Curr Genet (1991)20:239 243 Current Genetics �9 Springer-Verlag 1991

The fate of mitoehondrial DNAs of mt + and m t - origin in gametes and zygotes of Chlamydomonas

M.-C. Beckers 1, C. Munaut 1, A. Minet 2, and R. E Matagne 1

1 G6n6tique des Microorganismes, D6partement de Botanique (B22) et 2 Biochimie G+n6rale et Compar6e, Universit~ de Liege, Sart-Tilman, B-4000 Li6ge, Belgique

Received December 20, 1990/March 20, 1991

Summary. In order to study the mechanism responsible for the uniparental transmission of the mitochondrial ge- nome in crosses between Chlamydomonas reinhardtii and C. smithii, we have analyzed the fate of mitochondrial DNA during gametogenesis, zygospore differentiation and sporulation by hybridization experiments. Both mt§ and rot- gametes contain the same amount of mitochon- drial DNA and the two parental genomes persist for sev- eral days in the zygotes. The DNA of mr § origin is slowly eliminated during the period of zygote maturation. Light is required for total elimination of mt+ mitochondrial D N A in the zygospores. Using appropriate restriction enzymes, we have been unable to detect methylation of the mitochondrial D N A during gametogenesis or zygospore formation. The possibility that the mt+ mito- chondria themselves are specifically eliminated in the course of zygote maturat ion is discussed.

Key words: Mitochondrial D N A Uniparental inheri- tance - Chlamydomonas

Introduction

The mitochondrial genomes of the two interfertile species C. reinhardtii and C. smithii are made of 15.8 and 16.8 kb linear DNA molecules, respectively, present in 40-50 copies per cell (Boynton et al. 1987), The two DNAs are colinear with the exception of a 1 075 bp group I intron (~ intron) located in the cytochrome b gene of C. smithii (Colleaux et al. 1990). C. smithii also possesses an addi- tional HpaI restriction site located in the COXI gene (Fig. 1). These physical differences have been used to demonstrate that, in crosses, the mitochondriat DNA is most often inherited from the mating-type minus ( r o t ) or paternal parent only (Boynton et al. 1987; Matagne et al. 1988), in contrast to the chloroplast genome which is inherited from the mt+ parent.

Offprint requests to." R. E Matagne

Very little is known about the mechanism responsible for the uniparental transmission of the mitochondrial D N A in Chlamydomonas. The UV irradiation of mt+ or mt - gametes prior to mating, the mat-3 nuclear muta- tion, and the ploidy of the mating partners - factors which all influence the pattern of chloroplast D NA trans- mission dramatically - do not affect the pattern of mito- chondrial D N A inheritance (Boynton et al. 1987; Gill- ham et al. 1987; Matagne et al. 1988).

The mode of mitochondrial D N A transmission can, however, be influenced by the nature of the organellar genome. In crosses between wild-type and mutant cells in which the mitochondrial genome contains a deletion, transmission is almost strictly paternal when the wild parent is rot- whereas, in the reciprocal cross, the mater- nal transmission of mitochondrial D N A (i.e., the trans- mission of the wild-type genome) occurs more frequently (Matagne et al. 1989; Dorthu and Matagne, unpublished results). Differences in mitochondrial D N A inheritance patterns in reciprocal crosses have also been reported for crosses involving acriflavin-treated parental strains (Gill- ham et al. 1987; Alexander et al. 1974).

We have investigated the fate of mitochondrial DNAs of mt + and mr- origin during zygospore maturation and germination and find that the mt+ mitochondrial DNA is slowly eliminated during the maturat ion process, where- as the rot- D N A is conserved. We further demonstrate that light plays a role in mitochondrial D N A elimination and that the mitochondrial genome of the mt+ parent is not methylated during gametic differentiation or zygote formation.

Materials and methods

Strains and culture conditions. The wild-type strains mating-type plus (mt +) and minus (mt-) of C. re inhardtii are derived from strain 137c. The CC 1373 is the wild-type mt+ strain of C. smithii. The CC 1852 mr- strain (called hereafter C. smithii rot-) possesses the mito- chondrial DNA of C. smithii; it was isolated from a cross between C. reinhardtii rot- and C. smithii mt+ (Boynton et al. 1987). The algae were routinely grown on TAP agar medium (Gorman and

240

Levine 1965) under cool white fluorescent light (90 gE x m 2 x s -1) at 25~ for 10-15 days.

