFDA-based assays have been used to evaluate physiological activity of phytoplankton and plant cells in culture (FDA: Selvin et al., 1988, Steward et al., 1999, Garvey et al., 2007; CFDA: Lee and Rhee, 1997, Latour et al., 2004). However, as reviewed by Agusti and Sanchez (2002) and Garvey et al. (2007), application of vital fluorescent stains for phytoplankton cells is complicated by (1) interference between stain fluorescence and autofluorescence (e.g., by chlorophyll), (2) ambiguities in establishing the threshold staining intensities that separate live from dead cells in the continuum from brightly stained to unstained cells of varying metabolic state, (3) difficulties in the identification of the stained cells within mixed assemblages, composed of both autotrophic and heterotrophic cells, and (4) high between-species variability in staining efficiency.

Unfortunately, studies comparing performance of different fluorochromes, in particular those from different functional groups, for algal viability assessment are rare as well as proper calibration of staining-based estimates with an independent method of mortality assessment. Such intercalibrations have been done for heterotrophic bacteria (e.g., Fuller et al., 2000), but to the best of our knowledge, not for unicellular algae. Clearly, to facilitate method choice, interpret results and enable between-study comparisons, a comparative analysis of different methods for obtaining viability estimates is needed.

Our objective was to evaluate two fluorescent dyes, TO-PRO-1 iodide and 5-CFDA-AM, for algal viability assessment using unicellular marine algae Brachiomonas submarina Bohlin and Tetraselmis suecica Kyling as test species. The two dyes tested employ independent cellular criteria: membrane integrity (TO-PRO-1 iodide, 5-CFDA-AM) and cell enzymatic activity (5-CFDA-AM). Both test species possess a cell wall, which is delicate in B. submarina and robust in T. suecica (Domozych et al., 1981). As cell wall thickness constitutes a major problem when using staining of intracellular components, this choice of test species facilitates method evaluation. For both dyes and both species, estimates obtained by staining were calibrated using predetermined mixtures of viable and non-viable cells and validated using cell growth assay as reference measurements. In addition, effects of storage time for viability estimates using refrigerated and frozen samples were evaluated for both staining techniques.

Section snippets

Algal cultures

Algae T. suecica (Prasinophyceae; CCMP908; The CCMP National Center for Culture of Marine Phytoplankton, USA) and B. submarina (Chlorophyceae; CCAP 7/1A; Culture Collection of Algae and Protozoa, CCAP, UK) were grown in f/2 medium with constant illumination (90 μE cm 2 s 1) at 15 °C and 18‰ salinity in artificial seawater (ASW; Instant OceanTM, Aquarium Systems). Cultures were maintained in extended exponential growth through a semi-continuous harvesting regime (~ 30% exchange every other day). The

Calibration of staining methods using predetermined mixtures

There were significant linear relationships between the nominal viability of cells in the mixtures containing different proportions of live and dead cells and the viability assayed by both TO-PRO-1 iodide and 5-CFDA-AM (Table 1A). The slopes of the regressions for TO-PRO-1 were not significantly different from 1, while those for 5-CFDA-AM were (Table 1A). The regression slopes and intercepts were not different between the algal species stained with the same dye (TO-PRO-1 iodide: F1,6 = 0.861, p > 

Discussion

Through the use of a predetermined mixture of viable and radiation-killed cells, the performance of two fluorescent dyes for viability assessment was evaluated and the relationships between the two methods estimating percentage of non-viable cells in the mixture were established. Whereas both dyes captured the gradient in cell viability in both algae (Table 1), with a high correlation between the two methods (R2 > 0.97; Fig. 1), the overall correspondence in cell viability between the growth

Acknowledgments

We thank S. Svensson (Systems Ecology, Stockholm University) for technical assistance and Prof. M. Harms-Ringdahl (Department of Genetics, Microbiology and Toxicology, Stockholm University) for his kind permission to use γ-radiation facilities. This research was supported by Agricultural Sciences and Spatial Planning (Formas), AlfaLaval AB and Wallenius Water AB (Sweden). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References (37)

  • C.P.D. Brussaard et al.

