|Home | About | Journals | Submit | Contact Us | Français|
The orderly progression through the cell division cycle is of paramount importance to all organisms, as improper progression through the cycle could result in defects with grave consequences. Previously, our lab has shown that model eukaryotes such as Saccharomyces cerevisiae, Caenorhabditis elegans, and Danio rerio all retain high viability after prolonged arrest in a state of anoxia-induced suspended animation, implying that in such a state, progression through the cell division cycle is reversibly arrested in an orderly manner. Here, we show that S. cerevisiae (both wild-type and several cold-sensitive strains) and C. elegans embryos exhibit a dramatic decrease in viability that is associated with dysregulation of the cell cycle when exposed to low temperatures. Further, we find that when the yeast or worms are first transitioned into a state of anoxia-induced suspended animation before cold exposure, the associated cold-induced viability defects are largely abrogated. We present evidence that by imposing an anoxia-induced reversible arrest of the cell cycle, the cells are prevented from engaging in aberrant cell cycle events in the cold, thus allowing the organisms to avoid the lethality that would have occurred in a cold, oxygenated environment.
The cell division cycle is an intricate series of interrelated events that must occur in a tightly regulated manner during each iteration of the cycle in order to maintain high fidelity (Alberts et al., 2007 ). Abnormal progression through the cell cycle could have dire consequences, such as chromosome missegregation, aneuploidy, or cell death. To help maintain the fidelity of the cycle, organisms have evolved cell cycle checkpoint mechanisms that function to prevent cells from engaging in improper progression through the cycle (Hartwell and Kastan, 1994 ). While these checkpoint mechanisms are often sufficient to safeguard cell division cycle fidelity, the function of these checkpoints can be compromised by various combinations of genetic mutation and adverse environmental conditions (e.g., Moir and Botstein, 1982 ).
Many laboratories have used a combination of mutations and environmental manipulations to study the consequences of the cell cycle gone awry (e.g., Hartwell et al., 1970 ; Nurse and Thuriaux, 1980 ; Moir and Botstein, 1982 ; Toda et al., 1983 ). Perhaps the most celebrated of these efforts were the pioneering studies carried out by Hartwell and colleagues, using temperature-sensitive (ts) mutants in the budding yeast Saccharomyces cerevisiae that exhibited cell division cycle (cdc) phenotypes when shifted to the restrictive temperature (Hartwell et al., 1970 ). While this work yielded many insights into the workings of the cell cycle, other labs took a complementary approach using cold-sensitive (cs) mutants in the hopes of identifying additional genes that are involved in controlling cell cycle progression. Botstein and colleagues found many cs mutants in S. cerevisiae that exhibited characteristic arrest at specific points in the cell cycle when shifted to the cold (Moir et al., 1982 ). Thus, in budding yeast, considerable work has been done to examine the interaction of lowered temperature with certain genetic backgrounds that result in cell cycle defects.
Similarly, our lab is interested in the response of model organisms to environmental stresses and has previously shown that the model eukaryotes S. cerevisiae (Chan and Roth, 2008 ), Caenorhabditis elegans (Padilla et al., 2002 ), and Danio rerio embryos (Padilla and Roth, 2001 ) all enter into a reversible state of profound hypometabolism when subjected to extreme oxygen deprivation. We call this phenomenon anoxia-induced suspended animation, as all life processes that can be observed by light microscopy reversibly arrest, pending restoration of oxygen. Moreover, because these model eukaryotes retain high viability, it is probable that complex processes, such as progression through the cell cycle, are reversibly halted in an orderly manner. For example, the san-1 (suspended animation-1) gene, encoding a component of the spindle checkpoint, is required for C. elegans embryos to engage in anoxia-induced suspended animation (Nystul et al., 2003 ). As such, we sought to determine if we could exploit this conserved phenomenon of orderly cell cycle arrest to enhance survival of model systems that are prone to cell cycle errors when subjected to an environmental insult, in this case, lowered temperatures.
In this work, we report that wild-type C. elegans embryos are unable to survive a 24-h exposure to 4°C. This lethality is associated with extensive chromosome segregation defects in the cold embryos. Similarly, we show that cold-sensitive mutants of S. cerevisiae, as well as the wild-type strain BY4741 (the parental strain for the MATa deletion set), all can be made to exhibit cold-sensitive lethality associated with abnormal cell cycle progression when grown at low temperature. Further, we find that when these organisms are placed into a suspended state, they are protected from cold-induced insults and retain high viability, at least partly because they are prevented from proceeding through the cell cycle in an error-prone manner.
