J. Cell Set. ao, 341-355 (1976) 341Printed in Great Britain

THE LOCALIZATION OF [3H]THYMIDINE

INCORPORATION IN THE DNA OF

REPLICATING SPINACH CHLOROPLASTS

BY ELECTRON-MICROSCOPE

AUTORADIOGRAPHY

R. J. ROSE AND J. V. POSSINGHAMCSIRO Division of Horticultural Research,G.P.O. Box 350, Adelaide, South Australia 5001, Australia

SUMMARY

Electron-microscope autoradiography has been used to obtain information on the localizationof DNA labelled with pHJthymidine in chloroplasts known to be replicating and concomitantlysynthesizing and segregating DNA, in cultured leaf disks. The studies were made using bothMicrodol-X developer and a 'compact' developer which gave a smaller grain size. About 80 %of the grains were associated with the granal membranes and with presumptive DNA regions(3-nm fibril material in clear areas). Few grains occurred in association with the chloroplastenvelope. We suggest that the DNA of chloroplasts is associated with the grana lamellae andextends into the stroma. Some light-microscope autoradiographs of whole chloroplasts showspiral or helical-like labelling patterns. We interpret these patterns as demonstration of thepossibility that DNA occurs along the length of a continuous lamellar membrane system.Chloroplast fractionation experiments provided data consistent with the electron-microscopeautoradiographic studies as most of the label was associated with chlorophyll-containinglamellae.

We consider an association of chloroplast DNA molecules along the length of a continuouslamellar system would ensure an orderly segregation of DNA to daughter chloroplasts, duringthe binary fission of spinach chloroplasts by constriction division.

INTRODUCTION

It is now well established that in the higher plant chloroplast there are many copies(20-30) of the chloroplast genome (Wells & Birnstiel, 1969; Kirk, 1972). Thereplicating chloroplasts of light-grown spinach leaf disks actively synthesize DNA andduring the division cycle segregate it in similar amounts to the daughter chloroplastsRose, Cran & Possingham, 1974, 1975; Possingham & Rose, 1976. It is thereforepossible that the chloroplast DNA (cDNA) of higher plants is located in the chloro-plast in such a way as to permit such a segregation.

The organization of DNA in the chloroplasts of a higher plant has been studied bylight-microscope autoradiography and electron microscopy of serial sections(Herrmann, 1970; Herrmann & Kowallik, 1970; Kowallik & Herrmann, 1972). InBeta vulgaris chloroplasts they found several distinctly separate DNA nucleoids,separated from each other by lamellate thylakoid stacks. As the number of nucleoidswas much less than the number of chloroplast genomes, Kowallik & Herrmann (1972)

342 R. J. Rose and J. V. Possingham

concluded that each nucleoid contained 4-8 cDNA molecules. Earlier electron-microscope studies of Swiss chard (Kislev, Swift & Bogorad, 1965) and Avena sativa(Gunning, 1965) revealed a number of DNA areas in a single chloroplast section.There appears to be no evidence to-date for a regular or definite pattern of DNAthroughout higher plant chloroplasts as occurs in certain algae which have a con-tinuous ring-shaped nucleoid lying inside the peripheral thylakoids which loop aroundthe rim of the chloroplast (Bisalputra & Bisalputra, 1969; Gibbs, Cheng & Slankis,1974). In these algae it appears that there are associations between the cDNA andthylakoids which facilitate this arrangement. Close associations between thylakoidmembranes and DNA also occur in dinoflagellates (Kowallik & Haberkorn, 1971;Bibby & Dodge, 1974). In higher plant chloroplasts observations of cDNA membraneassociations have been made in spinach (Woodcock & Fernandez-Moran, 1968) andin Pelargonium (Knoth, Herrmann, Bottger & Borner, 1974), though the regularity ofthis relationship and which chloroplast membranes are involved has not been clearlydefined.

To obtain more information on the location of cDNA in relation to the segregationprocess in replicating spinach chloroplasts, we have studied the incorporation ofPHJthymidine (3H-TdR) into chloroplasts using light- and electron-microscopeautoradiography and chloroplast isolation and fractionation techniques.

