Current Genetics (1982) 6:111-117

~

~

~

© Springer-Veflag 1982

Quantitative Estimations of Chloroplast DNA in Bleached Mutants of Euglena

gracilis Younis Hussein* Philippe Heizmann, Paul Nicolas, and Victor Nigon Universit6 Claude Bernard Lyon I, D~partement de Biologic Gtn&ale, 43, Boulevard du 11 novembre 1918, F-69622 Villeurbanne Cedex, France

Summary. The level of chloroplast DNA has been estimated in bleached mutants o f Euglena by the increase of the renaturation rate of a radioactive chloroplast DNA probe in response to the addition of total mutant DNA. Two classes of bleached mutants differ from each other by their level of chloroplast DNA: - a few bleached mutants contain chloroplast DNA in amounts similar to those of wild type strains; all sequences of the wild type chloroplast DNA seem to be present but in non stoechiometric proportions. - most bleached mutants have about 100 to 1,000 times less chloroplast DNA than wild type cells. In these mutants chloroplast DNA sequences form two frequency classes: - one class has about 5 to 12% of the complexity of the wild type genome; these sequences are reiterated from 30 to 160 times per cell and hybridize with wild type ribosomal cistrons. They are expressed as chloroplast ribosomal RNA in all bleached mutants analyzed so far. - the second class shows at least 40% of the complexity of the wild type genome; these sequences are present in 2 to 7 copies per cell. Key words: Chloroplast DNA - Bleached mutants Ribosomal cistrons

Introduction

Although they are irreversibly devoid of functional chloroplasts, bleached mutants of Euglena gracilis contain residual chloroplast DNA. We showed indeed that restriction digests of total DNA from bleached mutants transfered onto Southern blots give positive hybridization signals with cloned chloroplast DNA fragments (Heizmann et al. 1981, 1982). In the present work we estimated quantitatively the amounts of chloroplast DNA conserved in these bleached mutants; the increase of the renaturation rate of a chloroplast DNA probe in response to the addition of total mutant DNA was measured, allowing to calculate the numer of chloroplast DNA molecules present in each mutant cell. The same method had already been used by Rawson and Boerma (1976) and by Chelm et al. (1977) to estimate the amounts of chloroplast DNA in wild type cells: depending on the physiological state of the cuRures, 200 to 3,000 molecules per cell were estimated. The bleached mutants analyzed by these authors were not considered as having significant amounts of chloroplast DNA. However the data presented here confirm qualitative results: bleached mutants do contain chloroplast DNA. Two distinct classes of mutants differ by their content and organization of chloroplast DNA: the cpmutants have high levels of chloroplast DNA; the p mutants have very low amounts of organdie DNA (Heizmann et al. 1981).

Materials and Methods * Present address: Department of Genetics, Faculty of Agricul-

ture, Assiut, Egypt Offprint requests to: P. Heizmann

Strains and Cultures. The origin of the various strains is given in

Table 1. The cultures were grown on a butyrate organotrophie medium under fluorescent light (Freyssinet et al. 1972) and harvested in late growth phase. O172-8083/82/0006/0111/$ 01.40

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena

112

1/%SS

1t%SS/

o~1,5 t

O.Ol,Ol

Fig. 2 0.011

Fig. 1

i/ /J/// 120

240

360

Time

(mllt)

N i

h

120

240

n 360

i Time

(rain) 1/%S8

1/%SS O.030 0,01150 0.01125

0.020

0.0110 0.010

0.01075 120

240

360

Time

(rain)

Fig. 3

n

i

L

120

240

360

i

Time

(min)

