Journal of Experimental Botany, Vol. 52, No. 361, pp. 1615-1623,
August 1, 2001
© 2001 Oxford University Press
Original Papers |
Ethylene induces cell death at particular phases of the cell cycle in the tobacco TBY-2 cell line
1 School of Environmental Science and Land Management, University College Worcester, Henwick Grove, Worcester WR2 6AJ, UK
2 Department of Biology, Biotechnical Faculty, University of Ljubljana, Vecna pot 111, 1000 Ljubljana, Slovenia
3 School of Biosciences, Cardiff University, PO Box 915, Cardiff CF10 3TL, UK
Received 24 November 2000; Accepted 26 March 2001
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
It was examined whether ethylene induces programmed cell death in a cell cycle-specific manner. Following synchronization of the tobacco TBY-2 cell line with aphidicolin and its subsequent removal, ethylene was injected into the head space of 300 cm3 culture flasks at 0 h or 3.5 h later and cells were sampled for 26 h. There were significant increases in cell mortality at G2/M in both the 0 h and 3.5 h ethylene treatments, and for the latter treatment, another peak in S-phase. The effect at G2/M was greater in the 3.5 h treatment, but was ameliorated by the simultaneous addition of silver nitrate (1.2 µM). In addition, the 3.5 h ethylene treatment resulted in a 1 h delay in the characteristic rise in the mitotic index following aphidicolin-induced synchrony. The addition of silver nitrate alone (1.2 µM), also delayed the entry of cells into mitosis but had no effect on cell cycle length compared with the controls (14 h throughout all treatments) but it induced a peak of mortality 2.5 h after its addition. Nuclear shrinkage was also a characteristic feature of dying cells at G2/M. Using Apoptag®, an in situ apoptosis detection kit, nuclear DNA fragmentation was observed in the TBY-2 cells which were often isolated on the end of a filament of normal cells. In the 3.5 h ethylene treatment, a marked increase was noted in the percentage of such cells at the G2/M transition compared with the controls. Hence, the data show cell death occurring at a major phase transition of the cell cycle and the observations of nuclear shrinkage, isolation of dying cells and nuclear DNA fragmentation suggest a programmed mechanism of cell death exacerbated by ethylene treatment.
Key words: Cell cycle, cell death, ethylene, tobacco.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Programmed cell death (PCD) and apoptosis are terms often used interchangeably. However, PCD has been defined in animal systems as a type of cell death which is part of an organism's life cycle, is initiated by specific physiological signals, and requires de novo gene transcription (Ellis et al., 1991
). Apoptosis was originally used to define particular features of cells undergoing PCD; ultrastructural and biochemical characteristics including chromatin condensation, membrane blebbing and DNA laddering (Kerr et al., 1972). However, whether all PCD manifests apoptotic features is questionable because some animal cells appear to undergo PCD physiologically but without any of the apoptotic hallmarks (Schwartz et al., 1993). In plants, PCD has been identified in a number of developmental processes (Jones and Dangl, 1996) including tracheary element formation, destruction of the suspensor, aerenchyma development, and floral organ abortion. However apoptotic features can be detected in some cell types but not others. For example, ethylene-induced aerenchyma formation clearly involves PCD (He et al., 1996) but it is unclear whether the type of PCD in this system can be termed apoptotic.
Recent work in fission yeast and higher eukaryotes has shown that PCD can be tightly linked to cell cycle checkpoints (Hirao et al., 2000; Tanaka et al., 2000). In fission yeast and vertebrate cells, a G2/M checkpoint is characterized by phosphoregulation of Cdc2 kinase, a key enzyme, which when bound to its partner cyclinB, drives cells into mitosis (Nurse, 1990). Positive regulation is by Cdc25 phosphatase and negative regulation is mainly by Wee1 kinase (Russell and Nurse, 1986, 1987). In fission yeast, Cdc25 is also the target of a molecular network that checks the integrity of DNA at the G2/M checkpoint prior to cell division; cell cycle progression is blocked when DNA is damaged because Cdc25 is inactivated (Rhind and Russell, 1998; Hirao et al., 2000). The latter is then unable to dephosphorylate Cdc2 kinase so that cells are held in G2 (Lopez-Girona et al., 1999; reviewed by Rhind and Russell, 1998). Only when the damage is repaired is Cdc25 released into the nucleus to activate Cdc2 kinase (Lopez-Girona et al., 1999). If the damage is too extensive, the cell exits into PCD which is regulated by the universal guardian of the cell cycle, p53 (Lane, 1992; Hirao et al., 2000).
