Age-associated change in macronuclear DNA content in Paramecium Caudatum*

Since clonal ageing in the Paramecium aurelia complex (for species nomenclature, see Sonnebom, 1975) was established (Sonneborn, 1954), many age-related changes have been described and attempts to analyse the underlying mechanisms have been made (for reviews, see Siegal, 1967; Nanney, 1974; Sonneborn, 1978; Smith-Sonneborn, 1981). One facet of clonal aging is developmental events that are evolutionally acquired and thus genetically programmed (Sonneborn, 1978); another facet is random events caused by environmental hazards and thus amenable to some protection (Smith-Sonneborn, 1981).

Although age-related changes are generally progressive, they are not always simple. Some are gradual as exemplified by fission rate (Sonneborn, 1954), duration of the period (Smith-Sonneborn & Klass, 1974), or endocytic capacity (Smith-Sonneborn & Rodermel, 1976), while others are abrupt as exemplified by lethality after autogamy (Sonneborn & Schneller, 1960; Rodermel & Smith-Sonneborn, 1977). Two phenomena that do not change as a function of age are copper tolerance (Nyberg, 1978) and DNA polymerase activity (Williams & Smith-Sonneborn, 1980). Among changes which decrease with age, the amount of macronuclear DNA in Paramecium tetraurelia and Paramecium bursaria is peculiar in that it decreases during the early period of the life-cycle (Schwartz & Meister, 1973, 1975; Klass & Smith-Sonneborn, 1976) and this drop may be followed by further fluctuation (Schwartz & Meister, 1975). A decrease in macronuclear DNA content during the early period of the life-cycle is also known in Tetrahymena thermophila (Doerder & DeBault, 1978) (for species nomenclature, see Nanney & McCoy, 1976).

The present investigation addresses the question whether the initial drop in macronuclear DNA is a common phenomenon in ciliates, by the study of changes in the amount of macronuclear DNA in Paramecium caudatum. The cycle of clonal ageing in P. caudatum was recently clarified: the life-span was 527 fissions, characteristics of senescence appeared at about 350 fissions. And nearly the same change in life-cycle, as a function of fission, was reproduced in subclones after they had been kept as stock cultures for more than a year (Takagi & Yoshida, 1980). Also studied in relation to the change in macronuclear DNA content were the macronuclear ploidy level, change in distribution of macronuclear DNA to daughter cells, and change in relative volume of the macronucleus to the whole cell. The results were reported in an abstract (Takagi & Kanazawa, 1980).

Cells of known ages were available from cultures of the stock Mi8a, mating type V of P. caudatum syngen 3 (Takagi & Yoshida, 1980). A single cell was isolated from the stock culture of a given age and allowed to divide about 13 fissions to give a test-tube culture of about 14 ml. Also used were F₁ clones, denoted NK clones, from a cross between Yt3G of mating type V and St37 of mating type VI, both of which were kindly provided by Dr K. Hiwatashi of Tohoku University. These clones were cultivated in daily isolation lines and surplus cultures were used for DNA measurements and for the test of sexual immaturity. The surplus culture of a given age was allowed to grow to a test-tube culture of about 14 ml. All of the test-tube cultures were starved for 3 days to exhaust food vacuoles before using them for measurement of DNA. This procedure was useful especially for old cells in which food vacuoles were often irregularly positioned. In other experiments, when distribution of DNA to daughter cells was determined, cells with food vacuoles were used.

The culture medium was 2 % lettuce juice in 2 mM-sodium phosphate buffer solution (pH 6·8), inoculated i day before use with Klebsiella aerogenes. The temperature was about 25 °C throughout the study except for the stock cultures, which were kept at 17 °C.

A cell suspension was placed on a coverslip, hot-air dried with a drier for 10 min, and fixed with 83:1 mixture of ethanol/acetic acid for 10 min. Hydrolysis in 1 M-HCI (60 °C) for 10 min was followed by a rinse with 1 M-HCI (0°C) for 1 h since this procedure was useful for bringing about bacterial discoloration without affecting the colour of the macronucleus.

The macronuclei were stained as described by Shibatani & Naora (1952), except that the Schiff’s reagent was diluted to 1/16 concentration to obtain an optimal range of the optical density; O.D. = 0·2–0·7. Cells were stained in this solution for 1 h, then bleached in dilute sulphurous acid, rinsed in distilled water, dehydrated by passing through ethanol series and xylols, and mounted in Eukit (O. Kindler Ltd).

Microspectrophotometric determinations were made by scanning the whole macronucleus with a spot-light of 2 μm in diameter at 565 nm using the Olympus MMSP. The DNA content of the micronucleus, which usually lies in a concavity of the macronucleus, could not be measured separately from the macronuclear DNA content except for some unusual cases when both nuclei existed apart from each other. Such measurements were used to estimate the degree of macronuclear polyploidy. The total extinction value for each measurement was corrected for background.

