- Email: [email protected]
A purified cellular extract accelerates the cell cycle in Physarum polycephalum
EXPERIMENTAL
CELL
RESEARCH
19 1,
332-336 (1990)
SHORT NOTE A Purified Cellular Extract Accelerates the Cell Cycle in Physarum polycephalum UWE HOBOHM,
ARMIN
Dept. Biology, University
HILDEBRANDT,
of Bremen, D-2800 Bremen, West Germany
plasmodia harvested, in this case, 3 h before mitosis [4, 171. Loidl and Grijbner applied plasmodial extracts during the fusion of microplasmodia and monitored the effect on the time of the first mitosis. They found a maximal accelerating effect in plasmodia collected about 90 min before mitosis [ 111. When they purified the accelerating activity lo-fold, it showed many of the characteristic properties of proteins [ 121. One main difficulty in identifying hypothetical control substances was the lack of an adequate biological assay. In order to test extracts from different phases of the cell cycle by using microinjection, it was necessary to develop an injection buffer which diffuses freely within the entire plasmodium. The results of our experiments with such a buffer indicate that a substance of 2.5 kDa, probably a peptide, which is produced rather late in the cell cycle of Physarum, is able to advance the cycle. Control processes shortly before mitosis are well known in other organisms, e.g., yeast, frog embryos, and HeLa cells. However, the participation of a low-molecular-mass compound has not been observed before (for review see [15]).
Plasmodia of the myxomycete Physarum polycephalum (strain Cl) were collected at different times during the cell cycle and extracts were prepared from homogenates using a buffer optimized for microinjection into plasmodial veins. These extracts were injected into plasmodia during the first 3 h of the cell cycle. The time of the following mitosis was monitored and compared with that of the buffer-injected controls. Extracts of plasmodia homogenized 45 min before late telophase accelerated the onset of mitosis in the injected plasmodium up to 70 min, i.e., an advance of lo-14% compared to the f3- to 10-h cell cycle duration of the controls. The accelerating activity vanished completely after heating, freezing, or protease digestion, thus indicating the peptide nature of the active agent. Purification of the active compound by means of gel filtration revealed a molecular mass of about 2500 Da. The active portion of the extract was further fractionated by HPLC and the activity determined in a single peak. (c‘ 1990
Academic
Press,
Inc.
INTRODUCTION
MATERIALS
The plasmodial stage of Physarum may serve as a unique model system for cell cycle investigations, because its million nuclei divide in almost perfect synchrony [16]. Several attempts have been made to detect hypothetical trigger or control substances: Blessing and Lempp, for example, layered cellular extracts onto the surface of the plasmodia. They described an acceleration of the cell cycle induced by extracts from donor plasmodia collected 45 min before metaphase. No effect was observed if the donor plasmodia had passed mitosis [3]. Similar experiments with lyophilized plasmodia from late phases of the cell cycle layered onto the surface of plasmodia in early phases again resulted in advances. A maximal accelerating effect was induced by
r To whom reprint 0014.4827/90
Copyright All
rights
requests should be addressed.
$3.00 0 1990 by Academic Press, of reproduction in any form
332 Inc. reserved.
AND LUDGER RENSING~
AND
METHODS
Strains, culture conditions, and preparation of plasmodial extracts. Macroplasmodia were cultivated in petri dishes on filter paper (Schleicher & Schtilll575, Dassel, West Germany) layered on a steel wire grid on a semidefined nutrient medium supplemented with hemin. Mitosis was determined in ethanol-fixed smears under a phase-contrast microscope [5]. The fusion of microplasmodia was defined as mitosis 0. Late telophase was used as a marker, i.e., when the time point of complete segregation of the two new nuclei was observed, which could be determined more precisely (up to +-1 min) than metaphase. Whole plasmodia were collected at specific stages of the cell cycle, frozen in liquid nitrogen, and stored at 80°C. The following preparative steps were done on ice. Frozen samples were thawed, suspended in 1.5~ (w/w) injection buffer (pH ll.O), vigorously stirred with a spatula for 1 min, adjusted to pH 7.0, and centrifuged for 30 min at 33,000g (4°C). Purified samples were stored liquid at ~18°C with 20% ethanol. Using a Speedvac, the ethanol was stepwise evaporated and the samples were filled with injection buffer up to the original volume for further use. We tested the haploid Screening of strains, buffers and injection. strains Cl-188, Cl-lB9, Cl-Mu, Cl-82, the diploid strains M-Go, F,
