Analysis of relationship between cell cycle stage and apoptosis induction in K562 cells by sedimentation field-flow fractionation

Journal of Chromatography B, 877 (2009) 1155–1161

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Analysis of relationship between cell cycle stage and apoptosis induction in K562 cells by sedimentation field-flow fractionation J. Bertrand a , B. Liagre b , G. Bégaud-Grimaud c , M.O. Jauberteau a , J.L. Beneytout b , P.J.P. Cardot c , S. Battu c,∗ a Laboratoire d’immunologie, EA 3842, “Homéostasie Cellulaire et Pathologies”, Université de Limoges, Faculté de Médecine, 2 rue du Docteur Marcland, 87025 Limoges Cedex, France b Laboratoire de Biochimie, EA 4021 “Biomolécules et Thérapies Anti-tumorales”, Université de Limoges, Faculté de Pharmacie, 2 rue du Dr Marcland, 87025 Limoges Cedex, France c Laboratoire de Chimie Analytique et Bromatologie, EA 3842, “Homéostasie Cellulaire et Pathologies”, Université de Limoges, Faculté de Pharmacie, 2 rue du Docteur Marcland, 87025 Limoges Cedex, France

a r t i c l e

i n f o

Article history: Received 27 November 2008 Accepted 27 February 2009 Available online 5 March 2009 Keywords: K562 cells Diosgenin Cell cycle Apoptosis Sedimentation field-flow fractionation Cell sorting

a b s t r a c t Recently, sedimentation field-flow fractionation (SdFFF) was used to study the specific kinetics of diosgenin-induced apoptosis in K562 cells. Here, we propose a new SdFFF cell separation application in the field of cancer research concerning the correlation between induction of a biological event (i.e. apoptosis) and cell status (i.e. cell cycle position). SdFFF isolated subpopulations depending on the cell cycle position allowing the study of apoptosis kinetics and extent. Results showed that cells in G0/G1 phases (F3 cells) underwent significant and earlier apoptosis than cells in the active part of the cell cycle (S/G2/M phases). Results shed light on the correlation between differences in apoptosis kinetics and cell cycle stage when exposure to the inducer began. SdFFF monitoring and size measurement also led to the description of different subpopulations demonstrating complex variations in density between fractions associated with differences in biological processes. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Within the field-flow fractionation (FFF) superfamily, sedimentation FFF (SdFFF) could be described with flow-FFF, hollow fiber-FFF, hybrid-FFF, magnetic FFF or DEP-FFF [1–13], as one of the most useful methods for cell separation and sorting [14–23]. SdFFF allows the macroscale preparation of functional cell populations for analytical and preparative applications [16–23]. Similar to many other cell separation techniques (centrifugation, elutriation, microfluidic devices, etc.), the principle of cell separation is based on physical criteria such as size and density [11,22,24–27]. The fundamental principle of FFF is based on the differential elution of species submitted to the combined action of: (1) a parabolic profile generated by flowing a mobile phase through a ribbon-like capillary channel and (2) an external field applied perpendicularly to the flow direction [28–30]. In SdFFF, a multigravitational external field is generated by rotation of the separation channel in a rotor basket, constituting one of the most complex devices used in FFF separation [25,26,29]. SdFFF could be defined as a gentle, non-invasive and tagless method. These advantages are based on the drastic limitation of cell-solid phase interaction by

∗ Corresponding author. Tel.: +33 5 5543 5979; fax: +33 5 5543 5858. E-mail address: [email protected] (S. Battu). 1570-0232/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2009.02.064

the use of: (1) a specific separation device: empty ribbon-like channel without stationary phase and (2) a device setup allowing the “Hyperlayer” elution mode, a size/density driven separation mechanism. Since the report of Caldwell et al. [14] on mammalian cells, FFF, SdFFF and related technologies have been used in many biological fields such as hematology, microorganism analysis, biochemistry/biotechnology and molecular biology, neurology and cancer research [4,20,22,31–43,7,44–48]. Over several years, we have studied the use of SdFFF in cancer research to study chemical induced apoptosis or differentiation in cancer cell lines. Different aspects have been evaluated including: (1) monitoring of the biological event [17,49–51]; (2) cell sorting of specific subpopulations such as pre-apoptotic [17], or differentiated cells [49] which can then be further used as models; (3) kinetics of the biological event using both the monitoring and cell separation capacities of SdFFF [19,21]. Cell sorting of specific phenotypes or immature cells from complex cancer cell populations such as neuroblastomas have been also performed [16,41]. Diosgenin is a well-known steroidal saponin, which can be found in several plant species. It has been described to have various effects in vivo [52–54] and in vitro, such as inducing megakaryocytic differentiation of human erythroleukemia cells at 10 ␮M, or exerting antiproliferative and pro-apoptotic actions (40 ␮M) on rheumatoid arthritis synoviocytes or cancer cell lines [17,19,21,49–51,55–63]. In a recent work, we used SdFFF to explore the specific biphasic