Fig. 4

Figs 1, 2, 3 and 4. Renaturation kinetics of a labeled chloroplast DNA probe in the presence of total DNA from wild type cells (Fig. 1), from ~ - bleached mutants W34ZUD (Fig. 2) and W3BUL (Fig. 3) and from c p - bleached mutants (Fig. 4). Single and double stranded DNA were separated on hydroxylapatite and their proportions estimated by scintillation counting. The results were expressed according to Wetmur and Davidson (1968) as the reverse of single stranded DNA proportion as a function of time. Regression straight lines were calculated for each reassociation. 1251 labeled chloroplast DNA (0.25/~g/ml; 1.5 • 106 CPM//~g) was self renatured (e) or reassociated with DNA from: Fig. 1. Micrococcus luteus (480 ~g/ml, *); etiolated Euglena (480 ug/ml, v); green heterotrophic Euglena (120/~g/ml, A; 240/~g/ml, =; 480 t~g/ml, *); purified chloroplast DNA (80 #g/ml, *). Fig, 2. W34ZUD (960/~g/ml, ~; 1,920 ~g/ml, n; 3,200/~g/ml o). Fig. 3. W3BUL (960 ug/ml, *; 1,440 ~g/ml, *; 1,920/~g/ml A). Fig. 4. green heterotrophic Euglena (480/~g/ml, *); Y1BXD (480/~g/ml, A); Y3BUD (480/~g/ml, v.)

Extraction o f Nucleic Acids. Total DNA were extracted by the

Renaturation Kinetics. DNA were sonicated to obtain fragments

phenol-chloroform-SDS procedure already used in previous work (Heizmann et al. 1981, 1982). Total RNA were obtained by phenol-SDS extraction according to Parish and Kirby (1966). Purified chloroplast DNA were obtained from chloroplasts floated according to Richards and Manning (1975). After phenol extraction they were further purified from nuclear DNA contaminants by two or three cycles of CsC1 equilibrium centrifugation.

3 0 0 - 4 0 0 base pairs long; their size was checked by agarose gel electrophoresis using hDNA restriction digests as size maxkers. Chloroplast DNA were about 400 base pairs long after iodination and were used as such. Renaturations were performed after denaturation in Tris 10 mM - pH 7.5 (100 °C - 10 min), addition of NaC1 to 0.3 M and transfer to 63 °C. Aliquots were taken during the course of renaturation, diluted in cold 0.03 M phosphate buffer (pH 7.2) and loaded on hydroxylapatite columns. Single stranded DNA were eluted with 0.12 M phosphate buffer and double stranded DNA were recovered in 0.4 M phosphate buffer; their radioactivity was determined with a Packard Autogamma counter. We represented our data as second order plots (Wetmur and Davidson 1968) which allow analysis by simple calculations of regression straight lines and regression coefficients, and provide good precision in the initial part of the renaturation kinetics. We verified experimentally that the slopes of the second order straight

Radioactive Labeling o f Nucleic Acids. RNA and denatured DNA were labeled with 125I sodium iodide (Amersham) according to Meeker et al. (1976). Specifice activities of 2 to 5 . 106 CPM//ag were obtained. Nick-translation of DNA and hybridization to Southern filters were performed according to Jeffreys and FlaveU (1977).

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena line plots are proportional to the absolute concentrations of chloroplast DNA in the reaction mixture, i.e. that the second order renaturation constant (k2) depends only on the relative concentration of chloroplast DNA. The proportion of chloroplast DNAin a mixture of total DNA can thus be estimated from the k2 values ( proportion of chloroplast DNA

= k2 in mixture ) k2 for pure ct DNA "

The amount of pure chloroplast DNA was calculated from the cellular content of total DNA (2.5-3.5 pg/cell according to our estimations) and from the proportions of total DNA which is chloroplastic. In the case of bleached mutants where only a fraction of the genome is repeated, this fraction was estimated by the intercept of the second order straight line representing the reassociation of the low repeated chloroplast DNA, with the axis of ordinates.