Yeast strains DBY640 (MATa gal− mal− ade2), DBY1252 (MATa ade2 cdc51-1), DBY1583 (MATa his4-539 ura3-52 lys2-801 ndc1-1), and DBY1802 (MATa his3-524-539 ura3-52 lys2-801) were generous gifts from Dr. David Botstein (Princeton University). cin1Δ and cin2Δ were constructed de novo in the BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) background and verified by PCR using standard methods (http://labs.fhcrc.org/gottschling/Yeast%20Protocols/index.html). Anoxic atmospheres were generated by either of two means: continuous perfusion in modified Pyrex crystallization dishes with nitrogen gas (N2; Air Gas Nor Pac, Seattle, WA) scrubbed with an Aeronex CE500KF14R inline gas purifier to remove residual oxygen or using Bio-Bag Environmental Chamber Type A (Becton Dickinson, Franklin Lakes, NJ) per manufacturer's instructions. Both methods yielded equivalent results. Yeast were placed into one of several VWR Low Temperature Incubators Model No. 2005 (VWR, West Chester, PA) that were set to the appropriate temperature for cold exposures. Temperature was verified at the beginning and end of each experiment using an alcohol thermometer immersed in a beaker of water that was thermally equilibrated within the incubator.
To obtain yeast for cell biological analysis, cells were grown overnight in 5 ml YPD. Cells were collected by light centrifugation, washed once in PBS, and then resuspended in a small volume of PBS. A small portion of the cells were plated at low density onto solid YPG media (1% yeast extract, 2% peptone, 3% glycerol, with benomyl where appropriate) to assess viability by colony formation. (Glycerol, i.e., nonfermentable, media was used in all yeast experiments in order to enable the yeast to undergo suspended animation in anoxia.) The remaining cells were split evenly and plated onto sterile nylon membranes (GE Water and Process Technologies, Trevose, PA) on YPG plates. One set of plates was made anoxic at low temperature, whereas the controls were incubated at the same low temperature in room air.
After the appropriate incubation periods, the cells were collected by centrifugation in PBS. Cells were fixed with 3.7% formaldehyde for 1 h at room temperature, followed by two washes using 100 mM potassium phosphate buffer, pH 7.5, and one wash using the same buffer supplemented with 1.2 M sorbitol. Cell walls then were digested using standard methods to form spheroplasts. Spheroplasts were applied to poly-l-lysine–treated slides and allowed to settle. After removing excess liquid, slides were incubated for 5–6 min in methanol at −20°C, followed by 30 s in acetone at room temperature. Slides were washed three times using PBS with 1% BSA. Slides were then incubated overnight at room temperature in the presence of primary antibody mAb YL1/2 against α-tubulin (Abcam, Cambridge, MA) at 1:250 dilution in PBS-BSA. Slides were washed four times with PBS-BSA before application of secondary antibody (anti-rat conjugated AlexaFluor 488) at 1:250 dilution, with incubation for 2 h at room temperature. Slides were washed four times with PBS-BSA, twice more with PBS, and stained with DAPI for 5 min before mounting in 1 mg/ml phenylenediamine, pH 9, in 90% glycerol. Images were acquired using a Zeiss MRm camera on a Zeiss Axioskop with AxioVision 4.6 software (Thornwood, NY).
As a population source for all experiments, wild-type Bristol strain N2 was continuously maintained at 20°C with care taken to ensure that the population never starved (Brenner, 1974 ). For experiments concerning embryos laid on a plate, five adults were allowed to lay eggs for 2 h in a spot of Escherichia coli (OP50) on a small nematode growth medium (NGM) plate. The adults were removed and the plate with embryos placed into the appropriate environment. In all viability experiments, the nematodes were allowed to recover at 20°C in room air, after exposure to the cold. Embryos were scored for hatching 24-h after exposure, and then followed to adulthood. Animals that could not be accounted for were not included in the total.
