Increase in macromolecular amounts during the cell cycle of tetrahymen a a contribution to cell cycle control*

In an exponentially multiplying cell system the cell division cycle of individual cells is to be considered as a set of regularly repeating events which occur in an ordered sequence. The significance of each of these events for the completion of the cycle can be studied by inhibiting single events or by experimental alteration of external conditions. By several different approaches causal relationships between parts of the cell cycle have been described. It was found e.g. that in Tetrahymena replication of macronuclear DNA can be separated in different ways from the cell division cycle, demonstrating that the completion of the division cycle is not directly dependent on the replication of DNA and vice versa (Hjelm & Zeuthen, 1967; Cleffmann, 1968; Frankel, Jenkins & DeBault, 1976; Zeuthen, 1978). Nevertheless, in the vast majority of cell cycles the coupling of replication and division is maintained and replication occurs invariantly at the same stage of the cell cycle. Therefore, some connexion between these 2 events must exist. The nature of this coupling mechanism is not known. It may consist of a more complex situation of the cell in which cell constituents reach specific ratios. To find such situations we describe in the present study the development of the main macromolecular components during the cell cycle in Tetrahymena. The results are based on cytophotometric and autoradiographic determinations in single cells. This allows one to relate different parameters for individual cells. The findings suggest that the initiation of DNA synthesis is under negative control of relative RNA and/or DNA content in the cell.

Cultures of Tetrahymena sp. strain HSM^* were grown in synthetic medium at 29 °C. By regular transfer the cells were kept in exponential growth phase : the time for doubling in cell number (= average generation time) was 210 min and constant within the range of 500 to 30000 cells per ml. The multiplication rate was determined by measuring cell densities in culture samples (Coulter Counter) or by averaging individual generation times of cell samples in capillaries. Cells for experiments described below were taken from cultures at about 1000 cells per ml. They were collected at late division, transferred to capillaries, and prepared for autoradiography or cytophotometry at the indicated times. They were usually dried onto gelatine-coated slides and fixed in ethanol acetic acid (3:1, 20 min).

Cell size was measured as the area of dried and stained cells taken by a planimeter on photographs. Since the data of Morrison & Thompkins (1973) as well as our own demonstrate that determinations of projected area, Coulter volume, and volume of fixed cells in suspension measured microscopically are in good correlation in samples of different developmental stages and exhibit the same pattern of development, area has been taken as the measure for cell size.

RNA content was determined cytophotometrically at 600 nm after staining with gallo-cyanine according to Kiefer, Kiefer & Sandritter (1966). For cytophotometry a Zeiss microscope photometer SMP 05 equipped with a scanning stage and a 0·5-μm diaphragm was used throughout this study. Since gallocyanine stains all nucleic acids parallel samples were taken from G₁ and G₂ cells treated with RNase and stained with gallocyanine. The average absorbence of these cells was subtracted from the readings from the experimental cells.

Protein content was determined cytophotometrically at 410 nm after staining with dinitrofluorbenzene (DNFB) according to Kimball et al. (1971).

RNA synthesis was determined by grain counting in autoradiographs after 15-min labelling with [³H]uridine at a final concentration of 2 μCi ³H-U per ml. The technical details have been described earlier. For discussion of the method see section Results.

Protein synthesis was determined as incorporation rate of PHjleucine. Samples of 20–30 cells were dried on slides after a 10-min pulse and radioactivity determined after solubilizing the cells.

RNA toprotein ratio of individual cells was determined by successive staining and measuring both parameters. Since absorption spectra of DNFB and gallocyanine are sufficiently different that DNFB contributes at 600 nm less than 1% to gallocyanine absorption, protein content was measured first, and after recording the location of the cells RNA content was determined as described above.

The relation of RNA synthesis to macronuclear DNA amount was investigated by first performing autoradiography for [³H]uridine incorporation and staining the same cells with Feulgen as described earlier (Cleffmann, 1968). Hydrolysis at 55 °C removes the autoradiographic emulsion together with the silver grains so that cytophotometry of Feulgen-positive material at 550 nm can be applied (Kimball et al. 1971).

