Plant Cell Reports (1985) 4:41-45

Plant Cell Reports © Springer-Verlag 1985

Chloroplast DNA synthesis in light and dark grown cultured Nicotiana tabacum cells as determined by molecular hybridization Gordon Cannon, Sabine Hcinhorst, Janusz Siedlecki, and Arthur Weissbach Department of Cell Biology, Roche Institute of Molecular Biology, Roche Research Center, Nutley, NJ 07110, USA Received December 14, 1984- Communicated by O. L. Gamborg

Abstract A simple method using molecular hybridization was devised to quantitatively measure chloroplast DNA synthesis in rive. Total cellular DNA isolated from ~icotiana tabacum suspension cells, labeled with H-thymidine, was hybridized to nitrocellulose membrane-bound cloned chloroplast DNA (ct DNA) fragments. Colorless, dark grown N. tabacum cells were found to contain approximately 3300-4800 chloroplast genome copies per cell, whereas light grown, green cells contain about 9500-12000 chloroplast genomes per cell. This difference in ct DNA levels suggests that the chloroplast genome is somewhat amplified during growth of the cells in the light. The hybridization technique was also used to measure the efficiency of hybridization between cloned spinach ct DNA and tobacco ct DNA. The two DNAs were found to cross-hybridize with an efficiency of 69-75%. Introduction The structure of the chloroplast genome is well established. It exists as a circular double stranded DNA molecule of approximately i0 daltons (Bedbrook and Kolodner 1979). Each chloroplast contains multiple copies of the genome and copy numbers from 8-200 per chloroplast have been reported for different species (Herrmann and Possingham 3980). It has also been demonstrated in numerous higher plants that cellular chloroplast DNA (ct DNA) content is dependent on age, tissue type and developmental stage of the cell containing the chloroplast (Scott and Possingham 1983). Little is known about at DNA synthesis or how it is regulated and relatively few studies have addressed the question of ct DNA synthesis in vivo. A major obstacle to these studies has been the lack of a method for easily quantitating ct DNA synthesis. Resolution of nuclear, mitochondrial and ct genomes on density gradients has been successful for some algae (Blamire et el. 1974) but this method is seldom suitable for separation of organellar genomes of higher plants (Lamppa and Bendich 1979). Likewise, isolation of intact chloroplasts after an in rive labeling period is difficult to use for quantitation because of varying yields from experiment to experiment and because contamination by nuclear and mitochondrial DNAs is often observed. To avoid these problems we have devised a method, based on molecular hybridization of total radiolabeled cellular DNA to cloned ct DNA, that permits us to specifically follow et DNA synthesis in growing suspension cultures of plant cells. Using this method we have been able to measure the chloroplast

Offprint requests to: A. Weissbach

genome copy number in cultured Nicotiana tabacum cells grown in the dark (colorless) or in the light (green). Also this paper reports the relative cross-hybridization efficiencies of tobacco and spinach ct DNAs, so that the ct DNA of either species can be used for accurate determination of chloroplast DNA by molecular hybridization on nitrocellulose membrane filters. Haterials