SHORT Tu, and the following buffers (components in g/liter bidest): phosphate buffer: 3.52 g KH,PO,, 7.26 g Na,HPO,. 2H,O; sodium borate buffer: 4.77 g Na,B,O, . lOH,O, 20.5 ml 1 M HCl; sodium citrate/ NaOH buffer: 12.53 g citric acid monohydrate, 159.6 ml 1 M NaOH; Ringer’s solution 1: 6.5 g NaCl, 0.14 g KCl, 0.12 g CaCl,, 0.2 g NaHCO,, 0.01 g NaH,PO,. H,O; Ringer’s solution 2: 8.6 g NaCl, 0.3 g KCl, 0.33 g CaCl,, 0.2 g NaHCO,, NaH,PO,. H,O; modified Ringer’s solution 2 without Ca’+; Tris buffer: 12.12 g Tris-HCl; imidazole-HCl buffer: 6.81 g imidazole (Merck, Darmstadt, West Germany), 24.3 ml 0.1 M HCl; 0.1 M Hepes buffer; 0.1 M Mes buffer; 0.1 M Pipes buffer; injection buffer [lo]: 2 mM Tris base, 0.2 mM ATP, 0.5 M DTT; injection buffer [27]: 2.25 g KCl, 0.61 g MgCl,, 0.012 g ATP, 1.6 g TrisHCl, 0.1 ml 4 X 10e3 M CaCl,. The buffers were adjusted to about 200 mOsm and different pH (6.5,7.0,7.5) before injection. During the screening experiments 5510 ~1 buffer was injected during 5 min at different locations into plasmodia. An injection was judged successful when no sign of injection and no injury were detected after 223 h and if the buffer was equally distributed inside the entire plasmodium. During extract injection experiments, one plasmodium 445 cm in diameter was cut into 4-6 pieces, and 3-5 ~1 of extract or of buffer was injected during 2-3 min into each piece. Extract (200 ~1) was incubated with ei.&gradation e.rperiments. ther 5 U carrier bound phosphatase (Sigma, Deisenhofen, West Germany), 5 U carrier bound RNase (Sigma), 5 U carrier bound proteinase K (Sigma) for 1 h, pH 7.0 (phosphatase, pH 8.0), at 37°C. During incubation, the mixture was gently shaken. The carriers were then pelleted and the supernatant was tested by means of injection. Crude extracts were fractionated using a Gel filtration and HPLC. 1.9 X 80.0.cm column filled with TSK-HW40S (Merck). The elution buffer was injection buffer/bidest (l/5); 20 ml/h was eluted and 5-ml fractions were collected and concentrated 5/l using a Speedvac centrifuge before injection. A 0.4 X 0.8 X 25.0.cm, 300.7~Cl8 column (Macherey & Nagel, Diiren, West Germany) was used for HPLC. A 200.~1 extract was protonated with trifluoroacetate (TFA) and absorption measured at 214 and 280 nm (LKB 2141). A gradient tridest/ 0.1% TFA (A) to acetonitrile/O.l% TFA (B) was used with the following program: 8 min B 0%, 140 min B increasing to 100%. Fractions were collected and concentrated to 200 ~1 using a Speedvac. The HPLC buffer was stepwise evaporated and redissolved in extract buffer before injection.
RESULTS
Injection of extracts into nonstarving macroplasmodia of Physarum is difficult: macroplasmodia tend to either block injected veins or develop visibly disturbed areas, which stop growth. We tested the 13 different buffers (see Materials and Methods) at different pH. Two of the buffers were described as suitable for microinjection into Physarum by Kukulies [lo] and Ueda [27]. All buffers were tested in seven strains of Physwum (Cl-lB8, Cl-lB9, Cl-Mu, Cl-82, M-Go, F, Tu) and revealed a strain-dependent sensitivity to injections of buffer. The best results were observed when strain ClMii was injected with a modified Ringer’s solution (pH 7.0). With this strain and buffer, signs of injection disappeared after 2-3 h and the plasmodium showed a continuous, solid surface of uniform color. In order to test the distribution of injected substances within the plasmodium, we used RNase (0.01 M) and [3H]thymidine. The injection of RNase resulted in ex-
333
NOTE I
I
I
,o 0
60
40
20
0
-20 250
200
150
100
SO
0
FIG. 1.
Cellular extracts from different cell cycle phases and their advancing/delaying effect on the third mitosis of the injected plasmodium. Abscissa: Phase of donor plasmodium at the time of harvesting (min before third mitosis). Ordinate: Resulting delay/advance of mitosis in injected plasmodia (min). All injections were administered during the first 3 h after the second synchronous mitosis after the fusion of microplasmodia.
cessive production of slime, disintegration of nuclei, and death of the entire plasmodium within i-1 h. After injection of [3H]thymidine into one part of the macroplasmodium, radioactivity was detected at the most distant part within 1 h. Furthermore, successful injections appeared to depend on optimal growth: if plasmodia grew as a thin network, injections were impossible. In order to evaluate the effects of injection experiments on cell cycle duration it was necessary to know the intra- and interplasmodial cell cycle synchrony of the strain used. In 87 plasmodia with a diameter up to 5 cm, the deviation between nuclei of distant parts at the time of late telophase was not more than 4.00 (k2.13) min. This degree of synchrony was higher than the lomin deviations reported by Sachsenmaier [20]. In order to test the interplasmodial synchrony, one plasmodium was cut into 2-4 pieces and monitored for the time of the following mitosis. Pieces which grew isolated for 250-300 min showed a difference of 2-10 min in the time of mitosis, whereas after a 9-h isolation there was a difference of lo-22 min. The interplasmodial synchrony in our experiments is, thus, somewhat less than that described by Loidl and Grobner [ll]. In the following experiments we injected extracts into 4-6 pieces of plasmodia and compared the time of mitosis to that of 4-6 buffer-injected controls. With this number of pieces an average phase difference of 10 min marks the 95% level of significance. The results of the injection experiments with whole extracts are shown in Fig. 1. Donor extracts obtained 45 min before late telophase (=35 min before metaphase) advanced the mitosis of the injected plasmodium up to 70 min. The advancing activity is present in the extract only during a short interval of 20-30 min. Significant advances could be demonstrated only if the plasmodia were injected during the first 3 h of the cell cycle at 25°C
334
SHORT
0
50
100
150
200
250
300
FIG. 2. Injection of active extracts (i.e., collected during the interval 30-50 min before telophase) at different phases of the acceptor plasmodium. Abscissa: Cell cycle phase of the plasmodium at the time of injection (min after second mitosis). Ordinate: Resulting delay/advance of mitosis in injected plasmodia (min).
