Cell cycle-arresting drugs, thapsigargin (Tg) and chloroquine (CQ), are employed to study endoplasmic reticulum (ER) stress and the autophagic process using cell lines without measuring the cell cycle of such cells. The potential cell cycle-dependent aspect of such processes in cell lines may impact upon the degree of ER stress and autophagy measured. ER stress is known to be caused by a build-up of misfolded proteins within the ER, which may then undergo ER phagy or reticulophagy. The cell cycle-dependent nature of all these processes is not well studied, so we investigated ER stress and autophagy by use of a combination of flow cytometric assays. These included cell cycle-dependent measurement of reticulophagy, misfolded protein levels and autophagic marker LC3-II in K562 and Jurkat cells. ER stress-inducing drug Tg caused significant reticulophagy in both cell types. This was cell cycle dependent in K562 cells only, with proliferating cells undergoing more reticulophagy. In contrast, autophagy-initiating drug CQ caused reticulophagy at higher doses in Jurkat cells, whereas K562 cells showed a cell cycle-dependent elongation of the ER, which was less pronounced in proliferating cells. The level of cellular misfolded protein in response to both drugs was high in K562 cells when either undergoing reticulophagy or elongation in a non-cell cycle-dependent manner, whereas the misfolded protein levels in Jurkat cells in response to both drugs were lower than those observed in K562 cells. Both cell lines employed in this study showed no increase of LC3-II above controls in response to Tg treatment. However, CQ induced a cell cycle-dependent increase of LC3-II in both cell types. Thus, the type of cell employed and the cell cycle dependent modulation of thebiological processes involved in ER stress and autophagy should be considered when designing studies in ER stress and autophagy.
- ER stress
- misfolded proteins
- flow cytometry
Endoplasmic reticulum (ER) stress has been implicated in numerous degenerative neurological diseases and cancer; elucidation of the mechanisms involved may determine drug targets for the treatment of such diseases [1, 2]. ER stress is known to cause autophagy and ultimately cell death via apoptosis, the mechanism of which is only beginning to be understood [3–5]. ER stress can be induced by a range of drugs, including thapsigargin (Tg) which acts by inhibition of ER ATPase located in the ER, resulting in the accumulation of misfolded proteins within the ER . Tg is also known to cause cell arrest in G1 phase of the cell cycle and has been shown to inhibit autophagic flux by inhibition of the translocation of Rab7, a protein required for the fusion of lysosomes with autophagosomes [6–8]. ER stress resulting from a build-up of misfolded proteins occurs once a threshold of misfolded protein accumulation has been reached. This then initiates the unfolded protein response (UPR) by the ER stress sensor signalling proteins, IRE1, PERK and ATF6 which initiate protein refolding and elongation of the ER until ER homeostasis is returned [3, 9, 10]. This process is a coping mechanism which reduces the mass of misfolded protein per unit volume of ER. If the high level of misfolded protein persists, then reticulophagy (phagy of the ER) occurs . Chloroquine (CQ) is also known to initiate autophagy and block completion of the autophagic process by increasing the pH within the lysosome, with resultant inhibition of the lysosome fusion process with autophagosomes . CQ was investigated for its ER stress-inducing qualities by comparison with Tg upon Jurkat T cells and K562 erythromyeloid cell lines. Thus measurement of changes in ER mass and misfolded protein levels would give a measure of the degree of ER stress within a cell.
The term autophagy or type II cell death is derived from the Greek roots “auto” (self) and “phagy” (eat) and was first observed by Porter in 1962 [13, 14]. Autophagy or macroautophagy is an intracellular degradation system that maintains cell homeostasis and is characterised by the formation of a double membrane around the cytosolic components to be degraded, forming an autophagosome of sequestered malfunctioning components ranging from misfolded proteins to organelles such as stressed ER [15, 16]. An autophagosome then fuses with lysosomes, giving rise to an autolysosome, where the intracellular components are degraded by hydrolases which produces energy, thus promoting cellular haemostasis [2, 15, 17–20]. The main biological autophagy marker is the microtubule-associated protein LC3-II or LC3B. LC3-I is normally located in the cytoplasm but when cleaved and lipidated by phosphatidylethanolamine is then incorporated into the autophagosome inner leaflet of the membrane in the form of LC3-II [21–23].
