Planta

Planta (1992)188:551 558

9 Springer-Verlag1992

Zygote formation in the homothallic green alga Chlamydomonas monoica Strehlow H. van den Ende*, M.L. van den Briel, R. Lingeman, P. van der Gulik, and T. Munuik

Department of Molecular Cell Biology,Department of Fundamental and Applied Ecology, University of Amsterdam, Kruislaan 318, 1098 SM Amsterdam, The Netherlands Received 19 February; accepted 17 July 1992 Abstract. Mating between cells of opposite mating type

within a clonal population of Chlamydomonas monoica results in thick-walled zygotes. Zygote formation was studied in cells from continuous cultures fed with culture medium containing nitrate concentrations sufficient or limiting for growth. The factors that were considered were cell density and nitrate content of the medium. The following results were obtained: (i) Zygotes were only formed by cells that had experienced a relatively low nitrogen level that did not limit cell division. (ii) Cells were competent to mate only during a limited period of time after their release from the mother cell wall. (iii) There was a correlation between zygote yields and the number of low-nitrogen cells that were able to execute a cell division under the conditions being tested. (iv) The zygote yield per cell division was independent of the cell density. These findings indicate that the strategy used by C. monoica ceils to find a mate is not dependent on random encounters. A possible explanation is that at least a large proportion of zygotes is formed by matings between cells originating from the same mother cell (siblings). Key words: Autogamy - Chlamydomonas- Gametogene-

sis - Mating - Sexual interaction

Introduction

Among the species of the unicellular, biflagellate, green alga Chlamydomonas, two mating systems can be distinguished: heterothallism and homothallism. In heterothallic species, clonal cells have one and the same mating type, either mating type plus or mating type minus (mt § or mt-). Mating type is genetically stable and inherits as one mendelian locus (Matagne 1987). In homothallic species, a single haploid cell can give rise to oppo* To whomcorrespondenceshouldbe addressed; FAX: 31 (20) 525 7715

site mating types during vegetative growth and-or gametogenesis (VanWinkle-Swift and Hahn 1986). In the heterothallic C. reinhardtii, zygote formation is the result of sexual interaction between cells which have been exposed to nitrogen stress and have differentiated into gametes (Martin and Goodenough 1975; Treier et al. 1989). Gametes of opposite mating type adhere specifically to each other by their flagella. Sexual adhesion induces intracellular signals that evoke responses necessary for fusion. The most conspicuous responses are the release of the cell wall and the activation of a mating structure, a special zone of the protoplast surface between the flagella, by which each gamete fuses with its partner (Snell 1990). When the conditions are optimal, zygote yields approaching 100 % can be obtained. At high cell densities, flagellar adhesion results in mass agglutination of the cells, involving hundreds of cells of different mating type clumping together. Apparently, it is a very effective strategy to stimulate the frequency and intensity of physical contact between potential partners. Similar sexual interactions are observed in the heterothallic C. eugametos. Particularly in this species it has been observed that cells are mating competent in the early part of the G1 phase of the cell cycle, up to the commitment point for cell division (Tomson et al. 1985; Zachleder et al. 1991). The homothallic C. monoica has been investigated genetically to large extent (VanWinkle-Swift and Thuerauf 1991), but the sexual process has not been subjected to extensive analysis. In this species, zygote formation is triggered by nitrogen deprival, but mass agglutination of gametes does not occur and maximal zygote yields are generally in the order of magnitude of 25 % of the total cell population (VanWinkle-Swift and Aubert 1982). We have set out to analyze zygote production in this species in order to assess to what extent the mating process differs from that of heterothallic species. Eventually, we hope to learn more about mating-type differentiation, in order to be able to determine whether it is strictly controlled as in yeasts (Klar 1987; Herskowitz 1988), or is subject to environmental variability.

