Planta
Planta (1985)164:459-472
9 Springer-Verlag 1985
Structure, composition, and distribution of plastid nucleoids in Narcissus pseudonarcissus P. Hansmann*, H. Falk, K. Ronai and P. Sitte Institut ffir Biologic II, Zellbiologie, Schfinzlestrasse 1, D-7800 Freiburg/Brsg., Federal Republic of Germany
The size, frequency and distribution of the nucleoids of chloroplasts (cl-nucleoids) and chromoplasts (cr-nucleoids) of the daffodil have been investigated in situ using the DNA-specific fluorochrome 4'6-diamidino-2-phenylindole. Chromoplasts contain fewer nucleoids (approx. 4) than chloroplasts ( > 10), and larger chromoplasts (cultivated form, approx. 4) contain more than smaller ones (wild type, approx. 2). During chromoplast development the nucleoid number decreases in parallel with the chlorophyll content. Each nucleoid contains 2-3 plastome copies on average. In chloroplasts the nucleoids are evenly distributed, whereas they are peripherally located in chromoplasts. The fine structure of isolated cl- and crnucleoids, purified either by Sepharose 4B-CL columns or by metrizamide gradients, was investigated electron microscopically. The cl-nucleoids consist of a central protein-rich core with 'naked' DNA-loops protruding from it. In cr-nucleoids, on the other hand, the total D N A is tightly packed within the proteinaceous core. The protein-containing core region of the nucleoids is made up of knotty and fibrillar sub-structures with diameters of 18 and 37 nm, respectively. After proteinase treatment, or increasing ion concentration, most of the proteins are removed and the D N A is exposed even in the case of cr-nucleoids, the stability of which proved to be greater than that of cl-nucleoids. The chemical composition of isolated plastid nucleoids has been determined qualitatively and quantitatively. Chromoplast-nucleoids contain, relative to the same D N A quantity, about six times Abstract.
* To whom correspondence should be addressed Abbreviations: cl-=chloroplast; cr-:chromoplast; D A P I =
4'6-diamidino-2-phenylindole; DNase = deoxyribonuclease; kDa = kilodaltons; MG = purified by metrizamide gradients; SC=purified by Sepharose CL-4B column gel filtration; SDSPAGE: sodium dodecylsulfate-polyacrylamide gel electrophoresis
as much protein as cl-nucleoids. Accordingly the buoyant density of cr-nucleoids in metrizamide gradients is higher than that of cl-nucleoids. In addition to D N A and protein, R N A could be found in the nucleoid fraction. No pigments were present. The cr- and cl-nucleoids have many identical proteins. There are, however, also characteristic differences in their protein pattern which are possibly related to the different expression of the genomes of chloroplasts and chromoplasts. Nucleoids of both plastid types contain some proteins which also occur in isolated envelope membranes (probably partly in the outer membrane) and thus possibly take part in binding the D N A to membranes. Key words:
(nucleoid) -
Chloroplast (nucleoid) - Chromoplast Narcissus - Plastid (envelope, DNA).
Introduction
In bacteria, mitochondria, and plastids, the genetic material (i.e. circular double-stranded DNA) is located within nucleoids. According to electron microscopical investigations in situ, the structure of these DNA-containing areas appears to be similar in all prokaryotic systems (Ris and Plaut 1962; Kellenberger etal. 1958; Bisset 1970; Nass and Nass 1962; Hermann and Kowallik 1970). In fact, the fine structure of nucleoids is so different from that of a eukaryotic cell nucleus that the systematic separation of living organisms into pro- and eukaryotes could be based solely on this criterion. Nucleoids in plastids have been investigated repeatedly by use of electron-microscopical, radioautographic, and cytochemical methods (Woodcock and Fernfindez-Morfin 1968; Kowallik and Herrmann 1972a; Knoth et al. 1974; Rose and Poss-
460
ingham 1976; James and Jope 1978; Scott and Possingham 1980; Kuroiwa etal. 1981; D6rle 1981; Possingham et al. 1983). It has been found that chloroplasts contain different numbers of nucleoids depending on their degree of differentiation. These nucleoids in turn contain variable numbers of plastome copies. Only recently has it been possible to isolate and characterize DNA-protein complexes (nucleoids) from chloroplasts. They are, after suitable isolation, transcriptionally active (Hallick et al. 1976; Briat et at. t979); they appear highly condensed with a beaded substructure, and have a complex protein composition (Briat et al. 1982). When isolated using strongly deproteinizing conditions, the DNA, as visualised after spreading preparations in the electron microscope, is in the form of a rosette with a central body holding the otherwise naked D N A together (Herrmann et al. 1974; Yoshida etal. 1978; Yoshida 1981). As has been shown earlier by electron microscopy of lysed chloroplasts, a large portion of the plastid D N A is membrane bound (Woodcock and Fern/mdezMorfin 1968; Herrmann etal. 1974; Rose and Possingham 1976; Rose 1979; Rose and Lindbeck 1982). It has been suggested that the central body of the nucleoids might be responsible for their binding to chloroplast membranes (Herrmann et al. 1974; Yoshida et al. 1978; Briat et al. 1982). In the present work, the comparative number and distribution of nucleoids in chloro- and chromoplasts of the daffodil have been investigated in situ by fluorescence microscopy. For further investigation the nucleoids from both plastid types have been isolated by two different methods, and characterized light and electron microscopically as well as biochemically. Chloroplasts and chromoplasts of the same plant offer the opportunity to investigate plastid nucleoids in quite dissimilar states of activity. This is particularly true in the case of the daffodil (Narcissus pseudonarcissus L.), the leaves and coronae of which provide suitable raw material for the mass isolation of chloroplasts and chromoplasts, respectively. Chloroplasts and chromoplasts of the daffodil differ fundamentally in pigmentation, fine structure, and composition (Liedvogel et al. 1976). The protein complement of the two plastid types differs significantly (Liedvogei and Sitte 1974; Hansmann and Sitte 1984). Although leaf chloroplasts and flower chromoplasts of N. pseudonarcissus share many enzyme activities associated with lipid metabolism (Liedvogel and Kleinig 1976; Kreuz and Kleinig 1981), it appears that in chromoplasts (in contrast to chloroplasts) no transcription or pro-
P. Hansmann et al. : Plastid-nucleoids of the daffodil
tein synthesis occurs (data not shown). As the plastome of both plastid types is apparently identical (Thompson 1980), modifications in the plastid DNA cannot be responsible for this differential gene expression, As will be shown in the present paper, there are, however, structural differences between the DNA-protein complexes of the two plastid types which are mirrored in the protein composition of the nucleoids. Materials and methods For mass isolation of chloroplasts and chromoplasts the daffodils (Narcissus pseudonarcissus L. cv. 'Golden Harvest') were bought from the local market. Wild-type material (collected near G6rardmer, France) or cultivated forms (cultivated outdoors) were used for in situ investigations. Ampholines (pH range 3-10), desoxyribonuclease (DNase) I, 4'6-diamidino-2phenylindole (DAPI), and illetrizamide were obtained from Serva, Heidelberg, FRG. Sepharose CL-4B was purchased from Pharmacia, Uppsala, Sweden. Staphylococcus aureus V8 protease was obtained from Miles Laboratories, USA. Proteinmolecular-weight-markers for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were from Boehringer, Mannheim, FRG, and Sigma, St. Louis, Mo., USA.