1156

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161

apoptosis kinetics in K562 cells (erythroleukemic cell line) after diosgenin exposure [19,64]. The association of washout (removal of the chemical inducer from the culture medium) and early SdFFF “Hyperlayer” elution effectively separated different apoptotic stages in the same population. These results suggested different sensitivities to apoptosis induction, and we hypothesized that apoptosis kinetics could depend on the cell cycle position when diosgenin exposure started [19]. The goal of this work was to examine the possible relationship between cell cycle position and apoptosis kinetics. To achieve this goal, SdFFF was used to sort subpopulations from crude K562 cells to analyse the cell cycle before any apoptosis induction, leading to an immediate cell cycle status distribution. At the same time, to avoid any distortion between cell cycle analysis and apoptosis induction, fractions were subcultured and incubated for 6 h with diosgenin. Then, early apoptosis was monitored (SdFFF) and quantified by flow cytometry. As previously described [20], results showed the capacity of SdFFF to sort cells depending on their position in the cell cycle, leading to the preparation of different subpopulations with regards to kinetics and the extent of apoptosis. These results led to a better understanding of the mechanism of diosgenin-induced apoptosis. 2. Materials and methods 2.1. Cell line, cell culture and treatment The K562 human erythroleukemia cell line was provided by Dr I. Dusanter-Fourt (INSERM U567-CNRS UMR8104, Paris, France). Cells were cultured in RPMI-1640 medium (Gibco, Invitrogen, Paisley, Scotland) supplemented with 10% fetal calf serum (Gibco), 1% sodium pyruvate, 1% HEPES [N-(2-hydroxy-ethyl)piperazine-N -2ethansulfonic acid] (Gibco), 100 ␮g/mL streptomycin and 100 U/mL penicillin. Cultures were maintained in a humidified atmosphere with 5% CO2 at 37 ◦ C. Cells were seeded at 2 × 105 cells/mL in 75 cm2 tissue culture flasks (Sarstedt, Marnay, France) and allowed to grow in culture medium for 24 h prior to SdFFF elution. Cells were harvested, counted by trypan blue dye exclusion method and resuspended in PBS (Gibco) at 2.5 × 106 cells/mL before SdFFF analysis. 2.2. SdFFF device, cell elution conditions and subpopulation preparation SdFFF separation device used in this study derived from those previously described and schematized [17,19]. The apparatus was composed of two 938 mm × 40 mm × 2 mm polystyrene plates, separated by a mylar® spacer in which the channel was carved. Channel dimensions were 818 mm × 12 mm × 0.175 mm with two 50 mm V-shaped ends. The measured total void volume (channel volume + connecting tubing + injection and detection device) was 1792 ± 2.00 ␮L (n = 6). Void volume was calculated after injection and retention time determination of a non-retained compound (0.10 g/L of benzoic acid, UV detection at 254 nm). The channel rotor axis distance was measured at r = 14.82 cm. A Waters 515 programmable HPLC pump (Waters Associates, Milford, MA, USA) was used to pump the sterile mobile phase. Sample injections were done by means of a Rheodyne® 7125i chromatographic injector (Rheodyne, Cotati, CA, USA). The elution signal was recorded at 254 nm by means of a Waters 486 Tunable Absorbance Detector (Waters Associates) and a M1111 (100 mV input) acquisition system (Keithley, Metrabyte, Tauton, MA, USA) operated at 4 Hz connected to a PC computer. A M71B4 Carpanelli engine (Bologna, Italy) associated with a Mininvert 370 pilot unit (Richard Systems, Les Ullis, France), controlled the rotating speed of the centrifuge baskets. Sedimentation fields were expressed in units of gravity, 1 g = 980 cm/s2 , and