552

Materials and methods Cultures and media. Chlamydomonas monoica Strehlow (UTEX 220) and Chtamydomonas euoametos Moewus. (UTEX 9 and UTEX 10) were obtained from the Collection of Algae at the University of Texas, Austin, Tex., USA. Chlamydomonas monoiea was cultivated in Bold's Basal medium (NaNO3, variable, but maximally 3 mM; CaC12-2H20, 0.17 mM; MgSO4-7H20, 0.3 mM; NaC1, 0.42 mM; KH2PO4, 1.28 raM; K2HPO4, 0.54 mM; FeEDTA, 0.07 mM; micronutrients (Wiese 1965), 1 ml" 1-1). Chlamydomonas eugametos was cultivated as described by Tomson et al. (1986). Continuous cultures. Continuous cultures were grown in 1-1 vessels containing 600 ml Bold's Basal Medium, aerated with air, and illuminated with a day-light tamp (HPI/T, 400 W; Philips, Eindhoven, The Netherlands; fluence rate 300 lmaol photons- m -2 - s -1) under a regime of 12 h light and 12 h dark, or, where indicated, 18 h light and 6 h dark. The cell densities were maintained at approx. 2-5 9 10s cells 9ml- 1 When the cultures were grown in medium in which the NaNO3 content exceeded 0.3 mM, the growth rate was not limited by the nitrogen supply. When the nitrogen content in the medium used was below this value, cell proliferation was limited by the supply of medium. In such cultures, a growth rate of 0.06 h - 1 was maintained by adjusting the flow of the medium. It was considered advantageous to use continuous cultures because they were a source of cells with a reproducible history, being either nitrogen-sufficient or nitrogen-limited. Induction and assay of zygote formation. Zygote formation was studied in samples taken from continuous cultures, 1-3 h after the onset of the light period, unless specified otherwise. Cell density of the samples was adjusted to 1-1.5- 105 cells, m l - i using cell-free supernatant. Large samples (10 ml) were incubated in 100-mt Erlenmeyer flasks or plastic Petri dishes (9 cm diameter) without agitation. After a specified period, subsamples were taken and fixed with glutaraldehyde in water (1.25 % final concentration), after which the cells and zygotes were counted using a hemocytometer. Small samples (0.3-1.0 ml) were incubated in 1-ml wells of plastic tissue culture clusters (Costar, Cambridge, Mass., USA; 11.3 mm well diameter, 48 wells per cluster). These were placed in humidified

H. van den Ende et al. : Zygote formation in Chtamydomonas monoica boxes at 20~ C, with continuous illumination (100 lamol'photons 9m - 2. s- 1) and without agitation. After a specified period, the content of each well was thoroughly rinsed into a preweighed test tube. After fixing the cells with glutaraldehyde (1.25% final concentration), the tubes were reweighed to determine the volume of the cell suspension and the cell density was determined by using a hemocytometer. The cells were then spun down in a table centrifuge and taken up into 0.2 ml Lugol solution (2 g KI and 1 g I2 in 100 ml water). Zygotes could then be distinguished from other cells by their clear unstained secondary walls. Their abundance was determined using a hemocytometer. The advantage of incubating small samples in tissue-culture wells was that each sample could be harvested completely, which decreased the variability of the results compared with subsampling from a large volume.

Results I n initial experiments with C. monoica, gametogenesis was i n d u c e d as described b y V a n W i n k l e - S w i f t a n d B a u e r (1982) by t r a n s f e r r i n g l o g a r i t h m i c - p h a s e cells f r o m agitated liquid b a t c h cultures in Bold's Basal m e d i u m (BB m e d i u m ) to the same m e d i u m with reduced N O 3 c o n t e n t (0.6 m M ) , i n which the cells were c o n t i n u o u s l y illumin a t e d with 100 g m o l p h o t o n s 9m - 1 . s-1 white light at 18-20 ~ C w i t h o u t agitation. M a t i n g c o m p e t e n c e was assessed by d e t e r m i n i n g the yield o f thick-walled zygotes f o r m e d after different periods o f time. I n cultures with reduced N O ~ c o n t e n t , small n u m b e r s o f c o n j u g a t i n g pairs o f cells were observed m o v i n g a b o u t vigorously. T h e y were c o n n e c t e d b y their flagellar tips, as has been described for C. eugametos a n d C. reinhardtii, a n d it was frequently observed t h a t the flagella were b e n t a r o u n d o n e o f the i n t e r a c t i n g cell bodies. I n C. euoametos it has been s h o w n t h a t the cell b o d y a r o u n d which the flagella are b e n t is the m t - p a r t n e r , so b y a n a l o g y , we

Fig. 1A, B. Mature zygotes of Chlamydomonas monoica. A Normal zygote, showing the transparent secondary wall (arrows). B Dumb bell-shaped zygote. • 2500 bar= 10 lam