Methods for monitoring autophagy began with the initial discovery of the process by the use of electron microscopy, which showed the presence of autophagosome and autolysosome or autophagolysosome [13, 14]. Biochemical techniques such as Western blotting can be used to quantitate the degree of autophagy in cells by measuring the autophagy marker proteins, LC3-II and LC3-I [21–23]. Fluorescently tagged LC3-II can also be imaged and flow cytometrically analysed through transfections with GFP-RFP through transfections with GFP-RFP, with the benefit that GFP fluorescence is dissipated by the acidic conditions prevailing in autolysosomes, whereas RFP is not, thus making LC3-II-GFP detection specific for autophagosomes and LC3-II-RFP specific for autolysosomes [22, 24–26]. The number and intensity of fluorescently labelled anti-LC3-II-positive puncta (autophagosomes-autolysosomes) can also be quantitated by time-consuming image analysis, whereas measuring the increase in median fluorescent values of LC3-II antigen level flow cytometrically makes the process significantly less burdensome, especially when combined with cell cycle analysis [25, 27–29]. Here we show how flow cytometry can be used to measure not only the autophagy marker LC3-II but also the cellular end products of ER stress which include reticulophagy and cellular levels of misfolded proteins and their cell cycle distribution. Most studies using cell lines have not generally focussed upon the cell cycle distribution of the autophagy marker even though autophagy-inducing drugs such as rapamycin are known to cause G1 cell cycle arrest. However a few studies have investigated the cell cycle distribution of LC3-II with variable results which seem to vary depending upon the cell line employed [17, 30, 31]. Thus this chapter investigates whether the autophagic biological marker, LC3-II, and end products of ER stress (ER mass and misfolded protein) showed a cell cycle-dependent nature during ER stress. Drugs such as Tg and CQ, which cause the build-up of misfolded proteins and induce ER stress, were then compared in terms of cell cycle-dependent reticulophagy, misfolded protein levels and LC3-II .
Here, we employed flow cytometric cell cycle analysis of live cells to measure reticulophagy combined with an assay employing fixed-cell immunofluorescence analysis of LC3-II developed in this laboratory and the cell cycle analysis of cellular misfolded proteins [31, 33, 34]. The use of ER Tracker was used as previously described but was now employed in a cell cycle-dependent manner to determine the relative change in ER mass compared to untreated control cells [35, 36]. We also determined the degree of misfolded protein aggregate formation in a cell cycle-dependent manner using the Proteostat probe (Enzo Life Sciences) that fluoresces when bound to misfolded proteins in fixed cells . Thus, we were able to investigate whether the observed drug-induced ER stress had a cell cycle-dependent nature by the measurement of changes in ER mass, misfolded proteins and the autophagy marker LC3-II.
2. Materials and methods
2.1. Induction of ER stress
Jurkat and K562 cells were grown in RPMI-1640 with
2.2. Cell cycle and ER Tracker
Jurkat and K562 cells with or without treatment were adjusted to 1 × 106/ml and loaded with Hoechst 33342 (15 μg/ml, Sigma Chemicals, UK) and ER Tracker Red dye (ERTR 100 nM, Invitrogen, UK), by incubating cells with the dyes for 1 and 0.5 h at 37°C, respectively. Dead cells were detected by DRAQ7 (2.5 μM, Biostatus, UK), incubated with cells for 10 min at 37°C. Live cells were analysed for ERTR Area MFI levels from Ho33343 Area and Width dot plots; see Figure 1. The percentage change in ER mass test samples was determined by comparing test and control values for each phase of the cell cycle comparing control values for each phase of the cell cycle: 30,000 events collected by flow cytometry, with Ho33342 excited at 350–60 nm (UV laser) and emission collected at 450/50 nm; ERTR Area excited at 488 nm (blue laser) and emission collected at 610/10 nm; and DRAQ7 excited at 633 nm (red laser) and emission collected at 780/60 nm (n = 3).
2.3. Misfolded protein labelling
Jurkat and K562 cells with or without treatment were pelleted and resuspended in 200 μl of Solution A fixative for 15 min at room temperature (RT) (Caltag, UK). Cells were then washed in PBS buffer (Invitrogen, UK) and cell pellets permeabilised with 0.25% Triton X-100 (Sigma Chemicals, UK) for 15 min at RT. Cells were then washed in PBS buffer and a 1:20,000 dilution of Proteostat (Enzo Life Sciences, UK) was made in Proteostat buffer (1 μl of Proteostat reagent in 20 ml of buffer) and 400 μl added to the cell pellet with 1 μg/ml DAPI and incubated for 30 min at RT. Flow cytometric analysis was performed with excitation of Proteostat with the blue 488 nm laser and excitation of DAPI with UV laser line and emissions collected at 610/20 nm and 440/40 nm, respectively (n = 3). Cell cycle analysis was performed by collecting 30,000 events with the DAPI 440/40 Area and Width parameter for doublet discrimination; see Figure 3. Then Proteostat Area MFI for cells in G1, S phase and G2m cycling cells was determined by comparison of control and test MFI for each phase of the cell cycle by the formula given in the manufacturer’s instructions, (MFItest − MFIcon)/MFItest × 100 = misfolded protein level.