For experiments with two-cell embryos, adult C. elegans were picked into a drop of sterile water containing 100 μg/ml ampicillin, 15 μg/ml tetracycline, and 200 μg/ml streptomycin on a glass plate. Adults were chopped with a razor blade, and two-cell embryos were transferred by mouth pipette. Thirty to 60 two-cell embryos were transferred to a small glass boat (custom-made to fit atmospheric chambers; Avalon Glass Works, Seattle, WA) filled with 3 ml of 1% agarose in M9 buffer. The glass boats were then placed in glass syringes for exposure to the appropriate environment. After exposure, agarose chunks containing the embryos were cut out of the boat and placed with embryos facing up onto a medium-sized NGM plate seeded with OP50. Embryos were scored for hatching 24-h after exposure, and hatched larvae were transferred to the surface of the plate and followed to adulthood. Animals that could not be accounted for were not included in the total.
Oxygen deprivation experiments were performed as described previously (Padilla et al., 2002 ; Nystul and Roth, 2004 ). To generate anoxic environments we used either an Anaerobic Bio-Bag Type A Environmental Chamber according to the manufacturer's instructions, or perfused an environmental chamber with 100% N2 gas at 80 ml/min. The chambers were designed by Dr. Wayne Van Voorhies (New Mexico State University, Las Cruces, NM; Van Voorhies and Ward, 2000 ). The environmental chamber is a 30-ml glass syringe (Popper and Sons, New Hyde Park, NY) fitted with a custom steel stopper lined with two Viton O-rings to ensure a tight seal. The stopper is bored through and has a steel luer lock on the exterior face so that a hose carrying compressed gas can be attached. A desired gas mixture is delivered to the chamber at a constant pressure and flow rate from compressed tanks by passing first through a rotometer (Aalborg, Orangeburg, NY) or a mass flow controller (Sierra Instruments, Monterey, CA) to monitor flow rate and then through a 250-ml gas washing bottle (Kimble Chase, Vineland, NJ) containing 250 ml of water to hydrate the gas. A one-eighth–inch outer diameter nylon or fluorinated ethylene propylene tubing (Cole-Parmer, Vernon Hills, IL) was used, and connections between tubing and regulators were made with brass John Guest-type fittings (Airgas, Nor Pac, Vancouver, WA). All other connections were made with either microflow quick-connect fittings or standard luer fittings (Cole-Parmer).
To generate reproducible hypothermic environments, we calibrated a dedicated incubator (Low Temperature Incubator VWR Model No. 2005 or Echotherm [Torrey Pines Scientific, San Diego, CA] Chilling Incubator IN35) for each experiment. Temperature measurements were continuously monitored and logged for confirmation using HOBO data loggers (Onset, Pocasset, MA).
To generate slides enriched for young embryos (chromosomes and microtubule-organizing centers (MTOCs) are large and easy to see in young embryos), a synchronous population of young adults was generated by bleaching (Epstein and Shakes, 1995 ). At the onset of egg laying, the synchronous adults were washed off and chopped with razor blades on a glass plate. The minced material was spread evenly on an NGM plate and exposed to the appropriate environment as described. After exposure, the embryos were washed off the plate in water, concentrated, and frozen onto slides using dry ice (Moore et al., 1999 ). Slides were frozen within 4 min of removing the embryos from the exposure environment.
To examine chromosome morphology, coverslips were flicked off the frozen slides and embryos fixed in prechilled methanol for 5 min at −20°C. Slides were rinsed twice in PBS and then placed in PBS-BSA for 30 min before incubating with mAb YL1/2 and mAb 414 (specific for the nuclear pore complex, Abcam) for 1 h at room temperature. The YL1/2 and 414 antibodies were used at 1:1000 and 1:500 dilution in PBS-BSA, respectively. Slides were washed three times in blocking solution and then incubated with secondary antibodies for 1 h at room temperature. Secondary antibodies (goat anti-rat Alexa Fluor 488 and goat anti-mouse Alexa Fluor 568) were diluted 1:200 in PBS-BSA. Slides were rinsed two times in PBS-BSA and a third time in PBS-BSA containing DAPI to stain DNA. For consistency, only the AB to ABa and ABp division was scored for missegregation phenotypes. This comprises the two- to three-cell transition in embryogenesis. All data points are the result of at least three independent slide preparations.
To score MTOC duplication events, slides were prepared as described and fixed in −20° methanol for 5 min. Primary antibody mAb YL1/2 and secondary antibody goat anti-rat 488 were used at the same dilutions as above. For each environmental condition and time point, the number of MTOCs in each blastomere was scored. Only embryos with two to six blastomeres were scored. All data points are the result of at least three independent slide preparations.