553

H. van den Ende et al. : Zygote formation in Chlamydomonas monoica

assume that this is also the case in C. monoica (Musgrave et al. 1985; VanWinkle-Swift and Aubert 1983). Interacting cells produced vis-/t-vis pairs, in which two mating cells were connected by a plasma bridge. In forming vis-~t-~eispairs, C. monoica resembles C. eugametos. However, vis-A-vis pairs were only rarely seen. This indicates that they have a short lifetime and soon fuse to form a single-celled zygote, which is hard to distinguish from non-fused cells until it has formed a thick secondary cell wall. Partially fused cells were also observed, which had either discarded their walls and were fusing laterally to produce round zygotes (Fig. 1A) or in which the cell wall was retained, and then the protoplast of one partner had partially moved into the cell wall of the other partner, giving rise to dumb-bell-shaped figures (Fig. 1B). Frequently, the secondary cell wall was deposited before the cells had rounded off completely, so that dumb-bellshaped zygotes were produced. Similar results were obtained when samples were taken from continuous cultures fed with media of different nitrate contents (0.6-3 mM), in which cells were growing at maximal rate. When such samples were incubated without agitation with continuous illumination, the maximal growth rate was maintained for a period which was dependent on the original NO3 content of the medium. An example is shown in Fig. 2. At the transition of the logarithmic to the stationary phase, sexual conjugation between cells was observed, although the exact timing was hard to assess. It resulted in a steep increase of thick-walled zygotes 1-2 d later. From the types of experiment, described below (Figs. 3, 4), it was established that the minimal period of zygote maturation is 31 h (data not shown). No additional zygotes were formed after prolonged incubation of the samples. Apparently, at the end of the growth phase the cells experi-

enced a nutrient deficiency by which they became mating competent and produced zygotes in a relatively short period. In contrast, when the samples were taken from a continuous culture that was nitrogen-limited (which occurred when the NO3 concentration of the culture medium was 0.3 mM or lower), they exhibited hardly, if any, cell proliferation (Fig. 2). In such samples, very few zygotes were formed, even though one would expect that they contained cells that experienced nitrogen stress. When samples derived from an N-limited continuous culture, and thus containing non-dividing cells, were supplemented with NO3, growth was resumed immediately and this was accompanied by mating, as was evident from the appearance of zygotes. The more NO~was added, the more zygotes were formed (Fig. 3). When a second supplement of NO~ was added after the first one had been exhausted and the growth rate had declined again, a second wave of cell divisions and zygote formation was observed (not shown). These results show that mating competence is induced by nitrogen deficiency, but that effective mating occurs only in conditions allowing cell proliferation, implying the presence of sufficient nitrogen for at least one cell division. These two requirements led to a more-detailed investigation of zygote formation in an NO~-limited continuous culture, since such a culture exhibits steady-state growth of cells which might be sufficiently N-deficient to become mating competent. In such cultures zygotes were indeed found, but in very low abundances. There are two explanations for this. In the first place, zygotes that are formed are washed out at a rate which is determined by the turnover of the culture medium. In the second place, the strong agitation of the culture inhibits pairing of the cells. It was found that zygote formation in samples from N-sufficient cultures (see Fig. 2) was strongly inhibited

3000

r

,

2000

O v-

12

a

a a.

8 (J 1000

4

b i

0

I

2

,

I

4

i

I

,

6

Time (days) F i g . 2A, B. Zygote formation in samples taken from continuous

cultures of C. monoica, fed with BB medium containing 0.9 m M (a) or 0.3 m M NOg- (b). A number of samples (0.5 ml) were taken at one time, 7 h after the onset of the light period. They were incubated

O_

-

_

2

v

4

6

Time (days)

without agitation at 100 ~tmol p h o t o n s , m -2 9s 1 irradiance. At different times, the samples were fixed with glutaraldehyde (1.25% final concentration) and examined for cell (A) and zygote (B) content. The means of replicate values (n = 2) are represented

554

H. van den Ende et al. : Zygote formation in Chlamydomonas monoica

300

40001 A

3000

[

~

c 2OO

~E=.2000

100

1000 ~

;

8

Time (days)

0~.