calculated as previously described [37]. Cleaning and decontamination procedures have been previously described [26]. The optimal elution conditions (“Hyperlayer” mode) were determined experimentally and were: flow injection through the accumulation wall of 100 ␮L K562 cell suspension (2.5 × 106 cells/mL); flow rate: 0.80 mL/min; mobile phase: sterile PBS, pH 7.4 (Gibco); external multi-gravitational field strength: 8.00 ± 0.01 g (219.7 ± 0.1 rpm). Fig. 1A summarizes the protocol used to prepare the different studied subpopulations. After 24 h incubation, K562 cells (2.5 × 106 cells/mL in sterile PBS pH 7.4) were eluted (Fig. 1B) resulting in the separation of four cell fractions collected and designated as follows: (1) TP (total peak) (elution time 3 min 30 s to 6 min 50 s); and (2) Fn (fraction number) with F1 (3 min 30 s to 4 min 20 s), F2: (4 min 25 s to 5 min 20 s) and F3 (5 min 30 s to 6 min 50 s). To obtain a sufficient quantity of cells for cell cycle studies, or culture and later apoptosis induction, monitoring and quantification, successive SdFFF cumulative fraction collections were performed (12–16). Cell cycle was immediately analyzed after SdFFF elution (Fig. 1). Otherwise, control and SdFFF eluted cells were incubated for 6 h with diosgenin in order to induce apoptosis which was monitored (SdFFF elution) and quantified (flow cytometry) (Fig. 1). 2.3. Cell cycle analysis After SdFFF sorting, cells (fractions and control, Fig. 1) were fixed and permeabilized in 70% ethanol in phosphate-buffered saline (PBS) at −20 ◦ C overnight, washed in PBS, treated with RNase (40 U/␮L, Boehringer Mannheim, Meylan, France) for 1 h at room temperature and stained with propidium iodide (PI) (50 ␮g/mL). Flow cytometry experiments were carried out using a FACS Vantage Diva SE (Becton Dickinson, USA) and were performed as previously described [61]. Data were analyzed with the Cell Quest Software (Becton Dickinson, USA). 2.4. Coulter counter A 256 channel Multisizer II Coulter Counter (Beckman Coulter, Fullerton, CA) was used to determine the mean cell population diameter. Cells: crude population or SdFFF collected fraction were diluted in Isoton® to a final volume of 15 mL. The counting conditions were: 500 ␮L sample volume, cumulating three successive assays. Results are displayed as the mean ± S.D. for three different experiments. 2.5. Apoptosis induction and quantification As described in Fig. 1, control and SdFFF eluted cells were cultured in 24 well plates for 6 h at a density of 1.5 × 105 cells/well with 1 mL culture medium/well, in a humidified atmosphere with 5% CO2 at 37 ◦ C. Cells were incubated (treated cells) or not (control cells) with 40 ␮M diosgenin ([25R]-5␣-spiroten-3␤-ol) (Sigma–Aldrich, Saint-Quentin Fallavier, France). The same amount of vehicle (<0.1% ethanol) was added to control cells. As previously described, early apoptosis, which is the most significant stage to follow apoptosis kinetics [19], was quantified by an ANNEXIN V-FITC Kit (Beckman Coulter, Paris, France). This assay determines the cell surface appearance of phosphatidylserines (PS), negatively charged phospholipids usually located in the inner leaflet of the plasma membrane. In the early phase of apoptosis, cell membrane integrity is maintained but cells lose the asymmetry of their membrane phospholipids. PS become exposed at the cell surface and form one of the specific signals for recognition and removal of apoptotic cells by macrophages. The ANNEXIN V-FITC Kit is an apoptosis kit based on the binding properties of Annexin V to PS and the DNA-

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161

1157

Fig. 1. (A) Preparation of different K562 cell populations. K562 populations were prepared ranging from controls (cells never exposed to diosgenin) to DIOS populations exposed to 40 ␮M diosgenin for 6 h. Cell cycle analysis was carried out prior any exposure to diosgenin. (B) Representative fractograms of K562 cells. Elution conditions: channel thickness: 175 ␮m; flow injection of 100 ␮L cell suspension (2.5 × 106 cells/mL); spectrophotometric detection at  = 254 nm; flow rate: 0.80 mL/min and external multi-gravitational field: 8.00 ± 0.01 g. ER corresponds to the End of channel Rotation: at this moment, the mean externally applied field strength is equal to zero gravity. RP, a residual signal, corresponds to Release Peak of reversible cell-accumulation wall sticking. Fraction collection: TP: 3 min 30 s to 6 min 50 s and/F1: 3 min 30 s to 4 min 20 s/F2: 4 min 25 s to 5 min 20 s/F3: 5 min 30 s to 6 min 50 s.

intercalating capabilities of propidium iodide (PI) which reflects membrane permeability as observed during necrosis or at the latter stage of apoptosis. After 6 h incubation, samples were prepared according to the manufacturer’s protocol and analyzed by flow cytometry (Beckman FC 500, Beckman-Coulter). Ten thousand cells were acquired, and data were analyzed using the computer program WinMDI (Ver. 2.8). 2.6. Statistical analysis The median and standard deviation (S.D.) were calculated using Excel software (Microsoft, Ver. 2002). Statistical analyses (Statview, Ver. 5.0) of differences between fractions were carried out by ANOVA test, and correlation between cell cycle repartition and apoptosis induction was carried out by Pearson test. A p-value of less than 0.05 (Fisher’s PLSD test) was considered to indicate significance. 3. Results and discussion SdFFF can be described as a macroscale, gentle, non-invasive and tagless method for cell separation in different fields such as oncology [17,19,21,49–51,63]. In a recent work [19], the early SdFFF “Hyperlayer” elution/washout process was used to explore specific biphasic apoptosis kinetics in K562 cells after exposure to diosgenin [64]. Then, effective separation of different apoptotic stages in the same population was achieved [19], explaining previous biological observations [64], of apoptosis occurring in two phases associated with a surprising phospho-ERK (active form of extracellular signal-regulated kinase) expression level in the treated population. These results suggested different sensitivities to apoptosis induction, which could depend on the cell cycle position when exposure to diosgenin began [19].