Macronuclear DNA content was measured in 13 subclones (Table 1). Cells 127–330 fissions old showed a relative DNA content, E⁶⁶⁵_t, from 150 to 169·6; cells 367–425 fissions old, 81·7–107·8; and cells 468–507 fissions old, 34·1–44·3. A slight decrease in DNA content during two-thirds of the life cycle was followed by a sharp drop during the last one-third of the life cycle (open circles in Fig. 1). The DNA content of cells 330 fissions old was significantly different from (Z-test, P < 0·05) but still 91·3 % of that of cells 127 fissions old, and the DNA content of cells 507 fissions old was 20·1 % of that of cells 127 fissions old.

However, DNA decline might have occurred before the age of 127 fissions. This was examined with young NK clones derived from a new cross, since cells younger than 127 fissions were not available from the stock Mi8a. Among 13 newly produced clones, NK-J, NK-K, NK-L and NK-M showed shorter life spans; they died at 52, 58, 183 and 228 fissions old, respectively. The remaining 9 clones lived for more than 300 fissions. Mean macronuclear DNA contents of these 9 clones at the average age of 16, 59, 99 and 152 fissions were shown in Table 2. Although the difference in mean DNA contents among groups of different ages was statistically significant (Z-test, P < 0·05), the average value at 152 fissions was still 90·3 % of that of cells 16 fissions old. In those clones with shorter life-spans, however, the averages at the last measurements were 62·5% (NK-K), 68·2% (NK-J), 69·3% (NK-L) and 84·2% (NK-M) of that of cells 16 fissions old. Thus we conclude that, in clones with long life spans, decrease in macronuclear DNA content is also slight during the period younger than 127 fissions.

Macronuclear DNA contents in two division products were compared between cells of two different fission ages, young cells 136 fissions old and old cells 436 fissions old. In Table 3 the DNA contents of daughter cells receiving more (A) or less (B) than half the parental DNA are given, together with the percentage deviation from the hypothetical value of equal distribution calculated from (A – B)/(A + B) × 100. The percentage deviation from the equal distribution or, in other words, the mean difference between daughters was as low as 5·2% on average in young cells, while it was as high as 27·6% on average in old cells.

In order to know if the decrease in macronuclear DNA content is accompanied by a decrease in macronuclear volume, we investigated age-associated changes in the size ratio of the macronucleus to the whole cell. The volume of the cell and macronucleus was approximated by the measurement of the weight of the cut-out of picture prints. All the pictures were enlarged 500 × on printing papers of the same lot number. Morphologically normal cells were selected for the measurements, especially at old ages. In spite of the sharp drop of macronuclear DNA content in old cells, the relative volume of the macronucleus to the whole cell remained almost constant throughout the life-cycle (Fig. 2). It is noticeable that the ratio of the macronucleus to the cell was unchanged at 330 fissions when both the macronucleus and cell volume increased.

The degree of macronuclear polyploidy in P. caudatum has been estimated at 50–160C (for a review, see Raikov, 1968). This might be underestimated if the measurements were done in old cells with decreased amount of macronuclear DNA and if micronuclear ploidy did not decline with age. Thus the macronuclear ploidy was estimated from cells younger than 162 fissions old. Only 18 cells were available for the estimate in which the macronucleus and micronucleus were apart enough to be measured separately. As shown in Table 4, the degree of macronuclear polyploidy was estimated as 407C on average, ranging from 142C to 648C, providing that the micronucleus was diploid (2C).

Changes in macronuclear DNA content throughout the clonal life-cycle have been reported in P. tetaurelia (Schwartz & Meister, 1973, 1975; Klass & Smith-Sonneborn, 1976), P. bursaria (Schwartz & Meister, 1973, 1975) and T. thermophila (Doerder &

DeBault, 1978). The decrease in macronuclear DNA content during the early period of the life-cycle was a common feature of all of these diverse species. In P. bursaria, the first decrease was followed by alternate changes of rise and fall to result in the two minima of DNA content being separated by a high maximum (Schwartz & Meister, 1975). In P. tetraurelia, the period of increase was not always followed by the second decline and the increase was sometimes small (Schwartz & Meister, 1975; Klass & Smith-Sonneborn, 1976). In T. thermophila, the first drop to 72% of the initial DNA content occurred during either 0–50 fissions or 60–130 fissions after conjugation and was unchanged thereafter (Doerder & DeBault, 1978). In contrast with these reports, an initial drop in macronuclear DNA in P. caudatum was not observed. Instead, the initial content of macronuclear DNA was almost unchanged for two-thirds of the life-cycle and then decreased rapidly to 20% of the initial value during the last one-third of the life-cycle. Thus no relation was suggested between the phylogenetic position of the ciliates and their changing pattern of macronuclear DNA content, although some genetic process was suggested to be involved in the decrease in macronuclear DNA content (Doerder & DeBault, 1978).