(Fig. 2). At 21°C on the other hand, extracts were still able to advance the next mitosis if injected 5.5 h after mitosis (11-h cell cycle). Upon filtration of the active extract through desalting columns, the activity was retarded, indicating that it possesses a molecular mass below 4000 Da. In order to characterize the active compound further, the extract was subjected to a number of treatments. After incubation with proteinase K the activity was destroyed, but the incubation with RNase and phosphatase had no effect. Furthermore, the activity was destroyed after heating (10 min at 9O”C), lyophilization, freezing, ammonium sulfate precipitation, and acetone precipitation. The activity was not lost during fluid storage (0°C or -18°C in 20% ethanol). Using preparative gel filtration, the active extracts were separated into 88 fractions. Such fractionation experiments produced reproducible elution patterns: we always found a yellow pigment in fraction 25, whereas fraction 24 was always the only fraction with cell cycle accelerating activity (Fig. 3). By comparison with the elution of reference proteins, the molecular mass was estimated to be 2500 t- 300 Da. We then analyzed fractions 23, 24, and 25 by using HPLC and found a distinct peak in fraction 24 (Fig. 4). After stepwise evaporation of the HPLC buffer and resuspension in extract buffer, this fraction was injected and induced a 17-min advance. No accelerating activity was found in other HPLC fractions. Comparing gel filtration fraction 24 from three cell cycle phases (75, 45, and 25 min before telophase) by HPLC demonstrated that the active fraction appeared only 45 min before telophase (Fig. 5), as expected from the injection experiments with whole extracts. DISCUSSION
The mitosis-advancing because of its sensitivity
agent is probably a peptide, to proteinase K and because
NOTE
the active fraction absorbed at 214 and 280 nm during HPLC. Gel filtration indicated a molecular mass of the active compound between 2000 and 3000. The sensitivity of the agent to freezing seems unusual for a short peptide; however, denaturation in short peptides can occur. The mellitin-monomer (26 residues) forms an LYhelix [ 11; endothelin, a vasoconstrictory-active peptide of MW 2492 Da, has two intramolecular disulfide bonds [29]; substance-P, a neurological linear peptide of MW 1348 Da, loses its activity at higher concentrations or pH above 8.0 within minutes [18]; and the rapid loss of activity of secretin, a peptide hormone with 27 residues, is partly caused by an a-p-isomerization of asparagine [2]. Freezing may lead to an irreversible exposition of hydrophobic residues or to denaturation induced by the dramatic change in pH and ionic strength in tiny, shrinking cavities of nonfrozen solvent during freezing [22]. Earlier attempts to identify sensitive phases and control substances within the cell cycle of Physarum have produced results which are compatible with ours. The mitotic delay after uv irradiation decreases with the cycle progression, and at 50 min before metaphase it is not possible to delay nuclei any more [14]. Similarly, the sensitivity toward cycloheximide disappeared if it was administered less than 30 min before metaphase [21] or lo-20 min before metaphase, respectively [9]. This “point of no return” in the processes leading to nuclear division was more accurately determined by the fusion experiments of Loidl and Sachsenmaier [13], because, in contrast to the inhibitor or uv-light experiments, side effects were excluded to a greater extent in their approach. Portions of macroplasmodia of different cell cycle phases were fused and the subsequent synchronous mitosis was compared to the nonfused portions of the same plasmodia, which served as controls. Normally,
FIG. 3. Upper panel: Elution pattern of gel filtration. [TSKHW40S (Merck) in a 1.9 X 80.0~cm column eluted with 20 ml/h using injection buffer/bidest (l/5)]. Abscissa, fraction number; and ordinate, absorption. Lower panel: Advancing activity of selected fractions. Abscissa, fraction number; and ordinate, resulting delay/advance of mitosis in injected plasmodia (min). Fraction 24 contains the maximal advancing activity.
SHORT
0%
45-24
335
NOTE
sis trigger substance may appear shortly before this point. In our experiments, the maximum mitosis-advancing activity appears 45 min before late telophase, i.e., about 15 min or less before prophase and the pointof-no-return. A control point for the cell cycle at the G2/M border has been described in Saccharomyces pombe, S. cereuisiae, Xenopus, Spisula, HeLa cells, and other organisms. Essential cell cycle proteins such as CDCS, MPF, and cyclin show maximal activity at this time (for review see [15]). An extensively discussed model of the Physarum cell cycle is that of the “unstable activator” [23-25, 281. Ac-
25-24
A 45-25
45-24
;
4
-
0%
FIG. 4. Different gel filtration fractions 23,24, and 25 of extracts 45 k 5 min before telophase were analyzed by HPLC (Cl8 column 0.4 X 0.8 X 25.0 cm (Macherey & Nagel) with a tridest-0.1% TFA/acetonitrileeO.l% TFA gradient which increased from 0 to 100% acetonitrile in 140 min). The absorbance was measured at 214 (shown below) and 280 nm (shown above). The increase in absorbance at 280 nm is a result of the increasing absorbance of acetonitrile. The dashed line shows the concentration of acetonitrile (W).
75-24 1.0 /
“older” nuclei were delayed by fusion with “younger” plasmodia and younger nuclei were advanced by fusion with older plasmodia. The fused plasmodia, thus, underwent mitosis at a time intermediate between the two controls. This was not the case if fusion was induced 30 min before prophase, i.e., 50 min before metaphase or later: at least a subgroup of the older nuclei divided undelayed, following their normal schedule and not in synchrony with their younger counterparts. Loidl and Sachsenmaier estimated that 30 min is required for a complete mixture of the cytosols; thus nuclei were determined for mitosis shortly before or during the first visible signs of mitosis [13]. Consequently a final mito-
/
/
100%
FIG. 5. HPLC of the gel filtration fraction 24 from extracts collected at 25 f 5, 45 f 5 and 75 f 5 min before telophase. (Cl8 column 0.4 X 0.8 X 25.0 cm (Macherey & Nagel) with a tridest-0.1% TFA/ acetonitrile-0.1% TFA gradient which increased from 0 to 100% acetonitrile in 140 min). The absorbance was measured at 214 (shown below) and 280 nm (shown above). The increase in absorbance at 280 nm is a result of the increasing absorbance of acetonitrile. The dashed line shows the concentration of acetonitrile (%).