The goal of this work was to study the possible relationship between cell cycle position and apoptosis kinetics. 3.1. SdFFF elution As previously reported [14,17,20,22,29,36,41,51,65], the elution mode of cell species is described as “Hyperlayer”. Cells are focused in a thin layer away from the accumulation wall enhancing subpopulation sorting, reducing cell-accumulation wall interactions while respecting cell functional integrity, short and long term viability, maturation and differentiation stages [17,20,26,36,41,51]. To achieve this goal, device setup and elution conditions were optimized to enhance the “Hyperlayer” elution mode, a size/density dependent cell elution order [14,29,30,65–72], which predicts that large and weaker cells are focused in faster streamlines to be eluted first. Cell shape and rigidity also influences cell elution order [65]. The different average velocities, and the retention order of different species are compared by means of the observed retention ratio Robs [73]. By using the same device setup (see Section 2), the “Hyperlayer” elution mode of K562 cells has been demonstrated in our previous work [19]. Fig. 1B shows representative fractogram obtained for control cells. Two major peaks were observed, the first corresponding to the non-retained species (void volume peak, Robs ≈ 1), the second corresponding specifically to the cell population with Robs = 0.467 ± 0.011 (n > 10). As described [50,51], the absolute value of Robs depends on conditions and culture time. Besides the field or flow dependence of Robs demonstrated in our previous study [19], the “Hyperlayer” elution mode description also predicted that samples were not in close contact with the accumulation wall. By using the following equation [41,70]: R=

6s ω

(1)

1158

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161

to be the less proliferative fraction compared to F1, with a significantly less percentage of S/G2/M cells (45.7 ± 4.4%, p < 0.05, t-test for paired data) (Fig. 2). As previously described for ES cells [20], SdFFF appeared to be an effective method to sort subpopulations according to their cell cycle position. Then, to avoid cell cycle progression, apoptosis was immediately induced after elution, by incubating fractions with 40 ␮M diosgenin, and was measured as early as 6 h. 3.3. Apoptosis monitoring and quantification

Fig. 2. Cell cycle distribution in K562 population. Cell cycle distribution was measured directly after SdFFF elution. Flow cytometric analyses of propidium iodide staining indicated the amounts of cells in G0/G1 (non-proliferative state) and S/G2/M (proliferative state) phases. Results were expressed as a mean percentage of cells ±S.D. (n = 3, *p < 0.05 is only indicated for the more interesting results).

in which R is the retention ratio (Robs values), ω is the channel thickness (175 ␮m), and s the distance of the center of the focused zone from the channel wall [68], s was 13.62 ␮m. The mean cell diameter was 13.23 ± 0.26 ␮m (n > 3, Coulter Counter® ). Thus, cell radii were less than the approximate average cell elevation value: r = 6.62 < s = 13.62 ␮m. Finally, after total cell elution, when the external field was turned off (mean external field = 0.00 g), we observed a residual signal (RP, Fig. 1B) corresponding to reversible particle release from the accumulation wall. These results confirms that K562 cells were eluted in “Hyperlayer” mode, as described previously [19]. The use of these optimal conditions will allowed K562 elution while respecting cell integrity and viability, without modifications of maturation or differentiation stage. Nevertheless, we did not know, at this stage, if subpopulation sorting was achieved, and in particular if we were able to sort cells according to their cell cycle stages before apoptosis induction (Fig. 1A). As shown in Fig. 1B, four cell fractions were collected: (1) the total peak fraction (TP) which corresponded to the total eluted population and (2) peak fractions 1, 2 and 3 (Fn ) which were the time-dependent collected fractions of the retained peak profile. The effectiveness cell sorting was first evaluated on the basis of cell cycle studies, and secondly, on the basis of apoptosis induction. 3.2. Cell cycle analysis G0/G1, S, G2 and M are different phases of the cell cycle. G0 is a phase of resting cells outside the cell cycle. Active cell cycle phases include G1 (gap 1), S (synthesis of DNA), G2 (gap 2) and M (mitosis). During the G1 phase, several growth factors exert their influence, such as active ERK (phospho-ERK is the active form of extracellular signal-regulated kinase) [64,74–76]. After appropriate signals, the cell enters the S phase and begins to replicate its chromosomes. During the G2 phase, DNA damage repair mechanisms operate. During the M phase, chromosomes segregate if they are intact, and cell division leads to two daughter cells. After this latter stage, the cell enters the resting state (G0) or remains in the cycling compartment G1 [74]. Thus, S/G2/M cells have a higher DNA (4N) content compared to G0/G1 cells (2N). As shown in Fig. 2, propidium iodide staining showed that control and TP cells behaved similarly, indicating that SdFFF elution respected cell cycle distribution. F2 population also showed a cell cycle distribution similar to control and TP fractions, with a majority of cells in the S/G2/M phases. Fig. 2 indicated that the percentage of G0/G1 cells was significantly higher (p < 0.05, t-test for paired data) in F3 (55.0 ± 4.6%) than in F1 (24.5 ± 6.5%). F3 also appeared

