Applying XTT,WST-1,and WST-8 to human chondrocytes: A comparison of membrane-impermeable tetrazolium salts in 2D and 3D cultures

Abstract

BACKGROUND:

Tetrazolium-based assays are optimized to assess proliferation/toxicity of monolayer or suspension cells in microtiter plates. With regard to tissue engineering and regenerative medicine the need for in vivo like 3D microtissues has an increasing relevance. Applying tetrazolium-based assays to 3D culture systems is technically more challenging. The composed microenvironment may influence the assay standards, e.g. equal distribution of tetrazolium.

OBJECTIVE:

Evaluation of membrane-impermeable tetrazolium salt-based assays with regard to spheroid culture (3D) of human chondrocytes.

METHODS:

Chondrocytes were isolated from human articular cartilage. XTT, WST-1, and WST-8 were applied to monolayer cells (2D, varying cell numbers) and spheroids (3D, different sizes) in 96well plates. Formazan formation was measured spectrophotometrically after different incubation periods. Evaluation was done using phase contrast microsopy (toxicity), analyzing the correlation of cell number and absorbance signals (Gompertz function), and document signal over background ratio.

RESULTS:

In monolayer culture the assays showed a correlation between seeded cell numbers and absorption data. Spheroid sizes are directly related to the starting cell number. A correlation between size and absorbance was only detectable starting from 10,000 cells/aggregate. Phase contrast microscopy of monolayer cells revealed strong toxicity effects of the WST-1 (4 h) and XTT (8 h) assay and no signs of toxicity using WST-8.

CONCLUSION:

The WST-8 assay is non-toxic and revealed the highest sensitivity in comparison to the XTT or WST-1 assay. There is evidence, that only cells of the outer rim of spheroids are able to convert membrane-impermeable tetrazolium salts to formazans.

Keywords

XTT WST-1 WST-8 human chondrocytes spheroid 3D culture 2D culture

1 Introduction

Articular cartilage covers the opposing bones in joints and is a highly specialized tissue whose remarkable properties of deformability, resistance to mechanical loading, and low-friction gliding are essential to joint function. However, defects resulting from traumata (e.g. acute physical injury) or osteoarthritis (OA) have a very limited ability to heal [1]. Besides traditional surgical procedures to repair these defects with moderate success, cell based therapies to regenerate cartilage defects are emerging [2 –4]. Human chondrocytes isolated from cartilage biopsies are a common source to generate cell-based implants (cell suspensions or in vitro formed tissues) for tissue regeneration of cartilage defects [5]. It remains still challenging to engineer tissue-like implants with biophysical, biochemical, and ultrastructural characteristics similar to that of native articular cartilage. Therefore, an important research demand for 3D spheroid culture and scaffold-based regeneration is to keep chondrocytes viable and differentiated [1]. Thus, monitoring proliferation, viability and toxicity are a major challenge in 2D and 3D chondrocyte-based research and therapeutical approaches. The determination of cell viability, proliferation and cytoxicity has become a key technology in a wide variety of biological approaches. The need for easy to use, fast, sensitive, quantitative, reliable and automated methods led to the development of different standard assays [6].

An important, accurate but time-consuming technique to quantify cell amount and cell proliferation is the direct cell counting using a hemocytometer. Simple and rapid methods include the incorporation of DNA synthesis markers like radiolabeled thymidine (³H-thymidine) into DNA [7]. Major disadvantages of this technique are the need to work with radioactive compounds and the production of radioactive waste. In addition, the incorporation of radioactive isotopes into cells during division may lead to DNA damage, cell cycle arrest, and cell apoptosis, with erroneous results. Another end-point method is based on the incorporation of 5-bromo-2’-deoxyuridine (BrdU) instead of thymidine to monitor DNA synthesis and cell proliferation. Detection is via immunohisto- and immunocytochemistry [7, 8]. Besides using extrinsic markers also intrinsic markers could be identified via specific antibodies, e.g. the proliferative cell nuclear antigen (PCNA) or the Ki67 protein. Techniques based on flow cytometry or the use of cell counters can provide an exact and highly reproducible quantification of cells but with excessively high cost of consumables. Furthermore, some of these techniques are not applicable to 3D cultures. For these reasons microculture tetrazolium assays are the technique of choice to estimate the number of viable cells using multiwell plates [9, 10]. The advantages of these methods are cheap, effective and simple real-time processes which require neither washing nor harvesting of cells. The complete assay from the onset of the microculture to spectrophotometrical data analysis by microplate readers is performed in the same 96well plate. An easy on-line computer processing consisting of data collection, calculation and report generation allows an automated handling of a high number of samples.