Testing the hypothesis that, in plants, exit into PCD is cell cycle-specific requires a cellular system free of developmental constraints and the consequent complexities of different cell types. In this laboratory, the tobacco TBY-2 cell line was used (first established from seedlings of Nicotiana tabacum cv. Bright Yellow in 1969) which can be induced to exhibit synchronous cell cycles following treatment with aphidicolin (Nagata et al., 1992; Francis et al., 1995). The intention was to induce cell death in the synchronized TBY-2 cells with a known inducer. Ethylene was chosen because it is known to induce PCD, for example, in cereal endosperms (Young and Gaille, 1999) and in roots, following hypoxia, where root cortical cell breakdown leads to aerenchyma formation (Drew et al., 1979; He et al., 1996). Hence, by commencing injection of ethylene at different times following the removal of aphidicolin, the TBY-2 cell line was a convenient tool with which to test the hypothesis that this plant growth regulator (pgr) induces cell death at specific points in the cell cycle.
Arguably, PCD is a phenomenon linked solely to developmental programmes but in Arabidopsis cell cultures, genes that are expressed in senescing organs also accumulated in the late, growth arrest phase of cell culture (Callard et al., 1996). Moreover, features associated with PCD, inter-nucleosomal cleavage and chromatin compaction, were observed in the same cell cultures at a stage corresponding to loss of cell viability (Callard et al., 1996). In this study, cell death in TBY-2 cells and two associated features of PCD, nuclear shrinkage and the generation of 3' ends in nuclear DNA together with a remarkable ethylene-induced enhancement of mortality and DNA fragmentation at G2/M, a major cell cycle checkpoint is reported.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material
The tobacco TBY-2 cell line was cultured in modified Linsmaier and Skoog medium (Linsmaier and Skoog, 1965
) and was subcultured at 7 d intervals as described before (Francis et al., 1995).
Synchronization and mitotic index measurements
Aphidicolin, a reversible inhibitor of DNA polymerase 
, was used to synchronize the cells using the exact protocol of Nagata et al. (Nagata et al., 1992). In the presence of this inhibitor, any cell replicating its nuclear DNA is arrested and all other cells are unable to enter S-phase. Hence, a 24 h exposure to, and subsequent removal of, aphidicolin will cause the vast majority of cycling cells to accumulate in late G1 and then be released into S-phase (for a full description of the synchronization method see Francis et al., 1995).
Following release from aphidicolin, 100 µl of cell suspension was taken at hourly intervals for 24 h, and mixed immediately with 20 µl Hoechst stain (Bisbenzimide Sigma, 100 µg cm-3 in 2% (v/v) Triton X-100). The mitotic index (the sum of prophase, metaphase, anaphase, and telophase mitotic figures as a percentage of all cells) was measured for a minimum of 200 cells per slide on random transects across the coverslip on each of three replicate slides per sampling time per treatment. Cells and mitotic figures were visualized using a fluorescence microscope (Olympus BH2, UV, 
=420 nm).
Cell viability and mortality index
Following release from aphidicolin, mortality indices were scored at 2 h intervals for up to 26 h. Cells (50 µl) were diluted 1 : 1 in a mixed solution of fluorescein diacetate (Sigma, 200 µg ml-1, FDA) and propidium iodide (Sigma, 120 µg ml-1, PI) in 3% (w/v) sucrose on ice (20 min) (Harris and Oparka, 1994). A minimum of 200 cells were scored as dead (red) or alive (green) on random transects across the coverslip on each of three replicate slides every 2 h per treatment.
Monitoring the expression of histone H4
At hourly intervals following synchronization total RNA was isolated using Tri-Reagent (Sigma). Formaldehyde gels for northern analysis were as described previously (Rogers et al., 1992) using 10 µg of total RNA. Equal loading was checked by ethidium bromide staining of agarose gels. RNA was transferred to nylon membranes by capillary blotting. Prehybridization and hybridization were performed at 40 °C in a solution containing 5xDenharts, 6xSSC, 0.1% SDS, 5% PEG, 0.1% tetrasodium pyrophosphate, and 100 µg ml-1 denatured E. coli DNA. Random primed probes were prepared as described previously (Feinberg and Vogelstein, 1983) using 196 bp from the coding region of an Arabidopsis thaliana H4 gene (Chaboute et al., 1987). Blots were washed in 2xSSC and 0.1% SDS at 40 °C.
Ethylene and ethylene/silver treatments
Preliminary experiments on synchronized cells established a concentration of ethylene in the head space that resulted in a significant, albeit not catastrophic, mortality rate in the TBY-2 cells. At 3.5 h following the release from synchrony, ethylene was injected into the head space through a layer of Nescofilm into a 300 ml culture flask containing the cell suspension in 95 ml of medium. The flask was then sealed with two layers of aluminium foil. This procedure was repeated each time cells were sampled at 8, 9 or 10 h following release from aphidicolin. In the head space, the concentration used (µl l-1) and mean percentage mortalities (%), pooled from all three sampling times, were: 100 µl l-1 (2.10±0.17%); 8850 µl l-1 (2.62±0.04%); 12 400 µl l-1 (7.20±0.96%); 17 700 µl l-1 (21.07±5.8%), and 35 000 µl l-1 (24.92±5.6%) compared with 3.72±0.63% for the controls. The threshold concentration that gave an appreciable but not an overwhelming rise in mortality was chosen to be 17 700 µl l-1. It was calculated that although this is a large gaseous amount of ethylene, its concentration in pure water at this partial pressure at 25 °C would be 73 µmol l-1, but its solubility in a concentrated and complex solution of the Linsmaier and Skoog (Linsmaier and Skoog, 1965) medium would be considerably less because of competition with the solutes therein.