2.4. Indirect immunofluorescence LC3-II labelling and cell cycle analysis
Jurkat and K562 cells with or without treatment were pelleted and resuspended in 200 μl of Solution A fixative for 15 min at RT (Caltag, UK). Cells were then washed in PBS buffer and cell pellets permeabilised with 0.25% Triton X-100 for 15 min at RT. Cells were washed in PBS and anti-LC3-II monoclonal antibody (1:400 dilution) (Cat. No. 3868, Cell Signalling Technology Inc., USA) incubated for 0.5 h at RT. Cells were then washed in PBS and labelled with 0.125 μg of secondary fluorescent conjugate Alexa Fluor-647 goat anti-rabbit IgG (Invitrogen, UK) for 0.5 h at RT. Cells were then washed in PBS buffer and resuspended in 400 μl of PBS with DAPI (1 μg/ml). Analysis of upregulated LC3-II-Alexa Fluor-647 MFI signal in treated samples (above control levels) was determined for each phase of the cell cycle by gating cells on a FSC versus SSC dot plot, with doublet discrimination achieved by use of the DAPI Area and Width parameters with gating for G1, S and G2m on the DAPI-A parameter; see Figure 5. Thirty thousand events were collected by flow cytometry. DAPI was excited at 350–60 nm (UV laser) and emission collected at 450/50 nm; LC3-II-AF-647 was excited at 633 nm (red laser) and emission collected at 660/20 nm (n = 3).
2.5. Flow cytometry
A Becton Dickinson LSRII with FACSDiva software (ver 6.3.1) fitted with blue (488 nm), red (633 nm), violet (405 nm) and UV (350–360 nm) lasers. The optical filters (fitted in 2005) used were for ERTR (610/10 nm), Ho33342 and DAPI (450/50 nm), DRAQ7 (780/60 nm), LC3B-AF-647 (660/20 nm) and Proteostat (610/10 nm).
Student t tests were performed in GraphPad software with P > 0.05 not significant (NS), P < 0.05*, P < 0.01**, n = 3.
3.1. Flow cytometric analysis of cell cycle-dependent reticulophagy
For estimating the degree of reticulophagy after treatment with Tg or CQ, known ER stress inducers and interceptor of autophagy, cells were labelled with Ho33342, ERTR and DRAQ7. Live cells were analysed for reticulophagy after flow cytometric analysis on a Becton Dickinson LSRII after UV, blue and red laser excitation and fluorescence collection at 460, 610 and 780 nm, respectively. The gating strategy for the analysis of live cell cycle-dependent reticulophagy of such cells is shown in Figure 1.
Tg induced a high degree of K562 cell cycle-dependent reticulophagy (29–76%), which was greater in S, G2m and G1 phases with increasing [Tg] (P < 0.05, 0.01, n = 3, Figure 2A), whereas Jurkat cells displayed a lower level of reticulophagy 35 to >55%, which was not cell cycle dependent (P < 0.01, n = 3, Figure 2B). The K562 response to CQ was very different to that of Tg with an ER elongation of >10 to >40%, with the highest degree of ER stress (smallest amount of elongation) present in G2m cells (NS, P < 0.05, <0.01, n = 3, Figure 2C). CQ treatment of Jurkat cells displayed a lower level of non-cell cycle-dependent ER elongation than that observed in K562 cells (<20%, P < 0.05, NS, n = 3, Figure 2D). However the higher dose CQ induced a reticulophagy (<20–>50%) which was also non-cell cycle dependent (P < 0.05, P < 0.01, n = 3, Figure 2D).
3.2. ER stress responses: protein misfolding
For estimating the overall level of misfolded proteins after ER stress induction, Tg or CQ cells were fixed, permeabilised and labelled with DAPI and the misfolded protein detection reagent, Proteostat . Flow cytometric analysis of the cell cycle and misfolded protein levels was performed on a Becton Dickinson LSRII after UV and blue laser excitation and fluorescence collected at 460 and 610, nm respectively. The gating strategy for the cell cycle analysis of misfolded proteins is shown in Figure 3.
Tg treatment of K562 cells showed a moderately high level of misfolded proteins which were neither dose nor non-cell cycle dependent (>40, P < 0.05, P < 0.01, n = 3, Figure 4A). Jurkat cells showed a low level of misfolded proteins when treated with Tg which was also not dose nor cell cycle dependent (<35, P < 0.01, NS, n = 3, Figure 4B). CQ treatment of K562 cells showed a variable level of misfolded proteins which was not cell cycle dependent (25–80, P < 0.01, n = 3, Figure 4C). In contrast Jurkat cells produced a lower degree of misfolded proteins when treated with CQ (25–40 P < 0.05, P < 0.01, n = 3)) which was cell cycle dependent in that cells in G2m had lower amounts of misfolded proteins (Figure 4D).