Images were acquired using a DeltaVision multiple wavelength fluorescent microscope (Applied Precision, Issaquah, WA). Stacks of 0.2-μm optical sections were collected for the indicated wavelengths and then deconvolved using Softworx software (Applied Precision). All nematode images in this report are two-dimensional projections of multiple optical sections.
To begin our studies, we sought to identify environmental conditions in which model eukaryotes exhibit lethality that is correlated with dysregulation of the cell cycle, with the idea that rescue from such lethality might be possible by imposing anoxia-induced suspended animation. Accordingly, we obtained a number of cold-sensitive cell cycle mutants in budding yeast from the Botstein lab and tested these strains to determine whether they were cold-sensitive for survival, as assayed by the ability to form colonies. Two such strains, DBY1252 and DBY1583, which bear the cdc51-1 (Moir et al., 1982 ) and the ndc1-1 (Thomas and Botstein, 1986 ) mutations, respectively, manifested such a conditional viability defect. cdc51-1 and ndc1-1 cells formed colonies as well as their respective parental strains (DBY640 and DBY1802) on rich medium with glycerol as the sole carbon source (a nonfermentable medium) at 30°C. In contrast, when cdc51-1 cells were shifted to 12°C for 4 d after plating, there was a 100-fold decrease in the ability to form colonies, despite being shifted back to 30°C to allow growth of any surviving cells at day 4 (see Figure 1A). Similarly, ndc1-1 cells that were shifted to 11°C for 7 d after plating showed a 60% decrease in colony formation, after allowing for recovery after the cold exposure (see Figure 1B). Neither parental strain exhibited a colony forming defect after experiencing the same shift to low temperature.
We also examined two deletion mutants that Botstein and colleagues had identified as cold-sensitive, cin1Δ and cin2Δ (Hoyt et al., 1990 ). These two mutants formed small colonies that were comparable in number to the BY4741 parental strain when shifted to a range of cold temperatures after plating on rich medium with glycerol as the sole carbon source (data not shown). However, we noted that cin1Δ and cin2Δ were also sensitive to the microtubule-depolymerizing drug benomyl (Stearns et al., 1990 ). It is possible then that benomyl and low temperature would act additively to disrupt microtubule function more than low temperature alone. We therefore determined whether it were possible to elicit a colony-forming defect similar to cdc51-1 and ndc1-1, by spotting cin1Δ and cin2Δ cells onto media with a range of benomyl concentrations and exposing these cells to various cold temperatures. Using this approach and follow-on colony formation assays, we found that cin1Δ cells showed a 90% decrease in colony formation when incubated at 16°C for 7 d on plates with 1.0 μg/ml benomyl (see Figure 1C). Similarly, cin2Δ cells showed an >80% decrease in colony formation after 4 d at 16°C on plates with 1.0 μg/ml benomyl (see Figure 1D). This concentration of benomyl was well tolerated by these mutants at a permissive temperature (25°C, see Supplemental Figure 1), but was lethal at the restrictive temperature.
Given our success at inducing a colony forming defect in cin1Δ and cin2Δ using low temperature combined with benomyl, we tested whether a similar effect could be obtained for an essentially wild-type strain, BY4741. After testing a range of benomyl concentrations combined with a range of low temperatures, we found that BY4741 cells showed a 90% decrease in colony formation on media with 10 μg/ml benomyl after a 7-d exposure to 11°C (see Figure 1E). In total, we have identified two mutants (cdc51-1 and ndc1-1) that exhibit a cold-sensitive colony forming defect, another two mutants (cin1Δ and cin2Δ) that show a similar defect in the presence of low doses of benomyl, and a wild-type strain (BY4741) that manifests the cold-sensitive colony forming defect in a higher dose of benomyl.
Botstein and colleagues had previously reported that cdc51-1 (Moir et al., 1982 ) and ndc1-1 (Thomas and Botstein, 1986 ), as well as cin1Δ and cin2Δ (Hoyt et al., 1990 ) all showed abnormal cell cycle progression at low temperature, as revealed by staining for DNA and tubulin. We likewise used DAPI and the anti-tubulin antibody YL1/2 to determine if similarly abnormal cell morphology occurred in the cells we had treated with low temperature. Previously, cdc51-1 cells were reported to tend to arrest as large-budded cells with undivided nuclei when shifted to 17°C (Moir et al., 1982 ). We verified this observation, albeit using a longer exposure to compensate for the slower growth rate on the nonfermentable glycerol medium, and at a slightly lower temperature of 12°C (see Table 1). After 4 d at 12°C, 46.2% of cdc51-1 cells exhibited the characteristic terminal large-budded cell phenotype (Moir et al., 1982 ), with an undivided nucleus in one of the cell bodies, far from the bud neck (see Table 1). In contrast, none of the large-budded cells in the parental strain showed any discernible abnormalities.
