- Email: [email protected]
Synthesis of EF-Tu and distribution of its mRNA between stroma and thylakoids during the cell cycle of Chlamydomonas reinhardii
209
Biochimica et Biophysica Acta, 1048 (1990) 209-216 Elsevier
BBAEXP 92038
Synthesis of EF-Tu and distribution of its m R N A between stroma and thylakoids during the cell cycle of Chlamydomonas reinhardii Eric Breidenbach 1, Stefan Leu 2, Allan Michaels 2 and Arminio Boschetti 1 z Institutfftr Biochemie, Universitiit Bern, (Switzerland) and 2 Department of Biology, Ben Gurion University, Beer Sheva (Israel)
(Received 4 September 1989) (Revised manuscript received 30 November 1989)
Key words: Chloroplast; Protein synthesis; Elongation factor EF-Tu; mRNA content; Cell cycle; (C. reinhardii)
In Chlamydomonas reinhardii the elongation factor EF-Tu is encoded in the chloroplast DNA. We identified EF-Tu in the electrophoretic product pattern of chloroplast-made proteins and showed that this protein is only synthesized in the first half of the light period in synchronized cells. The newly synthesized EF-Tu contributed little to the almost invariable content of EF-Tu in chloroplasts during the light period of the cell cycle. However, increasing cell volume and the lack of EF-Tu synthesis in the second half of the light period led to a decrease in the concentration of EF-Tu in chloroplasts. At different times in the vegetative cell cycle, the RNA was extracted from whole chloroplasts and from free and thylakoid-bound chloroplast polysomes. The content of mRNA of EF-Tu in chloroplasts and the distribution between stroma and thylakoids were determined. During the light period, the content of the mRNA for EF-Tu varied in parallel to the rate of EF-Tu synthesis. However, in the dark, some mRNA was present even in the absence of EF-Tu synthesis. Most of the mRNA was bound to thylakoids during the whole cell cycle. This suggests that synthesis of EF-Tu is associated with thylakoid membranes.
Introduction
In chloroplasts there are two populations of ribosomes and polysomes with respect to their localization [1-5]: The ribosomes and polysomes remaining in the stroma and the thylakoid-bound ribosomes and polysomes which are anchored to the membrane, either by the nascent peptide, by electrostatic interaction, by specific binding of ribosomes or by a combination of these interactions (see review, Ref. 6). It was postulated that membrane-bound polysomes synthesize membrane proteins [7-13]. However, these polysomes also contain mRNAs for soluble proteins and for extrinsic membrane proteins. These mRNAs were first detected by in vitro translation of extracted RNA and quantified by hybridization techniques [10,14-19]. To date the capacity of these polysomes to
Abbreviations: 32 kDa protein, 32 kDa herbicide-binding membrane protein of Photosystem II; CF1, coupling factor 1 of chloroplast ATP-synthase; LS, large subunit of the ribulose-l,5-bisphosphate carboxylase/oxygenase (EC 4.1.1.39); PAGE, polyacrylamide gel electrophoresis. Correspondence: Professor Arminio Boschetti, Institut fiJr Biochemie, Universit~it Bern, Freiestr. 3, CH-3012 Bern, Switzerland.
translate the mRNA for soluble proteins is still under discussion. The large subunit of ribulose-l,5-bisphosphate carboxylase/oxygenase (LS) was originally not found among the products of run-off translation of membrane-bound polysomes [10,14]. However, Klein et al. [19] and Hattori and Margulies [20] identified the LS as product of thylakoid-bound polysomes and Bhaya and Jagendorf [21] showed that the a- and /3-subunits of CF t were produced by both thylakoid-bound and free polysomes. The amount of membrane-bound ribosomes and polysomes varies during the cell cycle of synchronized C. reinhardii [3,22] and in greening peas [23]. Chloroplast proteins are synthesized in a cell cycle-dependent manner in synchronized C. reinhardii [17,24-27] or in Euglena gracilis [28,29]. Therefore, the partition of ribosomes and polysomes between stroma and thylakoids may be a central point in regulation of protein synthesis in chloroplasts. During light-dependent chloroplast formation of higher plants [30-35] and algae [36,37] or during the cell cycle of synchronized algae [17,26,29,43] the rate of protein synthesis varies much more than the abundance of the mRNAs for most of the investigated proteins. Many chloroplast mRNAs are present in high concentrations during the entire cell cycle and may be
0167-4781/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
210 stored in a silent form when protein synthesis is reduced. Chloroplast mRNA can be found in arrested polysomes. In amaranth seedlings, the translation of the LS was immediately arrested after a light-to-dark transition, but the mRNA remained bound to the polysomes for several hours [35]. The mRNA for the P-700 chlorophyll a protein of Photosystem I was almost exclusively confined to the membrane-bound polysomes in etiolated dark-grown barley seedlings and in chloroplasts of illuminated plants. Although similar levels of this mRNA were present in etioplasts and chloroplasts, no synthesis of P-700 chlorophyll a protein could be detected in etioplasts. A very short exposure to light induced P-700 chlorophyll a protein synthesis [33]. The mRNA for the 32 kDa membrane protein in the stroma of spinach chloroplasts was associated with proteins, but not with ribosomes [15]. In our previous work we analyzed the distribution of the mRNA for the 32 kDa protein and the LS between membrane-bound polysomes and stroma polysomes during the cell cycle of C. reinhardii. In the light, when chloroplast protein synthesis was stimulated, the mRNA for the soluble LS was enriched in membrane-bound polysomes in the same way as the mRNA for the 32 kDa membrane protein. In addition, the distribution for both mRNAs between thylakoids and stroma fluctuated in parallel during the cell cycle [17]. This contradicts the hypothesis that membrane-bound polysomes synthesize only membrane proteins and stroma polysomes synthesize soluble proteins [38]. One could argue that the subchloroplast distribution of the mRNA for the LS reflects an artefact because of the well known 'stickiness' of the growing peptide of the LS to membranes and other cell structures. To confirm our previous findings, we investigated the distribution of the mRNA for the soluble chloroplast elongation factor EF-Tu, which in addition is not directly involved in photosynthesis. In contrast to higher plants, the gene tufA coding for the chloroplast elongation factor EF-Tu is located on the chloroplast D N A in C. reinhardii [39] and in Euglena [40], and is homologous to EF-Tu of E. coli [41,42]. The mRNA for EF-Tu has been found to show a more pronounced fluctuation than other mRNAs investigated during the cell cycle of C. reinhardii [17,43]. Furthermore, the rate of synthesis and the abundance of EF-Tu were studied during the cell cycle, to establish possible involvement of EF-Tu in the regulation of chloroplast protein synthesis. Materials and Methods
Culture conditions. Cultures of Chlamydomonas reinhardii cw-15 (stock cc-277) were obtained from the Chlamydomonas Genetics Center, Duke University,
Durham, NC. Growth conditions and synchronization procedure were as previously described [22]. Determination of the cell volume. Cell size was measured electronically with a Coulter Counter, Model ZB (50 /~m orfice) particle counter connected to a Coulter channelyzer C-1000 (Coulter Electronics, U.K.). Labeling of cells in vivo with [35S]sulfate. Synchronously cultivated C. reinhardii were grown under a 14:10 hours l i g h t / d a r k regime. Samples were taken periodically from the culture and harvested by centrifugation, washed twice with culture medium without sulfate and brought to 5 . 1 0 7 cells/ml. The cells were preincubated for 10 min with 10/~g/ml cycloheximide and labeled in the presence of 10/~g/ml cycloheximide and [35S]sulfate for 60 min in the light or dark as previously described [17]. Isolation of chloroplasts. Chloroplasts were isolated according to the procedure of Mendiola-Morgenthaler et al. [44] with the modifications previously described [17]. Fractionation of chloroplasts. Isolated chloroplasts were fractionated as described [16], except that 10 mmol/1 ribonucleoside-vanadyl complexes were added as an RNase inhibitor before breaking the chloroplasts in the Yeda press [45]. Electrophoretic analysis of proteins. Labeled cells were broken by freezing and ultrasonication and the lysate was centrifuged at 30000 × g for 15 min. The supernatant, containing the soluble fraction, was immediately prepared for electrophoresis. The pellet, containing the membranes, was washed twice with 100 mmol/1 TrisHCI (pH 7.6)/1 mmol/1 MgC12, and solubilized in the same volume as the corresponding supernatant. Aliquots of labeled cells or cell fractions were solubilized in 3% SDS and 2% 2-mercaptoethanol at 80 ° C for 5 min and the proteins were separated on linear SDS-polyacrylamide gradient gels (10-20%) in the buffer system of Laemmli [46]. The gels were either stained with Coomassie brilliant blue R250, treated with Amplify (Amersham), dried and exposed to X-ray film (Hyperfilm-MP, Amersham), or the separated proteins were transferred electrophoretically onto nitrocellulose filter.