The unique chemical and biological properties of the assay components depend on the positively charged quaternary tetrazole ring core containing four nitrogen atoms. This central structure is surrounded by three aromatic groups that usually involve phenyl moieties (Fig. 1). Following mild reduction, tetrazolium structures convert from colourless or weakly coloured salts into brightly coloured formazan products by disruption of the positively-charged quaternary tetrazolium ring [11] (Fig. 2). Water soluble formazans of the XTT, WST-1, and WST-8 assays can be directly quantified using a scanning multiwell spectrophotometer. The amount of formazan dye generated is directly proportional to the number of living cells and the incubation time, and can be monitored by the absorbance [11]. Berridge et al. 2005 proposed a new mechanism of cellular reduction of tetrazolium salts with a net negative charge like XTT [12], WST-1 [13], and WST-8 [14]. These products are cell-impermeable [11] due to the two negative sulfonated groups and therefore require a photochemically stable intermediate electron acceptor, like 1-Methoxy-PMS [11, 15], to facilitate NADH/NADPH reduction [13] (Fig. 1). Electrons are transported from intracellular NADH to extracellular oxygen or PMS via an NADH:oxidoreductase at the inner surface of the plasma membrane. PMS radicals reduce e.g. WST-1 to a WST-1 radical and then to WST-1H₂ either by PMS radicals or superoxide [16].

Fig.1

Schematic model of the mechanism of cellular reduction of XTT, WST-1 and WST-8 substrates. All three tetrazolium salts are reduced by trans-plasma membrane electron transport via the intermediate electron carrier 1-Methoxy-PMS to a colored formazan. The main cellular reductive agent is NADH/NADPH. Due to the negative net charge of all three tetrazolium salts, these reagent compounds are unable to enter the cells.

Fig.2

Chemical structures of the tetrazolium salts XTT, WST-1 and WST-8 and of the intermediate electron acceptor 1-Methoxy-PMS.

The application of tissue spheroids are expanding, e.g. as in vitro tissue models in research and preclinical studies [17, 18], as in vitro platforms for testing drug effects and delivery systems [19], for cell-based therapies in regenerative medicine [20] or for spheroid-based bioprinting of human tissue and organ constructs [21, 22]. This implicates an increased requirement for a scalable production of uniformly sized multicellular aggregates and the necessity to determine the cell number of 3D aggregates. A special cell morphology and microenvironment paralleled by differences in cell behavior makes the analytics of spheroids quite challenging.

The aim of this project was to test three different microculture tetrazolium assays, the Cell Proliferation Kit II (XTT), the Cell Proliferation Reagent WST-1 and the Orangu^TM Cell Counting Solution (WST-8) for suitability with 2D and 3D cultured chondrocytes. We examined especially the correlation of cell numbers with absorbance and incubation time, and the sensitivity and cytotoxicity of the three selected tetrazolium salt assays in Monolayer (2D) and 3D-spheroid culture conditions using human chondrocytes.

2 Materials and methods

2.1 Isolation and cultivation of chondrocytes as monolayer (2D culture)

Human articular cartilage samples were obtained from femoral condyles of four patients undergoing knee surgery. An informed, written consent was obtained from all patients. The study was performed in accordance with the ethical guidelines of Clinical Hemorheology and Microcirculation [23]. Isolation of chondrocytes was performed as previously described [24]. Briefly, chondrocytes were isolated by mechanical mincing of the tissue with a scalpel followed by enzymatic treatment (collagenase, 350 U/ml in DMEM:HAM’s F12 (1 : 1), Biowest, Nuaillé, France). The closed tube was placed on a shaker (Thermomixer comfort, Eppendorf, Hamburg, Germany) at 300 rpm interval mixing and incubated at 37°C for 20 h. The isolated chondrocytes were centrifuged at 300 xg for 5 min. The supernatant was removed and the cell pellet was resuspended with 10 ml of DMEM:HAM’s F12 (1 : 1) with 4 mM L-glutamine (Biowest) and 10% human serum (German Red Cross, Cottbus, Germany). The chondrocytes were cultured as monolayers at 37°C and 5% CO₂. Cells were detached for subcultures using 0.05% trypsin/0.02% EDTA (Biowest), and plated at a defined ratio (1 : 3). Cells from passage two (P2) up to passage four (P4) were used for the experiments. A phase contrast microscope (CKX 41 with a DP 71 camera, Olympus, Hamburg, Germany) was used for observing the morphology of the cells.

2.2 Generation of spheroids (3D culture)

Spheroids were generated using a scaffold-free culture system developed by Ursula Anderer [24, 25]. To allow the formation of spheroids with different sizes, human chondrocytes were seeded in agarose coated 96-well plates in decreasing cell numbers performing serial dilutions (1 : 2) ranging from 300,000 to 590 cells/well. Spheroids were cultured in DMEM high glucose:HAM’s F12 (1 : 1) with 4 mM L-glutamine and 5% human serum. After 8 days in culture, the morphology was documented using phase contrast microscopy (CKX 41 with a DP 71 camera) and measuring the size of the spheroids (Cell^D-Imaging software for Life Science Microscopy, Soft Imaging Systems, Muenster, Germany). A schematic overview of cell isolation and spheroid formation is given in Fig. 3.