MATERIALS AND METHODS

Culture of leaf disks

The growth of the spinach plants (Spinacia oleracea), and subsequent leaf disk culture onsterile nutrient agar has been previously described (Possingham & Smith, 1972). Growth in thelight refers to a 14 h-day, at 20000 lux, with day and night temperatures of 25 and 22 °C,respectively. Growth in continuous darkness was at 25 °C.

Autoradiography

The disks were incubated in 50 /iCi [6—3H]thymidine (27 Ci/mM, Radiochemical Centre,Amersham, U.K.), added in 1 ml under sterile conditions to the 20 ml of nutrient agar mediumin 9-cm Petri dishes. The light-microscope autoradiographic techniques have been describedpreviously, as has the specificity of incorporation of 3H-TdR into DNA (Rose et al. 1975).

For electron-microscope autoradiography, tissue from the disks was fixed in 625 %glutaraldehyde, postfixed in osmium and embedded in Araldite (Cran & Possingham, 1972a).Sections were mounted on carbon-coated copper grids, and stained in uranyl acetate in 50 %ethanol, followed by lead citrate. The stained sections were then coated with a thin layer ofcarbon, and fixed to coverslips with double-sided tape. The grids were coated with an approxi-mate monolayer of diluted Ilford L-4 emulsion, by passing the coverslip through a film formedat the top of a 20-ml specimen tube (a modification of the Caro & van Tubergen (1962) loopmethod). The coverslips were attached to glass slides by double-sided tape, and the gridsexposed at 5 °C for 6—15 weeks in light-tight boxes containing silica gel. Development wascarried out with Microdol-X for 3 min or with the ' compact' development mixture of Lettr6 &Paweletz (1966), for 1 min; in both cases at approximately 20 °C. The 'compact' developermixture (containing Phenidone as the developing agent) gave backgrounds too high for routineuse with chloroplasts, which in general have a low level of labelling.

Grain locations over chloroplasts were determined using Microdol-X development. Photo-graphs of 131 chloroplasts from random locations in background-free areas were obtained by2 different operators. We used the central region of the filamentous grain as the estimated

Location of spinach choroplast DNA 343

position of the original latent image (Caro & van Tubergen, 1962). We also scored photographsas to whether the grains were touching different membranes, as done by Comings & Okada(1973), in their nuclear DNA replication studies.

[3H]thymidine incorporation into isolated chloroplast fractions

Five-day-old light-grown disks (30 per plate) were incubated 24 h in 3H-TdR as in the auto-radiography studies, and usually 180 with a total fresh weight of about 4 g used for chlorQplastisolation.

Chloroplasts were obtained by a razor chopping method designed to give a high proportionof intact chloroplasts free of contaminating nuclei (Spencer & Wildman, 1964). Disks werechopped in a medium containing 0-25 M sucrose, 25 % Ficoll, 5 % dextran, 25 mM Tris-HClpH 78, 1 /ig/ml Cleland's reagent, 2 mM CaCl3) 1 mM magnesium acetate, o-i % bovine serumalbumin and 05 mg/ml cold thymidine (Honda, Hongladarom & Laties, 1966). The resultingbrei was filtered twice through fine mesh (ioo-/tm) steel gauze, and 3-ml aliquots were layeredon the top of discontinuous sucrose gradients. The gradient consisted of 6 ml of 60 % sucrose,2 ml of 45 % sucrose and 6 ml of 20 % sucrose in modified Honda medium. After centri-fugation in a swinging bucket rotor for 20 min at 10000 rev/min (Beckman model L2-65B,SW 27.1 rotor), the chloroplast bands at the bottom and top of the 45 % sucrose interfaceswere collected. A double banding pattern was characteristic of chloroplasts isolated from disks,as many contain large amounts of starch.