In separate experiments, 5 cm3 of pure ethylene (SIB Analytical, Sandwich, Kent, UK) was injected into the head space of the 300 ml flask (17 700 µl l-1) at 0 h or 3.5 h after the release from aphidicolin. These are referred to from here on as the 0 h or 3.5 h ethylene treatments in which sampling was terminated at 9 h and 25 h, respectively.
Silver is a well-established inhibitor of ethylene action (Drew et al., 1981). In the ethylene+silver treatment, following the removal of aphidicolin, the washed cells were transferred to culture medium and 3.5 h later, silver nitrate was added to give a final concentration of 1.2 µM together with ethylene and injected as above. The concentration used here (1.2 µM AgNO3) was determined by preliminary experiments and was sufficiently high to produce an appreciable ameliorative effect on ethylene-induced cell death but was not completely toxic in the TBY-2 culture medium.
In a separate experiment, silver nitrate (1.2 µm) was added at 3.5 h, as before, to replicate flasks. This is referred to as the silver treatment in which measurements were terminated at 18 h and 26 h, for mortality and cell cycle, respectively.
Cell and nuclear area
Digital images of the FDA/PI stained cells were captured and measured using SigmaScan®. Cell and nuclear areas were measured in random transects for both living and dead cells for the controls and 3.5 h ethylene treatment, at 3, 6, 8, and 10 h following the release from aphidicolin.
Detection of 3'-OH DNA termini
The ApopTag® apoptosis detection kit was used. Cells from the 3.5 h ethylene and control treatments were sampled every 2 h. At each sampling time, 1 cm3 of cells was removed, centrifuged at 1000 rpm (10 min), the supernatant discarded, the pellet of cells fixed in 1% paraformaldehyde in PBS (v/v) and stored at 4 °C. Cells (70 µl) were placed on alcohol-cleaned slides, a coverslip added and the slide placed on dry ice until the coverslip became heavily frosted (5–10 min); it was then removed with a razor blade (Conger and Fairchild, 1953). The preparation was then allowed to dry at room temperature. Subsequently, the procedure outlined in the manual for indirect fluorescence was followed. The counter-staining procedure outlined in the ApopTag® manual was modified in that the cells were mounted in PBS, c. PI (120 µg cm-3). The cells were observed as above using a digital camera (Fujitsu HC-300Z). At each sampling time, the number of green (antibody labelled) or red (PI-stained) nuclei was scored and the results expressed as % labelled (approximately 300 cells per slide per sampling time per treatment).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
S-phase duration confirmed as 3.5 h
Histone H4 expression is a sensitive marker of S-phase. Its level of expression fell away 3–4 h following the release of the TBY-2 cells from aphidicolin (Fig. 1
). This fits well with the more sensitive measure of S-phase duration of 3.5 h as measured from the mitotic index curve (calculated as described in the legend to Fig. 2). This pattern of H4 expression is remarkably similar to that shown in a recent publication on synchronized TBY-2 cells which also exhibited an S-phase of 3.5 h (Sorrell et al., 1999).
|
|
) or in cells to which ethylene was added (
) at 0 h (•). The vertical bars represent ±SE; where no bars are shown the ±SE was less than the diameter of the symbol (n=3). The mitotic index data generated a curve with a single peak enabling some of the component phases of the cell cycle to be calculated (Quastler and Sherman, 1959
Ethylene and ethylene+silver treatments affect the component phases but not the overall length of the cell cycle
In the 0 h ethylene treatment (Fig. 2
), the experiment was terminated after 10 h enabling the measurement of S-phase, G2 and M-phase which were unaffected relative to the control: G2=5 h, S=3.5 h, M=2 h in both treatments.