Immunological detection and quantification of EF-Tu. The nitrocellulose filters containing electrophoretically separated proteins (Western blots) were treated with antisera raised against the EF-Tu of Escherichia coli (a gift of Erik Vijgenboom, Leiden University, The Netherlands). The bound antibodies were detected with an anti-rabbit IgG horseradish peroxidase conjugate (Sigma) and enzymatic staining with 4-chloro-l-naphthol. The relative amounts of EF-Tu present in different probes of the same Western blot were measured using a scanning spectrophotometer in the reflection mode (CAMAG). Determination of labeling of distinct proteins. The dried gels were soaked in 7% acetic acid and the respective
211 bands were cut out. The gel slices were bleached and dissolved in H202 (30%) and the radioactivity determined by liquid scintillation spectrometry. Chlorophyll determination was according to Vernon [47]. Cloned DNA probes. An internal PstI-EcoRI fragment of the tufA gene of C. reinhardii cloned in the vector pUC8 was obtained from S. Baldauf and J. Palmer. The psaB gene was subcloned into pGem from EcoRI-fragment 15 of C. reinhardii chloroplast D N A [48,43]. Quantification of mRNA. Extraction of R N A and Northern blot analysis were done as previously described [16]. The intensity of the respective hybridizing signal was determined and integrated in the linear range of the X-ray film with the Shimadzu high speed TLS scanner CS-920.
with [35S]sulfate for 60 min in the presence of cycloheximide [17]. Product analysis by fluorography of electrophoretically separated proteins showed a peptide of about 45 kDa in whole cells and in the soluble fraction ($30) (Fig. la). This protein was identified as the translational elongation factor EF-Tu by crossreaction with antibodies raised against EF-Tu of E. coil on Western blots (Fig. la). The rate of EF-Tu synthesis during the cell cycle was determined by pulse-labeling of synchronized cells harvested from the dark period of the cell cycle (hours 0-10) and from the light period (hours 10-24). In the following the dark hours will be referred to as D 1 - D 1 0 , and the light hours as L0-L14. Proteins of equal numbers of cells were electrophoretically separated and the radioactivity was measured in solubilized gel slices containing the soluble 45 kDa protein. The rate of synthesis of EF-Tu changed significantly during the cell cycle, from nearly undetectable amounts in the dark to one of the main products at the beginning of the light period (Fig. l b and Fig. 2). EF-Tu was not labeled in cells harvested in the middle of the dark period, whether incubated in the light or in the dark. At the onset of light the rate of EF-Tu synthesis was already elevated
Results
Synthesis of the elongation factor EF-Tu during the cell cycle Chloroplast-made proteins were labeled by incubation of synchronized cells of Chlamydomonas reinhardii
A
B
C
kDa
93
_
32 kDprotein 2 5 --
14--
E
c m ..... ................
s
c
m
s
c
m
s
L 1 (I
L2(I) L6(I) L13(I) D2(d) D6(d)
- " ' "I " I"I " Lll2(I) L4(I)
I
L8(I) D2[I) D6(I]
Fig. 1. Electrophoretic analysis of [aSS]sulfate-labeled cells of C. reinhardii. (a) In vivo labeled cells (c) and corresponding aliquots of membranes (m) or of soluble fraction (s) were separated by SDS-gel electrophoresis and transferred onto nitrocellulose. (A) Coomassie blue-stained protein patterns of lysed cells and cell fractions of C. reinhardii and of molecular mass markers; (B) Autoradiographof the Western blot of the samples in A; (C) Immunostainingof the Western blot with antibodies against EF-Tu. (b) Comparison of the content of EF-Tu in whole cells with the rate of its synthesis during one 24 h cell cycle.The cells were harvested at indicated hours in the dark period (hours 1-10 : D0-D10) or in the light period (hours 10-24 : L0-L14) and labeled in the presence of cycloheximidein the light (1) or in the dark (d) for 60 min. Equal numbers of cells (2.73-106 cells, corresponding to 1.9-6.8 #g chlorophyll) were separated on SDS-PAGE. The proteins were transferred onto nitrocellulose. (D) Autoradiograph of the Western blot; (E) Immunostainingof the same blot with antibodies against EF-Tu.
212 100 r
E~
8oi 60-
2
!
¥
4OL
2OF
D2
D4
D6
DIS LO L2
L4
L6
L8
LlO L12 L14
Fig. 2. Rate o f synthesis of E F - T u d u r i n g the cell cycle. Synchronized
cells from the dark period (D0-D10), and from the light period (L0-L14) were labeled for 60 min in the dark (*) or in the light (n). After SDS-PAGE, the incorporation of radioactivity into the band identified as EF-Tu was measured. All values of three different gels were standardized to the same cell density and the highest incorporation during the cell cycle was taken as 100%. In this set of experiments we did not determine the rate of EF-Tu synthesis between D6 and DI0. However, fluorographs of previous experiments confirmed that the respectiveband was scarcelylabeled at the end of the dark period [171.
and reached its maximum between L1 and L5. Towards the second half of the light period the rate of synthesis declined rapidly to less than 15% of the highest rate.
