In vitro tests are faster, easier to perform and to quantify, usually cheaper, and allow studies of isolated steps [1
]. In vitro, chemical substances such as drugs and pesticides have various cytotoxic mechanisms such as destruction of cell membranes and prevention of protein synthesis [2
]. To identify cell death caused by these toxicities, the fields of toxicology and pharmacology require inexpensive, reliable, and reproducible short-term cytotoxicity and cell viability assays.
In particular, in the case of a harmless method for assessing the viability of a target cell, more detailed information can be obtained by long-term monitoring of the viability. Furthermore, since it is not necessary to prepare the same number of samples for viability assay, the cost and labor can be reduced, and also the problem of non-uniformity between samples does not occur. Therefore, there should be a potential demand for a long-term, noninvasive, safe viability assay of the same sample.
The following conventional methods to distinguish between live and dead cells in culture have been established:
Dye exclusion test (DET)
Cells stained with a synthetic dye such as trypan blue (TB) are judged as dead cells [3
Colony formation assay
The number of live cells is evaluated by the number of colonies formed on an agar culture after an inoculation of diluted cell suspension and following a definite time of culture [4
Enzyme activity assay
Enzymatic reaction in living cells or enzymes leaking out of dead cells are used for viability assay [5
Flow cytometry analysis
Dead cells are labeled with a fluorescent dye [6
] and detected by fluorescence flow cytometry [7
The dead or living state of cells is diagnosed by deflection change of the probe light beam [9
The positions of alternative electro-orientation of living or dead cylindrical yeast cells are dependent on the applied frequency [10
However, these methods have difficulties such as requiring specialized techniques and equipment, damaging cells, inability to perform in situ measurement in the cultivation process, or only applicable for cells of a specific shape (non-spherical) [9
]. Therefore, in order to solve these problems, we propose a DET using natural edible pigments that are safe for humans and minimally invasive to cells, as an alternative to a DET using conventional synthetic dyes that are toxic to humans and invasive to cells.
Trypan blue (TB), a vital dye, has been conventionally used for the DET. TB is a widely used diazo dye for selectively coloring dead cells or tissues. The mechanism for TB to stain dead cells is based on its negative charge which prevents the incorporation of it into the living cells with the membrane negatively charged. Therefore, living cells are not stained, but dead cells with the compromised cell membrane are stained by TB [12
]. However, TB is known to cause cell health and environmental problems due to its potential teratogenic effects [15
]. It has also been pointed out that pore formation in cell membranes is possibly induced to increase membrane permeability [13
]. As other dyes for the DET, methylene blue [17
], eosin [19
], nile blue [20
], and amethyst violet [21
] have been used but it is known that the selective permeability of the plasma membrane is destroyed or severely impaired [3
], indicating that these dyes are toxic for cells.
In order to avoid these problems, six natural food pigments (Monascus, purple sweet potato, yellow gardenia, green gardenia, red beet, and Spirulina) and a traditional synthetic dye (TB) for viability assay are tested in this study, among which Monascus pigment (MP) is of particular interest. MP is extracted from the Monascus sp. which is a kind of filamentous fungus.