In many previous reports, we described the use of SdFFF to monitor the induction and kinetics of apoptosis in different cell lines [17,19,49,51]. The decrease in Robs between control and treated cells is proof of apoptosis induction by diosgenin. This shifting could be observed as early as 6 h incubation [17,19,49,51]. In contrast with previous studies in which monitoring was conducted on the crude control versus crude treated population [19], apoptosis monitoring was performed this time on each separated fraction (Fig. 3A). Fractograms of each fraction, control (C), and treated (D 40 ␮M), were compared in Fig. 3A. We observed, as for the crude populations [19], a decrease in Robs for each subpopulation (F1, F2, F3) which confirmed the ability of SdFFF to monitor the induction of biological events. Unfortunately, the Robs variation between control and treated cells (Robs = Robs control − Robs treated ) (Fig. 3B, dotted histograms) only showed slight differences between fractions. According to previous results [19], the use of Robs variation to monitor apoptotic kinetics between fractions was effective only after 24 and 48 h incubation, but not after 6 h. As described in many other works, early stages of apoptosis were associated with an increase in mean cell diameter of the crude population [17,19,49,51], while advanced apoptosis process led to a decrease in mean cell diameter [77]. Then, we measured the mean diameter of each fraction, and we calculated the variation of the mean cell diameter between treated and control cells (diameter, Fig. 3B). In agreement with previous results obtained on crude populations [17,19,49,51], we observed an increase in cell diameter for each population, in particular in F1. In contrast we observed, for the first time, a decreased cell size in F3 (Fig. 3B). The association of SdFFF monitoring (Robs ) and size measurement led to the description of different subpopulations with regards to their biophysical properties (Fig. 3B). As for cell cycle position (Fig. 2), F1 and F3 appeared as two opposite fractions. F1 appeared as proliferative with a majority of cells in the S/G2/M cell cycle phases (Fig. 2). The S/G2/M stage was associated with increased DNA content and density (4N chromosomes). This could explain the increase in retention while cell diameter increased, implicating the size/density balance. In contrast for F3, in which the majority of cells were in G0/G1 phase (2N chromosomes), the decrease in cell diameter only could explain the increased retention. This could demonstrate, according to the “Hyperlayer” mode description [14,29,30,65–72], a complex variation of size and density between fractions which would be associated with differences in biological processes. The pro-apoptotic properties of diosgenin have been deeply demonstrated in many cell types [17,21,49–51,55–63], as well as for K562 cells [19,64]. Then, diosgenin-dependant apoptosis could be quantified by using the ANNEXIN V-FITC Kit. Based on the binding properties of Annexin V to PS and the DNA-intercalating capabilities of propidium iodide (PI), this assay discriminated between late and early apoptosis. By using the late stage apoptosis parameter, we were unable to demonstrate any significant (p > 0.05, t-test for paired data) differences between fractions (data not shown). Late apoptosis appeared as a non-specific event linked to the basal apoptotic rate in cell populations after culture, sample preparation and

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161

1159

Fig. 3. SdFFF monitoring of apoptosis: (A) representative fractogram of fraction (Fn) incubated (D40 ␮M) or not (control: C) with 40 ␮M diosgenin for 6 h and (B) changes in cell diameter (diameter) and retention ratio (Robs ) between populations exposed or not to 40 ␮M diosgenin for 6 h.

SdFFF elution. In contrast (Fig. 4), early apoptosis appeared more sensitive, selective and informative to describe apoptosis kinetics [19]. Fig. 4 confirmed that 40 ␮M diosgenin induced apoptosis in K562 cells (control versus DIOS population, p < 0.05, t-test for paired data). Fig. 4 also showed that elution did not increase apoptosis (control versus TP fraction/DIOS versus TP DIOS). According to cell cycle analysis (Fig. 2) and size/elution balance (Fig. 3B), F2 behaved similarly to TP DIOS (Fig. 4). Again, F1 and F3 fractions behaved differently (Fig. 4). F3 was most apoptotic cell fraction (16.9 ± 1.5%, p < 0.05, t-test for paired data) compared to other fractions, and particularly to F1 (10.3 ± 0.6%). SdFFF has been already used to study chemical apoptosis induction in cancer cell lines (monitoring, subpopulation sorting,

Fig. 4. Early phase apoptosis analysis. Subpopulations were obtained and prepared as described in Figs. 1 and 2 and incubated, or not, for 6 h with 40 ␮M diosgenin. Early phase apoptosis was measured using ANNEXIN V-FITC kit and quantified by flow cytometry. Results were expressed as a mean percentage of cells ±S.D. (n = 3, *p < 0.05 is only indicated for the more interesting results).