Content of the elongation factor EF-Tu during the cell cycle The cell cycle-dependent synthesis of EF-Tu may play a key role in the regulation of protein synthesis in chloroplasts [43,49]. We therefore determined the content of this protein in chloroplasts during the cell cycle by immunostaining of Western blots after electrophoretic separation of proteins from an equal number of cells. Since C. reinhardii cells contain only one chloroplast, we concluded that the content of EF-Tu per chloroplast was almost constant during an entire cell cycle (Fig. lb). The weakening of the band at D6 1 --
-~
~
200
150
'~ 02
O
~
I D2
~ - .~U ~
~
~
5
0
>o
__
I DI6 8/ I , D4 D LQ L2
J L4
, L6
J L8
~ L L1Q L12
fo
Fig. 3. C o n t e n t and concentration of E F - T u d u r i n g the cell cycle. The
concentration of EF-Tu U7) was calculated by dividing the relative
content of EF-Tu per cell (*) by the cell volume (1:3)(all determinations of two independent experiments are shown).
reflects simply the reduction of the cell size, and hence of the amount of EF-Tu applied to the gel, after the cells had divided into four daughter cells between D2 and D6. The amounts of EF-Tu present at each time point were quantified by scanning the Western blots and expressed as relative content of EF-Tu per number of cells (Fig. 3). We estimated the concentration of EF-Tu by dividing the relative content by the relative cell volume, as determined with the Coulter Counter. There was a 4-fold decrease of the concentration of EF-Tu during the light period (Fig. 3). To test the method of quantification of EF-Tu by immunostaining, we separated a set of different dilutions of the soluble fraction ($30) from in vivo labeled cells on SDS-PAGE and blotted them onto nitrocellulose. The radioactivity of EF-Tu and the immunological quantification were linearly related at least over one order of magnitude (results not shown).
mRNA content in intact chloroplasts during the cell cycle The relative abundance of the m R N A for EF-Tu has been determined during the cell cycle by hybridization of Northern blots with a gene probe specific for EF-Tu (Fig. 4a). Some m R N A was detected in the dark. Its relative abundance increased to a maximum at L2, and then declined to almost undetectable levels at the end of the light period. The pronounced variation in the amount of this m R N A during the cell cycle differed significantly from the one we measured previously for the LS and for the 32 k D a protein [17] and psaB (Ref. 43; and Leu et al., unpublished data). The hybridization pattern obtained showed one main band at 2.1 kb, a weaker band at 1.7 kb and an additional faint band at 3 kb. At L2 and L6 considerable hybridization was obtained with low molecular mass RNA, probably degradation products of the m R N A for EF-Tu. To test the intactness of the extracted R N A we rehybridized some selected Northern blots with a specific gene probe for psaB encoding an apoprotein of the reaction center of Photosystem I [48] (Fig. 5). The degree of the hybridization signal in low molecular R N A seemed to be reduced. Therefore, the hybridization of the tufA probe with low molecular R N A points to a cell cycle specific instability of this transcript. Because of the complex hybridization pattern, the content of the m R N A for EF-Tu was quantified by densitometric scanning of the main transcript (2.1 kb) and compared to the rate of EF-Tu synthesis (Table I and Fig. 2). To compensate for the varying content of R N A in chloroplasts during the cell cycle, we determined the extractable R N A from isolated chloroplasts at each time and corrected our values. During the light period the rate of EF-Tu synthesis paralleled the m R N A content and therefore seemed to be regulated by m R N A abundance; while at D6, in spite of abundant m R N A , no EF-Tu synthesis was observed.
213
2.1kb
1.7kb
b
D6
D9~/4
LO
L2
L6
L13
--3kb --
c
rn
s
c
rn
s
c
m
s
c
m
s
c
m
s
c
m
21
kb
s
Fig. 4. Autoradiographs of Northern blots, hybridized with the nick translated gene probe for the elongation factor EF-Tu. (a) At indicated times during the cell cycle equal amounts of RNA from chloroplasts were separated by formamide/formaldehyde electrophoresis and blotted onto nitrocellulose; (b) equal amounts of RNA from chloroplasts (c), from the membrane fraction (m) and from the stroma (s) were separated electrophoretically and blotted onto nitrocellulose.
Distribution of mRNA between stroma and thylakoids The distribution of the m R N A for EF-Tu between stroma and thylakoids was determined and compared to the distribution of psaB m R N A and to the m R N A for LS and the 32 k D a protein previously studied. Our Northern blots, containing equal amounts of R N A per lane, demonstrated the presence of m R N A for the EF-Tu in the R N A extracted from stroma polysomes and from thylakoid-bound polysomes (Fig. 4b). However, in previous studies we showed that the amount of total R N A bound to thylakoid membranes increased
LO
at the beginning of the light period [17]. After 2 h in the light, about 80% of the R N A was bound to thylakoids, whereas only 20% remained in the stroma. Towards the end of the light period and in the dark, we extracted about equal amounts of R N A from thylakoids and the stroma fraction (Table II). Therefore, the distributions of m R N A for EF-Tu between stroma and thylakoids were calculated by multiplication of the relative contents of the m R N A in these fractions with the partition of total R N A between stroma and thylakoids and with the amount of EF-Tu m R N A in chloroplasts during the
L2
LO
2.9
__
L2
m
2.1 kb
C
FFI
S
C
m
S
c
m
S
C
m
S
Fig. 5. Autoradlograph of Northern blots hybridized with the gene probes tufA and psaB. A selected Northern blot with the RNA extracted from whole chloroplasts (c), from the membrane fraction (m) and from the stroma (s) was rehybridized with the gene probe of psaB: (a) Hybridized with the gene probe tufA; (b) Hybridized with the gene probe psaB.
214 TABLE I
Quantification of mRNA for EF-Tu in chloroplasts
t~
The relative content of m R N A for EF-Tu in whole chloroplasts was calculated. The amounts of extractable R N A from isolated chloroplasts from different times of the cell cycle were multiplied with the intensities of the corresponding hybridizing signals for EF-Tu m R N A on Northern blots, determined by densitometric scanning of the autoradiography (Fig. 4a) (e.g., D6: 6th hour of the dark period; L2: 2nd hour of the light period). The values at D6 were taken as 1. In order to compare the m R N A content with the rate of EF-Tu synthesis (Fig. 2), the highest content of m R N A was taken as 100%
c
08 e6
g u
04
r i
0.2F
0
Time
D6 D9 3 / 4 L0 L2 L6 LI3
Extracted R N A
Hybridizing
m R N A content
(rel. units)
signal (rel. units)
(rel. units)
(%)
1 1 0.9 1.3 2.4 2.8
1 0.9 1.7 2.0 0.4 < 0.1
1.0 0.9 1.6 2.6 1.0 < 0.3
39 35 61 100 38 <10
DI2
i D4
i D6
i D~,
I LO
i L2
i L4
t L6
~ L8
i LIO
1 L12
L14
Fig. 6. Distribution of m R N A for EF-Tu between thylakoids (*) and stroma (rl) during the cell cycle. The values have been calculated as described in the text and in Table II.
of thylakoids between free and membrane-bound polysomes in algea and higher plants. Less data are available about the distribution of mRNAs for soluble proteins. The mRNA for the LS was associated to a significant degree with membrane-bound polysomes, but other mRNAs for soluble proteins have not been tested until
cell cycle (Table II). By these corrections, it became obvious that about 70-80% of the mRNA was membrane-bound at all times of the cell cycle (Fig. 6). The Northern blots rehybridized with the gene probe specific for psaB showed that the mRNA for the respective membrane protein was mainly membrane-bound (Fig. 5b). Furthermore, the 5 S tRNA band visible in ethidium bromide stained RNA gels was observed only in the soluble fraction (not shown). These findings confirmed, that the distribution of the mRNAs was real and not due to an artefact of the fragmentation of chloroplasts. Discussion
There are several reports concerning the distribution of mRNAs for intrinsic or extrinsic membrane proteins
now.