The isolated chloroplasts were suspended in the modified Honda medium, pelleted at10000 rev/min and resuspended using a glass rod, in 17 ml of bursting medium (10 mMbicine-NaOH buffer, pH 76, 4 mM MgCl,), similar to that described by Douce, Holtz &Benson (1973). The chloroplasts were left in this medium in ice for 30 min to ensure detachmentof most chloroplast envelopes. While causing loss of the envelope, the osmolarity of the mediumis designed to cause minimal disruption of the lamellar system (Douce et al. 1973). Followingthe treatment in bursting medium, the suspension was centrifuged in a SW 27.1 rotor for10 min at 3500 g to obtain a lamellae-rich fraction, and for 40 min at 38000 g to obtain anenvelope-rich fraction (Joy & Ellis, 1975).

The 3 fractions (lamellae, envelope and supernatant) were collected on Whatman GF/Aglass-fibre filters, and washed with 25 ml of bursting medium. The chlorophyll was extractedwith 80% acetone and estimated spectrophotometrically (Arnon, 1949). The filters were thenwashed with 5 % TCA (containing cold thymidine) ether-ethanol and finally ether before airdrying and counting in a toluene-based scintillation fluid (Rose et al. 1975).

RESULTS

Electron-microscope autoradiographs of 3H-TdR incorporation into chloroplastsare shown in Figs. 2-7. These photographs show the type of incorporation patternswe obtained, using disks of different ages, different physiological conditions anddifferent isotope incorporation times; variables which did not in any consistent wayalter grain localization patterns. We found that starch grains serve as a useful controlagainst random silver grain formation as they are seldom labelled. It is also apparentthat the grains do not occur at the periphery of the organelle in association with thechloroplast envelope.

By quantifying the autoradiographic data (see Materials and methods), we obtainthe silver grain distribution pattern shown in Fig. 1. Most of the grains occur over theclear areas containing fibrils (about 3 nm) (e.g. Figs. 2-4) or grana lamellae (Figs. 2-7).If we score the autoradiographs in another way (Comings & Okada, 1973), 5 % of thegrains touch the chloroplast envelope, while 72% touch the lamellar system of granaand stroma membranes.

344 R- J- Rose andj. V. Possingham

Electron-microscope autoradiographs of constricting chloroplasts are shown inFigs. 8 and 9. Silver grains occur on either side of the constriction, as do presumptiveDNA areas (Fig. 9 in particular).

Choroplast envelope

Stroma lamellae

Grana lamellae

Clear areas containingfibrils

Fig. 1. Histogram showing the location of silver grains in relation to the differentchloroplast components. 332 grains scored.

Resolution with Microdol-X development is to a degree hampered by the size of thesilver grains, and a 'compact' developer was used in an attempt to overcome thisproblem. Although we commonly had high backgrounds with this developer, theresults were essentially similar to those obtained with Microdol-X. This is illustratedin Fig. 10 which shows a labelled chloroplast, while in Fig. 11 a labelled nucleus isshown, indicating that the round silver grains formed with compact developer are notsimply osmiophilic lipid bodies, which are commonly present in chloroplasts.

We could not tell from the electron-microscope autoradiography data whether thereis a regular or repeating arrangement of the cDNA along chloroplast lamellar mem-branes. There was no clear evidence that DNA is associated with particular grana ofthe lamellar membrane regions as grains frequently occurred throughout the lengthof a given section. However, we found spiral or helical-like labelling patterns (Fig. 12)in some light-microscope autoradiographs. We observed these patterns over individualchloroplasts on a number of occasions, but to-date have not been able to relate themto particular treatments.

As there are always some uncertainties with resolution when electron-microscopeautoradiography is used (Rogers, 1971) we isolated chloroplasts and separated theminto lamellae-rich and envelope-rich fractions, as an alternative means of localizing3H-TdR incorporation. We found that the label in these experiments is mainly

Location of spinach chloroplast DNA 345

associated with the chloroplast lamellae fraction, which is high in chlorophyll (Table 1).A much smaller proportion of the label is associated with the chloroplast envelopeand small membrane fragment fraction. Some of this latter incorporation is probablydue to contamination with chlorophyll-containing lamellae fragments. A proportionof the labelled DNA is liberated into the supernatant but this proportion is variable,probably because of variations in the severity of osmotic disruption.