kinetics) [17,19,49,51]. In particular, we used SdFFF to explore the specific biphasic apoptosis kinetics (after 24 and 48 h incubation) in K562 cells after exposure to diosgenin [19], associated with a surprising phospho-ERK expression in the treated population [64]. The association of washout and early SdFFF “Hyperlayer” elution processes resulted in effective separation of cells engaged in different apoptotic stages from the same treated population [19]. Then, while last eluted cells quickly (12–16 h) underwent strong apoptosis correlated with low cell proliferation, cells eluted in the first part of fractogram proliferated before undergoing late and strong apoptosis after 48 h subculture [19]. Here, SdFFF was used to better understand the possible link between the kinetics and the extent of a biological event (apoptosis) and cell status (cell cycle position), right at the start of induction. As differences in cell status lead to different biological responses to inducers, it could be of primary importance to determine this correlation in order to optimize drug efficacy. To achieve this goal, SdFFF was used to prepare subpopulations from crude K562 cells. Then, prior any apoptosis induction, cell cycle was analyzed. As previous studies have shown that diosgenin could affect cell cycle [58–60], this protocol avoided any cell cycle modification due to the inducer. Moreover, a short incubation time (6 h) was selected to minimize the elapsed time between cell cycle and apoptosis analyses. It also represented the minimal time to quantify (flow cytometry) and monitored (SdFFF) early step apoptosis (specific to induced apoptosis). Results showed that SdFFF can separate subpopulations based on their position in the cell cycle, producing biological tools regarding to apoptosis sensitivity. Fractions (F1, F2), which contained a majority of cells in S/G2/M phases (active part of cell cycle), led to the least apoptotic populations after 6 h incubation. In contrast, F3 cells which were principally in G1/G0 phase (less active part of cell cycle) became the most apoptotic cell population. Statistical analysis showed a correlation between G0/G1 cell cycle position and early

1160

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161

apoptosis induction with a correlation factor = 0.985 (p < 0.05), and a negative correlation between S/G2/M cell cycle position and early apoptosis induction with a correlation factor = 0.999 (p < 0.05). These results are in agreement with those previously described [19], where the less proliferative cells, found here to be in G0/G1 phase, led to massive apoptosis after 12–16 h incubation, while the most proliferative cells underwent to apoptosis only after 24 or 48 h incubation. Finally, the association of SdFFF monitoring and size measurement also led to the description of different subpopulations with regards to their biophysical properties demonstrating the complex variation in density between fractions which could be associated with differences in biological processes. 4. Conclusion In this study, we propose a new SdFFF cell separation application in the field of cancer research to examine the correlation between induction of biological event and cell status. To achieve this goal, a crude cell population was eluted and fraction collected to obtain specific subpopulations with regards to the cell status of interest. On the basis of our previous work concerning diosgenin-dependent apoptosis induction in K562 cells, our goal was to establish a possible correlation between apoptosis kinetics and cell cycle position when diosgenin exposure started. Then, subpopulations with different cell cycle stages were prepared by SdFFF and used to study apoptosis kinetics and extent as early as 6 h incubation. Results showed that K562 cells in the active part of the cell cycle (S/G2/M, F1 cells) underwent less apoptosis, while cells eluted later (F3, G0/G1), proceeded to apoptosis earlier. These results confirm the difference in apoptosis kinetics in the K562 population, difference which could be now linked to cell cycle position when induction began. SdFFF monitoring and size measurement allowed the description of complex variations in size/density balance associated with biological process kinetics. Acknowledgements Authors are grateful to Dr J. Cook-Moreau for English and style corrections in the preparation of this manuscript. The expenses of this work were defrayed in part by the “Ministère de l’Education Nationale, de la Recherche et de la Technologie”, the “Conseil Régional du Limousin” and by the “Ligue contre le Cancer” (Comité du Limousin). References [1] J.C. Bigelow, J.C. Giddings, Y. Nabeshima, T. Tsuruta, K. Kataoka, T. Okano, N. Yui, Y. Sakurai, J. Immunol. Methods 117 (1989) 289. [2] J.J. Chalmers, M. Zborowski, L. Sun, L. Moore, Biotechnol. Prog. 14 (1998) 141. [3] T. Ikeya, K. Kataoka, T. Okano, Y. Sakurai, React. Funct. Polym. 37 (1998) 251. [4] Y. Huang, J. Yang, X.B. Wang, F.F. Becker, P.R. Gascoyne, J. Hematho. Stem Cell 8 (1999) 481. [5] B.K. Gale, K. Caldwell, A.B. Frazier, Micro Total Analysis Systems 2000, Proceedings of the.mu.TAS Symposium, 4th, Enschede, Netherlands, May 14–18, 2000, p. 399. [6] X.-B. Wang, J. Yang, Y. Huang, J. Vykoukal, F.F. Becker, P.R.C. Gascoyne, Anal. Chem. 72 (2000) 832. [7] P. Reschiglian, A. Zattoni, B. Roda, L. Cinque, D. Melucci, B.R. Min, M.H. Moon, J. Chromatogr. A 985 (2003) 519. [8] P.S. Williams, L.R. Moore, D. Leigh, M. Zborowski, Abstracts of Papers, 225th ACS National Meeting, New Orleans, LA, USA, March 23–27, 2003, ANYL. [9] M. Zborowski, J.J. Chalmers, Methods Mol. Biol. 295 (2005) 291. [10] L. Cinque, P.S. Williams, M. Zborowski, Abstracts of Papers, 232nd ACS National Meeting, San Francisco, CA, USA, September 10–14, 2006, ANYL. [11] M. Radisic, R.K. Lyer, S.K. Murthy, Int. J. Nanomed. 1 (2006) 3. [12] X. Tong, L. Yang, J.C. Lang, M. Zborowski, J.J. Chalmers, Cytom. B: Clin. Cytom. 72 (2007) 310. [13] Y. Jing, M.L.R.T. Schneider, P.S. Williams, J.J. Chalmers, S.S. Farag, B. Bolwell, M. Zborowski, Exp. Hematol. 35 (2007) 662. [14] K.D. Caldwell, Z.Q. Cheng, P. Hradecky, J.C. Giddings, Cell Biophys. 6 (1984) 233.