Abundance and synthesis of EF-Tu during the cell cycle The chloroplast EF-Tu was found exclusively in the soluble fraction of lysed cells and chloroplasts and was one of the predominantly labeled chloroplast-made proteins in C. reinhardii (Fig. 1). The very high homology of EF-Tu of E. coli with chloroplast EF-Tu [41,42] made it possible to identify this protein immunologically with antibodies raised against the E. coli protein. The identified protein has a molecular mass of 45 kDa which corresponds well to data of EF-Tu isolated from spinach chloroplasts [50], Euglena and E. coli [41,50]. Synthesis of EF-Tu was light dependent and increased strongly at the beginning of the light period. The rate of synthesis paralleled the content of the mRNA for EF-Tu during the light period of the cell
TABLE II
Distribution of mRNA for EF-Tu between thylakoids and stroma during the cell cycle The subchloroplast distribution of the m R N A was calculated. The relative m R N A content in the respective fraction, determined by hybridization of Northern blots of equal amounts of RNA, was multiplied with the content of m R N A in chloroplasts (Table I) and the partition of the extractable R N A between membranes (m) and stroma (s), respectively [17]. Time
D6 D93/4 L0 L2 L6 L13
m R N A content in chloroplasts
Partition of R N A (rel. units)
Relative m R N A content (rel. units)
Distribution of m R N A (rel. units)
(rel. units)
m
s
m
s
m
s
1 0.9 1.6 2.6 1 <0.3
0.5 0.6 0.6 0.8 0.7 0.5
0.5 0.4 0.4 0.2 0.3 0.5
0.9 0.8 0.7 0.5 0.7 0.9
0.1 0.2 0.3 0.5 0.3 0.1
0.5 0.4 0.6 1.0 0.5 <0.1
<0.1 0.1 0.2 0.3 0.1 <0.1
215 cycle, but not in the dark (Table I and Fig. 2). Therefore in the light, expression of EF-Tu might be regulated by mRNA abundance. In the dark at D6 however, significant amounts of EF-Tu transcript were present, which were not translated even if the cells were transferred to the light. This finding points to a stage-dependent competence of EF-Tu synthesis during the cell cycle similar to our previous observations [17], where we showed that in synchronized C. reinhardii harvested in the dark period the synthesis of LS and of the 32 kDa protein could hardly be stimulated by light even though their mRNAs were present. Furthermore, individual chloroplast proteins showed different patterns of synthesis within the cell cycle [17,24,25,51]. In the late dark period some synthesis of EF-Tu might occur, explaining the increased concentration of EF-Tu observed at the beginning of the light period. In synchronized C. reinhardii the highest rate of EF-Tu synthesis preceded the peak of maximal chloroplast protein synthesis and also the highest rate of synthesis of the LS and of the 32 kDa protein [17]. Therefore, the abundance of EF-Tu during the cell cycle could influence chloroplast protein synthesis. In C. reinhardii the transcript levels of t u f a increased when chloroplast protein synthesis was inhibited with chloramphenicol or dark treatment [49]. The authors speculated that t u f a expression may be an important regulatory mechanism in the chloroplast. The lack of EF-Tu synthesis in the second half of the light period could limit this factor for chloroplast translation and therefore be responsible for the decreasing chloroplast translational activity at this time (Ref. 43; and Leu et al., unpublished data). Therefore we determined the abundance of EF-Tu per cell by quantification on Western blots. We found an almost constant amount of EF-Tu per cell throughout the cell cycle. However, increasing chloroplast volume of growing cells led to a significant reduction of the concentration of EF-Tu during the light period. We have no data about the exact volume of chloroplasts during the cell cycle. Since cell volume and chlorophyll content per cell increase in parallel during the light period, it seems reasonable to assume that the chloroplast volume increases also in parallel. Therefore, the decrease in the relative concentration of EF-Tu with increasing cell volume also reflects the decrease of the concentration in the chloroplast (Fig. 3). The concentration of EF-Tu and the activity of chloroplast protein synthesis [17] were not correlated in the first half of the light period. A further indication against a limiting function of EF-Tu comes from reports that in E. coli the EF-Tu exceeds in concentration the other elongation factors and the ribosomes by a factor of about 10 [52]. At this time we have no direct evidence for a regulatory role of EF-Tu on chloroplast protein synthesis.
The m R N A for EF-Tu and its distribution between thylakoids and stroma The complex hybridization pattern of the tufA specific gene probe with chloroplast RNA points to a complicated m R N A processing or turnover mechanism. It is yet unknown whether the tufA gene in chloroplasts of C. reinhardii is transcribed polycistronically as the tufB operon or the str operon containing the tufa gene in E. coli [53]. Although we have no proof that the high molecular transcript of 3 kb is the primary transcript from which the 2.1 kb m R N A is derived, it is worth to note that this presumptive primary transcript was exclusively found in the stroma (Fig. 4b). The main band of 2.1 kb was mainly associated with thylakoids and the molecular size of our main band corresponds well to published data [49,54]. Similar to our results, Silk and Wu [49] also found extensive tuf A degradation products in their hybridizations. Furthermore, our Northern blots indicated a pronounced degradation of the tufA transcript towards the middle of the light period, when EF-Tu synthesis rapidly declined. This correlates with the finding that transcription of t u f a occurs immediately at the beginning of the light period but that the abundance of m R N A decreases around the middle of the light period (Ref. 43; and Leu et al. unpublished data). The m R N A for EF-Tu was distributed in an almost constant ratio between thylakoids and stroma during the light period of the cell cycle. This differed from the mRNAs for the LS and the 32 kDa protein, which move to the membrane fraction during the light period, but were predominantly in the soluble fraction during the dark [17]. In contrast, the m R N A for psaB was bound to the membrane during the whole cell cycle. This m R N A was also exclusively found in membrane-bound form in etioplasts and chloroplasts of barley [19] and of Vicia faba [18]. At L2, when protein synthesis was maximal, the content of mRNA for EF-Tu was highest. About 80% of the present mRNA was associated with thylakoid-bound polysomes. Therefore, it seems reasonable to postulate that the actual site of chloroplast translation is the thylakoid membrane, independently of the nature of the protein. Indeed all chloroplast m R N A species investigated so far were found in membranebound form during the time of intensive protein synthesis [15,17-19].
Acknowledgements We thank J. Palmer and S. Baldauf for the cloned DNA probe, and E. Vijgenboom for the antisera. This work was partly supported by the Swiss National Foundation for Scientific Research to A.B. and E.B. (grant 3.506-0.83) and by the National Science Foundation to A.M. and S.L. (grant DCB-8607745).