0

-

•

2

-

4

6

8

Time ( days )

Fig. 3A, B. Zygote formation in samples taken from a continuous culture of C. monoiea after addition of various amounts of NO~, The culture was fed with BB medium containing 0.3 mM NO~, and growth was N-limited. A number of samples (10 ml) were taken at one time, approx. 2 h after the onset of the light period, and incubated as described under Fig. 2. After 2.5 d (arrow), varying

amounts of a solution of NaNO3 were added to some of the samples: a, 0.6 raM; b, 1.2 raM; c, 1.8 mM final concentration. At different times, the samples were fixed with glutaraldehyde (1.25% final concentration) and examined for cell (A) and zygote (B) content. The means of replicate values (n= 2) and their range are represented

by shaking the culture (Table 1). Nevertheless, in samples taken f r o m an N-limited culture at various times in a 24 h period, and incubated without agitation, zygotes were found after approx. 2 d, mainly in those samples taken during and shortly after the onset o f the light period. This result was optimized by adjusting the dark-light regime to 18 h light and 6 h dark, which improved the synchrony of cell division in the continuous culture. As is shown in Fig. 4, approx. 70% o f the cells divided by the end of the

dark period, and a clear o p t i m u m in zygote production was found in samples taken in the period when cell hatching took place. This result strongly indicates that nitrogen-limited cells do m a t e but only during a restricted period after their release as daughter cells. F r o m Fig. 4 it can be inferred that this period has a length of approx. 3 h. This result might also explain the steep rise in zygote production at the transition o f the logarithmic phase to stationary phase, as was shown in Fig. 2. Although zygote production in this type o f experiment was very variable, Fig. 4 was reproducible in showing an o p t i m u m in samples taken at the beginning of the light period, and a minimum in samples taken during the dark period. The occurrence o f a minimum is not fully understood. While it is envisaged that mating is a lightrequiring process (VanWinkle-Swift and Bauer 1982), the question can be asked why sporangia sampled during the dark period do not germinate and produce matingcompetent cells in the samples during the following-tight period. The maximal zygote yield in samples taken during the release of daugther cells was only 3 %. Apparently, the great majority of the cells did not take part in the mating process. To explain this, it was argued that the low yield o f zygotes in this experiment was a consequence o f the limited time interval in which cells were mating competent and had to find a partner in order to mate successfully. So, it was asked whether the mating frequency, and thus the zygote yield, could be enhanced by increasing the cell density, since it was expected that, in analogy to heterothallic species, the chance of cells finding a compatible partner is directly dependent on the cell density ( T o m s o n et al. 1986). To answer this question, a series o f samples were derived f r o m an N-sufficient continuous

Table 1. Zygote formation in Chlamydomonas monoica and C. eugarectos: effect of agitating the cell suspension. A cell suspension of C. monoica, containing 2 9 105 cell 9ml- 1 derived from a continuous culture, fed with BB medium containing 0.9 mM NO~, was incubated in portions of 10 ml in 100-ml Erlenmeyer flasks in identical conditions (100 gmol photons, m -2" s -1 at 20~ C). One set on flasks was agitated on a rotary shaker (80 rpm, maximal deviation 2.5 cm), the other set was not shaken. After 7 d, the cells were fixed with glutaraldehyde, and the cell and zygote densities determined as described in Materials and methods. Vis-fi-vis pair formation in C. eugametos was carried out under the same conditions using the protocol as described by Tomson et al. (1986). The means of replicate values (n = 3) and their range are represented Species

Conditions

Zygote yield (%)

C. monoica

not shaken shaken

0.8• 0.2i 0.02

C, e u g a m e t o s mt +

mr-

not shaken shaken not shaken shaken

23 43 66 69

~5 ~5 • 8 i4

H. van den Ende et al.: Zygote formation in Chlamydomonas monoica 80

555

4

1500

100

A

13

!

r

75

|

60

,r-

0 I'-

1000 A @

2~

.c= 40 C

.m

O

O O.

N

r ,,Q

5O

E O

C

r 0 B

500

25

f-

20

0

|

0

0

=

5

10

15

E

0

20

50

100

FLI.