Reminiscent of cold-sensitive tubulin mutants (Huffaker et al., 1988 ), we found that in almost all cases, microtubules were only found in the nucleated cell within a large-budded pair of the cdc51-1 strain; the other cell would be devoid of immunoreactivity as well as DAPI staining (see Figure 2B). This result was never observed in the parental strain (see Figure 2A) and provides an explanation for the undivided nucleus phenotype, as proper chromosome segregation and nuclear division should be impossible given such abnormal spindle orientations.
Similar to cdc51-1, ndc1-1 previously has been shown to accumulate as large-budded cells with undivided nuclei after shifting to 13°C, accompanied by abortive cell divisions that give rise to a significant population of cells with no discernible nuclear DNA (called aploid cells,Thomas and Botstein, 1986 ). At 11°C on glycerol media, we found a similar increase in the percentage of large-budded cells with undivided nuclei (27.7% of total cells examined, see Table 2), as well as an increase in the percentage of cells with either multiple or fragmented nuclei (8.0%) and aploid cells (3.1%). In total, 38.9% of cold-treated ndc1-1 cells showed obviously abnormal nuclear morphology, whereas only one out of the 278 parental strain cells we examined showed such a defect. Consistent with the nuclear morphology defect, tubulin staining in the ndc1-1 abnormal large-budded cells almost always consisted of a single tubulin aster near the single undivided nucleus, with an absence of staining in the anucleate cell body (see Figure 2C). In binucleate ndc1-1 cells, each nucleus was similarly accompanied by a tubulin aster (see Figure 2C).
Previously, both cin1Δ and cin2Δ were shown to accumulate as large-budded cells after shifting to 11°C, with most of these cells exhibiting the single undivided nucleus phenotype (Hoyt et al., 1990 ). In our hands, cin1Δ cells that were incubated for 7 d at 16°C in the presence of 1.0 μg/ml benomyl on glycerol showed an enrichment for cells with multiple or fragmented nuclei (34%), whereas such abnormal cells were not observed in the BY4741 parental strain (see Table 3). Similarly, cin2Δ cells that were incubated for 4 d at 16°C in 1.0 μg/ml benomyl showed an increase in the percentage of cells with multiple or fragmented nuclei (24.5%, see Table 4). In contrast, all of the examined BY4741 control cells were morphologically normal under the same incubation conditions.
Finally, we determined if BY4741 cells incubated under conditions that result in a cold-sensitive colony formation defect (7 d at 11°C on 10 μg/ml benomyl) also showed abnormal nuclear and tubulin morphology. Surprisingly, these cells only manifested a mild increase in the percentage of morphologically abnormal cells: 21 of 236 cells examined (8.9%) had multiple or fragmented nuclei, 1 of 236 was aploid, and 1 of 236 was a small-budded cell with a single nucleus in the bud (see Table 5). However, considering that the cold-sensitive colony formation defect in BY4741 is an effect of large magnitude, we consider it likely that some of the cold-induced cell cycle defects are more subtle than that which can be observed by examining gross nuclear DNA and tubulin morphology.
To determine if a similar phenomenon could be observed in a metazoan model system, we examined the embryos of wild-type N2 C. elegans exposed to cold temperatures. Similar to the yeast, >99% of C. elegans embryos exposed to 4°C for 24-h died as embryos or shortly after hatching into larvae (see Figure 3A). A more detailed accounting of developmental progression after cold exposure can be found in Supplemental Figure 2. We found that 78% of such embryos contained blastomeres that manifested profound chromosome segregation defects when examined after cold exposure (see Figure 3B). The most frequent of these defects were the “cut” phenotype, where the cleavage furrow is apparently bisecting the mitotic DNA mass (33%, see Figure 3C), and multiple nuclei (29%, see Figure 3E). In addition, we observed blastomeres either with multipolar spindles, anaphase bridging (see Figure 3D), or containing no nuclear DNA, at low frequency. Although a few embryos managed to complete embryogenesis, most of these died as young larvae, suggesting that prolonged exposure to low temperature during early embryogenesis had profoundly detrimental effects on the animals' continued development.