Gametogenesis and conditions of zygote maturation. Gametogenesis was induced by transferring vegetative cells to 100 ml of nitrogen free (N-free) liquid minimal medium (Surzicki 1971) at a density of 3-5 • 10 6 cells/ml. The cells were maintained under light with con- tinuous agitation for 18-20h. Gametes of opposite mating-types were mixed and transferred to test tubes I (approximately 6 ml/tube). Three to 5 h after mixing, pellicles of zygotes appeared at the surface of the liquid. The zygotes were maintained in the light (VanWinkle- Swift 1977) or transferred after 1 day to the dark. The maturation of zygotes was performed in liquid medium or on agar plates. In the latter case, the zygote pellicles were carefully transferred from test tubes to a cellophane membrane covering the agar (TAP or N-free medium) 6 h after mating. This procedure facilitates subsequent harvesting of the zygospores.

Zygote isolation and purification. The zygote pellicles, collected from liquid medium or from cellophane membranes, were suspend- ed in distilled water and treated with ultrasound (MSE sonicator, 20 sec) to destroy selectively the contaminating unmated cells. The zygotes were purified by four successive centrifugations (800g) and washings. The pellets (containing about 10 s zygotes) were stored at - 2 0 ~ until DNA isolation.

Induction of meiosis and sporulation. At the end of the maturation period, the zygotes were purified as described above and spread over a cellophane membrane covering a fresh agar plate (TAP medi- um). The plates were incubated in the light. After 2-3 days, the zygotes matured on agar gave rise to zoospores (meiotic products) which were collected by centrifugation and stored at - 2 0 ~

COB COX1

I I I C.r.( 15 .8 kb)

H S s H

l I ] I C . S . ( 1 6 , 8 kb) (~ H

t

R1

Fig. 1. Simplified maps of C. reinhardtii and C. smithii mitochon- drial genomes indicating the positions of the HpaI (H) and SstI (Ss) restriction sites. The two maps are colinear except for the 1 kb c~ intron present in the cytochrome b (COB) gene of C. smithii. An additional HpaI site is also present in the COX1 gene of C. smithii. R1 represents the 5.5 kb DNA fragment present in the pULG-R1 probe used in the hybridization experiments

Isolation of total DNA. The pellets of zygotes or zoospores (+ i 0 a) were suspended in 300 gl NET-SDS buffer and extracted as previ- ously described (Matagne et al. 1988). For zygotes, a further purifi- cation of the DNA was generally required. After the second ethanol precipitation, the nucleic acids were dissolved in 90 gl TE buffer (10 mM Tris-HC1, 1 mM EDTA, pH 8.0) then added with 30 gl 5 M NaC1 and 30 gl 10% CTAB (cetyltrimethylammonium bromide). After centrifugation, the pellet (containing polysaccharides) was discarded. The nucleic acids were recovered by ethanol precipitation and finally dissolved in 20-30 gl TE.

Hybridization analyses. The total DNA was digested with appropri- ate endonucleases and the resulting restriction fragments were sep- arated by electrophoresis on 0.7% agarose gel. In one experiment (Fig. 3) the total DNA was first digested by HpaI, precipitated by ethanol then dissolved and digested by a second enzyme in the appropriate buffer. After Southern blotting on nitrocellulose mem- branes, the mitochondrial DNA fragments were detected by using the pULG-R1 biotinylated probe as previously described (Matagne et al. 1988). Figure 1 gives the positions of HpaI and SstI restriction sites in the mitochondrial genome of C. reinhardtii and C_ smithii as well as the position of the 5.5 kb R1 fragment present in the probe (Remacle et al. 1990). To quantify the DNA band intensities, the membranes were scanned with an Ultroscan LKB 2202 densitome- ter. We verified that, in the range used, the measured intensities were proportional to the DNA amounts.