and Methods

Plant }~terial and Growth Conditions

Nicotiana tabacum var. Wisconsin 38 suspension cultures were grown on Murashige and 8koog media (Hurashige and Skoog 1962) supplemented with 800 mg/l casein hydrolysate, 0.5 mg/l folic acid, 0.5 mg/l biotin, i0 mg/l thiamine, 0.2 mg/l benzyladenine (BA), 0.6 mg/l a-naphthaleneacetic acid and 3 g/l sucrose. The cultures were grown in the dark in 500 ml erlenmeyer flasks on a rotary shaker (150 rpm). Nicotiana tabacum var. Samsun (referred to as the Yamada line) was a gift from F. Sate and Y. Yamada of Kyoto University and was grown as suspension cultures on modified Linsmaier-Skoog media (Nishida et al. 1980) containing 3% sucrose. The light grown suspension cultures were maintained on a rotary shaker (150 rpm) under I0000 lux of light provided by wide spectrum fluorescent lamps (Verilux). Dark grown suspension cultures were maintained on a similar shaker in total darkness. Autotrophic cultures were maintained under 5% CO 2 on similar media lacking sucrose. Determination of Nuclear DNA Content Nuclei were isolated from cultured N. tabacum cells by the method of Willmitzer and Wagner (1981). DNA was estimated by the fluorometric assay described by Vytasek (1982). No significant difference in DNA quantities per nucleus was observed in light or dark grown cells. P_rej~aration of Cloned Ct DNA Fragments Recombinant plasmids containing Pstl endonuclease restriction fragments of spinach (Spinacia oleraceae) ct DNA cloned into pBR322 were obtained from J. Palmer (Palmer and Thompson 1981). Plasmids containing tobacco (Nicotia~a tabacum) ct DNA fragments in the PstI site of pBR322 were obtained from R. Fluhr (Fluhr et al. 1983). The recombinant plasmid pNT-32 that contains the psbA gene coding for the tobacco chloroplast 32000 dalton protein was obtained from L. Bogorad (personal communia~tSon). Isolation of spinach ct DNA fragments was accomplished by digesting a mixture (20 vg) of plasmids that contained an equimolar amount of the following PstI spinach ct fragments: 17.3 Kbp, 13.5 Kbp, 12.3 Kbp, 8.9 Kbp, 8~i Kbp, 7.7 Khp, 2.7 Kbp and 1.9 Kbp with Pstl (see Figure i). The tobacco ct DNA fragment mixture was prepared by digesting

42 a mixture (28 ~g) of plasmids containing equimolar amounts of the following tobacco PstI ct DNA fragments: 20.2 Kbp, 19.2 Kbp, 8.2 Kbp and 6.9 Kbp (Figure i). The ct DNA fragments of either mixture were then freed of the vector DNA by preparative electrophoresis on a 0.6% agarose column in a continuous elution electrophoresis device (PREP-GEL, Bethesda Research Laboratories). The 1 Kbp ct DNA fragment from within the coding region of the Nicotiana tabacum psbA gene was prepared by digestion of pNT32 (200 ~g) with EcoRI and subsequent isolation of the 3.0 Kbp insert by electrophoresis on low temperature gelling agarose (Sea-Plaque - FMC Corporation) following the method of Schmitt and Cohen (1983). The 3.0 Khp fragment so obtained was further digested with restriction endonucleases fnuDII and XbaI to yield the desired 1 Kbp fragment (as based on sequence data in Zurawski et al. 1982),-which was then isolated by electrophoresis on low melting agarose as above. Preparation of Nitrocellulose Membranes Containing Ct DNA Mixtures of either tobacco or spinach cloned ct DNA were prepared by mixing the plasmids containing the proper ct DNA fragments (see above) in equimolar amounts. After brief alkali denaturation, the DNA was diluted to 1 ~g/ml with 6 x SSC (i x SSC consists of 0.15 M NaCI~ 0.015 M Na-Citrate) and adsorbed onto nitrocellulose membranes (Sartorius type SM PI2; 2~5 cm dia.; 0.45 ~m pore size) according to Gillespie and Spiegelman (1965). Each membrane filter had the equivalent of 5 ~g of ct DNA adsorbed to it. Radio-labeling of Ct DNA Fragments Equimolar mixtures of purified spinach or tobacco ct DNA fragments (see ~ o v e ) free of pBR322 DNA were radio-labeled with ~ P-deoxycytidine triphosphate by random primer extension using DNA polymerase Klenow fragment (Eoehringer Mannheim) as described ~ Summers (1975). Specific radioactivity of_the -P-labeled ct DNA was routinely 0.5-1.2 x i08 cpm/~g of DNA. The average length of the labeled et DNA fragments was approximately 500 bp as determined by electrophoresis on 2% agarose gels. Extraction and Purification of N. tabacum Cellular DNA Isolation of cellular DNA from N. tabacum suspension cultures was carried out by a modification of the procedure reported by Murray and Thompson (1980). Cells (typically 0.5 - 1.5 g fresh weight) were washed on course sintered glass filters with two volumes of breaking buffer [50 mM Tris-HCl (pH 8.0), 20 mM EDTA, 1 M NaCI], resuspended in 5 ml of the same buffer and passed through a French pressure cell at 20000 psi. The resulting extract was clarified by centrifugation at 20000 x g for 15 minutes. Two volumes of cold ethanol were added to the supernatant and the precipitate collected by centrifugation. The pellet was resuspended in a minimal volume of 50 mM Tris-HCl (pH 8.0) containing 2 mM EDTA and 0.5% sodium dodecylsulfate (SDS). Proteinase K (E. Merck, Darmstadt) was added to a final concentration of 150 ~ / m l and the solution was incubated at 37°C for two hours. It was then extracted with phenol:chloroform (i:I vol/vol) saturated with i0 mM Tris-HC! (pH 8.0) containing 1 mM EDTA and the aqueous phase was precipitated with two volumes of cold ethanol. The precipitate was resuspended in 2-3 ml of cetyltrimethylammonium bromide (CTAB) buffer containing 1% CTAB, 50 mM Tris-HCI (pH 8.0), i mM EDTA, 0.7 M NaCI and 1 mM B-mercaptoethanol. After extraction with chloroform:isoamyl alcohol (24:1 vol/vol), the aqueous phase was removed and precipitated with CTAB precipitation buffer [1% CTAB, 50 m~ Tris-HCl (pH 8.0), 2 mM EDTA, 1 mM B-mercaptoethanol]. The DNA obtained by this proCedure was collected by