216 References 1 Falk, H. (1969) J. Cell Biol. 42, 582-587. 2 Chua, N.-H., Blobel, G., Siekevitz, P. and Palade, G.E. (1973) Proc. Natl. Acad. Sci. USA 70, 1554-1558. 3 Chua, N.-H., Blobel, G., Siekevitz, P. and Palade, G.E. (1976) J. Cell Biol. 71,497-514. 4 Chen, J.L. and Wildman, S.G. (1970) Biochim. Biophys. Acta 209, 207-219. 5 Margulies, M.M. and Michaels, A. (1975) Biochim. Biophys. Acta 402, 297-308. 6 Margulies, M.M. (1986) in Regulation of Chloroplast Differentiation, pp. 171-190, Alan R. Liss, New York. 7 Herrin, D., Michaels, A. and Hickey, E. (1981) Biochim. Biophys. Acta 655, 136-145. 8 Herrin, D. and Michaels, A. (1985) FEBS Lett. 184, 90-95. 9 Minami, E.-I., Shinohara, K., Kuwabara, T. and Watanabe, A. (1986) Arch. Biochem. Biophys. 244, 517-527. 10 Minami, E.-I. and Watanabe, A. (1984) Arch. Biochem. Biophys. 235, 562-570. 11 Herrin, D. and Michaels, A. (1985) Arch. Biochem. Biophys. 237, 225-236. 12 Bhaya, D. and Jagendorf, A.T. (1984) Plant Mol. Biol. 3, 277-280. 13 Margulies, M.M., Tiffany, H.L. and Hattori, T. (1987) Arch. Biochem. Biophys. 254, 454-461. 14 Leu, S., Bolli, R., Mendiola-Morgenthaler, L. and Boschetti, A. (1984) Planta 160, 204-211. 15 Minami, E.-I., Shinohara, K., Kwakami, N. and Watanabe, A. (1988) Plant Cell Physiol. 29, 1303-1309. 16 Breidenbach, E., Jenni, E., Leu, S. and Boschetti, A. (1988) Plant Cell Physiol. 29, 1-7. 17 Breidenbach, E., Jenni, E. and Boschetti, A. (1988) Eur. J. Biochem. 177, 225-232. 18 Friemann, A. and Hachtel, W. (1988) Planta 175, 50-59. 19 Klein, R.R., Mason, H.S. and Mullet, J.E. (1988) J. Cell Biol. 106, 289-301. 20 Hattori, T. and Margulies, M.M. (1986) Arch. Biochem. Biophys. 244, 630-640. 21 Bhaya, D. and Jagendorf, A.T. (1985) Arch. Biochem. Biophys. 237, 217-223. 22 Bolli, R., Mendiola-Morgenthaler, L. and Boschetti, A. (1981) Biochim. Biophys. Acta 653, 276-287. 23 Fish, L.E. and Jagendorf, A.T. (1982) Plant Physiol. 69, 814-825. 24 Howell, S.H., Posakony, W.J. and Hill, K.R. (1977) J. Cell Biol. 72, 223-241. 25 Iwanij, V., Chua, N.-H. and Siekevitz, P. (1975) J. Cell Biol. 64, 572-585. 26 Herrin, D.L., Michaels, A.S. and Paul, A.-L. (1986) J. Cell Biol. 103, 1837-1845. 27 Boschetti, A., Breidenbach, E., Leu, S. and Clemetson-Nussbaum, J. (1987) in Algal Development (Molecular and Cellular Aspects), (Wiessner, W., Robinson, D. and Starr, R.C., eds.), pp. 114-122, Springer-Verlag, Berlin.
28 Brandt, P., Winter, J., Von Kessel, B., Kohnke, B. (1987) in Algal Development (Molecular and Cellular Aspects), (Wiessner, W., Robinson, D.G. and Starr, R.C., eds.), pp. 134-141, SpringerVerlag, Berlin. 29 Brandt, P., Breidenbach, E., Prestin, B. and Boschetti, A. (1989) J. Plant Physiol. 134, 420-426. 30 lnamine, G., Nash, B., Weissbach, H. and Brot, N. (1985) Proc. Natl. Acad. Sci. USA 82, 5690-5694. 31 Klein, R.R. and Mullet, J.E. (1986) J. Biol. Chem. 261, 11138-11145. 32 Klein, R.R. and Mullet, J.E. (1987) J. Biol. Chem. 262, 4341-4348. 33 Laing, W., Kreuz, K. and Apel, K. (1988) Planta 176, 269-276. 34 Deng, X.-W. and Gruissem, W. (1987) Cell 49, 379-387. 35 Berry, J.O., Carr, J.P. and Klessig, D.F. (1988) Proc. Natl. Acad. Sci. USA 85, 4190-4194. 36 Bouet, C., Schantz, R., Dubertret, G., Pineau, B. and Ledoigt, G. (1986) Planta 167, 511-520. 37 Monroy, A.F., McCarthy, S.A. and Schwartzbach, S.D. (1987) Plant Sci. 61-76. 38 Ellis, R.J. (1977) Biochim. Biophys. Acta 463, 185-215. 39 Watson, J.C. and Surzycki, S.J. (1982) Proc. Natl. Acad. Sci. USA 79, 2264-2267. 40 Montandon, P.E. and Stutz, E. (1983) Nucleic Acids Res. 11, 5877-5892. 41 Sreedharan, S., Beck, C.M. and Spremulli, L.L. (1985) J. Biol. Chem. 260, 3126-3231. 42 Spremulli, G.H. and Spremulli, L.L. (1987) Biochem. Biophys. Res. Commun. 148, 1490-1495. 43 Leu, S., Herrin, D. and Michaels, A. (1989) in Proceedings of the Brauschweig Symposium of Applied Plant Molecular Biology (Galling, G., ed.), Zentralstelle ftir Weiterbildung der Technischen Universit~it Braunschweig, Braunschweig, F.R.G. pp. 22-37. 44 Mendiola-Morgenthaler, L., Leu, S. and Boschetti, A. (1985) Plant Sci. 38, 33-39. 45 Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: a Laboratory Manual, pp. 188-189, Cold Spring Harbor Laboratory, New York, Cold Spring Harbor, NY. 46 Laemmli, U.K. (1970) Nature (Lond.) 227, 680-685. 47 Vernon, L.P. (1960) Anal. Chem. 32, 1144-1150. 48 Ktick, U., Choquet, Y., Schneider, M., Dron, M. and Bennoun, P. (1987) EMBO J. 6, 2185 2195. 49 Silk, G.W. and Wu, M. (1988) Curr. Genet. 14, 119-126. 50 Tiboni, O. and Ciferri, O. (1986) Methods Enzymol. 118, 296-308. 51 Herrin, D. and Michaels, A. (1984) in Recombinant DNA and Cell Proliferation (Stein, G. and Stein, S.L., eds.), pp. 87-106, Academic press, Orlando, FL. 52 Van der Meide, P.H., Vijgenboom, E., Talens, A., Bosch, L. (1983) Eur. J. Biochem. 130, 397-407. 53 Van Delft, J.H.M., Schmidt, D.S. and Bosch, L. (1987) J. Mol. Biol. 197, 647-657. 54 Montandon, P.E. and Stutz, E. (1989) Experientia 45, A-33.
Biochimica et Biophysica Acta, 1048 (1990) 209-216 Elsevier
BBAEXP 92038
Synthesis of EF-Tu and distribution of its m R N A between stroma and thylakoids during the cell cycle of Chlamydomonas reinhardii Eric Breidenbach 1, Stefan Leu 2, Allan Michaels 2 and Arminio Boschetti 1 z Institutfftr Biochemie, Universitiit Bern, (Switzerland) and 2 Department of Biology, Ben Gurion University, Beer Sheva (Israel)
(Received 4 September 1989) (Revised manuscript received 30 November 1989)
Key words: Chloroplast; Protein synthesis; Elongation factor EF-Tu; mRNA content; Cell cycle; (C. reinhardii)
In Chlamydomonas reinhardii the elongation factor EF-Tu is encoded in the chloroplast DNA. We identified EF-Tu in the electrophoretic product pattern of chloroplast-made proteins and showed that this protein is only synthesized in the first half of the light period in synchronized cells. The newly synthesized EF-Tu contributed little to the almost invariable content of EF-Tu in chloroplasts during the light period of the cell cycle. However, increasing cell volume and the lack of EF-Tu synthesis in the second half of the light period led to a decrease in the concentration of EF-Tu in chloroplasts. At different times in the vegetative cell cycle, the RNA was extracted from whole chloroplasts and from free and thylakoid-bound chloroplast polysomes. The content of mRNA of EF-Tu in chloroplasts and the distribution between stroma and thylakoids were determined. During the light period, the content of the mRNA for EF-Tu varied in parallel to the rate of EF-Tu synthesis. However, in the dark, some mRNA was present even in the absence of EF-Tu synthesis. Most of the mRNA was bound to thylakoids during the whole cell cycle. This suggests that synthesis of EF-Tu is associated with thylakoid membranes.