R e s u l t s

The f a t e o f mitochondrial D N A during zygote maturation

Zygotes ob ta ined in the cross C. reinhardtii m t + x C. smithi i m r - were m a i n t a i n e d in N-free l iquid m e d i u m under light for 1 day and were then transferred to dark- ness to induce ma tu ra t ion . The mi tochondr ia l D N A was analyzed at different stages of the zygote m a t u r a t i o n pro- cess. Figure 2 shows the results ob ta ined after hybridiza- t ion of the HpaI restr ict ion f ragments with the p U L G - R 1 probe. The 6.35 kb f ragment typical o f C. smithi i ( m t - or pa te rna l parent) was present at all stages of m a t u r a t i o n whereas the 7.7 kb f ragment of C. reinhardtii (mt § or ma te rna l parent) progressively disappeared f rom the 3rd d a y a f t e r mat ing. However, the e l imina t ion of the mt § mitochondr ia l D N A was no t complete, even after 38 days of i ncuba t i on (Fig. 2, lanes 1 - 4 ; Table 1).

In the reciprocal cross C. smithii m t + x C. reinhardtii m t - , the D N A of mt § origin was also progressively elim-

Fig. 2. Hybridization of the R 1 probe with HpaI re- stricted total DNA extracted from zygotes or gametes. Lanes 1-6, cross C.r.mt + x C.s.mt-; lanes 7-10, cross C.r.mt- x C.s.mt +. Lanes 1-4, DNA of zygotes incubat- ed for 1 day in the light followed by 0, 2, 4 and 37 days in the dark, respectively; lane 5, DNA of zygotes incu- bated for 1 day in the light, 4 days in the dark and i day in the light; lane 6, DNA of zoospores obtained 2 days after transfer to light from zygotes matured on agar plates (1 day in the light then 4 days in the dark); lanes 7-9, DNA of zygotes incubated in N-free liquid for I day, 3 days and 5 days in the light, respectively; lane lO, DNA of mt+ and rot- gametes, maintained separate- ly in N-free liquid for 5 days in the light then mixed in equal amounts just before DNA extraction

241

Table 1. Amount of mt + mitochondrial DNA (expressed as a per- centage of the amount of mt mitochondrial DNA) present in the zygotes obtained in the cross C. reinhardtii mt + x C. smithii mU. The zygotes were matured either in liquid or on solid-agar medium (see text). The amounts of DNA were measured by densitometry performed on the 7.7 and 6.35 kb fragments, respectively

No. days after mating Liquid medium Solid agar medium

1 110 104 2 105 - 3 64 50 4 56 - 5 28 20

38 17 -

inated whereas the mr- DNA was conserved. This elimi- nation, however, began as early as I day after mating (data not shown).

The analysis of the SstI fragments which differ by the presence (C.s.) or the absence (C.r.) of the ~ intron led to the same conclusion (data not shown).

In zygotes maintained in liquid medium, cytoplasmic and chloroplast de-differentiations are partially inhibited and very few zygotes are able to germinate (Cavalier- Smith 1976). We thus followed the fate of mt + and m t - mitochondrial DNAs in zygotes matured on solid-agar medium or in liquid medium. No major difference was observed between both types of zygote (Table 1).

Zygotes matured in liquid medium or on agar plates were transferred from darkness to light, 5 days after mat- ing. The day following the transfer, the zygotes had not germinated and were still resistant to ultrasound. Howev- er, they had lost their mt + mitochondrial DNA (Fig. 2, lane 5). The zygotes matured on agar plates germinated 2-3 days after the transfer. Again, only the mitochondri- al DNA of m t - origin could be detected in the haploid spores (Fig. 2, lane 6). When the dark period was reduced to 1 or 2 days, the complete elimination of the mt+ DNA was also observed the day following the transfer to light, despite the high amount of that DNA just before the transfer (data not shown).

The fate of mitochondrial DNAs was also followed in zygotes maintained in N-free medium (liquid or solid) under continuous light. Again, the amount of maternal DNA gradually decreased with time. It was entirely elim- inated after 5 days (Fig. 1, lanes 7-9). In contrast, no elimination of mitochondrial DNA was observed in gametes maintained in N-free liquid medium for the same period (Fig. 2, lane 10).

Absence of methylation o f mitochondrial DNA

In Chlarnydomonas, the chloroplast DNA of mt + origin is extensively methylated during gametic differentiation and early zygote formation, whereas the methylation of the corresponding m t - chloroplast DNA is much less pronounced (Royer and Sager 1979; Sager et al, 1981; Feng and Chiang 1984). Whether this methylation is re- sponsible for the protection of the mt+ chloroplast DNA and its preferential transmission to the meiotic progeny

remain as unresolved questions (Feng and Chiang 1984). Assuming that the mitochondrial genome is also methy- lated in the mt + gametes, it could be proposed that such a modification hinders DNA replication, resulting in a slow elimination of the mt + mitochondrial genome (for a review on the role of methylation, see Ehrlich and Wang 1981).