336
SHORT
cording to this model, a hypothetical mitotic activator is continuously synthesized, beginning after mitosis or at the end of the S phase. The activator “titrates” hypothetical receptors within the nuclei, which divide when all receptors are bound. The rate of synthesis is proportional to cell growth and adjusted in a way such that its concentration during the nonmitotic part of the cycle remains constant. This can be achieved if degradation plus binding to the receptor is exactly compensated by the synthesis rate. The interphase concentration could even be near zero. In this case, different concentrations of injected extract would not imply different concentrations of activator. The model is compatible with size control as well with the results of heat/shock, CHX-inhibitor, and fusion experiments [13, 19, 21, 261. This model predicts that the concentration of activator suddenly rises after complete saturation of all receptors, which then triggers (or parallels) the onset of mitosis. A higher rate of degradation or the doubling of receptor sites might account for the subsequent drop in activity. A similar titration model has been discussed for the system CDC25/CDC2 [7], WEEl/CDC25 [25], and cyclin/ MPF [6]. The kinetics of the appearance of the mitosis-advancing activity described here is compatible with this model: extracts from most interphase plasmodia have no effect, because they do not change the concentration of activator. Only during the short period before mitosis is the concentration of free activator high and effective in raising the concentration of younger acceptor plasmodia after injection. This model does not explain, however, why injections of the activator are more effective during the early phase (3 h) of the cell cycle. Furthermore, the cell cycle-accelerating peptide may be a product of the proteolytic degradation of cyclin (or an analogous protein in Physarum). Spisula cyclins are continuously synthesized during interphase and proteolytically degraded during mitosis [6]. The injected peptide could “compete” with the degradation of the cyclin analogon during interphase and advance the cycle in this way. A 16-residue peptide derived from CDC2 induced chromatin condensation and advanced nuclear-envelope breakdown, which was interpreted, among other possibilities, as competition in degradation [8]. The 2.5-kDa peptide may as well represent a protease inhibitor, which prevents the cyclin protease from splitting cyclin during interphase. Shortly before mitosis, higher calcium concentrations, or other events in the cell cycle, may lead to a dissociation between inhibitor and protease. Both assumptions would be compatible with our observation that the injection of protease inhibitors like pepstatin A, leupeptin, and aprotinin advance the cell Received May 21, 1990 Revised version received July 9, 1990
NOTE
cycle in Physarum, whereas the injection delays the cycle [data not shown].
of pronase E
We thank Dr. R. Frank, GBF, Braunschweig for experimental assistance and for helpful discussions and Dr. M. Vicker, Univ. Bremen, for critically reading the manuscript.
REFERENCES 1.
2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
Baldwin, R. L., and Eisenberg, D. (1987) in Protein Engineering (Oxender, D. L., and Fox, C. F., Eds.), pp. 127-146, A. R. Liss, New York. Beyermann, H., Grosman, M. I., Scratchard, T., Solomon, T. E., and Voskamp, D. (1981) Life Sci. 29,885-892. Blessing, J., and Lempp, H. (1978) Ezp. Cell Res. 113,435438. Brewer, E. N. (1968) Fed. Proc. 28,840-844. Daniel, J. W., and Rusch, H. P. (1961) J. Gen. Microbial. 25, 47-53. Draetta, G., Luca, F., Westendorf, J., Brizuela, L., Ruderman, J., and Beach, D. (1989) Cell 56,829-838. Figge, J., and Smith, T. F. (1988) Nature (London) 334, 109. Gautier, J., Norbury, C., Lohka, M., Nurse, P., and Maller, J. (1988) Cell 54, 433-440. Herdugo, G. G., and Tyson, J. J. (1984) Exp. Cell Res. 153,570573. Kukulies, J., Stockem, W., and Achenbach, F. (1984) Eur. J. Cell Biol. 35, 235-241. Loidl, P., and Griibner, P. (1982) Exp. Cell Res. 13'7, 469-472. Loidl, P., and Griibner, P. (1985) J. Cell Biol. 100, 1930-1933. Loidl, P., and Sachsenmaier, W. (1982) Eur. J. Cell Biol. 28, 175-184. Matsumoto, S. (1977) Cell Struct. Funct. 2, 101&107. Murray, A., and Kirschner, M. W. (1989) Science 246,614-621. Nygaard, 0. F., Guttes, S., and Rusch, H. P. (1960) Biochim. Biophys. Acta 38, 298-302. Oppenheim, A., and Katzir, Z. (1971) Exp. Cell Res. 68,224-226. Porter, R., and O’Connor, M. (1982) Ciba Found. Symp., 91-99. Rusch, H. P., Sachsenmaier, W., Behrens, K., and Gruter, V. (1966) J. Cell Biol. 31, 204-210. Sachsenmaier, W. (1981) in The Cell Cycle (John, P., Ed.), pp. 73-80, Cambridge Univ. Press, Cambridge. Scheffey, C., and Wille, J. J. (1978) Exp. Cell Res. 113,259-262. Taborsky, G. (1979) Peptides, pp. l-25, de Gruyter, Berlin. Tyson, J., and Sachsenmaier, W. (1978) J. Theor. Biol. 73,723727. Tyson, J., and Sachsenmaier, W. (1985) in The Cell Cycle (Edmunds, L., Ed.), pp. 253-260, Springer, New York. Tyson, J. (1987) J. Theor. Biol. 126, 381-391. Tyson, J., Garcia-Herdugo, G., and Sachsenmaier, W. (1979) Exp. Cell Res. 119, 87-94. Ueda, T., Von Olenhusen, K. G., and Wohlfarth-Bottermann, K. E. (1978) Cytobiology 18, 76-83. Wheals, A., and Silverman, B. (1982) J. Theor. Biol. 97, 505510. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yazaki, Y., Goto, K., and Masaki, T. (1988) Nature (London) 332, 411-413.