Introduction
In chloroplasts there are two populations of ribosomes and polysomes with respect to their localization [1-5]: The ribosomes and polysomes remaining in the stroma and the thylakoid-bound ribosomes and polysomes which are anchored to the membrane, either by the nascent peptide, by electrostatic interaction, by specific binding of ribosomes or by a combination of these interactions (see review, Ref. 6). It was postulated that membrane-bound polysomes synthesize membrane proteins [7-13]. However, these polysomes also contain mRNAs for soluble proteins and for extrinsic membrane proteins. These mRNAs were first detected by in vitro translation of extracted RNA and quantified by hybridization techniques [10,14-19]. To date the capacity of these polysomes to
Abbreviations: 32 kDa protein, 32 kDa herbicide-binding membrane protein of Photosystem II; CF1, coupling factor 1 of chloroplast ATP-synthase; LS, large subunit of the ribulose-l,5-bisphosphate carboxylase/oxygenase (EC 4.1.1.39); PAGE, polyacrylamide gel electrophoresis. Correspondence: Professor Arminio Boschetti, Institut fiJr Biochemie, Universit~it Bern, Freiestr. 3, CH-3012 Bern, Switzerland.
translate the mRNA for soluble proteins is still under discussion. The large subunit of ribulose-l,5-bisphosphate carboxylase/oxygenase (LS) was originally not found among the products of run-off translation of membrane-bound polysomes [10,14]. However, Klein et al. [19] and Hattori and Margulies [20] identified the LS as product of thylakoid-bound polysomes and Bhaya and Jagendorf [21] showed that the a- and /3-subunits of CF t were produced by both thylakoid-bound and free polysomes. The amount of membrane-bound ribosomes and polysomes varies during the cell cycle of synchronized C. reinhardii [3,22] and in greening peas [23]. Chloroplast proteins are synthesized in a cell cycle-dependent manner in synchronized C. reinhardii [17,24-27] or in Euglena gracilis [28,29]. Therefore, the partition of ribosomes and polysomes between stroma and thylakoids may be a central point in regulation of protein synthesis in chloroplasts. During light-dependent chloroplast formation of higher plants [30-35] and algae [36,37] or during the cell cycle of synchronized algae [17,26,29,43] the rate of protein synthesis varies much more than the abundance of the mRNAs for most of the investigated proteins. Many chloroplast mRNAs are present in high concentrations during the entire cell cycle and may be
0167-4781/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
210 stored in a silent form when protein synthesis is reduced. Chloroplast mRNA can be found in arrested polysomes. In amaranth seedlings, the translation of the LS was immediately arrested after a light-to-dark transition, but the mRNA remained bound to the polysomes for several hours [35]. The mRNA for the P-700 chlorophyll a protein of Photosystem I was almost exclusively confined to the membrane-bound polysomes in etiolated dark-grown barley seedlings and in chloroplasts of illuminated plants. Although similar levels of this mRNA were present in etioplasts and chloroplasts, no synthesis of P-700 chlorophyll a protein could be detected in etioplasts. A very short exposure to light induced P-700 chlorophyll a protein synthesis [33]. The mRNA for the 32 kDa membrane protein in the stroma of spinach chloroplasts was associated with proteins, but not with ribosomes [15]. In our previous work we analyzed the distribution of the mRNA for the 32 kDa protein and the LS between membrane-bound polysomes and stroma polysomes during the cell cycle of C. reinhardii. In the light, when chloroplast protein synthesis was stimulated, the mRNA for the soluble LS was enriched in membrane-bound polysomes in the same way as the mRNA for the 32 kDa membrane protein. In addition, the distribution for both mRNAs between thylakoids and stroma fluctuated in parallel during the cell cycle [17]. This contradicts the hypothesis that membrane-bound polysomes synthesize only membrane proteins and stroma polysomes synthesize soluble proteins [38]. One could argue that the subchloroplast distribution of the mRNA for the LS reflects an artefact because of the well known 'stickiness' of the growing peptide of the LS to membranes and other cell structures. To confirm our previous findings, we investigated the distribution of the mRNA for the soluble chloroplast elongation factor EF-Tu, which in addition is not directly involved in photosynthesis. In contrast to higher plants, the gene tufA coding for the chloroplast elongation factor EF-Tu is located on the chloroplast D N A in C. reinhardii [39] and in Euglena [40], and is homologous to EF-Tu of E. coli [41,42]. The mRNA for EF-Tu has been found to show a more pronounced fluctuation than other mRNAs investigated during the cell cycle of C. reinhardii [17,43]. Furthermore, the rate of synthesis and the abundance of EF-Tu were studied during the cell cycle, to establish possible involvement of EF-Tu in the regulation of chloroplast protein synthesis. Materials and Methods
Culture conditions. Cultures of Chlamydomonas reinhardii cw-15 (stock cc-277) were obtained from the Chlamydomonas Genetics Center, Duke University,
Durham, NC. Growth conditions and synchronization procedure were as previously described [22]. Determination of the cell volume. Cell size was measured electronically with a Coulter Counter, Model ZB (50 /~m orfice) particle counter connected to a Coulter channelyzer C-1000 (Coulter Electronics, U.K.). Labeling of cells in vivo with [35S]sulfate. Synchronously cultivated C. reinhardii were grown under a 14:10 hours l i g h t / d a r k regime. Samples were taken periodically from the culture and harvested by centrifugation, washed twice with culture medium without sulfate and brought to 5 . 1 0 7 cells/ml. The cells were preincubated for 10 min with 10/~g/ml cycloheximide and labeled in the presence of 10/~g/ml cycloheximide and [35S]sulfate for 60 min in the light or dark as previously described [17]. Isolation of chloroplasts. Chloroplasts were isolated according to the procedure of Mendiola-Morgenthaler et al. [44] with the modifications previously described [17]. Fractionation of chloroplasts. Isolated chloroplasts were fractionated as described [16], except that 10 mmol/1 ribonucleoside-vanadyl complexes were added as an RNase inhibitor before breaking the chloroplasts in the Yeda press [45]. Electrophoretic analysis of proteins. Labeled cells were broken by freezing and ultrasonication and the lysate was centrifuged at 30000 × g for 15 min. The supernatant, containing the soluble fraction, was immediately prepared for electrophoresis. The pellet, containing the membranes, was washed twice with 100 mmol/1 TrisHCI (pH 7.6)/1 mmol/1 MgC12, and solubilized in the same volume as the corresponding supernatant. Aliquots of labeled cells or cell fractions were solubilized in 3% SDS and 2% 2-mercaptoethanol at 80 ° C for 5 min and the proteins were separated on linear SDS-polyacrylamide gradient gels (10-20%) in the buffer system of Laemmli [46]. The gels were either stained with Coomassie brilliant blue R250, treated with Amplify (Amersham), dried and exposed to X-ray film (Hyperfilm-MP, Amersham), or the separated proteins were transferred electrophoretically onto nitrocellulose filter.