MP exhibits a red color due to a molecule of the pigment whose main component is monascorubrin [22
]. It does not have a significant pH dependence of the color (although it tends to precipitate in acidic solution), has excellent stainability to proteins, and is relatively stable to heat [22
]. MP is not easily affected by metal ions other than copper [23
], whereas it is unstable against light irradiation, especially in an acidic condition. MP has been used for more than 1000 years as a food pigment and a folk medicine in China because an efficient production method was established using fermentation of rice. MP is a cost-effective and reproducible substrate, has variation in colors, is highly safe, and shows good solubility in water and ethanol [24
]. Moreover, it possesses biological activities such as anticancer properties, anti-mutagenic activity, antibacterial activity, and potential anti-obesity activity [26
]. Regarding the quality assurance of MP, it is based on the component standards in the Japan’s Specifications and Standards for Food Additives (JSFA) [27
]. The details are described in the article of the example applied to Paramecium
Animal cells for various pathological experiments do not have motility or characteristic pigments, making it difficult to judge using only visual microscopic observation. The breast cancer cell line, MCF-7 cells, which is widely available as an in vitro model for cancer research, has been used [29
]. Cisplatin (cis-diamminedichloroplatinum (II)) is a widely used chemotherapeutic drug for the treatment of various types of cancer, including breast cancer [30
]. In particular, MCF-7 cells are widely used in studies of estrogen receptor (ER, positive breast cancer cells [32
To verify that this method is applicable to a wide variety of cells, first, we examined whether natural pigments are not toxic to human breast cancer cells MCF-7 cells in the presence or absence of cisplatin and if they can be replaced with conventional synthetic dyes for viability assay. Then, the effect of natural pigments on cells with different properties was investigated, and the results were compared between the unicellular green alga Euglena
] and the protozoan Paramecium
], which differ in cell structure.
3.1. Confirmation of Staining by Natural Pigment
Six natural food pigments (Monascus
, purple sweet potato, yellow gardenia, green gardenia, red beet, and Spirulina
) were added to the cell suspension (1%). For each sample with pigment added, a sample with cisplatin added (20 μM) and a control sample with no pigment added were prepared. The pigments that could be easily visually discriminated by bright field microscope observation after 12 h, and in which more cells were stained in the cisplatin-added sample than in the control sample, were made usable as a reagent for viability assay. Trypan blue was determined to be usable because we confirmed that it was a conventional reagent for viability assay and that it could be usable by other tests (Figure 2
, Figure 3
and Figure 5
). The data are shown in Supplementary Material
, and the results are summarized in Table 3
As a result of DET with six types of natural pigments, it was found that only MP could be used as a reagent for viability assay. A characteristic of MP that is different from other natural pigments is strong staining of intracellular proteins. The mechanism of staining with TB also results from binding to intracellular proteins [14
]. In the preceding papers on DET of Euglena
, long-term viability could be determined with anthocyanin pigments (purple sweet potato and red cabbage dye) in addition to MP [28
]. This was thought to be due to differences in cell structure. Therefore, it was suggested that high affinity for the intracellular protein is a more suitable dye condition for DET.
3.2. Magnification of Observation and Staining of Dead Cells
Cell death of MCF-7 cells was induced by microwave exposure. After discarding the medium, dead cells were treated with MP (1 or 5%) for 20 min, and then washed with PBS. Cells treated with no pigment were used as the “control”. Cells on the bottom of the dish were observed with a bright-field microscope (4× or 20× objective lens).
From Figure 1
, one can visually confirm the staining of the MW treated dead cell when observing the 1% MP with a 4× objective lens. In the observation of 5% MP dead cells with a 4× objective lens, even deeper staining was obtained, and it was found that in this concentration range, dead cells were deeply stained in proportion to the MP concentration. In addition, observation with a 20× objective lens enabled us to distinguish individual stained cells with higher visibility. However, with respect to living cells, at a concentration of 0.6%, although the morphology of the cells did not change in 24 h, the results showed that the cells deformed into spheres in 48 h (data not shown). By contrast, no morphological change was observed at a concentration of 0.4%. Therefore, in the case where MP is used instead of TB in conventional DET without performing long-term monitoring of viability assay, it is advantageous to use high concentration for the purpose of enhancing visibility. Nevertheless, when performing long-term monitoring of viability assay, it is necessary to adjust the MP concentration to 0.4%.
3.3. Confirmation of Staining of Dead Cells
Cell death of MCF-7 cells was induced by microwave exposure. After discarding the medium, dead cells were treated with 0.4% MP or 0.1% TB for 20 min, and then washed with PBS. Cells treated with no pigment were used as the “control”. Cells on the bottom of the dish were observed with a bright-field microscope (20× objective lens). Because of the light red background, it looks as if the cells are not clearly stained, however, staining of dead cells was sufficiently confirmed by visual observation with a microscope.
shows the result of confirming the staining of MW treated dead cells by 0.4% MP, and staining was clearly confirmed against the “control” containing no pigment. Clear staining was confirmed also with 0.1% TB.