A substantial effect of the 3.5 h ethylene treatment was to delay the initial rise in the first mitotic index peak (Fig. 3a ) because G2 lengthened to 6.5 h compared with 5 h in the controls (Fig. 3a). Given that S-phase was 3.5 h in the controls (Figs 1, 2), the 3.5 h ethylene treatment was given after the majority of cells had finished DNA replication. Note that compared with the control, ethylene had no effect on cell cycle length (interval between peaks: 14 h in both treatments; Fig. 3a).
|
) or (c) in cells to which silver nitrate (s) was added at 3.5 h to two replicate flasks (
,
). The vertical bars represent ±SE; where bars are absent, the±SE was less than the diameter of the symbol (n=3). The mitotic index data generated curves exhibiting two peaks from which the duration of the cell cycle and its component phases were calculated (Quastler and Sherman, 1959The silver ions in the ethylene+silver treatment would be predicted to block the effects of ethylene. This treatment lengthened G2 slightly (5.5 h compared with 5 h in control) and M-phase (3 h compared with 2 h in control) (Fig. 2b
) and S-phase were slightly shortened (3 h compared with 3.5 h in the control, Fig. 2 and as aligned beneath Fig. 5b). However, cell cycle length was unaffected by this treatment, notwithstanding the presence of a small peak in the curve at 14 h (Fig. 2b). This either indicates a very rapid cell cycle of 6 h, or, more likely, the presence of a small cycling population that has reached its first mitosis.
|
To test whether the silver ions in the above treatment were causing any toxic effects on the cell cycle, a further experiment was undertaken comprising the addition of silver nitrate alone (1.2 µm) at 3.5 h. The silver treatment delayed the rise in the mitotic index yet further until 6–7 h (Fig. 3c
) but, again, cell cycle length remained at 14 h. Neither S-phase nor M-phase were affected by this treatment, but there was a compensatory decrease in G1 to 2 h compared with 3.5 h in the control (Fig. 3c).
The treatments spanning full cell cycles, were ranked in terms of the area under the mitotic index curve as an approximate indication of the proportion of cycling cells (Fig. 3): silver<ethylene+silver<3.5 ethylene<control. In other words, the silver treatment caused the largest reduction in the proportion of rapidly cycling cells which, it is concluded, is the main toxic effect of this ion on the cell cycle. However, relative to the controls, neither the silver, nor any of the other treatments perturbed the rate at which cycling cells divided (14 h throughout).
Ethylene treatment caused cell cycle-specific rises in mortality at G2/M and early S-phase
Having established the duration of each phase of the cell cycle for each treatment, it was then examined whether mortality showed any treatment-induced cell cycle-specificity.
In the 0 h ethylene treatment (Fig. 4), the mortality index was no different to the control at 2 h, but increased by 2-fold (from 4% to 8%) at 4 h and rose again at 8 h and 10 h to 13% and 17%, respectively, at which point this experiment was terminated. Aligning the phases of the cell cycle below the x-axis indicated that the largest incidences of mortality coincided with G2/M-phase and rising into M-phase (Fig. 4). These data suggested a cell cycle-specific response to ethylene but it was possible that cells were differentially sensitive to ethylene depending on their location in the cell cycle. Hence, the next step was to add ethylene to the synchronized cells just as they completed S-phase (i.e. the 3.5 h treatment). This treatment resulted in a rise in mortality at 4 h to 8% compared with 4% in the controls (i.e. within 0.5 h of adding ethylene, Fig. 5a). It then rose steadily before reaching a peak at 10 h (30%) then fell before rising again at 14 h and 16 h (Fig. 5a). By aligning the phases of the cell cycle along the x-axis, the data are consistent with a peak in mortality, which rises sharply at G2/M (at 10 h), a major checkpoint of the cell cycle (Fig. 5a). The second rise in mortality began in late G1 and peaked in S-phase (16 h in Fig. 5a).
|
Note that the comparisons between treatments at each sampling time are not entirely valid because, for example, at 10 h (time of maximum recorded mortality), the majority of cells in the 3.5 h ethylene treatment were at the G2/M boundary whereas the strictly comparable cells in the control would reached this boundary at about 9 h (Fig. 5a
). Nevertheless, the cell cycle specificity of ethylene on mortality is striking on either a sampling time, or cell cycle phase comparison (Fig. 5a).
The ethylene+silver treatment ameliorated ethylene-induced increase in mortality at G2/M but induced increased mortality in S- and M-phases
Since silver ions block ethylene action, it was ascertained if these ethylene-induced rises in mortality could be suppressed by adding silver nitrate simultaneously. The level of mortality recorded at each sampling time was lower in the ethylene+silver compared with ethylene treatment (Fig. 5b cf. Fig. 5a). The initial ethylene-induced rise in the mortality index at 4 h (Fig. 5a) was partially suppressed by the silver treatment (Fig. 5b), but the major rise in mortality induced by ethylene at 10 h (at G2/M, Fig. 5a) was completely suppressed (e.g. compare the 8 h bar in Fig. 5b with the 10 h bar in Fig. 5a). However, the ethylene-induced peak at 16 h (Fig. 5a) was only partially suppressed by ethylene+silver treatment (at 16 h in Fig. 5b) because although the latter peak was lower than that in the ethylene treatment it was significantly higher than the control (Fig. 5b). The ethylene+silver treatment also resulted in a rise in mortality in M-phase in addition to its effect of lengthening M-phase (Figs 2b, 5b). To resolve whether the silver ions affected mortality though toxic effects, synchronized cells were treated with silver nitrate alone (1.2 µM) at 3.5 h. The highest level of mortality in this treatment was at 6 h with a smaller rise when the cells were in M-phase (Fig. 5c). Note that beyond 14 h, it was not possible to detect mortality levels in the silver treatment that were higher than the controls (data not shown).