Because low temperatures near the freezing point for water are known to cause the depolymerization of microtubules (Osborn and Weber, 1976 ), we visualized microtubule configuration using the YL1/2 antibody (see Figure 4). As expected, antibody staining revealed two MTOCs within each dividing blastomere in control two-cell embryos (see Figure 4A) maintained at 20°C. In contrast, embryos exposed to 4°C often contained blastomeres that possessed various numbers of excessive MTOCs (see Figure 4, B, C, and D). The severity of this ectopic MTOC phenotype is correlated with the duration of exposure to the cold. After 4 h at 4°C, <5% of blastomeres had at least five MTOCs. By 14 h at 4°C, 65% of blastomeres had at least five MTOCs, while 17% of blastomeres had at least 10. After 24-h at 4°C, over 80% of blastomeres had at least five MTOCs, whereas over 35% had at least 10 MTOCs (Figure 4E). Taken together, these results suggest that low temperature causes dysregulation of the cell cycle and a striking manifestation of such dysregulation is that the process of MTOC duplication becomes uncoupled from other aspects of the cycle, e.g., cytokinesis.
Given that the cold-induced lethality we observed is associated with defects in progression through the cell cycle and that we had previously shown that anoxia-induced suspended animation reversibly halts the cell division cycle, we reasoned that placing the two model systems into a suspended state might preserve viability by imposing an orderly arrest of the cycle. Indeed, we found that making the cdc51-1 cells anoxic on glycerol medium at 12°C resulted in a substantial rescue of the cold-induced lethal phenotype. Although <1% of cold room air cells survived to form colonies, 66% of cold anoxic cells successfully formed colonies upon being shifted back into permissive conditions (see Figure 5A). We observed a similar rescuing effect for the other four genotypes as well. For ndc1-1, cells exposed to 11°C in air formed colonies at 43% of the capacity of permissive temperature controls, whereas cells exposed to 11°C in anoxia retained 83% colony forming capacity (see Figure 5B). cin1Δ cells exposed to 16°C in air in the presence of 1.0 μg/ml benomyl formed colonies at only 6% of the capacity of controls, in contrast to cold anoxic cells that retained 59% colony forming capacity (see Figure 5C). cin2Δ cells exposed to 16°C in air in the presence of 1.0 μg/ml benomyl formed colonies at 8% of the capacity of controls, compared with 71% in anoxia-treated cells (see Figure 5D). Finally, BY4741 cells exposed to 11°C in air on 10 μg/ml benomyl formed colonies at 6% of the capacity of controls, whereas cells exposed to cold in anoxia retained 58% of colony-forming capacity (see Figure 5E). These data show that, when yeast from a variety of genotypes are exposed to cold temperatures (with or without benomyl), i.e., conditions that cause significant lethality, it is possible to prevent much of this lethality by imposing a state of anoxia-induced suspended animation. This rescuing effect of anoxia was quite robust, with paired Student's t test p < 0.005 when comparing cold-aerated with cold anoxic cells, in all cases.
To determine whether anoxia-induced suspended animation can impose a relatively orderly arrest of the cell cycle that prevents cold yeast cells from committing irrecoverable errors in cell cycle progression, we compared the cellular morphology of cold anoxic cells to that of cold-aerated cells. In all five genotypes examined, there is a large decrease in the percentage of cells with abnormal morphology within the population that underwent cold exposure while anoxic. This decrease is accompanied by a relative increase in the percentage of (normal) unbudded cells. For cdc51-1, 46.2% of cold-aerated cells showed the characteristic terminal-arrest phenotype, whereas only 1.9% of cold-anoxic cells showed the same morphology. Conversely, only 46.7% of cold-aerated cells had a normal unbudded morphology, whereas 80.4% of cold-anoxic cells were in this state (see Table 1). ndc1-1 showed a very similar pattern, with a large decrease in the percentage of abnormal large-budded cells (27.7% in cold-aerated conditions to none in cold anoxia) and a large increase in the percentage of unbudded cells (53.4–84.5%, see Table 2). cin1Δ showed a large decrease in the percentage of cells with multiple or fragmented nuclei (34% in cold-aerated conditions to none in cold anoxia) that is nearly matched by a large increase in the percentage of unbudded cells (61.6–93.3%, see Table 3). Similar trends were observed in cin2Δ (Table 4) and BY4741 (Table 5). However, since the prevalence of the cold-induced morphological defects in these two backgrounds is lower, the magnitude of the rescuing effects are also necessarily smaller. These data suggest that imposing a state of anoxia-induced suspended animation on yeast in the cold prevents the cells from attempting to divide in an error-prone manner that contributes to the lethality observed under cold, oxygen-replete conditions.