CELL
RESEARCH
19 1,
332-336 (1990)
SHORT NOTE A Purified Cellular Extract Accelerates the Cell Cycle in Physarum polycephalum UWE HOBOHM,
ARMIN
Dept. Biology, University
HILDEBRANDT,
of Bremen, D-2800 Bremen, West Germany
plasmodia harvested, in this case, 3 h before mitosis [4, 171. Loidl and Grijbner applied plasmodial extracts during the fusion of microplasmodia and monitored the effect on the time of the first mitosis. They found a maximal accelerating effect in plasmodia collected about 90 min before mitosis [ 111. When they purified the accelerating activity lo-fold, it showed many of the characteristic properties of proteins [ 121. One main difficulty in identifying hypothetical control substances was the lack of an adequate biological assay. In order to test extracts from different phases of the cell cycle by using microinjection, it was necessary to develop an injection buffer which diffuses freely within the entire plasmodium. The results of our experiments with such a buffer indicate that a substance of 2.5 kDa, probably a peptide, which is produced rather late in the cell cycle of Physarum, is able to advance the cycle. Control processes shortly before mitosis are well known in other organisms, e.g., yeast, frog embryos, and HeLa cells. However, the participation of a low-molecular-mass compound has not been observed before (for review see [15]).
Plasmodia of the myxomycete Physarum polycephalum (strain Cl) were collected at different times during the cell cycle and extracts were prepared from homogenates using a buffer optimized for microinjection into plasmodial veins. These extracts were injected into plasmodia during the first 3 h of the cell cycle. The time of the following mitosis was monitored and compared with that of the buffer-injected controls. Extracts of plasmodia homogenized 45 min before late telophase accelerated the onset of mitosis in the injected plasmodium up to 70 min, i.e., an advance of lo-14% compared to the f3- to 10-h cell cycle duration of the controls. The accelerating activity vanished completely after heating, freezing, or protease digestion, thus indicating the peptide nature of the active agent. Purification of the active compound by means of gel filtration revealed a molecular mass of about 2500 Da. The active portion of the extract was further fractionated by HPLC and the activity determined in a single peak. (c‘ 1990
Academic
Press,
Inc.
INTRODUCTION
MATERIALS
The plasmodial stage of Physarum may serve as a unique model system for cell cycle investigations, because its million nuclei divide in almost perfect synchrony [16]. Several attempts have been made to detect hypothetical trigger or control substances: Blessing and Lempp, for example, layered cellular extracts onto the surface of the plasmodia. They described an acceleration of the cell cycle induced by extracts from donor plasmodia collected 45 min before metaphase. No effect was observed if the donor plasmodia had passed mitosis [3]. Similar experiments with lyophilized plasmodia from late phases of the cell cycle layered onto the surface of plasmodia in early phases again resulted in advances. A maximal accelerating effect was induced by
r To whom reprint 0014.4827/90
Copyright All
rights
requests should be addressed.
$3.00 0 1990 by Academic Press, of reproduction in any form
332 Inc. reserved.
AND LUDGER RENSING~
AND
METHODS
Strains, culture conditions, and preparation of plasmodial extracts. Macroplasmodia were cultivated in petri dishes on filter paper (Schleicher & Schtilll575, Dassel, West Germany) layered on a steel wire grid on a semidefined nutrient medium supplemented with hemin. Mitosis was determined in ethanol-fixed smears under a phase-contrast microscope [5]. The fusion of microplasmodia was defined as mitosis 0. Late telophase was used as a marker, i.e., when the time point of complete segregation of the two new nuclei was observed, which could be determined more precisely (up to +-1 min) than metaphase. Whole plasmodia were collected at specific stages of the cell cycle, frozen in liquid nitrogen, and stored at 80°C. The following preparative steps were done on ice. Frozen samples were thawed, suspended in 1.5~ (w/w) injection buffer (pH ll.O), vigorously stirred with a spatula for 1 min, adjusted to pH 7.0, and centrifuged for 30 min at 33,000g (4°C). Purified samples were stored liquid at ~18°C with 20% ethanol. Using a Speedvac, the ethanol was stepwise evaporated and the samples were filled with injection buffer up to the original volume for further use. We tested the haploid Screening of strains, buffers and injection. strains Cl-188, Cl-lB9, Cl-Mu, Cl-82, the diploid strains M-Go, F,