Hence it is concluded that the major effect of silver ions was to reduce the proportion of rapidly cycling cells (Fig. 3c).
Mortality in the TBY-2 cells was characterized by ethylene-induced nuclear shrinkage together with DNA fragmentation at G2/M
Cell and nuclear areas were measured by image analysis to determine whether ethylene had any cell cycle-specific effects on these parameters. As mentioned above, between-treatment comparisons on a sampling time basis may be confounded because of alterations in the length of the component phases. For cell and nuclear areas in the 3.5 h ethylene treatment and the controls, treatments were compared at comparable cell cycle phases (Fig. 6). The first comparison is S-phase (the 3 h sample of control cells before ethylene was added, see Fig. 5a), the second is G2 (
2 h before mitosis: 6 h in the controls and 8 h for the ethylene treatment, see Fig. 5a) and G2/M 10 h for the controls (in M-phase) and 10 h for the ethylene treatment (at G2/M); note from Fig. 5a that the latter comparison was the closest that could be made.
|
The nuclear area data indicate that the 3.5 h ethylene treatment resulted in significant nuclear shrinkage for dead cells compared with live ones, particularly at the G2/M transition (P=0.02). In the controls, this effect was only marginally significant for cells in S-phase (P=0.07), but not at other phases (Fig. 6
).
Nuclear shrinkage is one widely recognized symptom of PCD. DNA laddering was not detected, but it is concluded that the maximum level of mortality (30%, see Fig. 5a) may have been insufficient to resolve laddering. Hence, an in situ technique was used to examine whether 3'-OH termini could be detected as an indicator of DNA fragmentation, another characteristic of PCD.
Essentially, the indirect fluorescence technique works because apoptosing nuclei exhibit DNA strand breaks which expose 3'-OH termini. Nuclear DNA exhibiting 3'-OH termini, which would have incorporated the digoxygenin nucleotides, were labelled with the fluorescein-conjugated antidigoxygenin antibody resulting in an indirect ‘green’ fluorescence in the nuclei of these cells (Fig. 7). DNA fragmentation was detected both in the controls (1–2%) and in the 3.5 h ethylene treatment (3% rising to 14%). However, there was a 6–7-fold increase in the percentage of such cells in the ethylene treatment at 8–10 h (12–14%) coinciding with the maximum level of mortality (Fig. 5a). Hence, a major ethylene-induced increase in nuclear DNA fragmentation occurred at the G2/M transition of the cell cycle. Characteristically, green nuclei were observed in cells that were scattered apart and were often joined to red ones (Fig. 7). In mammalian cells, this apparent isolation of labelled cells is another feature of apoptosis (Thompson et al., 1992).
|
2 indicated significant heterogeneity in the data at the P<0.01 level (x130).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The data from this study confirm the hypothesis that ethylene treatment resulted in increased levels of mortality in the TBY-2 cell line in a cell cycle-specific manner. Indeed, when ethylene was added to cells that had just completed S-phase, a substantial peak of mortality was detected at G2/M, a major checkpoint of the cell cycle, and another in the next S-phase. Moreover, the cell death in the TBY-2 cells shows some features of apoptosis, nuclear shrinkage, DNA fragmentation, and the spatial isolation of cells exhibiting the fragmentation. These features were prominent in the 3.5 h ethylene treatment, notably at the G2/M transition, which also fits with the highest recorded level of ethylene-induced mortality. However, viable cells divided at the same rate as the controls which, in the authors' view, indicates that this level of ethylene did not cause massive perturbation to the cell cycle.