In the present study, the DNA content of the macronucleus included the micro-nuclear DNA content. However, the contribution of micronuclear DNA content was so small, usually less than 1 % of the macronuclear DNA (Table 4), that it was regarded as negligible. A methodologically more serious problem may concern the cell cycle stage at which DNA measurements were done. There is little information available concerning the cell cycle stage of macronuclei in overstarved cells, although moderately starved cells with the mating ability of T. thermophila have been reported to have G₁ macronuclei (Wolfe, 1973; Doerder & DeBault, 1975). Our conclusion on the decreasing pattern of macronuclear DNA content, however, would not be affected unless overstarved cells at younger ages had 5 or G₂ macronuclei and those at older ages had Gj macronuclei. Such a shift in cell cycle in overstarved cell populations appears to be unrealistic, because the proportion of the decrease in both the mean DNA content of G₁ macronuclei during the period from 136 fissions 10436 fissions (Table 3) and the mean DNA content of overstarved macronuclei during the period from 127 fissions through 425 fissions (Table 1) was almost the same, i.e. 51% and 48%, respectively.

We reported previously that the present clone Mi8a began to show senescent characteristics at about 350 fissions, with an increasing appearance of morphologically abnormal cells and an increasing probability of cell death after cell division (Takagi & Yoshida, 1980). The present results indicate that a severe decrease in macronuclear DNA may be characteristic of senescence, although there are no data available concerning the change in macronuclear DNA during the first 16 fissions after conjugation. A severe loss of macronuclear DNA associated with senescence was shown also in some clones with short life-spans (Table 2).

The mechanism for loss of macronuclear DNA is not known. The decreasing amount of DNA content per cell division from 169·6 at 127 fissions to 154·8 at 330 fissions is calculated to be as little as 0·04–0·05% of the macronuclear DNA in cells of this period. On the other hand, the decreasing amount of DNA per cell division from 154·8 at 330 fissions to 34·1 at 507 fissions is 0·68 and this corresponds to 0·4–2% of the macronuclear DNA in cells of this period, which is 10–40 times larger than that in the previous period. Based on our estimate of macronuclear DNA ploidy as 407C at 162 fissions (Table 4) and supposing that the ploidy level decreases with the decrease in DNA content, the younger cells may lose their 0·16–0·18C through each cell division and older cells may lose their 1·5–1·60 through each cell division. A net decrease in macronuclear DNA after a cell division could be caused by an incomplete duplication of DNA and/or by a partial disappearance of parental DNA. The latter is in fact the case in some ciliates being exemplified by a chromatin extrusion body (Cleffman, 1968; Frenkel, 1975), whereas no such dramatic extinction has been seen in P. tetaurelia (Berger & Schmidt, 1978). But this does not mean that the former is the case in P. caudatum. The maximal loss of 2% of parental DNA through a cell division is too small to identify the alternatives experimentally, because the amount is within the range of unequal distribution of DNA to daughter cells. The mean difference in DNA content of 5·2% between daughters of young cells was almost the same as those reported in P. tetraurelia (Berger & Schmidt, 1978), strain H of T. pyriformis (McDonald, 1958), and in Protocrucia sp. (Ruthmann & Hauser, 1974), but smaller than those reported in strain HSM of Tetrahymena pyriformis (Cleffmann, 1968), T. thermophila (Doerder & DeBault, 1975), Euplotes eurystomus (Witt, 1977) and Bursaria truncatella (Ruthmann, 1964; Zech, 1966). These unequal distributions resulting from amitosis of the macronucleus would be soon spread through divisions unless there exists some regulatory mechanism working through the process of DNA synthesis (Berger, 1979; Doerder, 1979). Thus the shift of the mean difference from 5·2% to 27·6% may indicate that such a regulatory mechanism no longer functions in old cells. No increase of the standard deviation in the measurements of DNA in old clones (Fig. 1) might be due to selective elimination of such cells from the population that received less or more DNA content than a certain threshold.

The relative volume of the macronucleus to the whole cell remained almost un-changed not only at the age of 330 fissions, when both the macronucleus and the cell increased in size, but also during the subsequent period when macronuclear DNA content decreased sharply. The increase in cell size at 330 fissions appears realistic as a sign of the initiation of senescence. On the contrary, the decrease in cell size at 370 fissions might be caused by selective elimination of morphologically abnormal cells, which began to appear at about 350 fissions and were often bigger in size than normal cells, as exemplified typically by monsters (Takagi & Yoshida, 1980). In contrast with the report of Schwartz & Meister (1975), in which they implied a parallel change of the cell size with the DNA content in P. tetraurelia and a reversely parallel change of the cell size with the DNA content in P. bursaria, the present data on P. caudatum show no relation between cell size and DNA content. Instead, the cell size paralleled the macronuclear size irrespective of the DNA content. The unchanged macronucleus: cell size ratio throughout the life cycle suggests that macronuclear chromatin becomes more diffuse in aged cells. This phenomenon may have something to do with a recent discovery that cellular aging is modulated by drugs that alter chromatin structure (for a review, see Smith-Sonnebom, 1981).

We are grateful to Dr E-iti Yokomura of Nara Women’s University for his technical advice on Feulgen microspectrophotometry and his encouragement throughout this study. We also wish to thank Dr Joan Smith-Sonnebom of the University of Wyoming for her critical reading of the manuscript and helpful discussions.