Isolation ofplastids (All steps at 4 ~ C) Chromoplast isolation. For isolation of chromoplasts, daffodil coronae (125 g) from open flowers were homogenized at maximum speed in a Waring Blendor (3 • 3 s) in the presence of 300 ml buffer A (0.33 M sorbitol, 50 mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)/HC1 pH 8.0, 3 mM disodium ethylenediaminetetraacetic acetate (NazEDTA), i mM 2-mercaptoethanol, 0.2% (w/v) polyvinylpyrrolidone). After filtration through three layers of fine-mesh nylon cloth and centrifugation at 1000 g for 5 rain, the chromoplasts were pelleted from the supernatant at 17000 g for 20 rain. The pellet was resuspended in 19 ml buffer A and 1 m10.2 M MgC1 z, and 1.5 mg DNase I was added. The suspension was incubated for i h, and then the digestion of nuclear D N A stopped by addition of 40 ml buffer B (0.33 M sucrose, 50 m M Tris-HC1 pH 8.0, 20 mM NazEDTA). The chromoplasts were pelleted once more (20 min at 17000g) and further purified by a flotation step as detailed by Liedvogel et al. (1976), but using buffer C (25 m M Tris-HC1 pH 8.0, 10 mM N a z E D T A ) instead of the one previously described. The final pellet was resuspended in buffer D (0.33 M sucrose, 50 mM Tris-HC1 pH 8.0, 4 mM EDTA). The chromoplasts were stored frozen at - 7 0 ~ C prior to use. Chloroplasts were isolated according to Kolodner and Tewari (t975) with slight modifications. Fresh daffodil leaves (660 g) were washed and homogenized in 1.8 1 buffer E (0.6 M sorbitol, 50 m M Tris-HC1 pH 8.0, 3 mM Na2EDTA, 1 mIVi 2-mercaptoethanol, 0.1% (w/v) bovine serum albumin) with a Waring Blendor (5 x 3 s at maximum speed). The slurry was filtered through fine-mesh nylon tissue and the filtrate centrifugated at ~ 500 g for 10 min. The pellet was resuspended in approx. 80 ml buffer E, and after centrifugation at 130 g for 30 s, tile chloroplasts were pelleted from the supernatant at 4000 g for 90 s. Then, the pellet was resuspended in 28.5 ml buffer E and, after addition of 1.5 ml 0.2 M MgC12 and 1.5 mg DNase I, the suspension was incubated for 30 min for nuclear-DNA degradation. The reaction was stopped by adding 70 ml of
P. H a n s m a n n et al. : Plastid-nucleoids of the daflbdil buffer F (0.6 M sucrose, 50 m M Tris-HCl pH 8.0, 20 m M Na2EDTA), the chloroplasts were repelleted at 1500g for 15 min and resuspended in buffer F. The suspension was overlayered onto a continuous sucrose gradient (1.0 M-1.9 M sucrose in buffer C), and centrifuged at 80 000 g for 2 h in a swinging-bucket rotor. The lower green band was then removed and diluted 1:1 with buffer F. The chloroplasts were pelleted (20 min at 3000 g), and for storage at - 7 0 ~ resuspended in buffer G (0.6 M sucrose, 50 m M Tris-HC1 pH 8.0, 4 m M EDTA). Isolation of chloroplast and chromoplast integral envelope and thylakoid proteins was carried out as described by Hansm a n n and Sitte (1984).
Isolation of nucleoids Nucleoids from both chloroplasts and chromoplasts were obtained by two different procedures. The first was adapted from the method used by Hallick et al. (1976) for isolation o f a DNAprotein complex from chloroplasts of Euglena by Sepharose 4B column gel filtration (SC). All solutions were filter-sterilized (Millipore, Molsheim, France; SLHA 025BS). The chromoand chloroplasts were pelleted from the thawed suspensions (using only half the isolate) by centrifugation for 20 rain at 17 000 g and 4 000 g, respectively. The plastids were resuspended in 5 ml buffer H (25% (w/v) glycerol, 20 m M triethanolamineHC1 pH 7.5, 0.1 m M EDTA, 1 m M 2-mercaptoethanol, 1% (w/v) Triton X-100, 0~5mM phenylmethylsulfonylfluoride (PMSF)) and lysed for 5 min at 22 ~ C. After centrifugation (10 rain, 3000 g, 40 C), 1/40 vol. of 4 M (NH~)2SO 4 was added to the supernatant. The solution was loaded through a small Sepharose 4B-C1 pre-column, freshly prepared in each case and equilibrated with buffer I (buffer H plus 0.1 M (NH~)zSO4), onto the separation column equilibrated with buffer I (but with P M S F omitted). Elution was carried out with the same buffer. Nucleoid-peak fractions were detected by DAPI epifluorescence microscopy (see below). For the second isolation method, the plastids were pelleted as described above and lysed by resuspending the pellets in buffer J (0.9 M metrizamide, 20 m M triethanolamine-HC1 pH 7.5, 0.1 m M EDTA, 0.4 m M CaC12, 2% (w/v) Triton X-100, 10% (w/v) glycerol, 0.5 m M PMSF), and by incubation for 5 min at 22 ~ C. Onto this solution (approx. 0.8 M metrizamide) a discontinuous metrizamide gradient (0, 0.3, and 0.6 M metrizamide in buffer K (10% (w/v) glycerol, 20 m M triethanolamine-HC1 pH 7.5, 0.1 m M EDTA, 0.4 m M CaCI 2, 1% (w/v) Triton X-100) was overlayered. After centrifugation for 5 h at 80000 g in a swinging-bucket rotor, the nucleoids (0.3/0.6 M metrizamide step) detected by DAPI epifluorescence microscopy were removed.
Preparation of isolated nucleoidsfor microscopy Nucleoids purified by centrifugation in metrizamide gradients (MG) were dialysed against buffer K without Triton X-100 by placing a 60-lal drop on a membrane filter (Millipore VMWP, 50 nm average pore size) as described by Marusyk and Sergeant (1980). This nucleoid (MG) fraction, or else the alternatively isolated nucleoids (SC), were used directly or incubated in the same medium for 1 h either at 22~ with 50 lag m l - 1 Proteinase K (DNase inactivated, compare Young and Sinsheimer 1967) or with 0.5 M NaCI (high salt) at 4 ~ C for protein depletion. The DNA-protein complexes were then fixed for 30 rain at 4 ~ C by addition of 1/40 vol. of 20% (w/v) glutardialdehyde. Drops (20 lal) placed on membrane filters as described above were dialysed for 2 h against buffer L (0.15 M NaCI, 10 m M Tris-HCt pH 7.5, 0.1 m M Na2EDTA ).
461
Preparation of isolated nucleoidsfor biochemical analyses Nucleoids (SC) were pelleted from the peak fractions at 120000 g and 4 ~ C for 16 h in a fixed-angle rotor. The pellets were resuspended in double-distilled water and stored at - 7 0 ~ C. Nucleoids (MG) were either used directly for determination of the buoyant density (see below) or the nucleoid band was diluted 1:3 with buffer K and the mixture was layered onto a 2.5-ml 0.8 M sucrose cushion in buffer K. After centrifugation in a 12-ml fixed-angle rotor at 200000g and 4 ~ for 3 h, the pellet was resuspended in double-distilled water and stored at - 7 0 ~ C prior to use.
Fluorescence microscopy The observations were made with a Zeiss (Oberkochen, F R G ) IM 35 inverted epifluorescence microscope equipped with Neofluar phase optics and a HBO 50 lamp. Photographs using the filter combinations BP 365/FT 395/LP 397 (DAPI) and BP 436/FT 460/LP 470 (chlorophyll) on K o d a k Tri-X pan (400 ASA) film. For in situ staining of the nucleoids (James and J o p e / 9 7 8 ; Coleman 1978), small pieces of the epidermis either from green leaves (chloroplasts) or from the coronae (chromoplasts) of the daffodil were vacuum-infiltrated with buffer M (0.3 M sucrose, 10 gl ml i DAPI, 5 m M magnesium acetate) and observed in the same medium. Isolated chromoplasts were fixed in buffer N (I % glutardialdehyde, 50 m M Na-cacodylate buffer pH 7.2, 0.33 M sucrose) for 2 h, pelleted for 2.5 rain at 2 500 g and resuspended in buffer 0 (1.2 M sucrose, 50 m M Tris-HC1 pH 7.5, 20 m M N a z E D T A , 2 gg m1-1 DAPI). A 10-pl aliquot of this suspension was put directly onto a slide and observed. In the case of isolated chloroplasts the same procedure was applied, but using a higher sucrose concentration (0.6 M). Isolated nucleoids were first fixed and dialysed as described above, 1/10 vol. 20 m M spermidine-HC1 was added in order to favour adsorption to the cover slip, and 10 gl of the samples were mixed with 2 lal of a 2 gg m l - 1 DAPI solution on the slide.