113 Renaturation of Chloroplast DNA in the Presence of Total DNA from Green or Etiolated Wild Type Cells The addition of total DNA from wild type cells increases the reassociation rate of labeled chloroplast DNA; as with purified chloroplast DNA the reassociation second order plots are straight lines. The acceleration of the reassociations estimated from their slopes is proportional to the amount of total DNA from wild type cells added to the reaction mixture. It is not affected by the addition of heterologous DNA from Micrococcus luteus (Fig. 1). These kinetics allow to estimate the proportion of chloroplast DNA in total DNA and thus the number of chloroplast DNA molecules per cell. Our estimations are close to those of Rawson and Boerma (1976) and somewhat lower that those of Chelm et al. (1977) (Table 1).

Results

Self Renaturation of Chloroplast DNA The purified chloroplast DNA used as probe showed a single component of density 1.686 devoid of any contaminant detectable by analytical centrifugation. After in vitro labeling with 12Slodine this chloroplast DNA probe can reassociate with cold carrier chloroplast DNA. The second order plot representing the renaturation kinetics is a straight line (regression coefficient = 0 . 9 6 0.99). This means that iodinated chloroplast DNA reassociates in the presence of cold chloroplast DNA essentially as a single component following a second order kinetics (Fig. 1). The reassociation rate of the probe is consistant with the complexity of Euglena chloroplast DNA since the second order constant of the renaturation is 1.56 mole -1 • 1 • s -1 and the Cot1/2 is 0.4 mole • 1-1 • s. These data are close to those obtained by other authors for the same material (Stutz and Vandrey 1971; Rawson and Boerma 1976; Chelm et al. 1977) or for DNA of comparable complexities (Chlorella: Bayen and Rode 1973; Clamydomonas: Wells and Sager 1971; Howell and Walker 1976). As for other chloroplast DNAs our probe showed an instant reassociation affecting 8 to 10% of the sample, in deviation from perfect second order kinetics. After renaturation to a Cot of 4 mole • 1-1 . s, 86% of the radioactive hybrid remained acidoprecipitable after S 1 nuclease digestion and 90% of the renatured material was retained on hydroxylapatite. Progressive heating of the hydroxylapatite column produced a two step thermoelution of the hybrid corresponding to the melting of AT rich fragments at 7 0 - 7 5 °C and to the melting of GC rich fragments at about 80 °C in 0.12 M phosphate buffer. Our probe showed thus characteristics similar to those of reasonably pure Euglena chloroplast DNA.

Renaturation of Chloroplast DNA in the Presence of Total DNA from Bleached Mutants The addition of total DNA from bleached mutant strains W34ZUD or WaBUL for instance increases significantly the renaturation rate of the radioactive chloroplast DNA probe (Figs. 2 and 3). This increase however is low in comparison to that obtained with the DNA from wild type cells, indicating much lower levels of chloroplast DNA in the mutants. In addition, the renaturation plot is biphasic; the experimental data representing the reassociation kinetics according to a second order plot give two segments of straight lines instead of a single straight line. It appears that in the presence of DNA from the strain Wa4ZUD, a fraction representing 11.2% of the complexity of the chloroplast DNA probe reassociates 12 to 15 times more rapidly than the remaining sequences. These remaining sequences reassociate only slightly faster when mutant DNA is added; but this slight acceleration has been observed in all our experiments and should thus be significant. The absolute reassociation rate of these sequences is so low that their renaturation could not be driven to completion. We ignore if the whole chloroplast DNA probe is concerned by this acceleration, i.e. if all chloroplast DNA sequences are present in the mutant. Renaturations occur as if W34ZUD had only 11.2% of wild type chloroplast sequences in multiple copies; the slope of the second order plot allows to estimate the copy number of these sequences as in the preceding paragraph (90 copies per cell of W34ZUD). The same calculations were made to estimate the reiteration of the fragments reassociation less rapidly (3 copies per cell). Several bleached mutants were analyzed in the same way (Table 1). The two c p - mutants Y1BXD and Y3BUD had been shown to contain chloroplast DNAin amounts detectable

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena

114

Table 1. Amounts of chloroplast DNA estimated for various strains of Euglena Strain