SHORT Tu, and the following buffers (components in g/liter bidest): phosphate buffer: 3.52 g KH,PO,, 7.26 g Na,HPO,. 2H,O; sodium borate buffer: 4.77 g Na,B,O, . lOH,O, 20.5 ml 1 M HCl; sodium citrate/ NaOH buffer: 12.53 g citric acid monohydrate, 159.6 ml 1 M NaOH; Ringer’s solution 1: 6.5 g NaCl, 0.14 g KCl, 0.12 g CaCl,, 0.2 g NaHCO,, 0.01 g NaH,PO,. H,O; Ringer’s solution 2: 8.6 g NaCl, 0.3 g KCl, 0.33 g CaCl,, 0.2 g NaHCO,, NaH,PO,. H,O; modified Ringer’s solution 2 without Ca’+; Tris buffer: 12.12 g Tris-HCl; imidazole-HCl buffer: 6.81 g imidazole (Merck, Darmstadt, West Germany), 24.3 ml 0.1 M HCl; 0.1 M Hepes buffer; 0.1 M Mes buffer; 0.1 M Pipes buffer; injection buffer [lo]: 2 mM Tris base, 0.2 mM ATP, 0.5 M DTT; injection buffer [27]: 2.25 g KCl, 0.61 g MgCl,, 0.012 g ATP, 1.6 g TrisHCl, 0.1 ml 4 X 10e3 M CaCl,. The buffers were adjusted to about 200 mOsm and different pH (6.5,7.0,7.5) before injection. During the screening experiments 5510 ~1 buffer was injected during 5 min at different locations into plasmodia. An injection was judged successful when no sign of injection and no injury were detected after 223 h and if the buffer was equally distributed inside the entire plasmodium. During extract injection experiments, one plasmodium 445 cm in diameter was cut into 4-6 pieces, and 3-5 ~1 of extract or of buffer was injected during 2-3 min into each piece. Extract (200 ~1) was incubated with ei.&gradation e.rperiments. ther 5 U carrier bound phosphatase (Sigma, Deisenhofen, West Germany), 5 U carrier bound RNase (Sigma), 5 U carrier bound proteinase K (Sigma) for 1 h, pH 7.0 (phosphatase, pH 8.0), at 37°C. During incubation, the mixture was gently shaken. The carriers were then pelleted and the supernatant was tested by means of injection. Crude extracts were fractionated using a Gel filtration and HPLC. 1.9 X 80.0.cm column filled with TSK-HW40S (Merck). The elution buffer was injection buffer/bidest (l/5); 20 ml/h was eluted and 5-ml fractions were collected and concentrated 5/l using a Speedvac centrifuge before injection. A 0.4 X 0.8 X 25.0.cm, 300.7~Cl8 column (Macherey & Nagel, Diiren, West Germany) was used for HPLC. A 200.~1 extract was protonated with trifluoroacetate (TFA) and absorption measured at 214 and 280 nm (LKB 2141). A gradient tridest/ 0.1% TFA (A) to acetonitrile/O.l% TFA (B) was used with the following program: 8 min B 0%, 140 min B increasing to 100%. Fractions were collected and concentrated to 200 ~1 using a Speedvac. The HPLC buffer was stepwise evaporated and redissolved in extract buffer before injection.
RESULTS
Injection of extracts into nonstarving macroplasmodia of Physarum is difficult: macroplasmodia tend to either block injected veins or develop visibly disturbed areas, which stop growth. We tested the 13 different buffers (see Materials and Methods) at different pH. Two of the buffers were described as suitable for microinjection into Physarum by Kukulies [lo] and Ueda [27]. All buffers were tested in seven strains of Physwum (Cl-lB8, Cl-lB9, Cl-Mu, Cl-82, M-Go, F, Tu) and revealed a strain-dependent sensitivity to injections of buffer. The best results were observed when strain ClMii was injected with a modified Ringer’s solution (pH 7.0). With this strain and buffer, signs of injection disappeared after 2-3 h and the plasmodium showed a continuous, solid surface of uniform color. In order to test the distribution of injected substances within the plasmodium, we used RNase (0.01 M) and [3H]thymidine. The injection of RNase resulted in ex-
333
NOTE I
I
I
,o 0
60
40
20
0
-20 250
200
150
100
SO
0
FIG. 1.
Cellular extracts from different cell cycle phases and their advancing/delaying effect on the third mitosis of the injected plasmodium. Abscissa: Phase of donor plasmodium at the time of harvesting (min before third mitosis). Ordinate: Resulting delay/advance of mitosis in injected plasmodia (min). All injections were administered during the first 3 h after the second synchronous mitosis after the fusion of microplasmodia.
cessive production of slime, disintegration of nuclei, and death of the entire plasmodium within i-1 h. After injection of [3H]thymidine into one part of the macroplasmodium, radioactivity was detected at the most distant part within 1 h. Furthermore, successful injections appeared to depend on optimal growth: if plasmodia grew as a thin network, injections were impossible. In order to evaluate the effects of injection experiments on cell cycle duration it was necessary to know the intra- and interplasmodial cell cycle synchrony of the strain used. In 87 plasmodia with a diameter up to 5 cm, the deviation between nuclei of distant parts at the time of late telophase was not more than 4.00 (k2.13) min. This degree of synchrony was higher than the lomin deviations reported by Sachsenmaier [20]. In order to test the interplasmodial synchrony, one plasmodium was cut into 2-4 pieces and monitored for the time of the following mitosis. Pieces which grew isolated for 250-300 min showed a difference of 2-10 min in the time of mitosis, whereas after a 9-h isolation there was a difference of lo-22 min. The interplasmodial synchrony in our experiments is, thus, somewhat less than that described by Loidl and Grobner [ll]. In the following experiments we injected extracts into 4-6 pieces of plasmodia and compared the time of mitosis to that of 4-6 buffer-injected controls. With this number of pieces an average phase difference of 10 min marks the 95% level of significance. The results of the injection experiments with whole extracts are shown in Fig. 1. Donor extracts obtained 45 min before late telophase (=35 min before metaphase) advanced the mitosis of the injected plasmodium up to 70 min. The advancing activity is present in the extract only during a short interval of 20-30 min. Significant advances could be demonstrated only if the plasmodia were injected during the first 3 h of the cell cycle at 25°C
334
SHORT
0
50
100
150
200
250
300
FIG. 2. Injection of active extracts (i.e., collected during the interval 30-50 min before telophase) at different phases of the acceptor plasmodium. Abscissa: Cell cycle phase of the plasmodium at the time of injection (min after second mitosis). Ordinate: Resulting delay/advance of mitosis in injected plasmodia (min).