Immunological detection and quantification of EF-Tu. The nitrocellulose filters containing electrophoretically separated proteins (Western blots) were treated with antisera raised against the EF-Tu of Escherichia coli (a gift of Erik Vijgenboom, Leiden University, The Netherlands). The bound antibodies were detected with an anti-rabbit IgG horseradish peroxidase conjugate (Sigma) and enzymatic staining with 4-chloro-l-naphthol. The relative amounts of EF-Tu present in different probes of the same Western blot were measured using a scanning spectrophotometer in the reflection mode (CAMAG). Determination of labeling of distinct proteins. The dried gels were soaked in 7% acetic acid and the respective
211 bands were cut out. The gel slices were bleached and dissolved in H202 (30%) and the radioactivity determined by liquid scintillation spectrometry. Chlorophyll determination was according to Vernon [47]. Cloned DNA probes. An internal PstI-EcoRI fragment of the tufA gene of C. reinhardii cloned in the vector pUC8 was obtained from S. Baldauf and J. Palmer. The psaB gene was subcloned into pGem from EcoRI-fragment 15 of C. reinhardii chloroplast D N A [48,43]. Quantification of mRNA. Extraction of R N A and Northern blot analysis were done as previously described [16]. The intensity of the respective hybridizing signal was determined and integrated in the linear range of the X-ray film with the Shimadzu high speed TLS scanner CS-920.
with [35S]sulfate for 60 min in the presence of cycloheximide [17]. Product analysis by fluorography of electrophoretically separated proteins showed a peptide of about 45 kDa in whole cells and in the soluble fraction ($30) (Fig. la). This protein was identified as the translational elongation factor EF-Tu by crossreaction with antibodies raised against EF-Tu of E. coil on Western blots (Fig. la). The rate of EF-Tu synthesis during the cell cycle was determined by pulse-labeling of synchronized cells harvested from the dark period of the cell cycle (hours 0-10) and from the light period (hours 10-24). In the following the dark hours will be referred to as D 1 - D 1 0 , and the light hours as L0-L14. Proteins of equal numbers of cells were electrophoretically separated and the radioactivity was measured in solubilized gel slices containing the soluble 45 kDa protein. The rate of synthesis of EF-Tu changed significantly during the cell cycle, from nearly undetectable amounts in the dark to one of the main products at the beginning of the light period (Fig. l b and Fig. 2). EF-Tu was not labeled in cells harvested in the middle of the dark period, whether incubated in the light or in the dark. At the onset of light the rate of EF-Tu synthesis was already elevated
Results
Synthesis of the elongation factor EF-Tu during the cell cycle Chloroplast-made proteins were labeled by incubation of synchronized cells of Chlamydomonas reinhardii
A
B
C
kDa
93
_
32 kDprotein 2 5 --
14--
E
c m ..... ................
s
c
m
s
c
m
s
L 1 (I
L2(I) L6(I) L13(I) D2(d) D6(d)
- " ' "I " I"I " Lll2(I) L4(I)
I
L8(I) D2[I) D6(I]
Fig. 1. Electrophoretic analysis of [aSS]sulfate-labeled cells of C. reinhardii. (a) In vivo labeled cells (c) and corresponding aliquots of membranes (m) or of soluble fraction (s) were separated by SDS-gel electrophoresis and transferred onto nitrocellulose. (A) Coomassie blue-stained protein patterns of lysed cells and cell fractions of C. reinhardii and of molecular mass markers; (B) Autoradiographof the Western blot of the samples in A; (C) Immunostainingof the Western blot with antibodies against EF-Tu. (b) Comparison of the content of EF-Tu in whole cells with the rate of its synthesis during one 24 h cell cycle.The cells were harvested at indicated hours in the dark period (hours 1-10 : D0-D10) or in the light period (hours 10-24 : L0-L14) and labeled in the presence of cycloheximidein the light (1) or in the dark (d) for 60 min. Equal numbers of cells (2.73-106 cells, corresponding to 1.9-6.8 #g chlorophyll) were separated on SDS-PAGE. The proteins were transferred onto nitrocellulose. (D) Autoradiograph of the Western blot; (E) Immunostainingof the same blot with antibodies against EF-Tu.
212 100 r
E~
8oi 60-
2
!
¥
4OL
2OF
D2
D4
D6
DIS LO L2
L4
L6
L8
LlO L12 L14
Fig. 2. Rate o f synthesis of E F - T u d u r i n g the cell cycle. Synchronized
cells from the dark period (D0-D10), and from the light period (L0-L14) were labeled for 60 min in the dark (*) or in the light (n). After SDS-PAGE, the incorporation of radioactivity into the band identified as EF-Tu was measured. All values of three different gels were standardized to the same cell density and the highest incorporation during the cell cycle was taken as 100%. In this set of experiments we did not determine the rate of EF-Tu synthesis between D6 and DI0. However, fluorographs of previous experiments confirmed that the respectiveband was scarcelylabeled at the end of the dark period [171.
and reached its maximum between L1 and L5. Towards the second half of the light period the rate of synthesis declined rapidly to less than 15% of the highest rate.
Content of the elongation factor EF-Tu during the cell cycle The cell cycle-dependent synthesis of EF-Tu may play a key role in the regulation of protein synthesis in chloroplasts [43,49]. We therefore determined the content of this protein in chloroplasts during the cell cycle by immunostaining of Western blots after electrophoretic separation of proteins from an equal number of cells. Since C. reinhardii cells contain only one chloroplast, we concluded that the content of EF-Tu per chloroplast was almost constant during an entire cell cycle (Fig. lb). The weakening of the band at D6 1 --
-~
~
200
150
'~ 02
O
~
I D2
~ - .~U ~
~
~
5
0
>o
__
I DI6 8/ I , D4 D LQ L2
J L4
, L6
J L8
~ L L1Q L12
fo
Fig. 3. C o n t e n t and concentration of E F - T u d u r i n g the cell cycle. The
concentration of EF-Tu U7) was calculated by dividing the relative
content of EF-Tu per cell (*) by the cell volume (1:3)(all determinations of two independent experiments are shown).
reflects simply the reduction of the cell size, and hence of the amount of EF-Tu applied to the gel, after the cells had divided into four daughter cells between D2 and D6. The amounts of EF-Tu present at each time point were quantified by scanning the Western blots and expressed as relative content of EF-Tu per number of cells (Fig. 3). We estimated the concentration of EF-Tu by dividing the relative content by the relative cell volume, as determined with the Coulter Counter. There was a 4-fold decrease of the concentration of EF-Tu during the light period (Fig. 3). To test the method of quantification of EF-Tu by immunostaining, we separated a set of different dilutions of the soluble fraction ($30) from in vivo labeled cells on SDS-PAGE and blotted them onto nitrocellulose. The radioactivity of EF-Tu and the immunological quantification were linearly related at least over one order of magnitude (results not shown).
mRNA content in intact chloroplasts during the cell cycle The relative abundance of the m R N A for EF-Tu has been determined during the cell cycle by hybridization of Northern blots with a gene probe specific for EF-Tu (Fig. 4a). Some m R N A was detected in the dark. Its relative abundance increased to a maximum at L2, and then declined to almost undetectable levels at the end of the light period. The pronounced variation in the amount of this m R N A during the cell cycle differed significantly from the one we measured previously for the LS and for the 32 k D a protein [17] and psaB (Ref. 43; and Leu et al., unpublished data). The hybridization pattern obtained showed one main band at 2.1 kb, a weaker band at 1.7 kb and an additional faint band at 3 kb. At L2 and L6 considerable hybridization was obtained with low molecular mass RNA, probably degradation products of the m R N A for EF-Tu. To test the intactness of the extracted R N A we rehybridized some selected Northern blots with a specific gene probe for psaB encoding an apoprotein of the reaction center of Photosystem I [48] (Fig. 5). The degree of the hybridization signal in low molecular R N A seemed to be reduced. Therefore, the hybridization of the tufA probe with low molecular R N A points to a cell cycle specific instability of this transcript. Because of the complex hybridization pattern, the content of the m R N A for EF-Tu was quantified by densitometric scanning of the main transcript (2.1 kb) and compared to the rate of EF-Tu synthesis (Table I and Fig. 2). To compensate for the varying content of R N A in chloroplasts during the cell cycle, we determined the extractable R N A from isolated chloroplasts at each time and corrected our values. During the light period the rate of EF-Tu synthesis paralleled the m R N A content and therefore seemed to be regulated by m R N A abundance; while at D6, in spite of abundant m R N A , no EF-Tu synthesis was observed.