Copy number of high complexity fragments

Copy number of low complexity "ribosomal" fragments

Complexity of repeated fragments as % of the wild type chloroplast DNA complexity

Origin of the strain

Z green Z etiolated Y1BXD Y3BUD W3BUL W34ZUD W3sEmsD W36ZHD ZHB WZN2L 35.110/1 36.60/7 36.60/14

730 210 530 340 4 3 5 7 5 2 4 7 6

70 90 100 160 28 50 30 65 50

100 100 100 100 6.3 -+0.3 11.2 -+ 0.3 9.5 10.3 8.3 5.5 8.3 12.1 12.1

G6ttingen Institute; strain 1224-5/25 GSttingen Institute; strain 1224-5/25 Dr. Schiff; X ray irradiation of baeillaxis Dr. Schiff; UV irradiation of baciUaris Dr. Schiff; UV irradiation of bacillaris This laboratory; UV irradiation of Z This laboratory; EMS treatment of Z This laboratory; heat treatment of Z Dr. Rawson; heat treatment of Z Dr. Calvayrac; DCMU + N 2 treatment of Z This laboratory; UV irradiation of Z This laboratory; UV irradiation of Z This laboratory; UV irradiation of Z

Z green;

Rawson Chelm

590 2,900

100 100

Rawson and Boerma (1976)

Z etiolated; Rawson Chelm

217 1,400

100

Chelm et al. (1977)

Table 2. Fraction of the radioactive chloroplast DNA probe rapidly reassociated in the presence of mixtures of total bleached mutant DNA. The fraction rapidly reassociated with single DNAs is indicated in brackets W3BUL (6.3%) + WZN2L (5.5%) W3BUL (6.3%)+ W36ZHD (10.3%) W36ZHD (10.3%) + WZN2L (5.5%) W36ZHD (10.3%) + W34ZUD (11.2%)

6% 9.8% 9.7% 11.8%

b y CsC1 analytical ultracentrifugation (Schiffand Epstein 1965). The reassociation kinetics observed after addition o f total DNA from these mutants show that t h e y contain chloroplast DNA similar to those o f wild t y p e cells in quantities and complexity. However their second order plots are definitely n o t straight lines b u t curves asymptotic to straight lines (Fig. 4). These plots are intermediate between those observed with DNA from wild type cells and those from most other bleached mutants.

Reassociations with Mixtures of DNA from Bleached Mutants Reassociations o f the radioactive chloroplast DNA probe with mixtures o f DNA from various bleached mutants were performed; these experiments show that fraction o f the chloroplast DNA rapidly reassociated is only determined b y the DNA from that component o f the mixture

which rapidly hybridizes the highest amount o f the probe (Table 2). The repeated segments from the different bleached mutants must thus bear some homologies. All the repeated sequences conserved in WZN2L are present in WaBUL and in W36ZHD; all the repeated sequences from WaBUL are also present in W36ZHDetc . . . . The repeated chloroplast DNA sequences in the bleached mutants might be composed o f a c o m m o n segment representing 5 to 10% of the probe.

Nature of the Repeated Chloroplast DNA Sequences in Bleached Mutants In order to identify the repeated chloroplast DNA sequences in the bleached mutants, total mutant DNA were digested with EcoR1, fractionated b y agarose gel electrophoresis, transfered to nitrocellulose filters and hybridized with nick-translated purified chloroplast DNA. Rather short hybridizations were used (6 h; 1 /~g of radioactive DNA in 30 ml o f hybridization mixture), allowing only the most abundant sequences to reassociate (Fig. 5). All restriction fragments were labeled in wild type cells, YIBXD and Y3BUD due to their high level of reiteration. In all p - bleached mutants, only the ribosomal fragments EcoL and P were d e t e c t e d i n this experiment and thus identified as the most abundant sequences of chloroplast DNA in these mutants. In another experiment already reported elsewhere (Heizmann et al. 1982), total DNA from W34ZUD were

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena

115 onto Southem blots o f EcoR 1 digests o f wild type chloroplast DNA: the ribosomal fragments EcoB, F, L and P were labeled demonstrating that in W34ZUD ribosomal sequences are more frequent than others.