(Fig. 2). At 21°C on the other hand, extracts were still able to advance the next mitosis if injected 5.5 h after mitosis (11-h cell cycle). Upon filtration of the active extract through desalting columns, the activity was retarded, indicating that it possesses a molecular mass below 4000 Da. In order to characterize the active compound further, the extract was subjected to a number of treatments. After incubation with proteinase K the activity was destroyed, but the incubation with RNase and phosphatase had no effect. Furthermore, the activity was destroyed after heating (10 min at 9O”C), lyophilization, freezing, ammonium sulfate precipitation, and acetone precipitation. The activity was not lost during fluid storage (0°C or -18°C in 20% ethanol). Using preparative gel filtration, the active extracts were separated into 88 fractions. Such fractionation experiments produced reproducible elution patterns: we always found a yellow pigment in fraction 25, whereas fraction 24 was always the only fraction with cell cycle accelerating activity (Fig. 3). By comparison with the elution of reference proteins, the molecular mass was estimated to be 2500 t- 300 Da. We then analyzed fractions 23, 24, and 25 by using HPLC and found a distinct peak in fraction 24 (Fig. 4). After stepwise evaporation of the HPLC buffer and resuspension in extract buffer, this fraction was injected and induced a 17-min advance. No accelerating activity was found in other HPLC fractions. Comparing gel filtration fraction 24 from three cell cycle phases (75, 45, and 25 min before telophase) by HPLC demonstrated that the active fraction appeared only 45 min before telophase (Fig. 5), as expected from the injection experiments with whole extracts. DISCUSSION
The mitosis-advancing because of its sensitivity
agent is probably a peptide, to proteinase K and because
NOTE
the active fraction absorbed at 214 and 280 nm during HPLC. Gel filtration indicated a molecular mass of the active compound between 2000 and 3000. The sensitivity of the agent to freezing seems unusual for a short peptide; however, denaturation in short peptides can occur. The mellitin-monomer (26 residues) forms an LYhelix [ 11; endothelin, a vasoconstrictory-active peptide of MW 2492 Da, has two intramolecular disulfide bonds [29]; substance-P, a neurological linear peptide of MW 1348 Da, loses its activity at higher concentrations or pH above 8.0 within minutes [18]; and the rapid loss of activity of secretin, a peptide hormone with 27 residues, is partly caused by an a-p-isomerization of asparagine [2]. Freezing may lead to an irreversible exposition of hydrophobic residues or to denaturation induced by the dramatic change in pH and ionic strength in tiny, shrinking cavities of nonfrozen solvent during freezing [22]. Earlier attempts to identify sensitive phases and control substances within the cell cycle of Physarum have produced results which are compatible with ours. The mitotic delay after uv irradiation decreases with the cycle progression, and at 50 min before metaphase it is not possible to delay nuclei any more [14]. Similarly, the sensitivity toward cycloheximide disappeared if it was administered less than 30 min before metaphase [21] or lo-20 min before metaphase, respectively [9]. This “point of no return” in the processes leading to nuclear division was more accurately determined by the fusion experiments of Loidl and Sachsenmaier [13], because, in contrast to the inhibitor or uv-light experiments, side effects were excluded to a greater extent in their approach. Portions of macroplasmodia of different cell cycle phases were fused and the subsequent synchronous mitosis was compared to the nonfused portions of the same plasmodia, which served as controls. Normally,
FIG. 3. Upper panel: Elution pattern of gel filtration. [TSKHW40S (Merck) in a 1.9 X 80.0~cm column eluted with 20 ml/h using injection buffer/bidest (l/5)]. Abscissa, fraction number; and ordinate, absorption. Lower panel: Advancing activity of selected fractions. Abscissa, fraction number; and ordinate, resulting delay/advance of mitosis in injected plasmodia (min). Fraction 24 contains the maximal advancing activity.
SHORT
0%
45-24
335
NOTE
sis trigger substance may appear shortly before this point. In our experiments, the maximum mitosis-advancing activity appears 45 min before late telophase, i.e., about 15 min or less before prophase and the pointof-no-return. A control point for the cell cycle at the G2/M border has been described in Saccharomyces pombe, S. cereuisiae, Xenopus, Spisula, HeLa cells, and other organisms. Essential cell cycle proteins such as CDCS, MPF, and cyclin show maximal activity at this time (for review see [15]). An extensively discussed model of the Physarum cell cycle is that of the “unstable activator” [23-25, 281. Ac-
25-24
A 45-25
45-24
;
4
-
0%
FIG. 4. Different gel filtration fractions 23,24, and 25 of extracts 45 k 5 min before telophase were analyzed by HPLC (Cl8 column 0.4 X 0.8 X 25.0 cm (Macherey & Nagel) with a tridest-0.1% TFA/acetonitrileeO.l% TFA gradient which increased from 0 to 100% acetonitrile in 140 min). The absorbance was measured at 214 (shown below) and 280 nm (shown above). The increase in absorbance at 280 nm is a result of the increasing absorbance of acetonitrile. The dashed line shows the concentration of acetonitrile (W).
75-24 1.0 /
“older” nuclei were delayed by fusion with “younger” plasmodia and younger nuclei were advanced by fusion with older plasmodia. The fused plasmodia, thus, underwent mitosis at a time intermediate between the two controls. This was not the case if fusion was induced 30 min before prophase, i.e., 50 min before metaphase or later: at least a subgroup of the older nuclei divided undelayed, following their normal schedule and not in synchrony with their younger counterparts. Loidl and Sachsenmaier estimated that 30 min is required for a complete mixture of the cytosols; thus nuclei were determined for mitosis shortly before or during the first visible signs of mitosis [13]. Consequently a final mito-
/
/
100%
FIG. 5. HPLC of the gel filtration fraction 24 from extracts collected at 25 f 5, 45 f 5 and 75 f 5 min before telophase. (Cl8 column 0.4 X 0.8 X 25.0 cm (Macherey & Nagel) with a tridest-0.1% TFA/ acetonitrile-0.1% TFA gradient which increased from 0 to 100% acetonitrile in 140 min). The absorbance was measured at 214 (shown below) and 280 nm (shown above). The increase in absorbance at 280 nm is a result of the increasing absorbance of acetonitrile. The dashed line shows the concentration of acetonitrile (%).