The possible methylation of mitochondrial DNA was investigated in gametes and in 4 day-old zygotes by using restriction enzymes either sensitive or insensitive to methylation. We compared the restriction patterns ob- tained after cleavage with HpaII (cleaving CCGG sites only) and MspI (cleaving both CCGG and CCGG), with MboI (GATC only) and DpnI (GATC only).

The fragments obtained after digestion with HpaI alone or HpaI combined with HpaII or MspI indicate that both in gametes (mt + or rot-) and zygotes, the inner cytosine of the CCGG restriction site is unmethylated (Fig. 3, lanes 1-3) (In zygotes, the small size restriction fragments of maternal origin were poorly visible, proba- bly because of loss of DNA material in double-digestion experiments: Fig. 3, lanes 2-3.)

The double-digestion by HpaI and DpnI gave the same pattern as after the action of HpaI alone. In contrast, digestion with HpaI and MboI gave smaller fragments (Fig. 3, lanes 4 and 5) both in gametes and in zygotes. This indicates that the adenine of GATC is not methylat- ed.

D i s c u s s i o n

In Chlamydomonas, only the mitochondrial genome of m t - (paternal) origin is transmitted to the meiotic proge- ny. This uniparental paternal transmission does not re- sult from the specific elimination of the mt + genome dur- ing gametic differentiation since, at the time of mating and in the young zygotes, both mt + and rot- mitochon- drial DNAs are present in equal amounts. Rather, it re- sults from a slow elimination of the maternal DNA oc- curring during the maturation of the zygote. Although there are some structural and physiological differences between zygotes matured in liquid medium or on solid agar plates (Cavalier-Smith 1976), the same kinetics of elimination is observed in both cases.

When the zygotes are matured by incubation in the dark, the degradation of mt + DNA is not complete. An apparently total loss of that DNA is observed only when the zygotes have been transferred to light. By contrast, when the zygotes are matured under light in nitrogen-free medium, maternal mitochondrial DNA is undetectable after 5 days. Mitochondrial DNA elimination represents a response to a light signal unrelated to the induction of meiosis and sporulation since the zygotes cannot germi- nate on a medium deprived of nitrogen. Moreover, the elimination is specific to zygotes since in gametes (mt + or m t - ) incubated for 5 days under the same conditions (N-free medium, continuous light), no loss of mitochon- drial DNA is detected.

Although the technique of hybridization does not de- tect any maternal mitochondrial DNA in mature zygotes

242

Fig. 3A-C. Hybridization of the R1 probe with DNA from C.s.mt- gametes (A), C.r.mt + gametes (B) and 4 day-old zygotes (C) after cleavage with: (1) HpaI, (2) HpaI and HpaII, (3) HpaI and MspI, (4) Hpal and DpnI and (5) HpaI and MboI

and meiotic zoospores, the DNA elimination may not be complete: previous experiments have shown that in cross- es involving a mt - parent which carries a deletion in the mitochondrial genome, a few zygotes transmit the ge- nome of mt + origin (Matagne et al. 1989).

In the rare zygotes which divide mitotically to give stable diploid clones, the ~ intron typical of C. smithii spreads to all C. reinhardtii mitochondrial DNA mole- cules by a conversion-like mechanism common to group I introns (Boynton et al. 1987; Remacle et al. 1990; Col- leaux et al. 1990). We can conclude that this unidirection- al conversion does not take place in meiotic zygotes be- cause the intron-less SstI fragments originating from the C. reinhardtii m t - parent persist throughout the period of zygote maturation. We have now shown that mitochon- drial DNA of both parental origins coexist in the matu- rating zygote for several days. This suggests that the en- zymes required for the conversion mechanism are absent, or inactive, in meiotic zygotes, or else that the fusion of mitochondria (to allow physical contacts between dis- tinctive parental mitochondrial DNA molecules) does not take place in meiotic zygotes. However, mitochondri- al fusions must occur occasionally to account for the rare recombinant mitochondrial DNAs observed in a cross between C. smithii haploid and C. reinhardtii diploid strains (Matagne et al. 1988).