Identification of the Restriction Fragments Expressed in Bleached Mutants and of Their Products

Fig. 5. Identification of the more abundant sequences in bleached mutants of Euglena. Total DNA from various strains were digested by EcoR 1;their Southern blots were briefly hybridized with nick-translated chloroplast DNA (1 gg DNA/30 ml hybridization medium for 6 hours; specific activity of the probe: 40 • 106CPM/#g). In ~obleached mutants only ribosomal sequences (EcoL and P) were labeled. In mutant Y1BXD the typical pattern of the baeillarisstrain shows the doubled EcoL and EcoP bands (Helling et al. 1979). In mutant Y3BUD also derived from bacillaris, the original pattern is altered and resembles the Z strain pattern. Ref.: purified Z chloroplast DNA (0.1 /~g/slot); Z wild type, Y1BXD and Y3BUD: 3 #g DNA/slot; W3BUL, W34ZUD, W35EtnsD, W36ZHD, ZHB: 15 ttg DNA/slot. Fluorography with Ilford Fast Tungstate intensitying screens and Kodak Xar-5 films at -70 °C for 2 days (Ref, Zwt, Ytb, Y3b) and 5 days for ~o- mutants

reassociated with the chloroplast DNA probe to a Cot value of 35 for which reiterated sequences were essentially double stranded and could be separated from single stranded slowly reassociating sequences by hydroxylapatite chromatography. They were then rehybridized

Total RNA extracted from wild type and from bleached mutants during the exponential growth phase were labeled in vitro with 125Iodine and hybridized to Southern filters loaded with EcoR 1 digests o f chloroplast DNA. This allowed to visualize the EcoR x fragments o f the chloroplast genome which are activately transcribed in the mutants (Fig. 6). As observed previously (Rawson and Boerma 1976b; Heizmann et al. 1978) all DNA fragments are not active even in wild type cells: some sequences (fragments EcoM, N, S, U) seem to have no transcriptional activity at all. On the opposite, ribosomal genes (fragments EcoB, F, L and P) are actively transcribed in wild type cells as well as in all bleached mutants analyzed here. The ampli. fled ribosomal genes of these mutants are functional in spite of their modification by mutagenesis. Their product are typical ribosomal RNA with biosynthetic pathway and maturation characteristic of chloroplast ribosomal RNA (Fig. 7): the formation o f mature m16 RNA (MW = 0.55 • 106 daltons) is preceded by the synthesis and processing of a p16 precursor (MW = 0.62 • 106 daltons); cycloheximide enhances the transcription rate o f chloroplast ribosomal RNA while lincomycin inhibits its maturation in wild type cells (Heizmann 1974) as well as in all bleached mutants (Heizmann et al. 1976). In addition, variable species o f non ribosomal fragments are expressed in ~0- mutants depending on the strain considered.

Fig. 6. Identification of transcriptionaUy active segments of chloroplast DNA in bleached mutants. Total RNA from various strains were labeled in vitro with 12Siodine and hybridized to Southern blots of-EcoR1 digests of purified wild type chloroplast DNA. Positive hybridizations identify the fragments expressed in each strain. The reference filter (REIDwas hybridized with radioactive DNAo The ribosomal fragments are EcoB, F, L and P