centrifugation, resuspended into ELUTIP-d high salt buffer [i M NaCI, 20 mM Tris-HCl (pH 7.4), 1 m~ EDTA] and reprecipitated with 2 volumes of cold ethanol. The resulting precipitate was resuspended into ELUTIP-d low salt buffer [0.2 M NaCI, 20 mM Tris-HCl (pH 7.4), 1 mM EDTA] and passed over an ELUTIP-d column (Schleicher and Schuell). After washing the column with 2-3 volumes of low salt buffer, the DNA was eluted with 0.4 ml of ELUTIP-d high salt buffer and precipitated by addition of 2 volumes cold ethanol. The resulting pellet accounted for 20-40% of the DNA in the initial cell free extract. To rid the DNA of any possible contaminating RNA and to shear the DNA, the DNA pellet was resuspended into 0,22 ml of 0.3 M Na0H in a 1.5 ml Eppendorf tube (Brinkmann Instrument, Inc.) and placed in boiling water for 3 minutes. It was then cooled in an ice bath and neutralized by addition of 0.012 ml Tris-HCl (pH 8.0) and 0.02 ml of 3.3 M HCI. The final pH was between 7.0 and 7.5~ Electrophoresis on 2% agarose gels demonstrated the length of the DNA to range from 400-800 nucleotides long. Hybridization Nitrocellulose membranes containing a mixture of plasmids equivalent to 5 vg of ct DNA were prehybridized in 20 ml glass scintillation vials with sealable tops. Two ml of prehybridization fluid [50% formamide, 3 x SSC, i x Denhardt solution, 0.1% SDS, 1 mM EDTA (Denhardt 1966)] was added to each vial containing a membrane on which ct DNA had been bound and a blank membrane containing no DNA. The vials were then placed into a 37°C water bath for 4 hours. The prehybridization fluid was removed by aspiration and replaced with hybridization fluid (prehybridization fluid containing 50 ~g/ml denatured calf thymus DNA). For hybridizations between two oned ct DNAs, 0.i pg of the appropriate denatured -P-labeled ct DNA (approximately i0000 cpm) was added to,the scintillation vial. For hybridization between ~H-labeled total cell~lar DNA and cloned ct DNA, 0.7 ~g of the denatured H-cellular DNA was added tonthe v i a l . In addition, 0.01 pg (about i000 cpm) of B2P-labeled ct DNA fragments identical to those bound on the membrane was added to act as an internal control. The vials were resealed and placed at 37°C for 44 hours after which the membrane filters were removed and washed twice in 2 x SSC containing 0pl% SDS at room temperature followed by two washes with 0.2 x SSC containing 0.1% SDS at 37°C. The membrane filters were blotted and dried under an incandescent lamp and counted in Beckman RediSolv-NA scintillation fluid. The control hybridizations between 100% homologous DNAs usually showed a hybridization efficiency of about 60-70% and the observed experimental hybridizations were corrected accordingly. Observed hybridization values were also corrected for the percentage of the chloroplast genome for which no equivalent fragment existed on the nitrocellulose membrane. Corrections were carried out as described by the following equation:

~

Corrected Hybridization

(%)

Observed Hybridization of Experimental DNA (%) = ~Fraction of] F r a c t i o n of the |Internal | IChl°r°plast Genome |Control | X Present in Membrane 132p-DNA 1 [Bound DNA [Hybridized J

In all hybridizations, the quantity of input chloroplast DNA was never greater than one-tenth of chloroplast DNA bound to the membrane. The stringency of the hybridization conditions was calculated to be equivalent to 72°C in 100% aqueous media containing 0.5 ~! NaCI. The reproducibility of the hybridization was determined to be + 10%.

43 Results and Discussion Use of Molecular Hybridization to Determine Cross-Hybridization of Tobacco and >pinach Cloned Ct DNA Fluhr and Edelman (1981) have shown considerable conservation of nucleotide sequence arrangement between the chloroplast genomes of Niootiana tabacum and Spinaoia oleraoeae. However, small scale rearrangements can be a frequent occurence within the genomes of chloroplasts (Palmer and Thompson 1981). Such rearrangements or base substitutions could significantly effect cross-hybridization between the two chloroplast genomes. Since some of our earlier work required hybridization between tobacco total cellular DNA and cloned spinach ct DNA, we decided to evaluate the extent of cross-hybridization between positionally similar cloned fragments of the tobacco and spinach chloroplast genomes. Figure 1 contains the PstI endonuclease restriction site maps of the tobacco and spinach chloroplast genomes. Cloned DNA fragments from each of the chloroplast genomes (as indicated in the legend to Figure i) were combined to make an Sp,nach

PstI RESTRICTION MAP OF CHLOROPLAST GENOME Fig. i. PstI restriction endonuclease cleavage site maps o~ spinach (Crouse et al. 1978) and tobacco (Fluhr et al. 1983) chloroplast genomes. The tobacco ct DNA fragment mixture referred to in the text consisted of the following fragments: 20.2 Kbp, 19.2 Kbp, 9.3 Kbp, 8.2 Kbp and 6.9 Kbp. The spinach ct DNA fragment mixture consisted of the following: 17.3 Kbp, 13.5 Kbp, 12.3 Kbp, 8.9 Kbp, 8.1 Kbp, 7.7 Kbp, 2.7 Kbp and 1.9 Kbp fragments. The inverted repeat regions (IR) are denoted by the solid black arcs and genes coding for the ribosomal RNA species are indicated within the repeated regions. ...................................................... equimolar mixture of tobacco ct DNA fragments representing 46.0% of the tobacco chloroplast genome and an equimolar mixture of spinach ct DNA fragments representing 47.6% of the spinach chloroplast genome. The two cloned ct DNA mixtures have in common fragments that represent about 41.5% of the total chloroplast genome, most of which are located in the large single copy region of the genome (Figure i). The cloned ~ b a c c o ct DNA mixture was then used to prepare a ~ P-labeled probe that was subsequently hybridized to the plasmid mixture containing spinach ct DNA bound to a nitrocellulose membrane. The estimated percent of probe in common with the membrane bound