Similar to the yeast, placing nematode embryos into anoxia-induced suspended animation before cold exposure protected them against cold-induced lethality. Ninety-seven percent of the anoxic embryos survived the 4°C exposure, to successfully develop to adulthood (see Figure 6A and compare with Figure 3A). In addition, similar to the yeast, the blastomeres in the nematode embryos were prevented from progressing improperly through the cell cycle. Consistent with this idea, none of the cell cycle defects we observed in room air 4°C embryos were seen in the suspended embryos (see Figure 6B). We also found that the order of exposure to cold temperature and anoxia has an important effect on the likelihood of survival (see Figure 6C). Specifically, two-cell embryos that were first transitioned into anoxia before the temperature shift successfully completed embryogenesis >85% of the time. In contrast, two-cell embryos that were shifted into the cold before the onset of anoxia successfully completed embryogenesis at a rate of only 21%, suggesting that proper entry into the suspended state before cold challenge is necessary to retain high survivability. Thus, we have shown that anoxia-induced suspended animation can preserve the potential for life in nematode embryos that are faced with otherwise lethally cold temperatures.
In this article, we report that several mutant strains in S. cerevisiae that were previously reported to be cold sensitive (cdc51-1, ndc1-1, cin1Δ, and cin2Δ), as well as the wild-type strain BY4741, can be made to manifest irrecoverable cell cycle defects that are associated with a high incidence of lethality after prolonged exposure to low temperature. Such phenotypes had previously been noted in these four mutants (Moir et al., 1982 ; Thomas and Botstein, 1986 ; Hoyt et al., 1990 ), which is consistent with the fact that the gene products mutated in these mutants are associated with tubulin function. Ndc1 has a role in spindle pole body duplication (Chial et al., 1999 ), whereas Cin1 and Cin2 are both involved in the proper folding of β-tubulin (Hoyt et al., 1997 ). It is thus not surprising that these mutants are more sensitive to low temperature than wild type, as microtubule dynamics is itself a cold-sensitive process (Osborne and Weber, 1976 ). Although strain BY4741 is much less sensitive to low temperature (the addition of 10 μg/ml benomyl is required to sensitize this strain to low temperature), we nonetheless observed similar cell cycle errors in this wild-type strain. This result suggests that perhaps all yeast strains can be similarly rendered cold-sensitive, given the “right” combination of low temperature and drug concentration to induce irreversible cell cycle errors due to impaired tubulin function.
We also found that when these same strains are made cold under anoxic conditions on nonfermentable media, the cold-induced lethality is diminished to a degree that is highly statistically significant. That is, many more cells retain the ability to form colonies after the cold exposure, provided they were kept in an anoxic state in the cold. Whereas cold, oxygen-replete cells showed obvious cell cycle defects, including undivided nuclei in large-budded cells, multiple nuclei within the same cell, and cells devoid of nuclear DNA, these defects were almost completely abolished in cold, anoxic cells. These results are consistent with the idea that when the cells are put into a state of anoxia-induced suspended animation, their cell cycles are arrested in an orderly, reversible manner. This prevents the occurrence of irrecoverable cell cycle errors, as revealed by examination of DNA and tubulin morphology as well as assays for colony formation. On return to permissive growth conditions, the cells that were suspended in anoxia are shown to be protected from the cold lethality, whereas cells that were cold, but not suspended, show much higher lethality. As we noted, a wild-type strain can be made cold sensitive by the addition of a microtubule-destabilizing agent, and this cold sensitivity can be substantially rescued by anoxia-induced suspended animation. If it were possible for all yeast strains to be rendered susceptible to cell cycle error-driven cold lethality by an appropriate concentration of an agent like benomyl, our results suggest that, in general, it might be possible to prevent such cold lethality by imposing a suspended state in anoxia.