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena

116

NN~ N~g~ e

Discussion

We estimated the amounts of chloroplast DNA by measuring the increase of the renaturation rate of a labeled chloroplast DNA probe in the presence of large concentrations of total DNA. We observed significant accelerations of the renaturation kinetics in every experiments with wild type and with mutant DNA. This technique had been used by others to estimate the level of chloroplast DNA in Euglena cells (Rawson and Boerma 1976; Chelm et al. 1977). Rawson and Boerma did not consider their results as significant of the presence of chloroplast DNA in the mutant strain ZHB. Chelm et al. obtained rather important renaturation (28%) of a labeled chloroplast DNA probe at very high Cot (4,200) in the presence of total DNA from the mutant WsBHL; their data indicate that self-renaturation probably contributed to less than 10% reassociation of the probe in the conditions of the experiment. This should mean that a significant in. crease occured for the renaturation rate of the chloroplast DNA probe in response to the addition of W8BHL DNA. Anyway the positive hybridization obtained between cloned chloroplast DNA fragments and total mutant DNA demonstrate that bleached mutants contain chloroplast DNA (Heizmann et al. 1981, 1982). The increase observed here in the renaturation rate of our probe are thus certainly the testimony of the presence of chloroplast DNA in bleached mutants.

Repeated Ribosomal DNA in Bleached Mutants Our results show that in the mutants analyzed some sequences of chloroplast DNA are more repeated than others; they represent 5 to 12% of the probe and are repeated from 30 to 160 copies per cell depending onthe strain. However our chloroplast DNA probe contained traces of nuclear DNA undetectable by ultracentrifugation analysis but yielding some background in filter hybridization after overexposure of films with filters. The

Fig. 7. Characterization of the ribosomal and chloroplast nature of RNA produced by bleached mutants. Cells were incubated with H332po4 (30 ttCi/ml, 60 rain) (C) or in the presence of cycloheximide(CHI, 2 ttg/ml) or cycloheximide plus lincomycin (CHI + LM, 2 gg/ml and 2 mg/ml). Their RNA were extracted and analyzed by polyacrylamide gel electrophoresis and autoradiography. The labeling of the precursor p16 and mature m16 16S RNA in the presence of the drugs is characteristic of chloroplast ribosomal RNA (Heizmann 1974)

flottation of chloroplasts does not yield indeed perfectly intact organelles capable to undergo DNase treatment in order to eliminate DNA contaminants from the nucleus. It is probable that these nuclear contaminants participate by their repeated sequences to the initial rapid renaturation of the probe. The values obtained for the repeated chloroplast DNA segments may thus be considered as overestimated. Moreover, some sequence homologies exist between the chloroplast and the nuclear DNA of Euglena (Curtis and Rawson 1982). Total DNA (i.e. nuclear DNA) could thus accelerate selectively the renaturation of chloroplast ribosomal sequences in our liquid phase reassociation experiments through their multiple copies of nuclear ribosomal genes. However the preferential labeling of restriction fragments EcoP and L during short term filter hybridizations with total chloroplast DNA (Fig. 5) as well as the evolution of these ribosomal fragments during mutagenesis (Heizmann et al. 1982) clearly show their relative amplification in bleached mutants. Two non exclusive kinds of structural arrangements can be imagined for the chloroplast DNA molecules in these mutants: - they may exist as large circular molecules like in the wild type genome but with an increased number of ribosomal cistrons, some of them being eventually deleted or rearranged; ribosomal and adjacent fragments may replicate preferentially, independantly of other fragments. Several authors suggested in the past that even in wild type cells ribosomal fragments might be amplified at some stages of the growth or in some growth conditions (Manning and Richards 1972; Gibson and Hershberger 1975; Mielenz and Hershberger 1974). This hypothesis led us to speculate that the replication origin should be close to the ribosomal cistrons in order to allow for their preferential replication in bleached mutants. Electron microscopic analyses of chloroplast DNA molecules bearing replication loops and physical landmarks (R-loops with ribosomal RNA; restriction cuts) confirmed our speculation (Ravel-Chapuis et al. 1982).