Fig.3

Experimental design to apply three tetrazolium salt-based assays to cells in 2D and 3D culture. Chondrocytes were isolated from human knee condyles and expanded in monolayer culture. Reagents of the XTT, WST-1, and WST-8 assays were applied to 2D culture in different cell densities and to 3D cultures with different spheroid sizes. The resulting formazan products (orange to red) were quantified spectrophotometrically.

2.3 Performing the XTT, WST-1, and WST-8 assays in 2D culture

Chondrocytes were seeded in 96well plates in serial dilutions ranging from 40,000 cells/well to 940 cells/well in a total volume of 100 μl. Each cell concentration was plated in triplicate. After 24 h incubation the individual substrates were added according to the manufacturers’ instructions. For the XTT assay (Roche Diagnostics, Penzberg, Germany) [26], 50 μl of the reaction mixture was added per well (50 μl XTT labelling reagent premixed with 1 μl electron coupling reagent). For the WST-1 (Roche Diagnostics) [27], and WST-8 (Cell Guidance Systems, Cambridge, UK) [28] assay 10 μl of the ready to use reaction mixture was added to each well. The absorbance was measured spectrophotometrically after multiple time periods of incubation (2, 4, 6, 8, 10, and 12 h) using the microplate reader FLUOstar Omega (BMG Labtech, Ortenberg, Germany). Absorbance measurements were done using the following wavelengths: 490 nm for XTT, 415 nm for WST-1, and 450 nm for WST-8, with a reference wavelength of 655 nm. As control (blank) only medium was incubated with the appropriate reaction mixture (three wells). These readouts were used as background control which were subtracted from the absorbance values measured in the wells with cells. A schematic overview of the experimental design is given in Fig. 3.

2.4 Performing the XTT, WST-1, and WST-8 assays in 3D culture

Spheroids with different sizes formed on agarose-coated 96well plates were transferred to non-coated 96well plates prior to the addition of the reaction mixture to avoid any interference of the tetrazolium salts or the formazan products with the agarose. The performance of the assays and the spectrophotometrical measurement was in accordance to the procedure of monolayer cells (see 2.3.). The incubation periods for 3D cultures were 1, 2, 3, 4, 5, and 6 h.

2.5 Statistical analysis

The results are presented as mean±SEM (standard error of the mean) of cells from four independent donors each in triplicate measurements.

3 Results

Monitoring the proliferation, viability, and toxicity are major challenges in 2D and 3D culture systems. We tested three different tetrazolium assays, the Cell Proliferation Kit II (XTT), the Cell Proliferation Reagent WST-1 and the Orangu^TM Cell Counting Solution (WST-8) for suitability for 2D and 3D cultured chondrocytes isolated from human cartilage of the knee joint. Therefore, we addressed different issues which are expected to vary between 2D and 3D cultures and the distinct assays. These include the starting cell numbers to ensure a linear response as well as an appropriate signal over background ratio and an optimal incubation time, e.g. to prevent saturation by formazan formation.

3.1 Formazan formation of chondrocytes in 2D culture

The formazan formation of 2D cultured chondrocytes seeded on 96-well plates in a wide range of cell concentrations (940 –40,000 cells/well) was measured spectrophotometrically after multiple time periods of incubation with the XTT, WST-1, and WST-8 reaction mixture. Figure 4A-C present the cell number/well as a function of the absorbance at multiple incubation time points. At first sight, the sigmoidal curve progression (best to be seen in the WST-8 graph) exhibits that a 2 h incubation with XTT, WST-1 and WST-8 reagent is not suitable for the determination of cell quantities (signal saturation at low absorbance values). Choosing a 4 h incubation time as standard for all three assays (recommended in the assay manuals) resulted in a maximally measurable absorbance value of 2.5 for XTT, 2.3 for WST-1 and 2.8 for WST-8. The graph Fig. 4D plots the slope of the curves calculated with the Gompertz function (A-C) in correlation to the incubation time. The slope of WST-8 with a linearity of 0.9970 relates to a higher sensitivity of this tetrazolium salt assay compared to WST-1 (linearity 0.9913) and XTT (linearity 0.9678).

Fig.4

Kinetics of the metabolism of the three different tetrazolium salts XTT (A), WST-1 (B), and WST-8 (C) of adherent 2D cell culture. Human Chondrocytes at cell concentrations indicated in the figure were preincubated for 24 h before the addition of the tetrazolium salts. After 2, 4, 6, 8, and 12 h incubation periods the absorbance was measured with a spectrophotometer. The slopes of the curves were calculated with the Gompertz function. Linear regression analysis was performed (D), showing a higher sensitivity for the WST-8 assay in comparison to XTT and WST-1, represented by the steeper slope of the curve. The 2 h incubation time was excluded (black square).