3.3. ER stress responses: LC3-II-associated autophagy is cell cycle dependent
For estimating the level of autophagy occurring in cells treated with ER stress-inducing drugs, Tg and CQ cells were fixed, permeabilised and labelled with DAPI cells and anti-LC3-II-Alexa Fluor-647 which were collected on a Becton Dickinson LSRII after UV and red laser excitation and fluorescence collected at 460 and 660 nm, respectively. The gating strategy for the cell cycle analysis of LC3-II is shown in Figure 5.
Although Tg induced a high degree of ER stress in K562 cells and a lower order in Jurkat cells after 24 h, LC3-II did not increase above control levels (data not shown). However CQ did induce a cell cycle-dependent increase of LC3-II for both cell lines employed in this study (Figure 6A and B). Both cell types showed a similar level of LC3-II in that 25 and 75 μM CQ showed a dose and cell cycle-dependent increase above control levels (Figure 6A and B), whereas 50 μM CQ showed a lower degree of LC3-II upregulation than the other doses of CQ employed in this study (Figure 6A and B).
Flow cytometry can be used to investigate ER stress, a biological process not much analysed by this experimental approach to date. Here we measured the end products of ER stress, that is, reticulophagy and misfolded proteins rather second messengers associated with the process, for example, PERK, IRE-1 and ATF6. However the use of live cells and ER Tracker probes to measure reticulophagy has been previously used to estimate the degree of reticulophagy [35, 36], although not in a cell cycle-dependent manner. Analysis of these cells under fixed conditions allowed flow cytometry to estimate the level of misfolded proteins present in the cell (rather than exclusively in the ER) in a cell cycle-dependent manner. The assays employed using just two or three fluorescent probes allowed for a relatively easy flow cytometric analysis, without the need for colour compensation or correction of the bleed through of the different fluorophores into each other, thus avoiding false readings and incorrect conclusions from the acquired data sets. Gating strategies used were as simple as possible, but given the relatively complex nature of the gating needed, they all employed the same approach. This study employed known ER stress-inducing drug Tg, as well as by comparison CQ, and used two cell types to show potential different responses to these drugs. To this end Tg was shown to induce ER stress after 24 h in both cell lines with the ER in such cells undergoing a significant degree of reticulophagy. K562 cells were more affected than Jurkat T cells, whereas only K562 cells showed a reticulophagy, which was more pronounced in specific phases of the cell cycle with the different concentrations of the drug.
The drug CQ, a known initiator of autophagy and apoptosis (although the drug also blocks the process at the lysosome-autophagosome fusion step), was also tested for its ER stress-inducing qualities [12, 32, 33]. CQ appeared to induce ER stress in both cell lines, with Jurkat cells treated with high concentration of CQ displaying a high level of reticulophagy like that observed with Tg. However lower doses of CQ induced ER stress which was typified by an elongation of the ER in Jurkat cells which was again not cell cycle dependent. This mode of action of CQ causing an elongation of the ER was repeated in K562 cells which was cell cycle dependent. Here the degree of elongation reduced as the cell moved through the cell cycle. Thus G2m cells showed the least amount of ER elongation and hence were more stressed than cells in the G1 phase of the cell cycle.
Induction of ER stress by both Tg and CQ was further confirmed by the detection of misfolded proteins in a non-cell cycle-dependent manner above that found in untreated cells (except G2m phase Jurkat cells treated with CQ). K562 cells had a high level of misfolded proteins in response to both Tg and CQ even though Tg induced a reticulophagy and CQ an ER elongation. Similarly Jurkat cells responded to Tg and CQ with a moderate level of misfolded proteins, with most cells displaying a reticulophagy in this instance. Thus, although cells were displaying ER elongation as well as reticulophagy with the different drugs, misfolded proteins were detectable and the highest level found in cells undergoing ER elongation rather than reticulophagy, this perhaps being a reflection that cells with less ER have less space for misfolded proteins and thus a lower level of misfolded proteins.
Although Tg induced a high degree of reticulophagy and misfolded proteins in both cell lines, there was no evidence of autophagy as LC3-II levels did not increase above control levels. However CQ shown previously and in this study to induce ER stress did show an increase LC3-II in a cell cycle-dependent manner in both cell lines.
Thus, the degree and mode of action of these two drugs appear to be cell type dependent, with K562 cells displaying a cell cycle-dependent ER stress response to both drugs, whereas Jurkat cells did not. However, the induction of the autophagic response (the process being blocked at the lysosome fusion step) to CQ was cell cycle dependent in both cell lines. While Tg did not induce an autophagic response in either cell line after 24 h. The type of cell employed and the cell cycle-dependent modulation of these biological processes involved in ER stress and autophagy should be considered when designing studies in ER stress and autophagy. Flow cytometry makes the analysis of these cell cycle-dependent events in the ER stress process easily measureable.