Electron microscopy The whole-mount method as described by Miller and Beatty (1969), was modified as detailed by Suzuki et al. (1980). In a Lucite microcentrifugation chamber, 10-20 pl of the prepared samples were layered onto a sucrose cushion (1 M sucrose, 2 0 m M triethanolamine-HC1 pH 7.5, 0.1 m M Na2EDTA , 0.15 M NaCI) and the DNA-protein complexes were pelleted onto a carbon-coated grid (previously activated by glowdischarge) for 5 min at 10000 g and 4 ~ C in a 5-ml swingingbucket rotor. The grids were washed twice for 30 s in doubledistilled water, stained with 1% (w/v) uranyl acetate for 30 s, and dehydrated by a graded ethanol series (0, 25, 50, 90, and 100%; 10 s each). The grids were rotary-shadowed with platinum/palladium and observed in a Zeiss EM 10CR at 60 kV. For the direct-mounting technique (Griffith and Christiansen 1978), 9 gl of the pretreated samples were mixed with 1 gl of 20 m M spermidine-HC1 and a 5-gl aliquot was placed on activated carbon-coated grids. After adsorption of the DNAprotein complexes for 3 rain, the grids were washed, stained, dehydrated, and rotary-shadowed. For negative staining, samples prepared as described for the direct-mounting technique were air dried after removing most of the staining solution from the grid with filter paper and immediately observed in the electron microscope. In some cases fluorescence and electron microscopy were combined in order to identify the nucleoids unequivocally in the electron
462
P. H a n s m a n n et al. : Plastid-nucleoids of the daffodil
Fig. 1 a, b. Epifluorescence photographs of plastid-nucleoids of different tissues of Narcissus pseudonarcissus after staining with DAPI. a Coronae of open flowers (chromoplasts); b green leaves (chloroplasts). B a r s = 10 pm, x 1 200 Fig. 2a, b. Isolated plastids, nucleoids stained with DAPI. a chromoplasts; b chloroplasts. Bars = 10 gm, x 1200 Fig. 3a-d. Isolated nucleoids stained with DAPI. a from chromoplasts (MG); b from chromoplasts (SC); c from chloroplasts (MG); d from chloroplasts (SC). B a r s = 10 pm, x 1200 Fig. 4a, b. Nuclcoids out of chromoplasts purified by gel filtration, a DAPI epifluorescence photograph; b the same area shown by electron microscopy after shadowing with Pt-Pd. Bars = 2 ~tm, x 5000
microscope. The nucleoids were allowed to adsorb to carboncoated formvar films (on electron-microscope-grids containing an asymmetric centre mark) as described above. After washing the grids for 30 s in double-distilled water, they were carefully mounted onto a slide with a drop of 0.2 gg ml - i DAPI solu-
tion. Suitable positions of the preparation were rapidly photographed and noted. The grid was then removed, washed, stained, dehydrated, and shadowed as above. The meshes corresponding to the ones observed in the light microscope could easily be detected in the electron microscope.
P. Hansmann et al. : Plastid-nucleoids of the daffodil
Determination of the buoyant density of nucleoids ( MG )
463 Table 1. Number of nucleoids per plastid in Narcissus pseudo-
narcissus An aliquot (equivalent to 10-20 gg DNA) of the nucleoid band (see above) was diluted 1:4 with buffer K and layered onto a discontinuous metrizamide density gradient containing steps of 0.2, 0.3, 0.4, 0.5, and 0.6 M metrizamide in buffer K. After centrifugation for 14 h at 150000 g and 4 ~ C in a 12-ml fixedangle rotor, the tube was punctured at the bottom, and fractions of approx. 400 gl were collected. Nucleoid-containing fractions were detected by DAPI fluorescence microscopy. The density of each fraction was determined from the refractive index after correction for the buffer used as described by Rickwood (1978).
Electrophoresis of the proteins One-dimensional electrophoresis. The chloroplast, chromoplast, and different membrane pellets were resuspended in doubledistilled water. The protein concentration of all samples was determined according to Bradford (1976). For SDS-PAGE, 0.5 vol. buffer P (90 mM Tris-HC1 pH 7.6, 6% (w/v) SDS, 15% (w/v) 2-mercaptoethanol, 15% (w/v) glycerol, 0.1% (w/v) bromophenol blue) was added to the samples containing 10-30 gg protein, and the solutions were boiled for 2 min. The proteins were separated on 8-18 % polyacrylamide gradient gels (1.5 mm thick slab gels) with a 4% polyacrylamide stacking gel using the buffer system of Chua (1980). For molecular-weight determinations 10 and 15% polyacrylamide linear separating gels and myosin (M r 205 000), fl-galactosidase (116 000), phosphorylase (97400), bovine serum albumin (66000), ovalbumin (45 000), carbonic anhydrase (29 000), trypsin-inhibitor (21 500), cytochrome c (12 500), and aprotinin (6 500) were taken as standards. Fixed and washed gels were finally stained by the very sensitive silver staining procedure of Bfirck et al. (1983).
n.d. = N o data. a-d) Different stages of chromoplast development characterized by the chlorophyll content (red-fluorescence): a) much (prochromoplasts); b) medium; c) low; d) no (chromoplasts) Mean value
Extreme values
Plastids investigated
Leaves : chloroplasts
n.d.
10-30
n.d.
Coronae: a) b) c) d)
4.3 3.7 2.7 2.2
1-9 1-7 1-6 0-7
61 75 37 64
10-30
n.d.
0-10
200
10-30 0-10
n.d. 105
Wild type
Culture form Leaves: chloroplasts n.d. Coronae: chromoplasts 3.7 Isolated plastids : chloroplasts n.d. chromoplasts 3.2
Table 2. Sizes (gm) of the nucleoids from N. pseudonarcissus
Light microscope DAPIfluorescence
Electron microscope
Tissue: chloroplasts chromoplasts
0.3-1.0 0.3-1.5
-
-
Isolated plastids: chloroplasts chromoplasts
0.3 1.0 0.3-2.0
-
-
0.3-2.0 0.3-2.0 0.3-3.0 0.3-2.0
0.3 2.5 0.3-2.0 0.34.0 0.3-2.5
0.3 2.5 0.3-2.0 0.3-3.5 0.3-3.0
Two-dimensional electrophoresis. To the aqueous protein solutions (see above), 0.i vol. buffer Q (100 mM Tris-HCl pH 7.5, 70 mM MgC12, 20 gg m l - 1 DNase I) were added, and the mixture was incubated for 1 h at 4 ~ C for D N A degradation. After addition of 0.1 vol. 5% SDS and incubation for 10 min at 70 ~ C, solid urea, Ampholines, 2-mercaptoethanol, and Triton X-100 were added to the samples to bring the concentrations to 9.5 M, 2%, 5%, and 4%, respectively. This procedure is a modification of the method of Ames and Nikaido (1976) and Siegenthaler and Nguyen (1983). Isoelectric focusing was as described by O'Farrell (1975), but with reversed polarity (acidic reservoir on top) as proposed by Righetti et al. (1982). After focusing, the gels were equilibrated twice for 20 min with buffer R (30 mM Tris-HC1 pH 7.6, 2% SDS, 5% 2-mercaptoethanol, 5% glycerol) and loaded onto SDS-polyacrylamide gradient gels (see above) as described by O'Farrell (1975).
After After shadowing negative staining
In situ
Isolated cl-Nucleoids (MG) cr-Nucleoids (MG) cl-Nucleoids (SC) cr-Nucleoids (SC)
Peptide mapping by limited proteolysis. The protocol was essentially the same as the one described by Cleveland et al. (1977). Staphylococcus aureus V8 protease was used for digestion of the proteins.