The amount of formazan dye generated should be directly proportional to the number of living cells and the incubation time. Plotting incubation time against absorbance divided by incubation time should reveal a straight line with a constant y-value (Fig. 5A-C). This important aspect could be confirmed for the WST-8 assay up to a specific cell number per well (Fig. 5D triangles up, 1×10⁴ cells/well) and to a minor degree also for WST-1 followed by the XTT assay. The results of the linear correlation analysis (Fig. 5D) reveal an applicable range of cell numbers suitable to perform these three tetrazolium-based assays. Using the XTT assay 940 –10,000, in the WST-1 assay 940 –5,000, and for the WST-8 assay 2,500 –15,000 cells/well are ideal cell densities.

Fig.5

Principle of proportionality. Human Chondrocytes at cell densities indicated in the figure were preincubated for 24 h before addition of the three different tetrazolium salts XTT, WST-1 and WST-8. After 2, 4, 6, 8, 10 and 12 h incubation periods the absorbance was measured with a spectrophotometer. The time-dependent consumption of substrate shows an optimal incubation time of 4 h for all three tetrazoliums salts (A-C, colored boxes). The amount of the WST-8 formazan produced (C) is directly proportional to the amount of cells up to 10.000 cells/well (triangle) visualized by straight lines with a constant y-value. (D) Plotting the cells/well against the absorbance/time reveals the measurable cell densities for XTT (940–10,000 cells/well), WST-1 (940–5,000 cells/well) and WST-8 (2,500–15,000 cells/well) at 4 h of incubation time.

3.2 Formazan formation of chondrocytes in 3D culture

Aggregates initiated via different cell numbers (590 –300,000 cells) formed spheroids with different sizes (Fig. 6D). Formazan formation was measured spectrophotometrically after multiple time periods of incubation (1 h –6 h). In the 3D culture system the absorption values of spheroids with initial cell numbers less than 5,000 cells were maximally 0.1. According to these low absorption values in the range of the background signals (see 3.3.), the signal of spheroids based on 5,000 cells were selected as cut-off. This aspect was observed in all three assay system (Fig. 6A-C). This cut-off also correlates with the linearity between cell number per aggregate and spheroid size (Fig. 6D). Starting cell numbers under 5,000 cells showed no linear correlation to the spheroid diameter (Fig. 6D). Striking is the observation that the absorbance values measured in the 3D system in correlation to the initial cell number are very low. Spheroids formed by 15×10⁴ cells with an incubation time of 4 h yielded signals of 0.6 (XTT), 0.9 (WST-1), and 1.15 (WST-8) (Fig. 6A-C).

Fig.6

Kinetics of the metabolism of the three different tetrazolium salts XTT (A), WST-1 (B) and WST-8 (C) in 3D spheroid culture. Chondrocyte spheroids with different sizes represented via cells/spheroid were preincubated for 8 days before the addition of the tetrazolium salts. After 1, 2, 3, 4, 5, and 6 h incubation periods the absorbance was measured with a spectrophotometer. The cut off is set at 5000 cells/well. Spheroids with less cells are not measurable with the used assay system. The spheroid size (D) shows linear increase in correlation to the cell number.

Figure 7 illustrates the relationship of absorbance values obtained from cells in 2D and 3D culture. On the one hand there is a linear function of the absorbance values of 2D cultured chondrocytes (circles) with respect to the cell densities for the analyzed data (Fig. 7A-C, correspond to Fig. 5D with a change of the x-axis dimension). On the other hand there is also a correlation of the absorbance values of 3D cultured chondrocytes (squares) with respect to the initial cell number per spheroid (see Figs. 6, and 7A-C). Transmitting the absorbance values measured in the 3D system to the absorbance values measured in the 2D system results in a virtual adjusted cell number for the 3D spheroids (calculated via shown equations in Fig. 7A-C). This means e.g. using the WST-8 assay an absorbance value per incubation time of 0.17 in the 3D system corresponds with an initial spheroid cell number of 150,000. The same absorbance value of 0.17 would correlate to a cell number of 5,284 in the 2D system. These “correlating” cell numbers are displayed as colored columns in Fig. 7D.

Fig.7

The diagrams show the cell number [2D = circles, 3D = squares] in correlation to the absorbance at 4 h incubation time of the tetrazolium salts XTT (A), WST-1 (B) and WST-8 (C). The equations represent the linear function of the cell number in correlation to the absorbance of 2D cultured cells and can be used as function to calculate the approximate cell number of the absorption data measured in 3D. (D) Comparison of cell numbers calculated with the theoretic mathematical model [black] and with the linear function for XTT [green], WST-1 [blue] and WST-8 [orange].