[15] P.J.P. Cardot, Y. Denizot, S. Battu, in: J. Cazes (Ed.), Encyclopedia of Chromatography, M. Dekker, Inc., New York, 2005, p. 282. [16] G. Bégaud-Grimaud, S. Battu, P. Lazcoz, J.S. Castresana, M.O. Jauberteau, P.J.P. Cardot, Int. J. Oncol. 31 (2007) 883. [17] G. Bégaud-Grimaud, S. Battu, B. Liagre, D.Y. Léger, J.L. Beneytout, P.J.P. Cardot, J. Chromatogr. A 1128 (2006) 194. [18] I. Comte, S. Battu, M. Mathonnet, B. Bessette, F. Lalloue, P. Cardot, C. Ayer-Le Lievre, J. Chromatogr. B 843 (2006) 175. [19] J. Bertrand, B. Liagre, G. Bégaud-Grimaud, M.O. Jauberteau, P. Cardot, J.L. Beneytout, S. Battu, J. Chromatogr. B 869 (2008) 75. [20] L. Guglielmi, S. Battu, M. Le Bert, J.L. Faucher, P.J.P. Cardot, Y. Denizot, Anal. Chem. 76 (2004) 1580. [21] D.Y. Leger, S. Battu, B. Liagre, P.J.P. Cardot, J.L. Beneytout, J. Chromatogr. A 1157 (2007) 309. [22] P. Reschiglian, A. Zattoni, B. Roda, E. Michelini, A. Roda, Trends Biotechnol. 23 (2005) 475. [23] E. Urbankova, A. Vacek, J. Chmelik, J. Chromatogr. B 687 (1996) 449. [24] J. Bauer, J. Chromatogr. B 722 (1999) 55. [25] T. Chianea, N.E. Assidjo, P.J.P. Cardot, Talanta 51 (2000) 835. [26] S. Battu, J. Cook-Moreau, P.J.P. Cardot, J. Liq. Chromatogr. Relat. Technol. 25 (2002) 2193. [27] M.A. McClain, C.T. Culbertson, S.C. Jacobson, N.L. Allbritton, C.E. Sims, J.M. Ramsey, Anal. Chem. 75 (2003) 5646. [28] J.C. Giddings, Sep. Sci. 1 (1966) 123. [29] J.C. Giddings, Science 260 (1993) 1456. [30] J.C. Giddings, in: M.E. Schimpf, K. Caldwell, J.C. Giddings (Eds.), Field-Flow Fractionation Handbook, John Wiley & Sons, Inc., New York, 2000, p. 3. [31] S. Li, J. Nickels, A.F. Palmer, Biomaterials 26 (2005) 3759. [32] M. Andersson, K. Fromell, E. Gullberg, P. Artursson, K.D. Caldwell, Anal. Chem. 77 (2005) 5488. [33] C.I.J. Andersson, C. Arfvidsson, P.T. Kallio, K.-G. Wahlund, L. Buelow, Biotechnol. Lett. 25 (2003) 1499. [34] C. Arfvidsson, G. Wahlund Karl, Anal. Biochem. 313 (2003) 76. [35] Z. Chen, A. Chauhan, J. Colloid Interf. Sci. 285 (2005) 834. [36] S. Battu, W. Elyaman, J. Hugon, P.J.P. Cardot, Biochim. Biophys. Acta 1528 (2001) 89. [37] L. Farmakis, A. Koliadima, Biotechnol. Prog. (2005), ACS ASAP. [38] M.T. Garcia, R. Sanz, M.T. Galceran, L. Puignou, Biotechnol. Prog. 22 (2006) 847. [39] P. Gascoyne, J. Satayavivad, M. Ruchirawat, Acta Trop. 89 (2004) 357. [40] D. Kang, M.H. Moon, Anal. Chem. 77 (2005) 4207. [41] C. Lautrette, P.J.P. Cardot, C. Vermot-Desroche, J. Wijdenes, M.O. Jauberteau, S. Battu, J. Chromatogr. B 791 (2003) 149. [42] H. Lee, S.K.R. Williams, K.L. Wahl, N.B. Valentine, Anal. Chem. 75 (2003) 2746. [43] J. Luo, M. Leeman, A. Ballagi, A. Elfwing, Z. Su, J.-C. Janson, K.-G. Wahlund, J. Chromatogr. A 1120 (2006) 158. [44] A. Roda, D. Parisi, M. Guardigli, A. Zattoni, P. Reschiglian, Anal. Chem. 78 (2006) 1085. [45] R. Sanz, P. Cardot, S. Battu, M.T. Galceran, Anal. Chem. 74 (2002) 4496. [46] G. Yohannes, M. Sneck, S.J.O. Varjo, M. Jussila, S.K. Wiedmer, P.T. Kovanen, K. Oeoerni, M.-L. Riekkola, Anal. Biochem. 354 (2006) 255. [47] J. Yang, Y. Huang, X.-B. Wang, F.F. Becker, P.R.C. Gascoyne, Anal. Chem. 71 (1999) 911. [48] X.B. Wang, J. Yang, Y. Huang, J. Vykoukal, F.F. Becker, P.R. Gascoyne, Anal. Chem. 72 (2000) 832. [49] C. Corbière, S. Battu, B. Liagre, P.J.P. Cardot, J.L. Beneytout, J. Chromatogr. B 808 (2004) 255. [50] D.Y. Leger, S. Battu, B. Liagre, J.L. Beneytout, P.J.P. Cardot, Anal. Biochem. 355 (2006) 19. [51] D.Y. Leger, B. Liagre, P.J.P. Cardot, J.L. Beneytout, S. Battu, Anal. Biochem. 335 (2004) 267. [52] K. Higdon, A. Scott, M. Tucci, H. Benghuzzi, A. Tsao, A. Puckett, Z. Cason, J. Hughes, Biomed. Sci. Instrum. 37 (2001) 281. [53] J. Turchan, C.B. Pocernich, C. Gairola, A. Chauhan, G. Schifitto, D.A. Butterfield, S. Buch, O. Narayan, A. Sinai, J. Geiger, J.R. Berger, H. Elford, A. Nath, Neurology 60 (2003) 307. [54] J. Raju, J.M. Patlolla, M.V. Swamy, C.V. Rao, Cancer Epidemiol. Biomarkers Prev. 13 (2004) 1392. [55] J.L. Beneytout, C. Nappez, M.J. Leboutet, G. Malinvaud, Biochem. Biophys. Res. Commun. 207 (1995) 398. [56] D.Y. Leger, B. Liagre, J.L. Beneytout, Int. J. Oncol. 28 (2006) 201. [57] B. Liagre, P. Vergne-Salle, C. Corbiere, J.L. Charissoux, J.L. Beneytout, Arthritis Res. Ther. 6 (2004) 373. [58] D.Y. Leger, B. Liagre, C. Corbiere, J. Cook-Moreau, J.-L. Beneytout, Int. J. Oncol. 25 (2004) 555. [59] S. Moalic, B. Liagre, C. Corbiere, A. Bianchi, M. Dauca, K. Bordji, J.L. Beneytout, FEBS Lett. 506 (2001) 225. [60] C. Corbière, B. Liagre, F. Terro, J.L. Beneytout, Cell Res. 14 (2004) 188. [61] C. Corbiere, B. Liagre, A. Bianchi, K. Bordji, M. Dauca, P. Netter, J.L. Beneytout, Int. J. Oncol. 22 (2003) 899. [62] J. Raju, R.P. Bird, Cancer Lett. 255 (2007) 194. [63] C. Cailleteau, B. Liagre, S. Battu, C. Jayat-Vignoles, J.L. Beneytout, Anal. Biochem. 380 (2008) 26. [64] B. Liagre, J. Bertrand, D.Y. Leger, J.L. Beneytout, Int. J. Mol. Med. 16 (2005) 1095. [65] X. Tong, K.D. Caldwell, J. Chromatogr. B 674 (1995) 39. [66] K.D. Caldwell, T.T. Nguyen, T.M. Murray, M.N. Myers, J.C. Giddings, Sep. Sci. Technol. 14 (1979) 935.