Quantitative determinations DNA, RNA, and proteins were extracted according to Daniel and Baldwin (1964). D N A was determined by the diphenylamine method (Burton 1956), R N A by the orcinol method (Ogur and Rosen 1950), and protein according to Schacterle and Pollack (1973). The carotenoid content was calculated using the formula of Liaaen Jensen and Jensen (1971). Chlorophyll was estimated as described by Arnon (1949).
Results
Nucleoids from both chloroplasts and chromoplasts were measured and counted after specific staining in situ with the fluorochrome DAPI (Fig. 1, Tables 1, 2). The size of the fluorescent spots and the number of nucleoids per plastid varied considerably. Chloroplasts contained more than ten nucleoids which were evenly distributed throughout the whole plastid (Fig. 1 b). In larger
464
P. H a n s m a n n et al. : Plastid-nucleoids o f t h e daffodil
P. H a n s m a n n et al. : Plastid-nucleoids of the daffodil
465
Fig. 6a, b. Nucleoids (MG) from daffodil chromoplasts (a) and chloroplasts (h) after negative staining. Bars = 0.5 gm, x 40000
chromoplasts of the cultivars, more nucleoids were found than in the smaller wild-type chromoplasts. In contrast to the chloroplasts, the chromoplast nucleoids were always situated in close proximity to the peripheral membrane convolute (Fig. I a). During chromoplast development from prochromoplasts there was a synchronous decrease in chlorophyll content and number of nucleoids per plastid (Table 1). Isolated plastids showed no changes in the form of nucleoids (Fig. 2). Isolated chromoplasts, however, contained on average fewer nucleoids than in situ chromoplasts (Table 1), probably due to a partial loss of nucleoids from injured chromoplasts. During the isolation of the chloroplasts, intact and broken chloroplasts were separated. All intact chloroplasts contained nucleoids, whereas broken or injured ones contained only fragments of nucleoids or none whatsoever. Contamination with nuclear D N A could not be detected in either plastid fraction. Isolated nucleoids showed a compact structure irrespective of the method of purification (Fig. 3),
although they were slightly larger than the in situ nucleoids (Table 2). Presumably, the in situ-visualized fluorescent spots represented only the nucleoid core. In order to identify the nucleoids unequivocally in the electron microscopical preparations, fluorescence and electron microscopy were combined (Fig. 4). When observed in the electron microscope, isolated chromoplast (cr-)nucleoids (MG) consisted, after protein-free spreading and shadowing with Pt/Pd, of electron-dense material in polymorphic forms (Figs. 4, 5 a, c), with no naked D N A visible. In contrast, chloroplast (cl-)nucleoids (MG) revealed an electron-dense, central area with naked D N A projecting in the form of loops or supercoils (Figs. 5b, d). Following proteinase K treatment prior to electron microscopical preparation, the electron-dense material was extensively degraded in both cr- and cl-nucleoids, with naked D N A becoming progressively more exposed (Fig. 5e, f). The same effect could be achieved by increasing the ionic strength of the isolation buffer (Fig. 5g, h). In both nucleoid types, however, the D N A was
(
Fig. 5a-h. Electron micrographs of DNA-protein complexes (nucleoids M G ) from Narcissus pseudonarcissus chromoplasts (a, e, e, and g) and chloroplasts (h, d, f, and h) after protein-free spreading and shadowing, a-d Whole-mount method; e-h directadsorption technique, a, h General view of the nucleoid fraction; e, d nucleoids in detail; e, f after proteinase treatment; g, h after high-salt treatment, a, b Bars = 5 lam, x 4 000; c-h bars = 0.5 ~tm, x 40000
P. H a n s m a n u et al. : Plastid-nucleoids of the daffodil
466 Table 3. Enrichment and composition of daffodil plastidal nucleoids Purification step
Leaves/chloroplasts
Coronae/chromoplasts Pigment a
Protein
RNA
DNA
Pigment b
I. Tissue
1.3
23
1.5
1
II. Plastids
5.7
91
4.4
1
17
III. Nucleoids MG SC
-
I0 6.4
1.1 0.88
1 1
-
1.2
Protein
RNA
DNA
27
1.7
1
250
10.4
1
2.2 0.86
0.58 0.28
1 1
a Carotenoid content b Chlorophyll content
Table 4. Dependence of D N A recovery on the method of nucleoid purification and plastid type Purification method
Chromoplasts
Chloroplasts
MG SC
54% 22%
32% 26%
still held together by proteinase K- and salt-resistant structures, this being more pronounced in the case of cr-nucleoids. When isolated nucleoids were negatively stained with uranyl acetate (rather than rotary shadowed), knotty structures were visible over the whole complex (Fig. 6). Occasionally, smooth filaments of 18 nm diameter were visible which coiled around each other forming fibrils with a diameter of 37 nm. Nucleoids purified by Sepharose gel filtration were not appreciably different from nucleoids purified in metrizamide gradients. The density of isolated nucleoids in metrizamide was 1.21 g c m - 3 for cl-nucleoids, and 1.25 g c m - 3 for cr-nucleoids. As the buoyant density of DNA-protein complexes in metrizamide is known to increase with rising protein contents (Rickwood and MacGillivray 1977), it is to be concluded that cr-nucleoids have a higher protein/ D N A ratio than cl-nucleoids. This expectation was impressively born out by chemical analysis. Some data for the composition of whole tissues (leaves and coronae), isolated plastids, and enriched nucleoids were shown in Table 3. The reference value ( D N A quantity) changed greatly with enzymatic decomposition of nuclear D N A during plastid isolation. In the course of nucleoid purification, all pigments were lost and D N A was enriched with respect to both protein and RNA. This enrichment was more pronounced in the case of column purification, the yield however being smaller (Table 4). The protein/DNA ratio
Fig. 7. Analysis by SDS-polyacrylamide gel electrophoresis of daffodil plastidal proteins. Proteins from isolated chloroplasts (track A); chloroplast-nucleoid proteins (track B); chromoplast-nucleoid proteins (track 6); proteins from isolated chromoplasts (track D); standard proteins for molecular-mass determination (track E)
was about six times higher in cr-nucleoids than in cl-nucleoids. For qualitative protein investigations, nucleoids (SC) were mainly used as these can be isolated more effectively. Comparison of nucleoid proteins with the total protein pattern of the plastids from which they were isolated showed that nucleoid proteins do not belong to the major proteins of the plastids (Fig. 7). The nucleoids from chloroplasts and chromoplasts showed, besides many electrophoretically identical proteins, several
P. Hansmann et al. : Plastid-nucleoids of the daffodil
467 Table 5. Relative molecular mass (M r x 10 -3) of some major proteins of the different nucleoid fractions. Letters and numbers as in Fig. 8. Chloroplast-nucleoid proteins (C); chromoplastnucleoid proteins (D)
Fig. 8. Comparison of the proteins of different subplastidal fractions by SDS-polyacrylamide gel electrophoresis. Integral thylakoid proteins (track A) ; integral chloroplast envelope proteins (track B); chloroplast-nucleoid proteins (track C); chromoplast-nucleoid proteins (track D); integral chromoplast membrane proteins (track E). Black bars indicate proteins with identical electrophoretic mobilities
Protein number
C
D
Protein number
C
D
1 2 3 4 5 6 7 8 9
207 130 110 96 91 73 58 56 29
207 162 130 110 96 91 73 68 42
10 11 12 13 14 15 16 17
22 21 20 18 15
40 32 29 22 21 20 18 15
cr-nucleoid proteins 4, 5, 7, 8, and 12 with the pattern of corresponding envelope proteins provided further evidence for the identity of these respective proteins (Fig. 10). At least three cl-nucleoid proteins (main protein: P6, 73 kDa) could thus be regarded as integral envelope proteins. Of the cr-nucleoids, at least five were envelope proteins (major proteins P7 and PS, 73 and 68 kDa). Three of them were identical to corresponding envelope proteins from chloroplasts. P8 from cr-nucleoids was missing from the chloroplast fraction. As the nucleoid fraction was devoid of fatty acids, and as no DNA could be detected in the envelope membrane fraction, there was virtually no mutual contamination of nucleoid and envelope fractions. Discussion
specific bands. Proteins in the low-molecularweight range (
Planta (1985)164:459-472
9 Springer-Verlag 1985
Structure, composition, and distribution of plastid nucleoids in Narcissus pseudonarcissus P. Hansmann*, H. Falk, K. Ronai and P. Sitte Institut ffir Biologic II, Zellbiologie, Schfinzlestrasse 1, D-7800 Freiburg/Brsg., Federal Republic of Germany
The size, frequency and distribution of the nucleoids of chloroplasts (cl-nucleoids) and chromoplasts (cr-nucleoids) of the daffodil have been investigated in situ using the DNA-specific fluorochrome 4'6-diamidino-2-phenylindole. Chromoplasts contain fewer nucleoids (approx. 4) than chloroplasts ( > 10), and larger chromoplasts (cultivated form, approx. 4) contain more than smaller ones (wild type, approx. 2). During chromoplast development the nucleoid number decreases in parallel with the chlorophyll content. Each nucleoid contains 2-3 plastome copies on average. In chloroplasts the nucleoids are evenly distributed, whereas they are peripherally located in chromoplasts. The fine structure of isolated cl- and crnucleoids, purified either by Sepharose 4B-CL columns or by metrizamide gradients, was investigated electron microscopically. The cl-nucleoids consist of a central protein-rich core with 'naked' DNA-loops protruding from it. In cr-nucleoids, on the other hand, the total D N A is tightly packed within the proteinaceous core. The protein-containing core region of the nucleoids is made up of knotty and fibrillar sub-structures with diameters of 18 and 37 nm, respectively. After proteinase treatment, or increasing ion concentration, most of the proteins are removed and the D N A is exposed even in the case of cr-nucleoids, the stability of which proved to be greater than that of cl-nucleoids. The chemical composition of isolated plastid nucleoids has been determined qualitatively and quantitatively. Chromoplast-nucleoids contain, relative to the same D N A quantity, about six times Abstract.