Table 1. [3H]thymidine incorporation into different chloroplast fractions

Incorporation intoTCA-insoluble material, Chlorophyll,

Fraction cpm (% of total) fig

3500-g pellet, chloroplast lamellae T5657 (61 %) 834fraction high in chlorophyll

38000-g pellet, chloroplast envelope 2460 (9%) o-86and small membrane fragments

Supernatant, soluble components 7736 (30 %) —

DISCUSSION

Our electron-microscope autoradiography data indicate that in spinach chloroplaststhere is an association of cDNA with the chloroplast lamellae system, and in par-ticular with grana lamellae. As well, the fibril-containing areas of low electron densitywere heavily labelled. These latter areas are usually recognized on morphologicalgrounds as DNA-containing regions following the work of Ris & Plaut (1962) withChlamydomonas. More recently, Gibbs et al. (1974) have shown by electron-microscope autoradiography that all plastid DNA is localized in the nucleoid ofOchromonas danica. We suggest that our labelling patterns could result from theassociation of cDNA with lamellar membranes and with its extension from these intothe stroma, where it is present in the fibril regions. Our autoradiographic data areconsistent with observations on the location of the DNA regions based on morpho-logical criteria (Figs. 2-4, 8-10). It is also consistent with the conclusion of Woodcock& Ferndndez-Mor&n (1968) obtained by spreading osmotically disrupted chloroplastsin a protein monolayer.

It could possibly be argued that grain locations over lamellae are fortuitous becauseof the relatively large amount of these membranes in chloroplasts. However, in thespinach disks used in our experiments granal stacking is not extensive (which in partmay be related to the high light intensity at which the disks are grown) and althoughthe variation is large, on average about 20 % of the area of a chloroplast section isoccupied by lamellar membranes, whereas almost 50% of the grains occur here.Histogram analysis shows that there is not a uniform distribution of grains throughoutthe chloroplast, as the DNA regions recognized on morphological grounds contain alarge proportion of the grains, whereas starch regions which occupy a large area arerarely labelled.

We believe that the light-microscope autoradiographs showing helical-like labellingpatterns are of considerable significance. One possible interpretation is that this pattern

346 R. J. Rose andj. V. Possingham

represents incorporation along the lamellar system, and evidence is available whichfavours a continuous helical arrangement of the chloroplast lamellar system in higherplants (Spencer & Wildman, 1962; Paolillo, 1970; Wildman, Jope & Atchison, 1974;S. G. Wildman, personal communication, 1974). Such labelling patterns might arisewhen there is synchrony of synthesis of cDNA molecules and of new regions of thechloroplast lamellae.

The data we have obtained from the fractionation of isolated chloroplasts areconsistent with the electron-microscope autoradiography. Clearly, little DNA isassociated with the chloroplast envelope and spinach chloroplasts thus do not appearto segregate their DNA in the same way as proposed for bacteria, where the DNA isattached to the limiting membrane (Ryter, 1968).

Our fractionation data are in agreement with the report of Spencer & Whitfeld(1967) that primer DNA and DNA polymerase are tightly associated with a particulatefraction free of external chloroplast membranes. Our finding that some of the cDNAcan be released free of membranes by osmotic shock is supported by the work ofWong & Wildman (1972) who used osmotic shock as the basis of a method for theisolation of cDNA.

The data we have obtained suggest a segregation mechanism based on the associ-ation of the cDNA with the lamellar system. The cDNA molecules would be locatedat intervals along the lamellar system, probably being associated with the thylakoidsof the grana. It is possible, from the light-microscope autoradiographs, that the cDNAassociations could occur at regular intervals. If the lamellar system is indeed acontinuum, its breakage at a single, central point would ensure that similar amountsof DNA would distribute to each daughter chloroplast during binary fission byconstriction division (Cran & Possingham, 1972a, b). Such a system would accountfor the results of our light-microscope autoradiographic studies (Rose et al. 1974;Possingham & Rose, 1976) where on average DNA doubles during the chloroplastdivision cycle and distributes fairly equally to daughter chloroplasts. In principle, thetype of segregation mechanism would be similar to that in the brown alga, Sphacelaria(Bisalputra & Bisalputra, 1970). Here the peripherally located ring-like genophorebecomes elongated with the lengthening of the chloroplast and segregates into 2daughter loops, each being transmitted to the daughter half of the chloroplast. Weconsider that in spinach the difference lies in the organization of the cDNA whichforms associations along the length of the lamellar system.