Based on the fact that a spheroid consists of dense packed cells with diameter of 20 μm in a spherical shape, it is possible to calculate a theoretical number of cells on the surface of the spheroid. This means e.g. 150,000 initial cells for 3D culture result in spheroids with an average diameter of 754 μm, which represents a calculated cell number of 5,685 on the surface of the spheroid (Fig. 7D, black columns). In table 1 these data and calculations for XTT, WST-1 and WST-8 applied to cells in 3D culture are summarized.

Table 1

Summary of the 3D data. Cell numbers per spheroid in correlation to the spheroid size. The calculated cell number on spheroid surface is faced to the 2D cell numbers transmitted from the 3D absorption value (see also Fig. 7)

Cell number per spheroid	Spheroid diameter [μm]	Calculated cell number on spheroid surface^a	Cell number XTT^b [3D absorption ⟶ 2D cell number]	Cell number WST-1^c [3D absorption ⟶ 2D cell number]	Cell number WST-8^d [3D absorption ⟶ 2D cell number]
9,400	259	676	518	396	823
18,750	324	1,050	683	545	1,084
37,500	407	1,665	818	782	1,848
75,000	554	3,069	1,429	1,244	2,749
150,000	754	5,685	2,086	2,130	5,284
300,000	1238	15,326	4,102	4,608	11,807

$cell number on spheroid {surface}^{a} = \frac{π * d_{spheroid}^{2}}{((\frac{π}{4}) \times d_{cell}^{2})}$

${XTT}^{b} = \frac{(absorbance)}{4, 308 * e^{- 005}}; {WST - 1}^{c} = \frac{(absorbance)}{6, 417 * e^{- 005}}; WST - 8^{d} = \frac{(absorbance)}{3, 299 * e^{- 005}}$

3.3 Absorption values of background controls

Only medium incubated with the appropriate reaction mixture of the tetrazolium salts resulted in measurable absorption values after the different incubation times. After 4 h incubation time values between 0.14 (WST-8), 0.20 (WST-1), and 0.28 (XTT) were measured (Fig. 8). After an incubation time of 12 h the absorbance value for the XTT assay increased up to 0.44.

Fig.8

Absorbance values of background controls. Wells of 96well plates filled with medium only and after the addition of the individual tetrazolium salts showed considerable absorbance values after different incubation times. Absorbance was measured using the appropriate wavelengths: 490 nm for XTT, 415 nm for WST-1, and 450 nm for WST-8, with a reference wavelength of 655 nm (see 2.3.).

3.4 Cytotoxic effects of the tetrazolium assays

Incubation of chondrocytes in monolayer culture with tetrazolium salts of the XTT, WST-1, and WST-8 assay resulted in morphological changes of the cells. Cells contracted and detached from the surface. WST-1 and to a lesser extend XTT demonstrated toxic effects on the cells, already visible after 4 h of incubation, whereas WST-8 showed no signs of toxicity at all (Fig. 9).

Fig.9

Cytotoxic effects of the tetrazolium-based assays. Phase contrast images taken of 2D cultured chondrocytes after incubation with the tetrazolium salts XTT, WST-1 and WST-8 for 4 h, 8 h and 12 h. Size of the error bars is 200 μm.

4 Discussion

In the present study we tested the three different microculture tetrazolium salts XTT, WST-1 and WST-8 for their suitability for 2D and 3D cultured human chondrocytes, isolated from femoral condyles of patients undergoing knee surgery. The aim of this project was to identify the most suitable assay system to monitor proliferation, viability and toxicity for chondrocyte-specific research projects. To work on this question we first analyzed the assumed correlation of cell number and incubation time based on the spectrophotometrically measured absorbance values of chondrocytes cultured in a 2D and 3D cell culture system. Furthermore we examined possible cytotoxic effects of assay components in 2D monolayer culture via phase contrast microscopy.

In the 2D system we could demonstrate that the WST-8 assay is preferable to be used in comparison with the XTT and WST-1 assay. Absorbance values are directly proportional to the cell number which was also previously shown with other cell types from human origin (e.g. HeLa [human cervical cancer], HL60 [human acute promyelonic leukemia], and H1299 [human lung cancer], from animals (e.g. L929 [mouse connective tissue], Balb3T3 [mouse embryo] and also protozoans like Leishmania [10 , 30]. WST-8 showed the highest correlation between cell number and absorbance followed by XTT and WST-1. Furthermore, WST-8 showed a strong correlation also with the incubation time. Table 2 opposes assay specific parameters recommended in the manuals of the manufacturers to our own data. The measurable range of cell numbers differ strongly from the recommended cell numbers in the manuals [26 –28]. The greatest discrepancy shows WST-1 with a maximum of 5,000 cells measurable in comparison to 50,000 cells postulated in the manual. The ranges estimated for the XTT and WST-8 assay are in a tolerable limits to the measured cell numbers with only minor differences. Our results are in accordance with Ginouves et al. 2014 [10] illustrating that WST-8 is an improved version of WST-1 and an advanced development of XTT with a higher sensitivity and constant stability. This outstanding position of the application of the WST-8 assay on chondrocytes in 2D is further supported by the fact that WST-8 kit components showed the lowest background values and no cytotoxic effect on the cells.