J. Bertrand et al. / J. Chromatogr. B 877 (2009) 1155–1161 [67] K.D. Caldwell, in: M.E. Schimpf, K.D. Caldwell, J.C. Giddings (Eds.), Field-Flow Fractionation Handbook, John Wiley & Sons, Inc., New York, 2000, p. 79. [68] J. Chmelik, J. Chromatogr. A 845 (1999) 285. [69] M. Martin, P.S. Williams, in: F. Dondi, G. Guiochon (Eds.), Theoretical Advancement in Chromatography and Related Separation Techniques, Kluwer, Dordrecht, 1992, p. 513. [70] J. Plockova, F. Matulik, J. Chmelik, J. Chromatogr. A 955 (2002) 95.

[71] [72] [73] [74] [75] [76] [77]

M.R. Schure, K.D. Caldwell, J.C. Giddings, Anal. Chem. 58 (1986) 1509. P.S. Williams, S. Lee, J.C. Giddings, Chem. Eng. Commun. 130 (1994) 143. P.S. Williams, J.C. Giddings, Anal. Chem. 66 (1994) 4215. M.J. Iatropoulos, G.M. Williams, Exp. Toxicol. Pathol. 48 (1996) 175. G.L. Johnson, R. Lapadat, Science 298 (2002) 1911. W. Woessmann, N.F. Mivechi, Exp Cell Res. 264 (2001) 193. A.H. Wyllie, Br. Med. Bull. 53 (1997) 451.

1161