We do not consider the evidence obtained by Kowallik & Herrmann (1972) byserial section conflicts with our observations. The discrete DNA regions they haveobserved may well all be associated with grana (see their fig. 8). The arrangement ofcDNA found in serial section might depend on whether there is a periodicity alongthe membrane and this could vary from plant to plant. Essentially what is importantis the exact distribution of DNA molecules in relation to the lamellar system ofchloroplasts.

Although in spinach chloroplasts the cDNA is predominantly associated withlamellar membranes rather than the outer membrane, the situation may well bedifferent in other plastid types. Etioplasts have been shown to segregate their DNA to

Location of spinach chloroplast DNA 347

daughter plastids and observations have been made of DNA-membrane associations(Sprey, 1968; Sprey & Gietz, 1973). At least some of the DNA appears to be associatedwith the etioplast envelope (Herrmann & Kowallik, 1970). In a mutant of Pelargoniumzonale known to have few lamellar structures, associations between cDNA and theplastid envelope have also been reported (Knoth et al. 1974). It is at least possible thatchanges which occur in plastid membrane systems during their differentiation fromproplastids or etioplasts to chloroplasts (such as the invagination of the outer mem-brane) could lead to changes in the location of cDNA.

We are indebted to Mrs J. Nikandrow for valuable technical assistance throughout this study.We wish to thank Miss P. Y. Dyer of the Department of Biochemistry, University of

Adelaide, for instruction on emulsion application for E.M. autoradiography.

REFERENCESARNON, D. J. (1949). Copper enzymes in isolated chloroplasts: polyphenoloxidase in Beta

vulgaris. PI. Physiol., Lancaster 24, 1-15.BIBBY, B. T. & DODGE, J.D. (1974). The fine structure of the chloroplast nucleoid in Scrippsiella

sweeneyae (Dinophyceae). J. Ultrastruct. Res. 48, 153-161.BISALPUTRA, T. & BISALPUTRA, A. A. (1969). The ultrastructure of chloroplast of a Brown Alga

Sphacelaria sp. I. Plastid DNA configuration - the chloroplast genophore. J. Ultrastruct.Res. 29, 151-170.

BISALPUTRA, T. & BISALPUTRA, A. A. (1970). The ultrastructure of chloroplast of a Brown AlgaSphacelaria sp. III. The replication and segregation of chloroplast genophore. J. Ultrastruct.Res. 32, 417-429.

CARO, L. G. & VAN TUBERGEN, R. P. (1962). High-resolution autoradiography. I. Methods.J. Cell Biol. 15, 173-188.

COMINGS, D. E. & OKADA, T. A. (1973). DNA replication and the nuclear membrane. J. molec.Biol. 75, 609-618.

CRAN, D. G. & POSSINGHAM, J. V. (1972a). Variation of plastid types in spinach. Protoplasma74, 345-356.

CRAN, D. G. & POSSINGHAM, J. V. (19726). Two forms of division profile in spinach chloro-plasts. Nature, New Biol. 235, 142.

DOUCE, R., HOLTZ, R. B. & BENSON, A. A. (1973). Isolation and properties of the envelope ofspinach chloroplasts. J. biol. Chem. 248, 7215-7222.

GIBBS, S. P., CHENG, D. & SLANKIS, T. (1974). The chloroplast nucleoid in Ockromonas damca.1. Three-dimensional morphology in light- and dark-grown cells. J. Cell Set. 16, 557-577.