Table 2
Comparison of specifications recommended in the assay manual and measured parameters

Assay Incubation time recommended Cell number recommended [cells/well] Incubation time chondrocytes Cell number chondrocytes [cells/well]

XTT^[1] (•) 4–24 h 2,000–20,000 4 h 940–10,000

6 h 940–7,500

8 h 940–7,500

WST-1^[2] (■) 0,5–4 h 1,000–50,000 4 h 940–5,000

6 h 940–5,000

8 h 940–3,750

WST-8^[3] [▴] 1–4 h 1,000–25,000 4 h 2,500–15,000

6 h 2,500–10,000

8 h 2,500–10,000

Assay	Incubation time recommended	Cell number recommended [cells/well]	Incubation time chondrocytes	Cell number chondrocytes [cells/well]
XTT^[1] (•)	4–24 h	2,000–20,000	4 h	940–10,000
			6 h	940–7,500
			8 h	940–7,500
WST-1^[2] (■)	0,5–4 h	1,000–50,000	4 h	940–5,000
			6 h	940–5,000
			8 h	940–3,750
WST-8^[3] [▴]	1–4 h	1,000–25,000	4 h	2,500–15,000
			6 h	2,500–10,000
			8 h	2,500–10,000

^[1]XTT, Cat. No. 11465015001, Roche; ^[2]WST-1, Cat. No. 11644807001 Roche; ^[3]Orangu^TM, Cat. No. OR01 Cell guidance systems.

Comparing cell numbers seeded and absorbance values measured in the 2D system and in the 3D system we were able to uncover great discrepancies. These discrepancies could be based on the diverse microenvironments of cells in 2D and 3D culture. Cells in monolayer culture are spread individually on a polystyrene culture vessel and the surface of each cell is exposed to the culture medium. The spheroid, however, has a special cell arrangement consisting of a dense packing of cells in a spherical shape and only the cells on the spheroid surface were exposed to the culture medium and therefore also to the experimental environment containing the tetrazolium reagents. Therefore, we hypothesize that only cells of the outer rim of the spheroids are able to contribute to the reduction of the tetrazolium salts. A simple mathematical model (Fig. 10) illustrates this phenomenon. After a short aggregation time cells in a spheroid structure are tightly packed without gaps between the cells. Therefore, a diffusion of assay reagent through or in between the cells is impossible. Transmitting the absorbance values of 3D cultures to appropriate cell numbers in 2D results in a correlation with the calculated cell numbers on the spheroid surface (see also Fig. 7D) confirming our theory.

Fig.10

Theoretical mathematical model explaining the effect of the spheroid structure on tetrazolium salt assays. According to the measured absorption values, only cells in the outer rim were able to interact with the tetrazolium reagents to form measurable formazan products. The cell number on the surface of the spheroid can be estimated according to the special 3D morphology and cell arrangement (upper formula) and compared to the estimated cell number calculated via the transmission of spheroid absorption values related to the appropriate 2D cell numbers (lower formula).

There are only few publications which refer to tetrazolium salt-based assays in 3D culture systems. Carlson reported that the WST-1 assay performed acceptable results in measuring the “cell quantity” in a fibroblast-populated collagen matrix and that the outcome was not affected by the extracellular matrix [31]. Huyck et al. reported on the equal quality (dynamic range, background, linearity) of measuring the proliferation of cell layers of cancer cell lines embedded in dense 3D matrices (collagen, matrigel) compared with cells in 2D culture using the XTT assay [32]. In contrast, HeLa cells grown on collagen gels show higher signals than those grown as 2D culture. Bonnier et al. interpreted their results as a consequence of the different environments of the culture systems. The dye of the test system freely diffuses through the collagen gel. The cells in 3D were all around embedded in matrix and substrate and cells in 2D were only exposed on one side to the substrate in the medium [33]. Here, the observed signal differences were attributed to the differences in the cell surface area which is accessible by the dye for diffusion. Taken our results and the literature data together it appears that also the type of 3D culture with special microenvironmental characteristics influences the outcome of tetrazolium-based proliferation assays.