* To whom correspondence should be addressed Abbreviations: cl-=chloroplast; cr-:chromoplast; D A P I =
4'6-diamidino-2-phenylindole; DNase = deoxyribonuclease; kDa = kilodaltons; MG = purified by metrizamide gradients; SC=purified by Sepharose CL-4B column gel filtration; SDSPAGE: sodium dodecylsulfate-polyacrylamide gel electrophoresis
as much protein as cl-nucleoids. Accordingly the buoyant density of cr-nucleoids in metrizamide gradients is higher than that of cl-nucleoids. In addition to D N A and protein, R N A could be found in the nucleoid fraction. No pigments were present. The cr- and cl-nucleoids have many identical proteins. There are, however, also characteristic differences in their protein pattern which are possibly related to the different expression of the genomes of chloroplasts and chromoplasts. Nucleoids of both plastid types contain some proteins which also occur in isolated envelope membranes (probably partly in the outer membrane) and thus possibly take part in binding the D N A to membranes. Key words:
(nucleoid) -
Chloroplast (nucleoid) - Chromoplast Narcissus - Plastid (envelope, DNA).
Introduction
In bacteria, mitochondria, and plastids, the genetic material (i.e. circular double-stranded DNA) is located within nucleoids. According to electron microscopical investigations in situ, the structure of these DNA-containing areas appears to be similar in all prokaryotic systems (Ris and Plaut 1962; Kellenberger etal. 1958; Bisset 1970; Nass and Nass 1962; Hermann and Kowallik 1970). In fact, the fine structure of nucleoids is so different from that of a eukaryotic cell nucleus that the systematic separation of living organisms into pro- and eukaryotes could be based solely on this criterion. Nucleoids in plastids have been investigated repeatedly by use of electron-microscopical, radioautographic, and cytochemical methods (Woodcock and Fernfindez-Morfin 1968; Kowallik and Herrmann 1972a; Knoth et al. 1974; Rose and Poss-
460
ingham 1976; James and Jope 1978; Scott and Possingham 1980; Kuroiwa etal. 1981; D6rle 1981; Possingham et al. 1983). It has been found that chloroplasts contain different numbers of nucleoids depending on their degree of differentiation. These nucleoids in turn contain variable numbers of plastome copies. Only recently has it been possible to isolate and characterize DNA-protein complexes (nucleoids) from chloroplasts. They are, after suitable isolation, transcriptionally active (Hallick et al. 1976; Briat et at. t979); they appear highly condensed with a beaded substructure, and have a complex protein composition (Briat et al. 1982). When isolated using strongly deproteinizing conditions, the DNA, as visualised after spreading preparations in the electron microscope, is in the form of a rosette with a central body holding the otherwise naked D N A together (Herrmann et al. 1974; Yoshida etal. 1978; Yoshida 1981). As has been shown earlier by electron microscopy of lysed chloroplasts, a large portion of the plastid D N A is membrane bound (Woodcock and Fern/mdezMorfin 1968; Herrmann etal. 1974; Rose and Possingham 1976; Rose 1979; Rose and Lindbeck 1982). It has been suggested that the central body of the nucleoids might be responsible for their binding to chloroplast membranes (Herrmann et al. 1974; Yoshida et al. 1978; Briat et al. 1982). In the present work, the comparative number and distribution of nucleoids in chloro- and chromoplasts of the daffodil have been investigated in situ by fluorescence microscopy. For further investigation the nucleoids from both plastid types have been isolated by two different methods, and characterized light and electron microscopically as well as biochemically. Chloroplasts and chromoplasts of the same plant offer the opportunity to investigate plastid nucleoids in quite dissimilar states of activity. This is particularly true in the case of the daffodil (Narcissus pseudonarcissus L.), the leaves and coronae of which provide suitable raw material for the mass isolation of chloroplasts and chromoplasts, respectively. Chloroplasts and chromoplasts of the daffodil differ fundamentally in pigmentation, fine structure, and composition (Liedvogel et al. 1976). The protein complement of the two plastid types differs significantly (Liedvogei and Sitte 1974; Hansmann and Sitte 1984). Although leaf chloroplasts and flower chromoplasts of N. pseudonarcissus share many enzyme activities associated with lipid metabolism (Liedvogel and Kleinig 1976; Kreuz and Kleinig 1981), it appears that in chromoplasts (in contrast to chloroplasts) no transcription or pro-
P. Hansmann et al. : Plastid-nucleoids of the daffodil
tein synthesis occurs (data not shown). As the plastome of both plastid types is apparently identical (Thompson 1980), modifications in the plastid DNA cannot be responsible for this differential gene expression, As will be shown in the present paper, there are, however, structural differences between the DNA-protein complexes of the two plastid types which are mirrored in the protein composition of the nucleoids. Materials and methods For mass isolation of chloroplasts and chromoplasts the daffodils (Narcissus pseudonarcissus L. cv. 'Golden Harvest') were bought from the local market. Wild-type material (collected near G6rardmer, France) or cultivated forms (cultivated outdoors) were used for in situ investigations. Ampholines (pH range 3-10), desoxyribonuclease (DNase) I, 4'6-diamidino-2phenylindole (DAPI), and illetrizamide were obtained from Serva, Heidelberg, FRG. Sepharose CL-4B was purchased from Pharmacia, Uppsala, Sweden. Staphylococcus aureus V8 protease was obtained from Miles Laboratories, USA. Proteinmolecular-weight-markers for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were from Boehringer, Mannheim, FRG, and Sigma, St. Louis, Mo., USA.