The duration of S-phase in synchronized cells of 3.5 h (Sorrell et al., 1999) was confirmed through careful analysis of the mitotic index curves coupled with histone H4 expression patterns. The maximum level of mortality observed was 30% at the G2/M transition and was induced when ethylene was applied to cells which had just completed S-phase and entered early G2. In animal cells, p53, a critical sensor of cell viability, up-regulates a number of gene products in G2. One of these is p21, a cyclin-dependent kinase inhibitor (CKI) which suppresses Cdc2's catalytic activity at the G2/M checkpoint (Levine, 1997). Another up-regulated protein is GADD45 which is induced upon DNA damage and is involved, indirectly, with DNA nucleotide repair (reviewed by Levine, 1997). These repressors prevent cell division until the cell acquires mitotic competence. Hence, it may be that in the ethylene treatment, cell viability depends upon a block at the G2/M transition imposed by a CKI-like protein, until DNA repair is complete; a CKI was identified in plants recently (Wang et al., 1997). Note that in the system used here, ethylene treatment delayed the entry of viable cells into mitosis by 1 h which could be a lag in which the cells recover from DNA damage inflicted by the ethylene treatment. As noted by Rhind and Russell, in fission yeast and vertebrate cells, the DNA damage checkpoint causes cells to arrest in G2 (Rhind and Russell, 1998). It would follow that in the 3.5 h ethylene treatment, for TBY-2 cells at the G2/M transition, about one-third fail to satisfy this DNA checkpoint, exit the cell cycle and enter a death process. The generation of 3'-OH ends and its enhancement by ethylene is interpreted as evidence for damage at the level of nuclear DNA. Moreover, nuclear shrinkage, another associated feature of PCD, was induced by ethylene at G2/M.
Note that the 3.5 h treatment given to cells that had just completed S-phase was much more effective at inducing mortality than the 0 h treatment given to cells that had just initiated S-phase. More recently, ethylene was injected at 4.5 h (1 h into G2) and again a delay in the rise of the mitotic index was observed, but the peak in mortality at G2/M (15%) was less than reported here (30%) although the first mitotic index peak was higher (35% [C Evett, unpublished data] compared with the peak here of 27%).
In tobacco, a 47 kDa protein kinase was identified that seemed to activate a signal transduction pathway that led to hypersensitive cell death in tobacco cell suspensions (Suzuki et al., 1999). Moreover, an inhibition of serine/threonine protein phosphatases induced rapid cell death in that system (Suzuki et al., 1999). Hence, this is a clear example of phosphoregulation that led to cell death. It would be interesting to resolve whether this 47 kDa protein kinase or related phosphatases could be regulated by ethylene in a cell cycle-specific manner.
A second peak of ethylene-induced mortality was also observed in S-phase. In yeast and vertebrates, a major cell cycle checkpoint for DNA damage operates in late G1 (Tanaka et al., 2000). Hence it could be that the observed rise in mortality in S-phase may be a consequence of a failure, in this case, of about 20% of the cells to satisfy this DNA damage checkpoint so that they exit into a death process in S-phase; note that the mortality level was also rising in late G1. To test this hypothesis, data on DNA fragmentation at these sampling times are required.
The ethylene-induced peak of mortality at G2/M was suppressed by silver ions, which fits with their known effect of blocking ethylene action (Drew et al., 1981). Indeed, at all sampling times, the level of mortality in the ethylene+silver treatment was reduced compared with that in the ethylene one. However, the ethylene+silver treatment resulted in a significant increase in mortality at M- and S-phase compared with the controls. Silver is a toxic metal (Woolhouse, 1983) and is perhaps exerting a toxic effect at mitosis as found when synchronized TBY-2 cells were treated with zinc at 100 µM (Francis et al., 1995). This was partly confirmed by the data from the silver treatment, which indicated a rise in mortality in M-phase, which was not detected in the ethylene treatment. Notably the silver+ethylene and silver treatments reduced the population of rapidly cycling cells to a greater extent than ethylene. Whether in response to toxic metals dying TBY-2 cells also exhibit apoptotic symptoms has to be resolved. Current work is examining the effects of alternative inhibitors of ethylene action.
The concentration of ethylene used in these experiments (17 700 µl l-1) was chosen because it induced measurable, but not overwhelming, levels of mortality (maximum 30%) in this system. Clearly, cells in a plant are never exposed to such high concentrations of gaseous ethylene. So, what is the physiological relevance of the data reported here? First, as emphasized earlier, although this much ethylene was released into the head-space of the flask, only a very small proportion would enter the liquid phase given the extremely low solubility of this gas in water. Hence, on this basis, the amount of ethylene taken up by the TBY-2 cells would be much more similar to that in the whole plant. Second, concentrations of ethylene lower than 17 700-µl l-1 had a negligible effect on mortality. Third, the concentration of silver nitrate used (1.2 µM) is known to inhibit ethylene action in whole plant studies (Drew et al., 1981). Note that this concentration was sufficient to ameliorate ethylene-induced mortality, most notably in cells sampled at G2/M of the cell cycle. Clearly, if unusually large amounts of ethylene had entered the TBY-2 cells, then 1.2 µM silver nitrate would have been ineffective in ameliorating ethylene action. Fourth, in the ethylene treatment, a substantial population of viable cells divided at the same rate as the controls. Again, if unusually large amounts of ethylene reached the cells then massive mortality or profound cell cycle perturbation, or both would be likely. Fifth, a clear parallel exists with metal toxicity studies particularly regarding the concentration required to produce a given degree of inhibition in cultured cells (Davies et al., 1991). For example, inhibitory concentrations of metals used in studies utilizing plant cells in the culture are much greater than those used in whole plant work (Steffens et al., 1986).