The timing of the macronuclear DNA replication phase (S-phase) of Tetrahymena has been determined in a number of previous papers (e.g. Stone & Cameron, 1964). Under the culture conditions applied in this study DNA is replicated between 55 and 135 min after cell division, which amounts to 26–65% °f the cell cycle. Initiation and termination of replication may vary between cells by ± 15 min. All available data show that in Tetrahymena the amount of DNA in the macronucleus is doubled during S-phase. Furthermore it was found that DNA is regularly lost during division (Cleffmann, 1968). On the average 3% of the DNA of a G₂ macronucleus is extruded from the dividing macronucleus. Therefore, the sum of the DNA in the daughter nuclei is less than the DNA in the macronucleus of the mother.

Cytophotometric determination of RNA in individual cells of known cell cycle age show that the rate of increase is not constant throughout the cycle (Fig. 1). During phase RNA increase is very low or even absent. During S-phase RNA increase accelerates. During C₂ the RNA increase on the average remains at the same high rate as at the end of S-phase. Repetitions of the experiment reveal essentially the same pattern in particular with respect to the low rate of RNA accumulation during and the increasing rate during S. The increase of RNA during a complete cycle does not result in a doubling of the content at the beginning of the cycle.

The increased accumulation of RNA during S and G₂ may depend on the number of templates which increases during Since the macronuclear DNA amount is variable in Tetrahymena it is possible to test whether the amount of RNA produced per cell is related to the macronuclear DNA content. Therefore, G₁ cells were incubated in [³H]uridine. After autoradiography the amount of label incorporated in acid-insoluble material was determined by counting grains per cell. Thereafter cells were stained with Feulgen and the amount of DNA measured cytophotometrically. Uridine incorporation alone is not an unequivocal measure for transcription rate. Several parameters like rate of uptake, pool size, turnover, selfabsorption may influence the results.

In the present experiment, however, cells of the same physiological state are compared. Therefore, the intensity of labelling is taken to indicate RNA-synthetic rate.

Fig. 2 shows that uridine labelling and DNA amount are positively correlated. The correlation between the parameters is rather weak due to the scatter of data. Therefore, only qualified conclusions can be drawn. Two repetitions of the same experiment, however, produced essentially the same results. This suggests that one of the factors determining the rate of RNA synthesis is the amount of DNA. This holds true not only for the changing DNA content during progress of the cell division cycle as stated above but also for DNA content varying between individual cells.

The rates of protein increase differ from the rates of RNA increase (Fig. i). Protein accumulation is lowest during S-phase and higher in G₁and particularly G₂. This pattern is due to changing rate in protein synthesis which decreases by about one third during early S-phase (Fig. 3). The same pattern as for protein increase is found for the increase in cell size (Fig. 1). In this case the retardation of increase during S-phase is even more pronounced. Cell size and protein content develop at similar rates during the cell cycle. This is also demonstrated by a close correlation (r = + 0.85) between cell size and cellular protein for individual cells (n = 52). Again protein content is not completely doubled during the cell cycle. Whereas the factor by which RNA accumulates is 1.6, it is 1.9 for cell size and 1.7 for protein.

The increase of RNA and the increase of protein (or size) during the cell cycle are inverse with respect to their relative rate. This leads to changing ratios of the 2 components. These ratios were measured directly in individual cells during the cell cycle. The RNA : protein ratio at the beginning of the cell cycle was set to 1. In Fig. 4 it is shown that the ratio decreases during the G₁-phase and reaches a minimum during early S. It increases during S-phase and remains approximately constant during G₂ at a level which is characteristic for the beginning of the cycle.

There are some reports on the growth pattern of Tetrahymena within the cell cycle. Determination of cell volume (Cameron & Prescott, 1961) and ³H-amino acid incorporation into protein (Prescott, 1960) revealed a linearity of growth except for the short period prior to division. Using the cartesian diver for a survey of growth of single cells in dry mass it was found that individual cells may exhibit different patterns such as linear or exponential increase (Lôvlie, 1963). The pattern of cell growth varies even more if one considers different cell systems. Every cell type, therefore, has to be analysed separately, distinguishing between different parameters of cell growth. The data presented here for Tetrahymena show that protein and RNA increase during the cell cycle at different rates. This is particularly obvious during G₁. Whereas the RNA content of the cell remains almost constant, protein and cell size increase considerably. During S, on the other hand, the increase in RNA is pronounced and the accumulation of protein slows down. The pattern leads to a relative decrease of RNA as referred to protein content and cell size. This could also be demonstrated directly in single cells. The absolute rates of increase of these 2 compounds are of such magnitude that they sum up to an almost linear increase in macromolecular dry weight throughout the cell cycle (Reuter, 1974).