3.4. Immersion Time and Staining of Dead Cells
Cell death of MCF-7 cells was induced by microwave exposure. After discarding the medium, dead cells were treated with MP (1 or 5%) for the indicated times, and then washed with PBS. Cells treated with no pigment were used as the “control”. Cells on the bottom of the dish were observed with a bright-field microscope (20× objective lens). Because of the light red background, it looks as if the cells are not clearly stained, however, staining of dead cells was sufficiently confirmed by visual observation with a microscope.
shows the state of staining of MW treated dead cells at each elapsed time from the addition of the pigment to the cell suspension. On the one hand, in the 0.4% MP, almost all cells were stained in 5 min, and staining of all cells was confirmed in 10 min. The time until this dyeing is within a practical range. On the other hand, almost all cells of the TB (0.1%) were stained in 1 min. There is no rigorous comparison of visibility, but TB is at a higher concentration, considering that the “pure MP” concentration is 5% of the amount of MP added to the medium (i.e., pure MP concentration is 0.02%). Therefore, it is reasonable to judge that the higher pigment concentration in TB resulted in a shorter staining time.
3.5. Staining of Cells in a Mixed Solution of MP and Cisplatin
The MCF-7 cells were treated with MP (1 or 5%) in the presence or absence of cisplatin for 14 h. (a) Cells on the bottom of the dish and (b) floating cells in the medium were washed with PBS. Cells treated with no pigment were used as the “control”. Cells were observed with a bright-field microscope (20× objective lens).
MP stained dead cells when 20 μM cisplatin was present in the live cell suspension, as well as when it was not (Figure 4
). Living MCF-7 attached to the bottom of the dish, whereas dead cells tended to float away from it. The detachment of dead cells in MCF-7 could be due to the weakening of cell adhesion due to cleavage of β-catenin because β-catenin forms a complex with cadherin involved in cell adhesion [37
]. Regardless of the presence or absence of cisplatin, the “control” containing no pigment and the sample with 1% MP showed very few floating cells, and 5% MP showed many cells. Regarding cell staining, at 24 h and 48 h, 1% MP stained only floating cells, and 5% MP stained both cells on the dish and floating cells, regardless of the presence or absence of cisplatin. Therefore, it was confirmed that MP was not affected by the presence of cisplatin for 48 h and stably functions as a reagent of viability assay. This is thought to be due to the fact that MP is less affected by metal ions other than copper [23
], and the platinum complex cisplatin did not affect staining. High concentrations of MP were toxic to MCF-7, as almost all cells were stained with 5% MP, in the presence or absence of cisplatin. From the previous report [38
], the cell proliferation of MCF-7 at MP concentrations of 0.001%, 0.01%, and 0.1% were 100%, 140%, and 0%, respectively. The cell proliferation values shown here are read from the graphs in the report. When 1% and 5% of the MP concentration in Figure 4
are converted into “pure MP concentration”, they are 0.05% and 0.25%, respectively. Therefore, the tendency about the MP concentration and the health condition of the cell coincide.
3.6. Confirmation of Noninvasiveness of Pigment
The MCF-7 cells were treated with 0.4% MP or 0.1% TB for 47 h. (a) Cells on the bottom of the dish and (b) floating cells from the medium were washed with PBS. Cells treated with no pigment were used as the “control”. Cells were observed with a bright-field microscope (20× objective lens). Due to the light red background, it looks as if the cells are not clearly stained, however, staining of dead cells was sufficiently confirmed by visual observation with a microscope.
shows a bright-field microscopic image of the living cells after 47 h at the MP concentration (0.4%) at which the viability assay can be performed. Almost all cells attached to the bottom of the dish were not stained by 0.4% MP. This was the same for 0.1% TB. The number of cells at 0.4% MP was increased against the “control” containing no pigment, while that at 0.1% TB was comparable to that of the “control”. The number of floating cells was almost the same as that of the “control” for both pigments, and the number was small. In addition, the staining of floating cells is not so clear for both pigments as compared with the cells attached to the bottom of the dish. In particular, the 0.4% MP floating cells did not appear to be stained, but this was due to the defocus, and it was confirmed that they could be visually discriminated.