GUNNING, B. E. S. (1965). The fine structure of chloroplast stroma following aldehydeosmium-tetroxide fixation. J. Cell Biol. 24, 79-93.

HERRMANN, R. G. (1970). Multiple amounts of DNA related to the size of chloroplasts.1. An autoradiographic study. Planta 90, 80—96.

HERRMANN, R. G. & KOWALLIK, K. V. (1970). Multiple amounts of DNA related to the sizeof chloroplasts. 11. Comparison of electron-microscopic and autoradiographic data.Protoplasma 69, 365-372.

HONDA, S. I., HONGLADAROM, T. & LATIES, G. G. (1966). A new isolation medium for plantorganelles. J. exp. Bot. 17, 460-472.

JOY, K. W. & ELLIS, R. J. (1975). Protein synthesis in chloroplasts. IV. Polypeptides of thechloroplast envelope. BioMm. biophys. Ada 378, 143-151.

KIRK, J. T. O. (1972). The genetic control of plastid formation: Recent advances and strategiesfor the future. Subcellulat Biochem. 1, 333-361.

KISLEV, N., SWIFT, H. & BOGORAD, L. (1965). Nucleic acids of chloroplasts and mitochondriain Swiss chard. J. Cell Biol. 25, 327-344.

348 R. J. Rose andj. V. Possingham

KNOTH, R.( HERRMANN, F. H., BSTTGER, M. & BORNEB, TH. (1974). Stxuktur und Funktionder genetischen Information in den Plastiden. XI. DNA in normalen und mutiertenPlastiden der Sorte 'Mrs. Parker' von Pelargonium zonale. Biochem. Physiol. Pflanzen 166,129-148.

KOWALLIK, K. V. & HABERKORN, G. (1971). The DNA-stmctures of the chloroplast ofProTocentrum micans (Dinophyceae). Arch. Mikrobiol. 80, 252-261.

KOWALLIK, K. V. & HERRMANN, R. G. (1972). Variable amounts of DNA related to the size ofchloroplasts. IV. Three-dimensional arrangement of DNA in fully differentiated chloroplastsof Beta vulgaris L. J. Cell Sci. 11, 357-377.

LETTRE, H. & PAWELETZ, N. (1966). Probleme der elektronenmikroskopischen Autoradio-graphie. Naturwissenschaften 53, 268-271.

PAOLILLO, D. J. Jr. (1970). The three-dimensional arrangement of intergranal lamellae inchloroplasts. J. Cell Sci. 6, 243-255.

POSSINGHAM, J. V. & ROSE, R. J. (1976). Chloroplast replication and chloroplast DNA synthesisin spinach leaves. Proc. R. Soc. B (in press).

POSSINGHAM, J. V. & SMITH, J. W. (1972). Factors affecting chloroplast replication in spinach.J. exp. Bot. 23, 1050-1059.

Ris, H. & PLAUT, W. (1962). Infrastructure of DNA-containing areas in the chloroplast ofChlamydomonas. J. Cell Biol. 13, 383-391.

ROGERS, A. W. (1971). Recent developments in the use of autoradiographic techniques withelectron microscopy. Phil. Trans. R. Soc. Ser. B 261, 159—171.

ROSE, R. J., CRAN, D. G. & POSSINGHAM, J. V. (1974). Distribution of DNA in dividing spinachchloroplasts. Nature, Lond. 251, 641-642.

ROSE, R. J., CRAN, D. G. & POSSINGHAM, J. V. (1975). Changes in DNA synthesis during cellgrowth and chioroplast replication in greening spinach leaf disks. J. Cell Sci. 17, 27-41.

RYTER, A. (1968). Association of the nucleus and the membrane of bacteria: a morphologicalstudy. Bad. Rev. 32, 39-54.

SPENCER, D. & WHITTELD, P. R. (1967). DNA synthesis in isolated chloroplasts. Biochem.biophys. Res. Commun. 28, 538-542.

SPENCER, D. & WILDMAN, S. G. (1962). Observations on the structure of grana-containingchloroplasts and a proposed model of chloroplast structure. Aust. J. biol. Sci. 15, 599-610.