The rate of increase of cellular RNA is subject to several limiting factors. The present study suggests that one of them is connected to macronuclear DNA amount. During S this amount is doubled and RNA accumulation increases markedly at this time. An increase in RNA synthesis in connexion with the S-phase has also been found in previous studies (Cleffmann, 1965 ; Jauker, Seyfert & Sgonina, 1975; Keiding & Andersen, 1978). One of these reports (Jauker et al. 1975) demonstrates that increase in RNA-synthetic rate is always coupled to S-phase under different growth conditions.

This gene dose effect on transcription rate does not only occur during the cell cycle of the average cell but also in individual cells since in cells with lower macronuclear DNA content less RNA is produced. This is not surprising but has not yet been shown on the level of individual cells. It is evident from Fig. 2 that the correlation between DNA amount and uridine incorporation is not proportional in the sense that a doubling in DNA does not result in a doubling in incorporation. From this and from the finding that incorporation rate may be higher during S (Jauker et al. 1975) than during G₂ it is evident that the amount of DNA is not the only factor determining the transcriptional activity and the gain in cellular RNA.

There is no definite information on the regulation of the particular pattern of protein accumulation in Tetrahymena. It is demonstrated here that the specific pattern of protein increase is a result of changing rate of protein synthesis.

Earlier findings (Jauker, 1977), this study, and recent unpublished results from this laboratory demonstrate that RNA increase as well as protein and volume increase do not result in a complete doubling within the cell cycle of exponentially growing Tetrahymena cultures. The mode of regulation of this unbalanced growth has been found and discussed with respect to RNA by Jauker (1977).

Several lines of evidence point out that the reduction of RNA/protein ratio during G₁ and at the beginning of S is not only a temporal coincidence but that a low RNA concentration is causally related to initiation of DNA replication: Decrease in macronuclear DNA as it occurs during successive cell generations and which is accompanied by reduction in transcription induces extra replication periods as soon as the DNA amount has reached a lower threshold (Cleffmann, 1968). Experimental inhibition of DNA synthesis by hydroxyurea releases replication of DNA immediately after division excluding a period (Worthington, Salamone & Nachtwey, 1976.) It was further shown by Seyfert (1977) that low macronuclear DNA content is connected to an early beginning of replication. These findings show that the lower the number of templates the sooner replication is initiated. This relationship is not confined to Tetrahymena but has also been found for tissue culture cells (Cress & Gemer, 1977). The amount of DNA in fact acts through transcriptional activity as was shown by Jauker (1975,who experimentally reduced transciiptions at the end of a cell cycle and found a significantly advanced replication phase.

All these findings suggest that replication is initiated when RNA in relation to protein reaches a lower threshold. This point will be attained if at increasing protein the net increase of RNA is low. Concentration of RNA or a certain fraction of it falls below the level that prevents initiation of DNA synthesis. If only a fraction of RNA is responsible for this mechanism it is likely to be mRNA because of its turnover rate. The inhibiting effect may be established via nucleotide pools. When RNA synthesis in relation to the cell size and overall nucleotide metabolism is low, pool size may increase to a level that is inducing DNA replication.

The negative control of DNA replication by low RNA to protein ratios fulfills all requirements for the model of regulation by an unstable inhibitor as described by Fantes et al. (1975). The amount of overall RNA and therefore most likely also of certain RNA fractions remains constant throughout G₁. Its concentration decreases reaching a lower threshold at the beginning of S. The difference between the present case and their model is, that the threshold reaction is not the initiation of mitosis but that of replication. Replication in turn causes a rise in concentration so that the cycle starts again at this point. Since it is known that is the most variable part of the cell cycle, the timing of the beginning of S also determines the length of the cell cycle.