Isolation ofplastids (All steps at 4 ~ C) Chromoplast isolation. For isolation of chromoplasts, daffodil coronae (125 g) from open flowers were homogenized at maximum speed in a Waring Blendor (3 • 3 s) in the presence of 300 ml buffer A (0.33 M sorbitol, 50 mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)/HC1 pH 8.0, 3 mM disodium ethylenediaminetetraacetic acetate (NazEDTA), i mM 2-mercaptoethanol, 0.2% (w/v) polyvinylpyrrolidone). After filtration through three layers of fine-mesh nylon cloth and centrifugation at 1000 g for 5 rain, the chromoplasts were pelleted from the supernatant at 17000 g for 20 rain. The pellet was resuspended in 19 ml buffer A and 1 m10.2 M MgC1 z, and 1.5 mg DNase I was added. The suspension was incubated for i h, and then the digestion of nuclear D N A stopped by addition of 40 ml buffer B (0.33 M sucrose, 50 m M Tris-HC1 pH 8.0, 20 mM NazEDTA). The chromoplasts were pelleted once more (20 min at 17000g) and further purified by a flotation step as detailed by Liedvogel et al. (1976), but using buffer C (25 m M Tris-HC1 pH 8.0, 10 mM N a z E D T A ) instead of the one previously described. The final pellet was resuspended in buffer D (0.33 M sucrose, 50 mM Tris-HC1 pH 8.0, 4 mM EDTA). The chromoplasts were stored frozen at - 7 0 ~ C prior to use. Chloroplasts were isolated according to Kolodner and Tewari (t975) with slight modifications. Fresh daffodil leaves (660 g) were washed and homogenized in 1.8 1 buffer E (0.6 M sorbitol, 50 m M Tris-HC1 pH 8.0, 3 mM Na2EDTA, 1 mIVi 2-mercaptoethanol, 0.1% (w/v) bovine serum albumin) with a Waring Blendor (5 x 3 s at maximum speed). The slurry was filtered through fine-mesh nylon tissue and the filtrate centrifugated at ~ 500 g for 10 min. The pellet was resuspended in approx. 80 ml buffer E, and after centrifugation at 130 g for 30 s, tile chloroplasts were pelleted from the supernatant at 4000 g for 90 s. Then, the pellet was resuspended in 28.5 ml buffer E and, after addition of 1.5 ml 0.2 M MgC12 and 1.5 mg DNase I, the suspension was incubated for 30 min for nuclear-DNA degradation. The reaction was stopped by adding 70 ml of
P. H a n s m a n n et al. : Plastid-nucleoids of the daflbdil buffer F (0.6 M sucrose, 50 m M Tris-HCl pH 8.0, 20 m M Na2EDTA), the chloroplasts were repelleted at 1500g for 15 min and resuspended in buffer F. The suspension was overlayered onto a continuous sucrose gradient (1.0 M-1.9 M sucrose in buffer C), and centrifuged at 80 000 g for 2 h in a swinging-bucket rotor. The lower green band was then removed and diluted 1:1 with buffer F. The chloroplasts were pelleted (20 min at 3000 g), and for storage at - 7 0 ~ resuspended in buffer G (0.6 M sucrose, 50 m M Tris-HC1 pH 8.0, 4 m M EDTA). Isolation of chloroplast and chromoplast integral envelope and thylakoid proteins was carried out as described by Hansm a n n and Sitte (1984).
Isolation of nucleoids Nucleoids from both chloroplasts and chromoplasts were obtained by two different procedures. The first was adapted from the method used by Hallick et al. (1976) for isolation o f a DNAprotein complex from chloroplasts of Euglena by Sepharose 4B column gel filtration (SC). All solutions were filter-sterilized (Millipore, Molsheim, France; SLHA 025BS). The chromoand chloroplasts were pelleted from the thawed suspensions (using only half the isolate) by centrifugation for 20 rain at 17 000 g and 4 000 g, respectively. The plastids were resuspended in 5 ml buffer H (25% (w/v) glycerol, 20 m M triethanolamineHC1 pH 7.5, 0.1 m M EDTA, 1 m M 2-mercaptoethanol, 1% (w/v) Triton X-100, 0~5mM phenylmethylsulfonylfluoride (PMSF)) and lysed for 5 min at 22 ~ C. After centrifugation (10 rain, 3000 g, 40 C), 1/40 vol. of 4 M (NH~)2SO 4 was added to the supernatant. The solution was loaded through a small Sepharose 4B-C1 pre-column, freshly prepared in each case and equilibrated with buffer I (buffer H plus 0.1 M (NH~)zSO4), onto the separation column equilibrated with buffer I (but with P M S F omitted). Elution was carried out with the same buffer. Nucleoid-peak fractions were detected by DAPI epifluorescence microscopy (see below). For the second isolation method, the plastids were pelleted as described above and lysed by resuspending the pellets in buffer J (0.9 M metrizamide, 20 m M triethanolamine-HC1 pH 7.5, 0.1 m M EDTA, 0.4 m M CaC12, 2% (w/v) Triton X-100, 10% (w/v) glycerol, 0.5 m M PMSF), and by incubation for 5 min at 22 ~ C. Onto this solution (approx. 0.8 M metrizamide) a discontinuous metrizamide gradient (0, 0.3, and 0.6 M metrizamide in buffer K (10% (w/v) glycerol, 20 m M triethanolamine-HC1 pH 7.5, 0.1 m M EDTA, 0.4 m M CaCI 2, 1% (w/v) Triton X-100) was overlayered. After centrifugation for 5 h at 80000 g in a swinging-bucket rotor, the nucleoids (0.3/0.6 M metrizamide step) detected by DAPI epifluorescence microscopy were removed.
Preparation of isolated nucleoidsfor microscopy Nucleoids purified by centrifugation in metrizamide gradients (MG) were dialysed against buffer K without Triton X-100 by placing a 60-lal drop on a membrane filter (Millipore VMWP, 50 nm average pore size) as described by Marusyk and Sergeant (1980). This nucleoid (MG) fraction, or else the alternatively isolated nucleoids (SC), were used directly or incubated in the same medium for 1 h either at 22~ with 50 lag m l - 1 Proteinase K (DNase inactivated, compare Young and Sinsheimer 1967) or with 0.5 M NaCI (high salt) at 4 ~ C for protein depletion. The DNA-protein complexes were then fixed for 30 rain at 4 ~ C by addition of 1/40 vol. of 20% (w/v) glutardialdehyde. Drops (20 lal) placed on membrane filters as described above were dialysed for 2 h against buffer L (0.15 M NaCI, 10 m M Tris-HCt pH 7.5, 0.1 m M Na2EDTA ).
461
Preparation of isolated nucleoidsfor biochemical analyses Nucleoids (SC) were pelleted from the peak fractions at 120000 g and 4 ~ C for 16 h in a fixed-angle rotor. The pellets were resuspended in double-distilled water and stored at - 7 0 ~ C. Nucleoids (MG) were either used directly for determination of the buoyant density (see below) or the nucleoid band was diluted 1:3 with buffer K and the mixture was layered onto a 2.5-ml 0.8 M sucrose cushion in buffer K. After centrifugation in a 12-ml fixed-angle rotor at 200000g and 4 ~ for 3 h, the pellet was resuspended in double-distilled water and stored at - 7 0 ~ C prior to use.
Fluorescence microscopy The observations were made with a Zeiss (Oberkochen, F R G ) IM 35 inverted epifluorescence microscope equipped with Neofluar phase optics and a HBO 50 lamp. Photographs using the filter combinations BP 365/FT 395/LP 397 (DAPI) and BP 436/FT 460/LP 470 (chlorophyll) on K o d a k Tri-X pan (400 ASA) film. For in situ staining of the nucleoids (James and J o p e / 9 7 8 ; Coleman 1978), small pieces of the epidermis either from green leaves (chloroplasts) or from the coronae (chromoplasts) of the daffodil were vacuum-infiltrated with buffer M (0.3 M sucrose, 10 gl ml i DAPI, 5 m M magnesium acetate) and observed in the same medium. Isolated chromoplasts were fixed in buffer N (I % glutardialdehyde, 50 m M Na-cacodylate buffer pH 7.2, 0.33 M sucrose) for 2 h, pelleted for 2.5 rain at 2 500 g and resuspended in buffer 0 (1.2 M sucrose, 50 m M Tris-HC1 pH 7.5, 20 m M N a z E D T A , 2 gg m1-1 DAPI). A 10-pl aliquot of this suspension was put directly onto a slide and observed. In the case of isolated chloroplasts the same procedure was applied, but using a higher sucrose concentration (0.6 M). Isolated nucleoids were first fixed and dialysed as described above, 1/10 vol. 20 m M spermidine-HC1 was added in order to favour adsorption to the cover slip, and 10 gl of the samples were mixed with 2 lal of a 2 gg m l - 1 DAPI solution on the slide.