The mechanisms contributing to the selective elimina- tion of paternal chloroplast DNA and of maternal mito- chondrial DNA in meiotic zygotes must be different. Ma- ternal chloroplast inheritance results from the elimina- tion of mt - chloroplast DNA within the first hours fol- lowing zygote formation (Kuroiwa et al. 1982; Munaut et al. 1990). This elimination has been explained by the action of a specific nuclease (Kuroiwa 1985). A similar explanation is probably not valid for the mitochondrial genome because the mt+ mitochondrial DNA elimina- tion during zygote maturation is slow and progressive. Moreover, the methylation of mt + chloroplast DNA ob-

served during gametic differentiation and zygote forma- tion (Royer and Sager 1979) does not take place for the mitochondrial DNA. Thus, if chloroplast DNA methyla- tion serves to protect mt + plastid DNA from digestion, the protection of rot- mitochondrial DNA must occur through some other mechanism.

By ultrastructural analyses Brand and Arnold (1987) have observed that, in mature zygotes, the volume of the chondriome (=mitochondrial inventory) was half the volume found in young zygotes. Moreover, every mature zygote contained a few degenerated mitochondria. The specific loss of mt+ mitochondrial DNA during matura- tion could thus be related to the loss of the mt + mito- chondria themselves, through a mechanism which has yet to be defined. Disintegration of organelles, rather than organellar DNA degradation per se, has been offered as an alternative explanation for maternal mitochondrial inheritance in higher plants (Vaughn et al. 1980).

To our knowledge, this is the first report on the specific elimination of one parental mitochondrial ge- nome during zygote maturation. Very recently, it has been suggested that the uniparental transmission of mito- chondrial DNA in Physarum polycephalum results from an active degradation of one parental genotype which is probably completed by the second division after gamete fusion (Meland et al. 1991).

Acknowledgements. We thank J. Monfort for technical assistance and F. Hayet for typing the manuscript. This research was support- ed by the Fonds de la Recherche Fondamentale et Collective (Bel- gium), grant 2.4521.89.

References

Alexander NJ, Gillham NW, Boynton JE (1974) Molec Gen Genet 130:275 -290

Boynton JE, Harris EH, Burkhart BD, Lamerson PM, Gillham NW (1987) Proc Natl Acad Sci USA 84:2391-2395

Brand N, Arnold CG (1987) Endocyt C Res 4:275-288 Cavalier-Smith T (1976) Protoplasma 87:297-315 Colleaux L, Michel-Wolwertz M-R, Matagne RF, Dujon B (1990)

Mol Gen Genet 223:288-296 Ehrlich M, Wang RYH (1981) Science 212:1350-1357 Feng TY, Chiang KS (1984) Proc Natl Acad Sci USA 81:3438-

3442 Gillham NW, Boynton JE, Harris EH (1987) Curr Genet 12:41-47 Gorman DS, Levine RP (1965) Proc Natl Acad Sci USA 54: 1665-

1669 Kuroiwa T (1985) Microbiol Sci 2:267-270 Kuroiwa T, Kawano S, Nishibayashi S, Sato C (1982) Nature

298:481-483 Matagne RF, Rongvaux D, Loppes R (1988) Mol Gen Genet

214:257-262

243

Matagne RF, Michel-Wolwertz M-R, Munaut C, Duyckaerts C, Sluse F (1989) J Cell Biol 108:1221-1226

Meland S, Johansen S, Johansen T, Hangli K, Hangli F (1991) Curr Genet 19:55-60

Munaut C, Dombrowicz D, Matagne RF (1990) Curr Genet 18:259 - 263

Remacle C, Bovie C, Michel-Wolwertz M-R, Loppes R, Matagne RF (1990) MoI Gen Genet 223:180-184

Royer HD, Sager R (1979) Proc Natl Acad Sci USA 76:5794-5798 Sager R, Grabowy C, Sano H (1981) Cell 24:41-47 Surzycki S (1971) Methods Enzymol 23:67-73 VanWinkle-Swift KP (1977) J Phycol 13:225-231 Vaughn KC, De Bonte LR, Wilson KG (1980) Science 208:196-198

C o m m u n i c a t e d by K. P. VanWink le -Swi f t