Although we could show a very good correlation of the WST-8 assay with cell proliferation the assay system has some limitations and should therefore be used cautiously. The extracellular conversion of tetrazolium salts to formazan products is based on the activity of intracellular mitochondrial oxidoreductases and the reduction of NADH/NADPH [11, 34]. Weir et al. pointed out that these readings can be affected by certain extrinsic (low atmospheric oxygen or high density culture) or intrinsic (defects in oxygen-sensing pathways) factors. Weir et al. could show that the WST-1 assay can be modulated by the level of cell oxygenation [34]. Considering the same reaction pathway for XTT and WST-8 the same complications are expectable. Therefore, it is recommended to consider if the cell metabolism of the selected cell type is influenced by different parameters like cell oxygenation, metabolic interfering or toxic reagents before applying the WST-8 assay.

5 Conclusion

In this study we could show that WST-8 is the best cell proliferation/quantification test reagent for 2D cultured human chondrocytes with respect to the XTT and WST-1 assay. Using cells in a 3D environment there are obvious limitations of these proliferation assays. Based on the specific cell arrangement only cells on the spheroid surface contribute to the conversion of the tetrazolium salts to formazan products. However, WST-8 may be used to monitor specific modifications in cells on the surface of 3D tissue-like structures.

Footnotes

Acknowledgments

The work was supported by grants of the “Gesundheitscampus Brandenburg” (Land Brandenburg). We are grateful to Matthias Suckow, Institute of Applied Chemistry, University of Technology Cottbus-Senftenberg, for assistance in mathematical calculations.

References

Correa

and Lietman

S.A.

, Articular cartilage repair: Current needs, methods and research directions, Seminars in Cell & Developmental Biology62 (2017), 67–77.

Saris

D.B.

, Vanlauwe

, Victor

, Almqvist

K.F.

, Verdonk

, Bellemans

and Luyten

F.P.

, Treatment of symptomatic cartilage defects of the knee: Characterized chondrocyte implantation results in better clinical outcome at 36 months in a randomized trial compared to microfracture, Am J Sports Med37 (2009), 10S–19S.

Peterson

, Vasiliadis

H.S.

, Brittberg

and Lindahl

, Autologous chondrocyte implantation a long-term follow-up, Am J Sports Med38 (2010), 1117–1124.

Niemeyer

, Porichis

, Steinwachs

, Erggelet

, Kreuz

P.C.

, Schmal

, Uhl

, Ghanem

, Südkamp

N.P.

and Salzmann

, Long-term outcomes after first-generation autologous chondrocyte implantation for cartilage defects of the knee, Am J Sports Med42 (2014), 150–157.

Anderer

, Savkovic

, Zelltherapien in der Regenerativen Medizin. In: Sack

Ulrich

, Hrsg. Zelluläre Diagnostik und Theraie. Berlin: DE GRUYTER, 2016, pp. 288–345.

Cook

J.A.

and Mitchell

J.B.

, Viability measurements in mammalian cell systems, Analytical Biochemistry179 (1989), 1–7.

Goodlad

R.A.

, Quantification of epithelial cell proliferation, cell dynamics, and cell kinetics in vivo, WIREs Dev Biol6 (2017), e274. doi: 10.1002/wdev.274

Böhlen

L.M.

, Limat

, Hunziker

, Epidermal Equivalents From Autologous Outer Root Sheath Cells for Chronic Wounds. In: Horch

, Munster

, Achauer

, editors. Cultured human keratinocytes and tissue engineered skin substitutes. Stuttgart: Thieme, 2001, pp. 330–334.

Marshall

N.J.

, Goodwin

C.J.

and Holt

S.J.

, A critical assessment of the use of microculture tetrazolium assays to measure cell growth and function, Growth Regul5(2) (1995), 69–84.

10.

Ginouves

, Carme

, Couppie

and Prevot

, Comparison of Tetrazolium Salt Assays for Evaluation of Drug Activity against Leishmania spp, Journal of Clinical Microbiology52(6) (2014), 2131–2138.

11.

Berridge

M.V.

, Herst

P.M.

and Tan

A.S.

, Tetrazolium dyes as tools in cell biology: New insights into their cellular reduction, Biotechnology Annual Review11 (2005), 127–152.

12.

Scudiere

D.A.

, Shoemaker

R.H.

, Paull

K.D.

, Monks

, Tierney

, Nofziger

T.H.

, Currens

M.J.

, Seniff

and Boyd

M.R.

, Evaluation of a Soluble Tetrazolium/Formazan Assay for Cell Growth and Drug Sensitivity in Culture Using Human and Other Tumor Cell Lines, Cancer Research48 (1988), 4827–4833.

13.

Ishiyama

, Miyazono

, Sasamoto

, Ohkura

and Ueno

, A highly water-soluble disulfonated tetrazolium salt as a chromogenic indicator for NADH as well as cell viability, Talanta44 (1997), 1299–1305.

14.