0 150

Initial cell n u m b e r s (Co) 9 10 "3

Time (h)

Fig. 4. Zygote formation in samples taken from an N-limited continuous culture of C. monoica. The culture was fed with BB medium containing 0.3 mM NO~, thus growth was N-limited, and was subjected to a light regime of 16 h light and 8 h darkness. Samples (0,5 ml) were taken at various times; they were either fixed directly and examined for the presence of sporangia, or incubated as described under Fig. 2. After 3 d, the samples were fixed with glutaraldehyde (1.25% final concentration) and examined for cell and zygote content. The means of replicate values (n = 2) and their range are represented

Fig, 5. Effect of initial cell density on zygote formation in C. monoica. Samples (0.8 ml) were taken from a continuous culture which was fed with BB medium containing a non-limiting concentration of 0.9 mM NOff. Initial cell numbers (Co) were adjusted with cell-free supernatant and the samples were then incubated as described under Fig. 2. After 8 d, samples were fixed with glutaraldehyde (1.25% final concentration) and examined for cell and zygote content. The means of replicate values (n = 2) and their range are represented

culture (NOff concentration 0.9 raM), which had the same volume, but in which the initial cell densities were varied by diluting them with cell-free medium. The suspensions were incubated without agitation and under constant illumination, favouring zygote formation, which was expected to take place shortly after incubation (Fig. 2). After 7 d (long enough to allow all samples to attain their maximal cell densities and zygotes formed to mature) the samples were analyzed for final cell density and zygote content. The results (Fig. 5) show that zygote formation was favoured by low, rather than by high, initial cell densities. Thus, zygote formation appears not to be stimulated by increasing the chance o f cell-cell collisions. However, the favourable effect o f low cell densities might be explained by the fact that the lower the original cell density in a given sample is, the more cell divisions will occur before the maximal cell density (determined by nutrient availability) is attained. Since it was concluded above that zygote formation is correlated with the n u m b e r o f cell divisions that can occur in a given batch of cells, it is evident that at low initial cell densities m o r e zygotes will be produced. To reinforce this explanation of the data in Fig. 5, the following model is presented. Suppose that the growth o f a population o f cells, containing C(t) cells and Z(t) zygotes at time t, is continuous between t = 0 and t = e (the end of the experiment), and that the cells divide asynchronously at a rate per unit o f time equal to Ix. The p a r a m e t e r Ix m a y depend

on the nutrient conditions. Then the total n u m b e r o f daughter cells formed between t and t + dt is g-C-dr I f a constant fraction "2x" o f the daughter cells forms zygotes, the net growth of the cell population is dC = ( 1 - 2 x ) ' l x ' C - d t Consequently, the growth equation of the cell population is d C / d t = (1 - 2 x ) .Ix. C

(Eq. 1)

The n u m b e r o f new zygotes formed between t and t + dt will be dZ=

x-g.C'dt

and the growth equation o f the zygote population is d Z / d t = x - Ix" C

(Eq. 2)

Elimination of dt results in the relation dZ = [ x / ( 1 - 2 x ) ] 9 Integration over the period t = 0 to t = e with Z(0) = 0 and C(0) = Co results in the relation Zo = [x/(1 - 2x)]{C e - Co}

(Eq. 3)

The parameter "x" can be estimated using this relation. Generally, the variance of the data will depend on the units used and the numbers counted. In order to meet the

556

H. van den Ende et al. : Zygote formation in Chlamydomonas rnonoica 3

2-

1-

() N

O-

eB

-1

-2

~

-3

o

.' ....

2

!

|

3

4

5

In[(Ce-Co)/CO] Fig. 6. Effect of initial ceil density on zygote formation in C. monoica. Data of Fig. 5 replotted, using Eq. 3a in text. The average of the final cell numbers (C~ = 861 :t: 105) was used. Regression of mean values yielded a slope of 1.6. De-,~iations of the means were not taken into account