spinach ct DNA was 89.7 as determined by comparison of published maps of the two ct genomes. Conversely, ~ cloned spinach ct DNA mixture was labeled wfth and used as a probe for hybridization with the plasmid mixture containing tobacco ct DNA immobilized on nitrocellulose. The estimated percent of probe in common with the membrane bound tobacco DNA was 87.3. Cont~@l hybridizations were conducted between each of the ~-P-labeled DNA mixtures and its unlabeled plasmid counterpart bound to a nitrocellulose membrane. An average direct value of 60-70% hybridization between identical DNAs (data not shown) was observed using the hybridization conditions detailed in " ~ t e r i a l s and Methods." As seen in Table 1 the tobacco ct DNA mixture and the spinach ct DNA mixture cross-hybridized to each other with about 75% efficiency (74.6% for the spinach ct DNA to tobacco ct DNA and 78.8% for the tobacco ct DNA to spinach ct DNA). This hybridization ...................................................... Table i. Cross-hybridization of spinach and tobacco ct DNA. Corrected DNA Bound to Efficiency of b Nitrocellulose Hybridization Probe Membrane (%) Plasmids containing Cloned tobacco cloned spinach • a • a ct DNA mlxture ct DNA mlxture 78.8 Plasmids containing Cloned spinach cloned tobacco • a . a ct DNA mlxture ct DNA mlxture 74.6 Plasmid with 8.9 Kbp spinach ct DNA fragment 1 Kbp tobacco ct containing the DNA fragment from entire coding within the coding region for the region for the 95.4 32Kd protein 32Kd protein asee "Materials and }iethods" for composition of mixtures anJ hybridization conditions. The probes (0.i ng~ i0 cpm) did not contain the vector pBR322 ~NA. Calculated as described in "Materials and Methods." .................... - ................................ value lends support to the suggestion by Fluhr and Edelman (1981) that numerous small scale rearrangements have occurred in the tobacco and spinach chloroplast genome and reflects a significant difference in the nucleotide sequence of two chloroplast genomes. Cross-Hybridization of the Chloroplast psbA Gene Codinz for the 32000 Dalton Protein To further test the reliability of the hybridization procedure, the 1 Kbp DNA fragment from within the cQding region of the N. tabacum psbA gene%~oding for th~ 32000 dalton chloroplast protein was --P-labeled and hybridized to a nitrocellulose bound 8.9 Kbp spinach at DNA fragment that contains the entire coding region for the 32000 dalton protein (Zurawski et al. 1982). The coding sequence of the chloroplast gene (psbA) coding for the chloroplast 32000 dalton protein of photosystem II is highly conserved in higher plants. The nucleotide sequence for this gene in Nicotiana debneyi is 95.7% homologous to that in spinach (Zurawski et al. 1982). Since the restriction endonuclease pattern with EooRI, Fn~DII, XbaI and PstI of the M. tabacum psbA gene is identical to that of N~ debneyi (unpublished data) it is likely that the psbA gene of N. tabacum also shares the same homology to the corresponding gene in the spinach chloroplast genome. As seen in Table i, the tobacco probe hybridizes to the membrane bound spinach ct DNA fragment with 95% efficiency (corrected value), indicating the good correlation between sequence homology and hybridization efficiency observed under

44 these conditions.

Table 3. £ells. --

H~bridization of Total Cellular Nicotiana*abacmm Wisconsin 38 DNA to Membrane Bound Cloned Ct DNA from Tobacco or Spinach The use of nitrocellulose bound plasmids containing cloned spinach ct DNA fragments to measure ct DNA levels in tobacco cells ~as evaluated by extracting total cellular DNA from ~H-thymidine labeled suspension cultures of Niootiana tabaoum Wisconsin 38 and hybridizing it to both spinach and tobacco cloned ct DNA bound to nitrocellulose membranes. Table 2 shows that 8.3% (corrected value) of the total ..........................

~ .............

Cells Wisconsin 38 Yamada Yamada Yamada Yamada

genome copy number in tobacco

Growth Conditions Ct Genome/Cell Dark, + Sucrose, Air 4800 Dark, + Sucrose, Air 3300 Light, + Sucrose, Air i0000 Light, - Sucrose, 5% CO 2 9500 Light, + Sucrose, 5% CO 2 12000

Plant Tissue Leaf Roots

2600a-5800 b