    Flow cytometric analysis of phytoplankton viability following viral infection

    Aquat. Microb. Ecol.

    (2001)
  • H.M. Davey et al.

    Red but not dead? Membranes of stressed Saccharomyces cerevisiae are permeable to propidium iodide

    Environ. Microbiol.

    (2011)
  • D.S. Domozych et al.

    Development of cell wall in Tetraselmis: role of the Golgi apparatus and extracellular wall assembly

    J. Cell Sci.

    (1981)
  • J. Dorsey et al.

    Rapid analytical technique for the assessment of cell metabolic activity in marine microalgae

    Cytometry

    (1989)
  • G. Duffy et al.

    A membrane-immunofluorescent-viability staining technique for the detection of Salmonella spp. from fresh and processed meat samples

    J. Appl. Microbiol.

    (2000)
  • D.J. Franklin et al.

    Mortality in cultures of the dinoflagellate Amphidinium carterae during culture senescence and darkness

    Proc. R. Soc. B

    (2004)
  • N.M. Franklin et al.

    Development of multispecies algal bioassays using flow cytometry

    Environ. Toxicol. Chem.

    (2004)
  • M.E. Fuller et al.

    Development of a vital fluorescent staining method for monitoring bacterial transport in subsurface environments

    Appl. Environ. Microbiol.

    (2000)
  • Cited by (30)

    • Kinetics of biofilm formation by pathogenic and spoilage microorganisms under conditions that mimic the poultry, meat, and egg processing industries

      2019, International Journal of Food Microbiology
      Citation Excerpt :

      Intracellular non-specific esterases hydrolyze the internalized CFDA, producing carboxyfluorescein, which is retained by live cells with an intact plasma membrane. Hence, the conversion to carboxyfluorescein by the cells indicates the integrity of the plasma membrane and the esterase activity (Gorokhova et al., 2012). Although SS and PP are not autofluorescence materials at the range used (450–490 nm) (Jun et al., 2010), TSB and EY residues emit fluorescence (Lee et al., 2008; Yang et al., 2017), explaining the difficulty in discerning the bacterial cells in the last biofilm formation stages.

    • Biofilm formation by Staphylococcus aureus and Salmonella spp. under mono and dual-species conditions and their sensitivity to cetrimonium bromide, peracetic acid and sodium hypochlorite

      2018, Brazilian Journal of Microbiology
      Citation Excerpt :

      Before and after disinfections treatment all coupons of SS and PP were stained with 5(6)-carboxyfluorescein diacetate (CFDA, 10 μg/mL), and dried in a level II cabinet; the CFDA excess was rinsed with SDS. The coupons were observed in a Nikon Eclipse E400 Epifluorescent Microscope using 100× oil immersion lens and filter BA 515 B-2ª at 450–490 nm; at least eighteen fields were observed.35 With this technique, it is possible to observe the presence of metabolically active living cells, the CFDA is a lipophilic substrate that is moderately permeable to most cell membranes.35

    • Genotoxicity of ferric oxide nanoparticles in Raphanus sativus: Deciphering the role of signaling factors, oxidative stress and cell death

      2016, Journal of Environmental Sciences (China)
      Citation Excerpt :

      This cellular conversion indicates the integrity of the plasma membrane. Since only an intact membrane prevents leakage of the polar dye into the medium and maintain cytoplasmic milieu, which is needed to support esterase activity (Gorokhova et al., 2012). An elevated level of esterase in our study is supported by similar responses in Lemna minor plants exposed to lead, cadmium, chromium, zinc, copper and mercury.

    • Flow cytometric applicability to evaluate UV inactivation of phytoplankton in marine water samples

      2015, Marine Pollution Bulletin
      Citation Excerpt :

      Working with stains, handling and preparation procedures can affect cells and lead to low fluorescence intensity. Microscopy analysis may underestimate live CFDA-AM stained T. suecica cells compared to microplate growth assay due to lowered metabolic activity and production of coatings caused by stress like sample handling (Gorokhova et al., 2012). To avoid stress induced cellular changes in our study, T. suecica were stained in growth medium without any centrifugation or washing steps.

    View all citing articles on Scopus
    View full text