Similar to the yeast, nearly all wild-type C. elegans embryos that are exposed to 4°C for 24-h fail to survive to adulthood, despite being returned to normal growth temperatures after the bout of cold. In fact, most never complete embryogenesis. Cell biological analysis of cold two-cell embryos showed that a variety of cell biological defects occur at low temperature. These defects include the cut phenotype, multiple nuclei, blastomeres with no DNA, anaphase bridging, and multipolar spindles. Instead of a uniform arrest at a specific point in the cell cycle, there is apparently a combination of different defects in nuclear division and chromosome segregation that result in a variety of terminal phenotypes. Thus, low temperature is a particularly destructive stress to young nematodes, as the cold almost invariably terminally disrupts embryogenesis or subsequent development in a number of ways.
Perhaps our most striking cell biological observation is the increased incidence of excessive MTOCs with increasing duration of cold exposure. At four hours in the cold, 43% of embryos are already affected, as these embryos contain at least one blastomere with an excessive MTOC. By 24-h, over 90% of embryos are similarly affected. Also, the numbers of excessive MTOCs increases dramatically over time, such that by 24-h in the cold, 80% of blastomeres have at least five MTOCs, whereas 37% have at least 10 MTOCs. Not surprisingly, the occurrence of excessive MTOCs coincides with aberrant DNA configurations that presage the eventual death of the cold-exposed embryos, because the subcellular architecture has been hopelessly distorted. The profound perturbations in tubulin configuration also help to explain the various other cell biological phenotypes that were already mentioned.
The manifold reduplication of MTOCs in cold embryos suggest that the cold induces a loss of synchronization between MTOC duplication and other aspects of the cell cycle, e.g., cytokinesis. One possible explanation for this result is that MTOC duplication is relatively less energetically costly than cytokinesis, such that in an energy-limited state in the cold, MTOC duplication continues to occur while there is insufficient energy to carry out cytokinesis. An alternative explanation is that, in the cold, there exist active mechanisms that are sufficient to arrest the majority of processes that comprise cell cycle progression (in an orderly manner), but MTOC duplication is not among the processes that are so governed. Our experiments do not address this issue, but it is thought that the maintaining of proper control of MTOC duplication is intimately linked to orderly progression of the cell division cycle (Azimzadeh and Bornens, 2007 ). We propose that quite independently of the questions addressed in this article, the application of our methodology to induce desynchronization of the centrosome cycle might be exploited to further elucidate the molecular mechanisms associated with this crucial component of the cell cycle.
We found that nematode embryos that are kept in an anoxic, suspended state in the cold do not reduplicate their MTOCs, thus avoiding this irrecoverable type of cell cycle error. As a result, when the nematodes are returned to permissive growth conditions, the anoxic, cold embryos survive while the vast majority of oxygenated, cold embryos do not complete embryogenesis and virtually none of these ultimately progresses to adulthood. These results are consistent with the view that by imposing anoxia, the embryos are unable to generate sufficient energy to attempt MTOC duplication in the cold. In effect, synchronization of the cell cycle is maintained in the suspended state because all processes that we can observe by microscopy have been reversibly and uniformly halted. This forced maintenance of cell cycle synchrony results in much higher survivorship compared with embryos where synchrony is not maintained.
Although a striking correlation is observed between the prevention of aberrant cell cycle progression and improved survival following anoxia in the cold, we note that not all of the yeast or nematode embryos exposed to the cold in room air manifested discernible cell cycle defects. Yet the large majority of these yeast or nematodes fail to survive. Thus, cold-induced damage can be subtle, but still lethal. For example, while examination of nuclear DNA by DAPI staining can reveal gross abnormalities such as multiple nuclei, more subtle defects, such as missegregation of single chromosomes, would not be detectable. Considering our results, we conclude that the suspended state in anoxia probably prevents many different forms of cold-induced damage, whether they be subtle or gross. Collectively, the prevention of these many possible errors helps to enable survival in otherwise lethally cold temperatures.
The authors thank the Botstein (Princeton University) and the Gottschling (Fred Hutchinson Cancer Research Center) labs for strains and/or reagents. We also thank Harold Frazier, Summer Lockett, and Dana Miller for their critical reading of the manuscript. This work was supported by National Institutes of Health Grant R01GM48435 to M.B.R., as well as a National Science Foundation Pre-doctoral Fellowship and an Achievement Rewards for College Scientists award to J.P.G.
This article was published online ahead of print in MBoC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E09-07-0614) on May 12, 2010.