0

O

-1

N

t'~ -2

-3

9 II

~

.//-o -4

O

D

9

I

A 9

Discussion

D -

i 2

i...... 3

The data derived from Fig. 5, using Eq. 3a, appeared to correlate to a large extent with this model, as shown in Fig. 6. This supports the hypothesis that zygote formation is correlated with the number of cell divisions occurring in a given batch of cells. Another parameter which might affect zygote production is the N O ~ concentration in the sample medium. Earlier in this article it was demonstrated that cells which had experienced N deficiency produced zygotes when they were able to execute cell divisions, even while a nitrogen source was available. In an experiment similar to that that described above, but in which the samples were taken from a nitrogen-limited continuous culture (NO~- concentration 0.3 mM), both the initial cell numbers and the initial N O 3 concentration in the samples were varied. Figure 7 shows that varying the N O 3 concentration between 0.35 and 0.94 m M was had no effect on zygote formation. Only at N O ~ concentrations above 1.7 m M was the linear relationship between the number o f cell divisions and zygote production no longer apparent (data not shown). The m a j o r conclusion o f these experiments is that the relation described in Eq. 3 essentially holds under all conditions that were tested, implying that the fraction x is independent of both Co and Cr (Fig. 7). In other words, the fraction of daughter cells that mate to form zygotes, is independent of the cell density. This implies that the interaction between these daughter cells, and in the first place the frequency by which they come into physical contact with each other, is not determined by the number o f cells that is present in a given sample.

-

! 4

Inl(Ce-Co)/Co Fig. 7. Effect of initial cell density at different nitrate concentrations on zygote formation in C. monoiea. Samples were taken from a continuous culture which was fed with BB medium containing a growth-limiting concentration of 0.3 mM NO~. Initial celt numbers (Co) were adjusted with cell-free supernatant and 10 lal of a solution of NaNO3 was added to each sample (o, 0.35 mM; 9 0.47 raM; ~, 0.59 mM; m, 0.7 raM; [], 0.94 mM final concentrations). Samples were incubated as described under Fig. 2. After 8 d, they were fixed with glutaraldehyde (1.25% final concentration) and examined for cell and zygote content. The final cell numbers (Ce) were 394 ~:44, 5175: 42, 632 ~ 92, 740 9 38, 990 + 101, respectively (means • SD; n = 6). Data are plotted as in Fig. 6. Regression of mean values yielded a slope of 1.2. Deviations of the means were not taken into account requirements of regression analysis (variance independency), Eq. 3 can be rewritten in terms o f the dimensionless quantities Zo/Co and (Co-Co)/Co and subsequently log-transformed, which results in the relation In Z J C o = In [ x / ( 1 - 2 x ] + l n [(C~- Co)/Cd

(Eq. 3a)

The yield of zygotes in a suspension o f C. monoiea cells is expected to be dependent on a number o f variables. Firstly, zygotes are generated as a function o f the number of gametes that arise in a population in response to an environmental trigger. Gametes are generally considered as differentiated cells, determined to undergo sexual cell fusion. However, gametes are not necessarily mating competent. This is most clearly illustrated by C. reinhardtff: vegetative cells differentiate into gametes when derived o f nitrogen, but become mating competent only when they are illuminated (Treier et al. 1989). So the zygote yield in a group o f cells provides information a b o u t their mating competence, rather than a b o u t the degree of gametogenesis. Secondly, the formation of zygotes is dependent on the efficiency by which matingcompetent cells produce zygotes. It is well known f r o m previous work with C. eugametos that even while mating occurs between two cells in the form o f flagellar adhesion, the degree o f signalling between them m a y be insufficient to induce cell-wall hydrolysis, mating-structure activation and-or cell fusion. G a m e t e activation culminating in cell fusion comprises a variety of physiological and ultrastructural changes in the cell that depend on a prolonged flagellar contact which is intensified in time by an increased receptiveness of the flagellar surface (Demets et al. 1990). Thirdly, the formation of zygotes