336
SHORT
cording to this model, a hypothetical mitotic activator is continuously synthesized, beginning after mitosis or at the end of the S phase. The activator “titrates” hypothetical receptors within the nuclei, which divide when all receptors are bound. The rate of synthesis is proportional to cell growth and adjusted in a way such that its concentration during the nonmitotic part of the cycle remains constant. This can be achieved if degradation plus binding to the receptor is exactly compensated by the synthesis rate. The interphase concentration could even be near zero. In this case, different concentrations of injected extract would not imply different concentrations of activator. The model is compatible with size control as well with the results of heat/shock, CHX-inhibitor, and fusion experiments [13, 19, 21, 261. This model predicts that the concentration of activator suddenly rises after complete saturation of all receptors, which then triggers (or parallels) the onset of mitosis. A higher rate of degradation or the doubling of receptor sites might account for the subsequent drop in activity. A similar titration model has been discussed for the system CDC25/CDC2 [7], WEEl/CDC25 [25], and cyclin/ MPF [6]. The kinetics of the appearance of the mitosis-advancing activity described here is compatible with this model: extracts from most interphase plasmodia have no effect, because they do not change the concentration of activator. Only during the short period before mitosis is the concentration of free activator high and effective in raising the concentration of younger acceptor plasmodia after injection. This model does not explain, however, why injections of the activator are more effective during the early phase (3 h) of the cell cycle. Furthermore, the cell cycle-accelerating peptide may be a product of the proteolytic degradation of cyclin (or an analogous protein in Physarum). Spisula cyclins are continuously synthesized during interphase and proteolytically degraded during mitosis [6]. The injected peptide could “compete” with the degradation of the cyclin analogon during interphase and advance the cycle in this way. A 16-residue peptide derived from CDC2 induced chromatin condensation and advanced nuclear-envelope breakdown, which was interpreted, among other possibilities, as competition in degradation [8]. The 2.5-kDa peptide may as well represent a protease inhibitor, which prevents the cyclin protease from splitting cyclin during interphase. Shortly before mitosis, higher calcium concentrations, or other events in the cell cycle, may lead to a dissociation between inhibitor and protease. Both assumptions would be compatible with our observation that the injection of protease inhibitors like pepstatin A, leupeptin, and aprotinin advance the cell Received May 21, 1990 Revised version received July 9, 1990
NOTE
cycle in Physarum, whereas the injection delays the cycle [data not shown].
of pronase E
We thank Dr. R. Frank, GBF, Braunschweig for experimental assistance and for helpful discussions and Dr. M. Vicker, Univ. Bremen, for critically reading the manuscript.
REFERENCES 1.
2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
Baldwin, R. L., and Eisenberg, D. (1987) in Protein Engineering (Oxender, D. L., and Fox, C. F., Eds.), pp. 127-146, A. R. Liss, New York. Beyermann, H., Grosman, M. I., Scratchard, T., Solomon, T. E., and Voskamp, D. (1981) Life Sci. 29,885-892. Blessing, J., and Lempp, H. (1978) Ezp. Cell Res. 113,435438. Brewer, E. N. (1968) Fed. Proc. 28,840-844. Daniel, J. W., and Rusch, H. P. (1961) J. Gen. Microbial. 25, 47-53. Draetta, G., Luca, F., Westendorf, J., Brizuela, L., Ruderman, J., and Beach, D. (1989) Cell 56,829-838. Figge, J., and Smith, T. F. (1988) Nature (London) 334, 109. Gautier, J., Norbury, C., Lohka, M., Nurse, P., and Maller, J. (1988) Cell 54, 433-440. Herdugo, G. G., and Tyson, J. J. (1984) Exp. Cell Res. 153,570573. Kukulies, J., Stockem, W., and Achenbach, F. (1984) Eur. J. Cell Biol. 35, 235-241. Loidl, P., and Griibner, P. (1982) Exp. Cell Res. 13'7, 469-472. Loidl, P., and Griibner, P. (1985) J. Cell Biol. 100, 1930-1933. Loidl, P., and Sachsenmaier, W. (1982) Eur. J. Cell Biol. 28, 175-184. Matsumoto, S. (1977) Cell Struct. Funct. 2, 101&107. Murray, A., and Kirschner, M. W. (1989) Science 246,614-621. Nygaard, 0. F., Guttes, S., and Rusch, H. P. (1960) Biochim. Biophys. Acta 38, 298-302. Oppenheim, A., and Katzir, Z. (1971) Exp. Cell Res. 68,224-226. Porter, R., and O’Connor, M. (1982) Ciba Found. Symp., 91-99. Rusch, H. P., Sachsenmaier, W., Behrens, K., and Gruter, V. (1966) J. Cell Biol. 31, 204-210. Sachsenmaier, W. (1981) in The Cell Cycle (John, P., Ed.), pp. 73-80, Cambridge Univ. Press, Cambridge. Scheffey, C., and Wille, J. J. (1978) Exp. Cell Res. 113,259-262. Taborsky, G. (1979) Peptides, pp. l-25, de Gruyter, Berlin. Tyson, J., and Sachsenmaier, W. (1978) J. Theor. Biol. 73,723727. Tyson, J., and Sachsenmaier, W. (1985) in The Cell Cycle (Edmunds, L., Ed.), pp. 253-260, Springer, New York. Tyson, J. (1987) J. Theor. Biol. 126, 381-391. Tyson, J., Garcia-Herdugo, G., and Sachsenmaier, W. (1979) Exp. Cell Res. 119, 87-94. Ueda, T., Von Olenhusen, K. G., and Wohlfarth-Bottermann, K. E. (1978) Cytobiology 18, 76-83. Wheals, A., and Silverman, B. (1982) J. Theor. Biol. 97, 505510. Yanagisawa, M., Kurihara, H., Kimura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yazaki, Y., Goto, K., and Masaki, T. (1988) Nature (London) 332, 411-413.