Various groups have shown how plant growth regulators (pgrs) interface with the G2/M and G1/S phase checkpoints of the plant cell cycle (Zhang et al., 1996; Wang et al., 1997; Riou-Khamlichi et al., 1999; Francis and Sorrell, 2000). The work reported here has demonstrated how another pgr, ethylene, induced cell death maximally, at G2/M in the TBY-2 cell line. Some characteristics of apoptosis such as nuclear blebbing and DNA laddering have yet to be detected. However, it should be noted that DNA laddering is not always a feature of plant cells undergoing PCD (Jones and Dangl, 1996; Buckner et al., 2000). A programmed mechanism is favoured in order to explain the cell cycle-specific cell death observed here, given the generation of 3'-OH ends, nuclear shrinkage and the isolation of dead/dying cells adjacent to living ones.
|
Acknowledgments |
|---|
RJH thanks UC Worcester for a research grant, CBO thanks the University of Wales and UC Worcester for a studentship and BVH thanks the Royal Society for a fellowship.
|
Notes |
|---|
4 To whom correspondence should be addressed. Fax: +44 292 087 4305. E-mail: francisd{at}cf.ac.uk
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Buckner B, Johal GS, Janick-Buckner D. 2000. Cell death in maize. Physiologia Plantarum 108, 231–239.
Callard D, Axelos M, Mazzolini L. 1996. Novel molecular markers for late phases of the growth cycle of Arabidopsis thaliana cell-suspension cultures are expressed during organ senescence. Plant Physiology 112, 705–715.[Abstract]
Chaboute ME, Chaubet N, Philipps G, Ehling M, Gigot C. 1987. Genomic organization and nucleotide-sequences of 2 histone H-3 and 2 histone H-4 genes of Arabidopsis thaliana. Plant Molecular Biology 8, 179–191.
Conger AD, Fairchild LM. 1953. A quick-freeze method for making smear slides permanent. Stain Technology 28, 281–283.[Web of Science][Medline]
Davies MS, Francis D, Thomas JD. 1991. Rapidity of cellular changes induced by zinc in a zinc-tolerant and non-tolerant cultivar of Festuca rubra L. New Phytologist 117, 103–108.
Drew MC, Jackson MB, Giffard SC. 1979. Ethylene-promoted adventitious rooting and development of cortical air-spaces (aerenchyma) in roots may be adaptive responses to flooding in Zea mays L. Planta 147, 83–88.[Web of Science]
Drew MC, Jackson MB, Giffard SC, Campbell R. 1981. Inhibition by silver ions of gas space (aerenchyma) formation in adventitious roots of Zea mays L. subjected to exogenous ethylene or to oxygen deficiency. Planta 153, 217–224.[Web of Science]
Ellis RE, Yuan J, Horvitz HR. 1991. Mechanisms and functions of cell death. Annual Review of Cell Biology 7, 663–698.[Web of Science]
Feinberg AP, Vogelstein B. 1983. A technique for radiolabelling DNA restriction fragments to high specific activity. Analytical Biochemistry 132, 6–13.[Web of Science][Medline]
Francis D, Sorrell DA. 2000. The interface between the cell cycle and plant growth regulators. Plant Growth Regulation 33, 1–12.
Francis D, Davies MS, Braybrook C, James NC, Herbert RJ. 1995. An effect of zinc on M-phase and G1 on the plant cell cycle in the synchronous TBY-2 tobacco cell suspension. Journal of Experimental Botany 46, 1887–1894.
Harris N, Oparka KJ. 1994. Plant cell biology, 1st edn. Oxford: Oxford University Press.
He C-J, Morgan PW, Drew MC. 1996. Transduction of an ethylene signal is required for cell death and lysis in the root cortex of maize during aerenchyma formation induced by hypoxia. Plant Physiology 112, 463–472.[Abstract]
Hirao A, Kong YY, Matsuoka S, Wakeham A, Ruland J, Yoshida H, Liu D, Elledge SJ, Mak TW. 2000. DNA damage-induced activation by the checkpoint kinase Chk2. Science 287, 1824–1827.
Jones AM, Dangl JL. 1996. Logjam at the styx, programmed cell death in plants. Trends in Plant Science 1, 114–119.[Web of Science]
Kerr JFR, Wyllieb AH, Currie AR. 1972. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. British Journal of Cancer 26, 239–257.[Web of Science][Medline]
Lane DP. 1992. Cancer. P53, guardian of the genome. Nature 358, 15–16.[Medline]
Levine AJ. 1997. p53, the cellular gatekeeper for growth and division. Cell 88, 323–331.[Web of Science][Medline]
Linsmaier EM, Skoog F. 1965. Organic growth factor requirements of tobacco tissue. Physiologia Plantarum 18, 100–127.