H. van den Ende et al.: Zygote formation in Chlamydomonas monoica in a given population of cells is dependent on the mt+/ rot- ratio (Tomson et al. 1986). In C. monoica, it is presently impossible to assess this ratio visually or physiologically. From this study, a number of conclusions can be derived. Thefirst conclusion is that in C. monoica mating competence is triggered by a relatively low nitrate concentration in the medium. In this respect, it resembles heterothallic Chlamydomonas species, as well as many other green algae. The second conclusion is that zygotes are only generated under conditions allowing cell proliferation. In a cell culture, these two conditions occur when the population proceeds from a phase of active proliferation to a phase where cell-cycle arrest occurs due to lack of a nitrogen source. During this transition, the attainment of mating competence, the actual sexual interaction and cell fusion takes place. Since in fact zygote formation is correlated with the number of low-nitrogen cells that are able to execute a cell division under the prevailing conditions, the third conclusion is that cells are competent to mate only during a limited period of time after their release. The major evidence for this is that zygotes are mainly formed in samples taken from a synchronized culture at the time that daughter cells are released. This occurs at the onset of the light period. A similar conclusion was drawn by Armbrust et al. (1990) concerning the induction of spermatogenesis in the centric diatom Thallasiosira weisflogii. The fourth conclusion is that the fraction of daughter cells, originating from nitrogen-stressed cells, that generates zygotes is independent of the cell density. We favour the following explanation for this unexpected result: we propose that sexual interaction occurs mainly between cells arising from the same mother cell (siblings), during or directly after cell hatching. This hypothesis is supported by preliminary genetic data, obtained by analyzing crosses between two compatible strains of different genotype (K. VanWinkle-Swift, Department of Biological Sciences. Northern Arizona University, Flagstaff, Ariz., USA, personal communication). It has two important implications: (i) siblings may be of different mating type; (ii) siblings that preferably mate with each other soon after cytokinesis do not require a strategy to find their partner, because they are already in close proximity. This would also explain why mass agglutination of mating cells, as occurs in mating cell suspensions of heterothallic species, is not observed in C. monoica (Kates and Jones 1974; Martin and Goodenough 1975; Matsuda et al. 1990). The question then arises why the fraction of newly born cells per cell division that is involved in effective zygote production (the fraction "x") is so low (ranging between 0.05 and 0.1 in Fig. 6). Assuming that all cells attain mating competence after a nutrient signal (as in C. reinhardtii; Kates and Jones 1974), it is possible that the reason for this low value is that the mt+/mt - ratio among daughter cells derivatives from unity. A more apparant explanation, however, is that the efficiency of the mating process is suboptimal: two compatible cells that interact by adhesion via their flagella may have a limited chance of fusing together and producing a zygote because of a

557 relatively weak binding of their flagella. This idea is based upon the observation that compared with C. eugametos (Table 1), zygote formation is very sensitive to agitation of the cell suspension. In both C. eugametos and C. reinhardtii, it has been demonstrated that the initial weak forces between the flagella of interacting cells are strengthened by the increase of the amount and-or a conformational change of the responsible surface receptors (Demets et al. 1988, 1990; Goodenough 1989; Tomson et al. 1990). Such a process may not be operative in C. monoica. Cell proliferation and cellular differentiation are often considered to be opposing phenomena (Schuele and Evans 1991). This is certainly so for metazoan cells which are able to make decisions about cell division or cellular differentiation based mainly on environmental signals. Also in the yeast Saccharomyces cerevisiae, peptide pheromones activate a signal-transduction pathway that leads to cellular differentiation and arrest of the cell division cycle. In C. monoica, it seems that both processes converge since mating competence and cell-cycle progression occur together. In this respect, this alga resembles C. euoametos in which the mating-competent stage is also part of the G1 phase of the cell cycle (Zachleder et al. 1990). It is, however, very different from C. reinhardtii in which mating-competent cells ("gametes") clearly represent a differentiated stage showing cell-cycle arrest. It is possible to unify these varying patterns by assuming that mating competence is a more or less transient property in different species. Thus, on one hand, in cells of C. monoica low nitrogen availability activates a signal transduction pathway that by the next cell division results in daughter cells featuring mating competence. These cells, however, quickly revert to undifferentiated cells in the presence of nutrients, by which a cell-cycle block is relieved. In the absence of nutrients the cells might lose their competence while entering a stage outside the normal mitotic cycle, which is called Go in mammalian cells (Broach 1991). In C. reinhardtii, on the other hand, the gametic stage might be more stable, resulting in a longer-lived state of mating competence, but even there reversion to vegetative cells occurs in the end, especially in the presence of suitable nutrients (Martin and Goodenough 1975). We are grateful to Dr. Karen VanWinkle-Swift, Department of BiologicalSciences,Northern Arizona University,Flagstaff,Ariz., USA, who carefullyread and commentedupon the manuscript. References

Armbrust, E.V., Chisholm, S.W., Olson, R.J. (1990) Role of light and the cell cycle on the induction of spermatogenesis in a centric diatom. J. Phycol. 26, 470~78 Broach, J.R. (1991) RAS genes in Saccharomyces eerevisiae: signal transduction in search of a pathway. Trends Genet. 7, 28-33 Demets, R., Tomson, A.M., Stegwee, D. van den Ende, H. (1988) Cell-cell adhesion in conjugating Chlamydomonas gametes: a self-enhancingprocess. Protoplasma 145, 27-36 Demets, R., Tomson, A.M., Stegwee, D., van den Ende, H. (1990) Cell-cell coordinationin conjugatingChlamydomonas gametes. Protoplasma 155, 188-199