Y. Hussein et al.: Chloroplast DNA in Bleached Mutants of Euglena The case o f the c p - mutants (Y1BXD and Y3BUD) is interesting since chloroplast DNA o f these mutants is still present in large amounts and can thus be isolated for direct analysis; it also shows the characteristic modifications introduced b y mutagenesis (non stoechiometry among fragments; modifications o f the ribosomal cistrons for Y3BUD (Fig. 5)). We present some characteristics o f the light DNA component (p = 1.686) isolated from Y3BUD in the accompanying paper. At last chloroplast ribosomal RNA were detected in all bleached mutants analyzed (more than 38 so far) (Heizmann et al. 1976). This fact is certainly related with the systematic amplification o f the ribosomal cistrons during mutagenesis. On the opposite the expression o f nonribosomal genes seems to be highly variable among the various mutants: in most cases the transcriptional inactivity o f a fragment is not merely due to its absence since chloroplast DNA fragments in bleached mutants are only seldom absent or visibly modified; it is rather the consequence o f a limited modification affecting slightly that fragment or its control b y another sequence. The high variability of nonribosomal gene activity contrasts strikingly with the uniform amplification and expression of the ribosomal DNA in bleached mutants o f Euglena. The behaviour o f the organelle genome during mutagenesis is probably a consequence o f the structure and organization o f the chloroplast replicon in Euglena.

Acknowledgements. We thank Dr. Jacques Grange (Unit~ de Virologic de I'INSERM) for analytical ultracentrifugations of our chloroplast DNA probes, Mr. P. M. Malet and Mrs. C. FaureSehwob for technical assistance. This work was partly supported by a C.N.R.S. grant (A.T.P. n ° 8068).

117 Freyssinet G, Heizmann P, Verdict G, Trabuchet G, Nigon V (1972) Physiol Veg 10:421-442 Gibson WH, Hershberger CL (1975) Arch Bioehem Biophys 168: 8-14 Heizmann P (1974) Biochimie 56:1357-1364 Heizmann P, Salvador G, Nigon V (1976) Exp Cell Res 99:253260 Heizmann P, Verdict G, Younis H (1978) In: Akoyunoglou G, Argyroudi-Akoyunoglou JH (eds) Chloroplast Development. Elsevier/North-Holland, Amsterdam, pp 623-628 Heizmann P, Doly J, Hussein Y, Nicolas P, Nigon V, Bernardi G (1981) Bioehim Biophys Acta 653:412-415 Heizmann P, Hussein Y, Nicolas P, Nigon V (1982) Curt Genet 5:9-15 Howell SH, Walker LL (1976) Biochim Biophys Acta 418:249256 Jeffreys AJ, Flavell RA (1977) Cell 12:429-439 Manning JE, Richards OC (1972) Biochem 11:2036-2043 Meeker RR, Thomas JJ, Tewari KK (1976) Plant Physio158:71 76 Mielenz JR, I-Iershberger CL (1974) Biochem Biophys Res Comm 58:71-76 Parish JH, Kirby KS (1966) Biochim Biophys Acta 129:554-562 Ravel-Chapuis P, Heizmann P, Nigon V (1982) CR Acad Sci (in press) Rawson JRY, Boerma C (1976) Proc Natl Acad Sci USA 73: 2401-2404 Rawson JRY, Boerma C (1976b) Bioehem 15:588-592 Richards OC and Manning JE (1975) In: Lefort-Tran M, Valencia R (eds) Les Cycles CeUulaires. Edition du C.N.R.S, pp 213221 Schiff JA, Epstein HT (1965) In: Locke M (ed) Reproduction: molecular, subcellular and cellular. Academic Press, New York, pp 131-189 Stutz E, Vandrey JP (1971) Proc. Ilnd Intern. Congr. on Photosynth. Forti G, Avron M, Melandri A (eds) Junk W, The Hague, pp 2601-2609 Wells R, Sager R (1971) J Mol Bio158:611-622 Wetmur JG, Davidson N (1968) J Mol Biol 31:349-370

References Bayen M, Rode A (1973) Eur J Biochem 39:413-420 Chelm BK, Hoben PJ, Hallick RB (1977) Biochem 16:782-785 Curtis SE, Rawson JRY (1982) Plant Physiol 69:69-71

Communicated b y R. J. Schweyen Received July 5, 1982