Lopez-Girona A, Furnari B, Mondesert O, Russell P. 1999. Nuclear localization of Cdc25 is regulated by DNA damage and a 14-3-3 protein. Nature 397, 172–175.[Medline]
Nagata T, Nemoto Y, Hasezewa S. 1992. Tobacco BY-2 cell line as the ‘HeLa’ cell line in the cell biology of higher plants. International Review of Cytology 132, 1–30.
Nurse P. 1990. Universal control mechanism regulating onset of M-phase. Nature 344, 503–508.[Medline]
Quastler H, Sherman FG. 1959. Cell population kinetics in the intestinal epithelium of the mouse. Experimental Cell Research 17, 420–438.[Web of Science][Medline]
Rhind P, Russell P. 1998. Mitotic DNA damage and replication checkpoints in yeast. Current Opinions in Cell Biology 10, 749–758.
Riou-Khamlichi C, Huntley R, Jacqmard A, Murray JAH. 1999. Cytokinin activation of Arabidopsis cell division through a D-type cyclin. Science 283, 1541–1544.
Rogers HJ, Harvey A, Lonsdale DM. 1992. Isolation and characterization of a tobacco gene with homology to pectate lyase which is specifically expressed during microsporogenesis. Plant Molecular Biology 20, 493–502.[Web of Science][Medline]
Russell P, Nurse P. 1986. cdc25+ functions as an inducer in the mitotic control of fission yeast. Cell 45, 145–153.[Web of Science][Medline]
Russell P, Nurse P. 1987. Negative regulation of mitosis by wee1+, a gene encoding a protein kinase homologue. Cell 49, 559–567.[Web of Science][Medline]
Schwartz LM, Smith SW, Jones MEE, Osborne BA. 1993. Do all programmed cell deaths occur via apoptosis? Proceedings of the National Academy of Sciences, USA 90, 980–984.
Sorrell DA, Combettes B, Chaubet-Gigot C, Murray JAH. 1999. Distinct cyclin D genes show mitotic accumulation or constant levels of transcript in tobaccco Bright Yellow-2 cells. Plant Physiology 119, 343–351.
Steffens JC, Hunt DF, Williams BG. 1986. Accumulation of non-protein metal binding polypeptides (
-glutamyl-cysteinyl)n-glycine in selected cadmium-resistant tomato cells. Journal of Biological Chemistry 261, 13879–13882.
Suzuki K, Yano A, Shinshi H. 1999. Slow and prolonged activation of the p47 protein kinase during hypersensitive cell death in a culture of tobacco cells. Plant Physiology 119, 1465–1472.
Tanaka H, Arakawa H, Yamaguchi T, Shiraishi K, Fukuda S, Matsui K, Takei Y, Nakamura Y. 2000. A ribonuceotide reductase gene involved in a p53-dependent cell-cycle checkpoint of DNA damage. Nature 404, 42–49.[Medline]
Thompson HJ, Strange R, Scedin PJ. 1992. Apoptosis in the genesis and prevention of cancer. Cancer Epidemiology, Biomarkers and Prevention 1, 597–602.
Wang H, Fowke LC, Crosby WL. 1997. A plant cyclin-dependent kinase inhibitor gene. Nature 386, 451.[Medline]
Woolhouse HW. 1983. Toxicity and tolerances in the response to metals in plants. Encyclopaedia of plant physiology, Vol. 12C. In: Lange OL, Nobel PS, Osmond CB, Ziegler H, eds. Physiological plant ecology. III. Chemical and biological environment. Berlin: Springer-Verlag, 245–300.
Young TE, Gaille DR. 1999. Analysis of programmed cell death in wheat endosperm reveals differences in endosperm development between cereals. Plant Molecular Biology 39, 915–926.[Web of Science][Medline]
Zhang K, Letham DS, John PCL. 1996. Cytokinin controls the cell cycle at mitosis by stimulating the tyrosine dephosphorylation and activation of p34cdc2-like H1 histone kinase. Planta 200, 2–12.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. N. Gushwa, D. Hayashi, A. Kemper, B. Abram, J. E. Taylor, J. Upton, C. F. Tay, S. Fiedler, S. Pullen, L. P. Miller, et al. Thermotolerant Guard Cell Protoplasts of Tree Tobacco Do Not Require Exogenous Hormones to Survive in Culture and Are Blocked from Reentering the Cell Cycle at the G1-to-S Transition Plant Physiology, August 1, 2003; 132(4): 1925 - 1940. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Dambrauskas, S. J. Aves, J. A. Bryant, D. Francis, and H. J. Rogers Genes encoding two essential DNA replication activation proteins, Cdc6 and Mcm3, exhibit very different patterns of expression in the tobacco BY-2 cell cycle J. Exp. Bot., February 1, 2003; 54(383): 699 - 706. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||