Tominaga

, Ishiyama

, Ohseto

, Sasamoto

, Hamamoto

, Suzukic

and Watanabec

, A water-soluble tetrazolium salt useful for colorimetric cell viability assay, Anal Commun36 (1999), 47–50.

15.

Hisada

and Yagi

, 1-Methoxy-5-methylphenazinium methyl sulfate. A photochemically stable electron mediator between NADH and various electron acceptors, J Biochem82(5) (1977), 1469–1473.

16.

Herst

P.M.

and Berridge

M.V.

, Cell surface oxygen consumption: A major contributor to cellular oxygen consumption in glycolytic cancer cell lines, Biochimica et Biophysica Acta1767 (2007), 170–177.

17.

Zscharnack

, Krause

, Aust

, Thümmler

, Peinemann

, Keller

, et al., Preclinical good laboratory practice-compliant safety study to evaluate biodistribution and tumorigenicity of a cartilage advanced therapy medicinal product (ATMP), Journal of Translational Medicine13 (2015), 160.

18.

Proctor

W.R.

, Foster

A.J.

, Vogt

, Summers

, Middleton

, Pilling

M.A.

, et al., Utility of spherical human liver microtissues for prediction of clinical drug induced liver injury, Arch Toxicol (2017). DOI: 10.1007/s00204-017-2002-1

19.

Mehta

, Hsiao

A.Y.

, Ingram

, Luker

G.D.

and Takayama

, Opportunities and challenges for use of tumor spheroids as models to test drug delivery and efficacy, Journal of Controlled Release164 (2012), 192–204.

20.

Anderer

and Libera

, In vitro engineering of human autogenous cartilage, J Bone Miner Res17(8) (2002), 1420–1429.

21.

Itoh

, Nakayama

, Noguchi

, Kamohara

, Furukawa

, Uchihashi

, et al., Scaffold-free tubular tissues created by aBio-3D printer undergo remodeling and endothelialization whenimplanted in rat aortae, PLoS One10(9) (2015), e0136681. doi: 10.1371/journal.pone.0136681

22.

Koudan

E.V.

, Korneva

J.V.

, Karalkin

P.A.

, Gladkaya

I.S.

, Gryadunova

A.A.

, Mironov

V.A.

, Khesuani

Y.D.

and Bulanova

E.A.

, The scalable standardized biofabrication of tissue spheroids from different cell types using nonadhesive technology, 3D Printing and Additive Manufacturing4(1) (2017), 53–60.

23.

Anonymous, Ethical guidelines for publication in Clinical Hemorheology and Microcirculation: Update 2016, Clin Hemorheol Microcirc63(1) (2016), 1–2.

24.

Martin

, Lehmann

, Anderer

, Generation of scaffold free 3-D cartilage-like microtissues from human chondrocytes. In: Daskalaki

, editor. Medical Advancements in Aging and Regenerative Technologies: Clinical Tools and Alications. USA:IGI Global, 2013, pp. 169–194.

25.

Anderer

and Libera

, In vitro engineering of human autogenous cartilage, J Bone Miner Res17(8) (2002), 1420–1429.

26.

Cell Proliferation Kit II (XTT), Roche, Cat. No. 11465015001.

27.

Cell Proliferation Reagent WST-1, Roche, Cat. No. 11644807001.

28.

Orangu^TM Cell Counting Solution. (WST-8), Cell Guidance Systems, Cat. No. OR01.

29.

Ishiyama

, Miyazono

, Sasamoto

, Ohkura

and Ueno

, A highly water-soluble disulfonated tetrazolium salt as a chromogenic indicator for NADH as well as cell viability, Talanta44 (1997), 1299–1305.

30.

Tominaga

, Ishiyama

, Ohseto

, Sasamoto

, Hamamoto

, Suzukic

and Watanabe

, A water-soluble tetrazolium salt useful for colorimetric cell viability assay, Anal Commun36 (1999), 47–50.

31.

Carlson

M.A.

, Technical note: Assay of cell quantity in the fibroblast-populated collagen matrix with a tetrazolium reagent, European Cells and Materials12 (2006), 44–48.

32.

Huyck

, Ampe

and Van Troys

, The XTT cell proliferation assay applied to cell layers embedded in three-dimensional matrix, ASSAY and Drug Development Technologies10(4) (2012), 382–392.

33.

Bonnier

, Keating

M.E.

, Wróbel

T.P.

, Majzner

, Baranska

, Garcia-Munoz

, Blanco

and Byrne

H.J.

, Cell viability assessment using the Alamar blue assay: A comparison of 2D and 3D cell culture models, Toxicology in Vitro29 (2015), 124–131.

34.

Weir

, Robertson

, Leigh

I.M.

and Panteleyev

A.A.

, The reduction of water-soluble tetrazolium salt reagent on the plasma membrane of epidermal keratinocytes is oxygen dependent, Analytical Biochemistry414 (2011), 31–37.