Electron microscopy The whole-mount method as described by Miller and Beatty (1969), was modified as detailed by Suzuki et al. (1980). In a Lucite microcentrifugation chamber, 10-20 pl of the prepared samples were layered onto a sucrose cushion (1 M sucrose, 2 0 m M triethanolamine-HC1 pH 7.5, 0.1 m M Na2EDTA , 0.15 M NaCI) and the DNA-protein complexes were pelleted onto a carbon-coated grid (previously activated by glowdischarge) for 5 min at 10000 g and 4 ~ C in a 5-ml swingingbucket rotor. The grids were washed twice for 30 s in doubledistilled water, stained with 1% (w/v) uranyl acetate for 30 s, and dehydrated by a graded ethanol series (0, 25, 50, 90, and 100%; 10 s each). The grids were rotary-shadowed with platinum/palladium and observed in a Zeiss EM 10CR at 60 kV. For the direct-mounting technique (Griffith and Christiansen 1978), 9 gl of the pretreated samples were mixed with 1 gl of 20 m M spermidine-HC1 and a 5-gl aliquot was placed on activated carbon-coated grids. After adsorption of the DNAprotein complexes for 3 rain, the grids were washed, stained, dehydrated, and rotary-shadowed. For negative staining, samples prepared as described for the direct-mounting technique were air dried after removing most of the staining solution from the grid with filter paper and immediately observed in the electron microscope. In some cases fluorescence and electron microscopy were combined in order to identify the nucleoids unequivocally in the electron
462
P. H a n s m a n n et al. : Plastid-nucleoids of the daffodil
Fig. 1 a, b. Epifluorescence photographs of plastid-nucleoids of different tissues of Narcissus pseudonarcissus after staining with DAPI. a Coronae of open flowers (chromoplasts); b green leaves (chloroplasts). B a r s = 10 pm, x 1 200 Fig. 2a, b. Isolated plastids, nucleoids stained with DAPI. a chromoplasts; b chloroplasts. Bars = 10 gm, x 1200 Fig. 3a-d. Isolated nucleoids stained with DAPI. a from chromoplasts (MG); b from chromoplasts (SC); c from chloroplasts (MG); d from chloroplasts (SC). B a r s = 10 pm, x 1200 Fig. 4a, b. Nuclcoids out of chromoplasts purified by gel filtration, a DAPI epifluorescence photograph; b the same area shown by electron microscopy after shadowing with Pt-Pd. Bars = 2 ~tm, x 5000
microscope. The nucleoids were allowed to adsorb to carboncoated formvar films (on electron-microscope-grids containing an asymmetric centre mark) as described above. After washing the grids for 30 s in double-distilled water, they were carefully mounted onto a slide with a drop of 0.2 gg ml - i DAPI solu-
tion. Suitable positions of the preparation were rapidly photographed and noted. The grid was then removed, washed, stained, dehydrated, and shadowed as above. The meshes corresponding to the ones observed in the light microscope could easily be detected in the electron microscope.
P. Hansmann et al. : Plastid-nucleoids of the daffodil
Determination of the buoyant density of nucleoids ( MG )
463 Table 1. Number of nucleoids per plastid in Narcissus pseudo-
narcissus An aliquot (equivalent to 10-20 gg DNA) of the nucleoid band (see above) was diluted 1:4 with buffer K and layered onto a discontinuous metrizamide density gradient containing steps of 0.2, 0.3, 0.4, 0.5, and 0.6 M metrizamide in buffer K. After centrifugation for 14 h at 150000 g and 4 ~ C in a 12-ml fixedangle rotor, the tube was punctured at the bottom, and fractions of approx. 400 gl were collected. Nucleoid-containing fractions were detected by DAPI fluorescence microscopy. The density of each fraction was determined from the refractive index after correction for the buffer used as described by Rickwood (1978).
Electrophoresis of the proteins One-dimensional electrophoresis. The chloroplast, chromoplast, and different membrane pellets were resuspended in doubledistilled water. The protein concentration of all samples was determined according to Bradford (1976). For SDS-PAGE, 0.5 vol. buffer P (90 mM Tris-HC1 pH 7.6, 6% (w/v) SDS, 15% (w/v) 2-mercaptoethanol, 15% (w/v) glycerol, 0.1% (w/v) bromophenol blue) was added to the samples containing 10-30 gg protein, and the solutions were boiled for 2 min. The proteins were separated on 8-18 % polyacrylamide gradient gels (1.5 mm thick slab gels) with a 4% polyacrylamide stacking gel using the buffer system of Chua (1980). For molecular-weight determinations 10 and 15% polyacrylamide linear separating gels and myosin (M r 205 000), fl-galactosidase (116 000), phosphorylase (97400), bovine serum albumin (66000), ovalbumin (45 000), carbonic anhydrase (29 000), trypsin-inhibitor (21 500), cytochrome c (12 500), and aprotinin (6 500) were taken as standards. Fixed and washed gels were finally stained by the very sensitive silver staining procedure of Bfirck et al. (1983).
n.d. = N o data. a-d) Different stages of chromoplast development characterized by the chlorophyll content (red-fluorescence): a) much (prochromoplasts); b) medium; c) low; d) no (chromoplasts) Mean value
Extreme values
Plastids investigated
Leaves : chloroplasts
n.d.
10-30
n.d.
Coronae: a) b) c) d)
4.3 3.7 2.7 2.2
1-9 1-7 1-6 0-7
61 75 37 64
10-30
n.d.
0-10
200
10-30 0-10
n.d. 105
Wild type
Culture form Leaves: chloroplasts n.d. Coronae: chromoplasts 3.7 Isolated plastids : chloroplasts n.d. chromoplasts 3.2
Table 2. Sizes (gm) of the nucleoids from N. pseudonarcissus
Light microscope DAPIfluorescence
Electron microscope
Tissue: chloroplasts chromoplasts
0.3-1.0 0.3-1.5
-
-
Isolated plastids: chloroplasts chromoplasts
0.3 1.0 0.3-2.0
-
-
0.3-2.0 0.3-2.0 0.3-3.0 0.3-2.0
0.3 2.5 0.3-2.0 0.34.0 0.3-2.5
0.3 2.5 0.3-2.0 0.3-3.5 0.3-3.0
Two-dimensional electrophoresis. To the aqueous protein solutions (see above), 0.i vol. buffer Q (100 mM Tris-HCl pH 7.5, 70 mM MgC12, 20 gg m l - 1 DNase I) were added, and the mixture was incubated for 1 h at 4 ~ C for D N A degradation. After addition of 0.1 vol. 5% SDS and incubation for 10 min at 70 ~ C, solid urea, Ampholines, 2-mercaptoethanol, and Triton X-100 were added to the samples to bring the concentrations to 9.5 M, 2%, 5%, and 4%, respectively. This procedure is a modification of the method of Ames and Nikaido (1976) and Siegenthaler and Nguyen (1983). Isoelectric focusing was as described by O'Farrell (1975), but with reversed polarity (acidic reservoir on top) as proposed by Righetti et al. (1982). After focusing, the gels were equilibrated twice for 20 min with buffer R (30 mM Tris-HC1 pH 7.6, 2% SDS, 5% 2-mercaptoethanol, 5% glycerol) and loaded onto SDS-polyacrylamide gradient gels (see above) as described by O'Farrell (1975).
After After shadowing negative staining
In situ
Isolated cl-Nucleoids (MG) cr-Nucleoids (MG) cl-Nucleoids (SC) cr-Nucleoids (SC)
Peptide mapping by limited proteolysis. The protocol was essentially the same as the one described by Cleveland et al. (1977). Staphylococcus aureus V8 protease was used for digestion of the proteins.