SPENCER, D. & WILDMAN, S. G. (1964). The incorporation of amino acids into protein bycell-free extracts from tobacco leaves. Biochemistry, N. Y. 3, 954—959.

SPREY, B. (1968). Zum verhalten DNS-haltiger areale des Plastidenstromas bei derPlastidenteilung. Planta 78, 115-133.

SPREY, B. & GLETZ, N. (1973). Isolierung von etioplasten und elektronenmikroskopischeAbbildung membranassoziierter Etioplasten-DNA. Z. Pflanzenphysiol. 68, 397—414.

WELLS, R. & BIRNSTIEL, M. (1969). Kinetic complexity of chloroplastal and mitochondrialdeoxyribonucleic acid from higher plants. Biochem. J. 112, 777-786.

WILDMAN, S. G., JOPE, C. & ATCHISON, B. A. (1974). Role of mitochondria in the origin ofchloroplast starch grains. Description of the phenomenon. PI. Physiol., Lancaster 54, 231-237.

WONG, F. Y. & WILDMAN, S. G. (1972). Simple procedure for isolation of satellite DNAs fromtobacco leaves in high yield and demonstration of minicircles. Biochim. biophys. Acta 259,5-i2-

WOODCOCK, C. L. F. & FERNANDEZ-MORAN, H. (1968). Electron microscopy of DNA con-formations in spinach chloroplasts. J. molec. Biol. 31, 627-631.

(Received 14 July 1975)

Figs. 2, 3. Electron-microscope autoradiographs of chloroplasts. Microdol-Xdevelopment.

Fig. 2. Disks excised and incubated in *H-TdR for 24 h in the light. Autoradio-graph exposed 6 weeks, x 36000.

Fig. 3. As for Fig. 2, but autoradiograph exposed 15 weeks. Arrow indicates a cleararea containing 3-nm fibrils, x 45000.

Location of spinach chloroplast DNA

350 R- J. Rose and J. V. Possingham

Figs. 4-7. Electron-microscope autoradiographs of chloroplasts. Microdol-Xdevelopment.

Fig. 4. Disks grown in darkness for 5 days, transferred to the light for 1 day, andthen incubated in 'H-TdR for 24 h in the light. Autoradiograph exposed 10 weeks,x 25000.

Fig. 5. As for Fig. 4. x 18000.Fig. 6. Disks excised and incubated in 3H-TdR for 72 h in the light. Autoradio-

graph exposed 12 weeks, x 29000.Fig. 7. As for Fig. 6. x 17000.

Location of spinach chloroplast DNA 351

352 R. J. Rose and J. V. Postingham

Figs. 8, 9. Electron-microscope autoradiographs of constricting chloroplasts.Microdol-X development.

Fig. 8. Disks excised and incubated in 3H-TdR for 24 h in the light. Autoradio-graph exposed 15 weeks. X 14 000.

Fig. 9. Disks grown in darkness for 5 days, transferred to the light for 1 day, andthen incubated in 3H-TdR for 24 h in the light. There is a single silver grain on eitherside of constriction. Note fibril-containing clear areas. Autoradiograph exposed10 weeks, x 25000.

Location of spinach chloroplast DNA 353

354 R- J- R°se andj. V. Possingham

Fig. 10. Electron-microscope autoradiograph of chloroplast. Compact development.Disks excised and incubated in 3H-TdR for 24 h in the light. The grey halo aroundmost of the black grains distinguishes them from lipid droplets. Autoradiographexposed 12 weeks, x 33 000.Fig. 11. Electron-microscope autoradiograph of a nucleus. Compact development.Disks excised and incubated in 3H-TdR for 24 h in the light. Autoradiographexposed 12 weeks, x 11000.Fig. 12. Light-microscope autoradiograph of a cell, showing grain patterns which arelocated over chloroplasts. Disks were grown in the dark for 5 days, before beingincubated in 3H-TdR for 24 h in the light. Autoradiograph exposed 13 days, x 900.

Location of spinach chloroplast DNA 355

t • • , .