558 Goodenough, U.W. (1989) Cyclic AMP enhances the sexual agglutinability of Chtamydomonas flagella. J. Cell Biol. 109, 247-252 Herskowitz, I. (1988) Life cycle of the budding yeast Saccharomyces cerevisiae. Microbiol. Rev. 52, 536-553 Kates, J.R., Jones, R.F. (1974) The control of gametic differentiation in liquid cultures of Chlamydomonas. J. Cell. Comp. Physiol. 63, 157-164 Klar, AJ.S. (1987). Differential parental DNA strands confer developmental asymmetry on daughter cells in fission yeast. Nature 326, 566-470 Martin, N.C., Goodenough, U.W. (1975) Gametic differentiation in Chlamydomonas reinhardtii. J. Cell Biol. 67, 587-605 Matagne, R.F. (1987) Chloroplast gene transmission in Chlamydomonas reinhardtii. A model for its control by the mating locus. Curr. genet. 12, 251-256 Matsuda, Y., Saito, T., Koseki, M., Shimada, T. (1990) The Chlamydomonas non-synchronous and synchronous gametogenesis as analyzed by the activities of cell body-agglutinin and cell wall lytic enzyme. Plant Physiol. (Life Sci. Adv.) 9, 1-6 Musgrave, A., de Wildt, P., Schuring, F., Crabbendam, K, van den Ende, H. (1985) Sexual agglutination in Chlamydomonas eugametos before and after fusion. Planta 166, 234-243 Schuele, R., Evans, R.M. (1991) Cross-coupling of signal transduction pathways: zinc finger meets leucine zipper. Trends Genet. 7, 377-381 Snell, W.J. (1990) Adhesion and signaling during fertilization in multicellular and unicellular organisms. Curr. Opin. Ceil Biol. 2, 821-832 Tomson, A.M., Demets, R., Bakker, N.P.M., Stegwee, D., van den Ende, H. (t985) Gametogenesis in liquid cultures of Chlamydomonas eugametos. J. Gen. Microbiol. 312, 1553-1560

H. van den Ende et al. : Zygote formation in Chlamydomonas monoica Tomson A.M., Demets, R., Sigon C.A.M., Stegwee, D., van den Ende, H. (1986) Cellular interactions during the mating process in Chlamydomonas euoametos. Plant Physiol. 81, 522-526 Tomson, A.M., Demets, R., Musgrave, A., Kooijman, R., Stegwee, D., van den Eude, H. (1990) Contact activation in Chlamydomonas gametes by increased binding capacity of sexual agglutination. J. Ceil Sci. 95, 293-301 Treier, U., Fuchs, S., Weber, M., Wakarchuk, W.W., Beck, C.F. (1989) Gametic differentiation in Chlamydomonas reinhardtii. Arch. Microbiol. 152, 572-577 VanWinkle-Swift, K.P., Aubert, B. (1983) Uniparental inheritance in a homothallic alga. Nature 303, 167-169 VanWinkle-Swift, K.P., Bauer, J.C. (1982) Self-sterile and maturation-defective mutants of the homothallic alga Chtamydornonas monoica (Chlorophyceae). J. Phycol. 18, 312-317 VanWinkle-Swift, K.P., Hahn, J.H. (1986) The search for matingtype limited genes in the homothallic alga Chlamydomonas monoica. Genetics 113, 601-19 VanWinkle-Swift, K.P., Thuerauf, DJ. (1991) The unusual sexual preferences of a Chlamydomonas mutant may provide insight into mating-type evolution. Genetics 127, 103-115 Wiese, L. (1965) On sexual agglutination and mating substances (gamones) in isogamous heterothallic Chlamydomonads. I. Evidence of the identity of the gamones with the surface components responsible for sexual flagellar contact. J. Phycol. 1, 46-54 Zachleder, V., Jakobs, M., van den Ende, H. (1991) Relationship between gametic differentiation and the cell cycle in the green alga Chlamydomonas eugametos. J. Gen. Microbiol. 137, 1333-1339