Quantitative determinations DNA, RNA, and proteins were extracted according to Daniel and Baldwin (1964). D N A was determined by the diphenylamine method (Burton 1956), R N A by the orcinol method (Ogur and Rosen 1950), and protein according to Schacterle and Pollack (1973). The carotenoid content was calculated using the formula of Liaaen Jensen and Jensen (1971). Chlorophyll was estimated as described by Arnon (1949).
Results
Nucleoids from both chloroplasts and chromoplasts were measured and counted after specific staining in situ with the fluorochrome DAPI (Fig. 1, Tables 1, 2). The size of the fluorescent spots and the number of nucleoids per plastid varied considerably. Chloroplasts contained more than ten nucleoids which were evenly distributed throughout the whole plastid (Fig. 1 b). In larger
464
P. H a n s m a n n et al. : Plastid-nucleoids o f t h e daffodil
P. H a n s m a n n et al. : Plastid-nucleoids of the daffodil
465
Fig. 6a, b. Nucleoids (MG) from daffodil chromoplasts (a) and chloroplasts (h) after negative staining. Bars = 0.5 gm, x 40000
chromoplasts of the cultivars, more nucleoids were found than in the smaller wild-type chromoplasts. In contrast to the chloroplasts, the chromoplast nucleoids were always situated in close proximity to the peripheral membrane convolute (Fig. I a). During chromoplast development from prochromoplasts there was a synchronous decrease in chlorophyll content and number of nucleoids per plastid (Table 1). Isolated plastids showed no changes in the form of nucleoids (Fig. 2). Isolated chromoplasts, however, contained on average fewer nucleoids than in situ chromoplasts (Table 1), probably due to a partial loss of nucleoids from injured chromoplasts. During the isolation of the chloroplasts, intact and broken chloroplasts were separated. All intact chloroplasts contained nucleoids, whereas broken or injured ones contained only fragments of nucleoids or none whatsoever. Contamination with nuclear D N A could not be detected in either plastid fraction. Isolated nucleoids showed a compact structure irrespective of the method of purification (Fig. 3),
although they were slightly larger than the in situ nucleoids (Table 2). Presumably, the in situ-visualized fluorescent spots represented only the nucleoid core. In order to identify the nucleoids unequivocally in the electron microscopical preparations, fluorescence and electron microscopy were combined (Fig. 4). When observed in the electron microscope, isolated chromoplast (cr-)nucleoids (MG) consisted, after protein-free spreading and shadowing with Pt/Pd, of electron-dense material in polymorphic forms (Figs. 4, 5 a, c), with no naked D N A visible. In contrast, chloroplast (cl-)nucleoids (MG) revealed an electron-dense, central area with naked D N A projecting in the form of loops or supercoils (Figs. 5b, d). Following proteinase K treatment prior to electron microscopical preparation, the electron-dense material was extensively degraded in both cr- and cl-nucleoids, with naked D N A becoming progressively more exposed (Fig. 5e, f). The same effect could be achieved by increasing the ionic strength of the isolation buffer (Fig. 5g, h). In both nucleoid types, however, the D N A was
(
Fig. 5a-h. Electron micrographs of DNA-protein complexes (nucleoids M G ) from Narcissus pseudonarcissus chromoplasts (a, e, e, and g) and chloroplasts (h, d, f, and h) after protein-free spreading and shadowing, a-d Whole-mount method; e-h directadsorption technique, a, h General view of the nucleoid fraction; e, d nucleoids in detail; e, f after proteinase treatment; g, h after high-salt treatment, a, b Bars = 5 lam, x 4 000; c-h bars = 0.5 ~tm, x 40000
P. H a n s m a n u et al. : Plastid-nucleoids of the daffodil
466 Table 3. Enrichment and composition of daffodil plastidal nucleoids Purification step
Leaves/chloroplasts
Coronae/chromoplasts Pigment a
Protein
RNA
DNA
Pigment b
I. Tissue
1.3
23
1.5
1
II. Plastids
5.7
91
4.4
1
17
III. Nucleoids MG SC
-
I0 6.4
1.1 0.88
1 1
-
1.2
Protein
RNA
DNA
27
1.7
1
250
10.4
1
2.2 0.86
0.58 0.28
1 1
a Carotenoid content b Chlorophyll content
Table 4. Dependence of D N A recovery on the method of nucleoid purification and plastid type Purification method
Chromoplasts
Chloroplasts
MG SC
54% 22%
32% 26%
still held together by proteinase K- and salt-resistant structures, this being more pronounced in the case of cr-nucleoids. When isolated nucleoids were negatively stained with uranyl acetate (rather than rotary shadowed), knotty structures were visible over the whole complex (Fig. 6). Occasionally, smooth filaments of 18 nm diameter were visible which coiled around each other forming fibrils with a diameter of 37 nm. Nucleoids purified by Sepharose gel filtration were not appreciably different from nucleoids purified in metrizamide gradients. The density of isolated nucleoids in metrizamide was 1.21 g c m - 3 for cl-nucleoids, and 1.25 g c m - 3 for cr-nucleoids. As the buoyant density of DNA-protein complexes in metrizamide is known to increase with rising protein contents (Rickwood and MacGillivray 1977), it is to be concluded that cr-nucleoids have a higher protein/ D N A ratio than cl-nucleoids. This expectation was impressively born out by chemical analysis. Some data for the composition of whole tissues (leaves and coronae), isolated plastids, and enriched nucleoids were shown in Table 3. The reference value ( D N A quantity) changed greatly with enzymatic decomposition of nuclear D N A during plastid isolation. In the course of nucleoid purification, all pigments were lost and D N A was enriched with respect to both protein and RNA. This enrichment was more pronounced in the case of column purification, the yield however being smaller (Table 4). The protein/DNA ratio
Fig. 7. Analysis by SDS-polyacrylamide gel electrophoresis of daffodil plastidal proteins. Proteins from isolated chloroplasts (track A); chloroplast-nucleoid proteins (track B); chromoplast-nucleoid proteins (track 6); proteins from isolated chromoplasts (track D); standard proteins for molecular-mass determination (track E)
was about six times higher in cr-nucleoids than in cl-nucleoids. For qualitative protein investigations, nucleoids (SC) were mainly used as these can be isolated more effectively. Comparison of nucleoid proteins with the total protein pattern of the plastids from which they were isolated showed that nucleoid proteins do not belong to the major proteins of the plastids (Fig. 7). The nucleoids from chloroplasts and chromoplasts showed, besides many electrophoretically identical proteins, several
P. Hansmann et al. : Plastid-nucleoids of the daffodil
467 Table 5. Relative molecular mass (M r x 10 -3) of some major proteins of the different nucleoid fractions. Letters and numbers as in Fig. 8. Chloroplast-nucleoid proteins (C); chromoplastnucleoid proteins (D)
Fig. 8. Comparison of the proteins of different subplastidal fractions by SDS-polyacrylamide gel electrophoresis. Integral thylakoid proteins (track A) ; integral chloroplast envelope proteins (track B); chloroplast-nucleoid proteins (track C); chromoplast-nucleoid proteins (track D); integral chromoplast membrane proteins (track E). Black bars indicate proteins with identical electrophoretic mobilities
Protein number
C
D
Protein number
C
D
1 2 3 4 5 6 7 8 9
207 130 110 96 91 73 58 56 29
207 162 130 110 96 91 73 68 42
10 11 12 13 14 15 16 17
22 21 20 18 15
40 32 29 22 21 20 18 15
cr-nucleoid proteins 4, 5, 7, 8, and 12 with the pattern of corresponding envelope proteins provided further evidence for the identity of these respective proteins (Fig. 10). At least three cl-nucleoid proteins (main protein: P6, 73 kDa) could thus be regarded as integral envelope proteins. Of the cr-nucleoids, at least five were envelope proteins (major proteins P7 and PS, 73 and 68 kDa). Three of them were identical to corresponding envelope proteins from chloroplasts. P8 from cr-nucleoids was missing from the chloroplast fraction. As the nucleoid fraction was devoid of fatty acids, and as no DNA could be detected in the envelope membrane fraction, there was virtually no mutual contamination of nucleoid and envelope fractions. Discussion
specific bands. Proteins in the low-molecularweight range (