3.7. Immersion Time and Staining of Dead Cells
The MCF-7 cells were treated with 0.4% MP or 0.1% TB for (a) 24 h or (b) 48 h. Then, the WST-8 assay was performed. Cells treated with no pigment were used as the “control”. Samples were measured for absorbance (450 nm) and the average value of “the absorbance ± standard error (n
= 6)” was calculated. Differences between values were analyzed using a Tukey test (* p
< 0.05 and ** p
< 0.01). The results of the Tukey test are shown in Figure 6
. Although no significant decrease in cell number was observed at 0.1% TB, the absorbance shows significantly lower values than the “control” and 0.4% MP.
Since 0.4% MP was the upper limit concentration without cell morphological change, the toxicity test was conducted only at this concentration. The WST-8 (2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium) assay was performed at 24 h and 48 h after pigment addition to examine the health of live cells in the dye solution (Figure 6
). At both times, the absorbance of 0.1% TB was significantly lower than the other samples, suggesting a decrease in cell number or a decrease in enzyme activity (Figure 6
). Microscopic observation showed that the number of cells in 0.1% TB was comparable to that of the other samples (control, 0.4% MP), and that almost no cells were stained, as shown in Figure 5
. Therefore, on the one hand, the enzymatic activity is considered to be remarkably suppressed in the living cells of 0.1% TB. On the other hand, 0.4% MP is noninvasive to cells due to its higher absorbance than the “control”.
As a result of DET using six types of natural pigments, only MP stained dead cells of MCF-7 with high visibility. A unique property of MP is that it has a high affinity for intracellular proteins. Since MP can be applied to DET even in Euglena
, the abovementioned properties of MP were expected to be versatile in that they could target more cells of different species. The MP concentrations shown below are that of the commercial products and the concentration of pure MP is 5% of that. By adjusting the MP concentration to 0.4% (pure MP concentration 0.02%), all the MW treated dead cells were stained within 10 min after the addition of the pigment, and long-term viability assay over 48 h was realized. At this concentration there is no morphological change of the cells and no inhibition of cell growth or reduction of NADH, so it is noninvasive. Dead cells became more intense and the visibility increased in proportion to the concentration of MP. In fact, almost all cells were stained at 1% (Figure 1
and Figure 4
). In this case, long-term noninvasive DET could be difficult because the morphological change occurred on the second day at 0.6%, but it would have the advantages of higher safety and less environmental impact than conventional DET with TB. Furthermore, when the MP concentration is 0.4% and the TB concentration is 0.1%, the cost of MP is estimated to be about 1/10 of TB, which is economical. As a similar method, a technique with erythrosine B (EB, also known as Red No. 3) used as a food additive was developed [39
]. This synthetic colorant is an edible tar dye that does not pass through biological membranes and is compatible with automatic cell counters. Because EB was suspected of being carcinogenic, the FDA (Food and Drug Administration) banned the use in food (1990) [40
]. Although the ban on EB was removed afterwards, most of the EB in the USA has been replaced with Allura Red (also known as Red 40). However, Allura Red is banned in many European countries because it is an azo dye [42
]. Therefore, it is difficult to monitor the noninvasive viability assay of a synthetic food colorant over a long period of time because of the concern regarding the influence on the human body. At a concentration of TB of 0.1%, no increase in dead cells was observed, but NADH activity was significantly suppressed, indicating that a heavy load was imposed on the cells. Therefore, MP was confirmed to be less invasive and less costly than TB. In addition, even when a metal complex anticancer agent such as cisplatin coexists, stable staining of dead cells is maintained. For example, a further practical use of this technique is expected in cooperation with a technique for distinguishing normal cells from cancer cells based on the difference in staining properties of edible natural or synthetic dyes under a multiphoton laser microscope [44