213
2.1kb
1.7kb
b
D6
D9~/4
LO
L2
L6
L13
--3kb --
c
rn
s
c
rn
s
c
m
s
c
m
s
c
m
s
c
m
21
kb
s
Fig. 4. Autoradiographs of Northern blots, hybridized with the nick translated gene probe for the elongation factor EF-Tu. (a) At indicated times during the cell cycle equal amounts of RNA from chloroplasts were separated by formamide/formaldehyde electrophoresis and blotted onto nitrocellulose; (b) equal amounts of RNA from chloroplasts (c), from the membrane fraction (m) and from the stroma (s) were separated electrophoretically and blotted onto nitrocellulose.
Distribution of mRNA between stroma and thylakoids The distribution of the m R N A for EF-Tu between stroma and thylakoids was determined and compared to the distribution of psaB m R N A and to the m R N A for LS and the 32 k D a protein previously studied. Our Northern blots, containing equal amounts of R N A per lane, demonstrated the presence of m R N A for the EF-Tu in the R N A extracted from stroma polysomes and from thylakoid-bound polysomes (Fig. 4b). However, in previous studies we showed that the amount of total R N A bound to thylakoid membranes increased
LO
at the beginning of the light period [17]. After 2 h in the light, about 80% of the R N A was bound to thylakoids, whereas only 20% remained in the stroma. Towards the end of the light period and in the dark, we extracted about equal amounts of R N A from thylakoids and the stroma fraction (Table II). Therefore, the distributions of m R N A for EF-Tu between stroma and thylakoids were calculated by multiplication of the relative contents of the m R N A in these fractions with the partition of total R N A between stroma and thylakoids and with the amount of EF-Tu m R N A in chloroplasts during the
L2
LO
2.9
__
L2
m
2.1 kb
C
FFI
S
C
m
S
c
m
S
C
m
S
Fig. 5. Autoradlograph of Northern blots hybridized with the gene probes tufA and psaB. A selected Northern blot with the RNA extracted from whole chloroplasts (c), from the membrane fraction (m) and from the stroma (s) was rehybridized with the gene probe of psaB: (a) Hybridized with the gene probe tufA; (b) Hybridized with the gene probe psaB.
214 TABLE I
Quantification of mRNA for EF-Tu in chloroplasts
t~
The relative content of m R N A for EF-Tu in whole chloroplasts was calculated. The amounts of extractable R N A from isolated chloroplasts from different times of the cell cycle were multiplied with the intensities of the corresponding hybridizing signals for EF-Tu m R N A on Northern blots, determined by densitometric scanning of the autoradiography (Fig. 4a) (e.g., D6: 6th hour of the dark period; L2: 2nd hour of the light period). The values at D6 were taken as 1. In order to compare the m R N A content with the rate of EF-Tu synthesis (Fig. 2), the highest content of m R N A was taken as 100%
c
08 e6
g u
04
r i
0.2F
0
Time
D6 D9 3 / 4 L0 L2 L6 LI3
Extracted R N A
Hybridizing
m R N A content
(rel. units)
signal (rel. units)
(rel. units)
(%)
1 1 0.9 1.3 2.4 2.8
1 0.9 1.7 2.0 0.4 < 0.1
1.0 0.9 1.6 2.6 1.0 < 0.3
39 35 61 100 38 <10
DI2
i D4
i D6
i D~,
I LO
i L2
i L4
t L6
~ L8
i LIO
1 L12
L14
Fig. 6. Distribution of m R N A for EF-Tu between thylakoids (*) and stroma (rl) during the cell cycle. The values have been calculated as described in the text and in Table II.
of thylakoids between free and membrane-bound polysomes in algea and higher plants. Less data are available about the distribution of mRNAs for soluble proteins. The mRNA for the LS was associated to a significant degree with membrane-bound polysomes, but other mRNAs for soluble proteins have not been tested until
cell cycle (Table II). By these corrections, it became obvious that about 70-80% of the mRNA was membrane-bound at all times of the cell cycle (Fig. 6). The Northern blots rehybridized with the gene probe specific for psaB showed that the mRNA for the respective membrane protein was mainly membrane-bound (Fig. 5b). Furthermore, the 5 S tRNA band visible in ethidium bromide stained RNA gels was observed only in the soluble fraction (not shown). These findings confirmed, that the distribution of the mRNAs was real and not due to an artefact of the fragmentation of chloroplasts. Discussion
There are several reports concerning the distribution of mRNAs for intrinsic or extrinsic membrane proteins
now.
Abundance and synthesis of EF-Tu during the cell cycle The chloroplast EF-Tu was found exclusively in the soluble fraction of lysed cells and chloroplasts and was one of the predominantly labeled chloroplast-made proteins in C. reinhardii (Fig. 1). The very high homology of EF-Tu of E. coli with chloroplast EF-Tu [41,42] made it possible to identify this protein immunologically with antibodies raised against the E. coli protein. The identified protein has a molecular mass of 45 kDa which corresponds well to data of EF-Tu isolated from spinach chloroplasts [50], Euglena and E. coli [41,50]. Synthesis of EF-Tu was light dependent and increased strongly at the beginning of the light period. The rate of synthesis paralleled the content of the mRNA for EF-Tu during the light period of the cell
TABLE II
Distribution of mRNA for EF-Tu between thylakoids and stroma during the cell cycle The subchloroplast distribution of the m R N A was calculated. The relative m R N A content in the respective fraction, determined by hybridization of Northern blots of equal amounts of RNA, was multiplied with the content of m R N A in chloroplasts (Table I) and the partition of the extractable R N A between membranes (m) and stroma (s), respectively [17]. Time
D6 D93/4 L0 L2 L6 L13
m R N A content in chloroplasts
Partition of R N A (rel. units)
Relative m R N A content (rel. units)
Distribution of m R N A (rel. units)
(rel. units)
m
s
m
s
m
s
1 0.9 1.6 2.6 1 <0.3
0.5 0.6 0.6 0.8 0.7 0.5
0.5 0.4 0.4 0.2 0.3 0.5
0.9 0.8 0.7 0.5 0.7 0.9
0.1 0.2 0.3 0.5 0.3 0.1
0.5 0.4 0.6 1.0 0.5 <0.1
<0.1 0.1 0.2 0.3 0.1 <0.1
215 cycle, but not in the dark (Table I and Fig. 2). Therefore in the light, expression of EF-Tu might be regulated by mRNA abundance. In the dark at D6 however, significant amounts of EF-Tu transcript were present, which were not translated even if the cells were transferred to the light. This finding points to a stage-dependent competence of EF-Tu synthesis during the cell cycle similar to our previous observations [17], where we showed that in synchronized C. reinhardii harvested in the dark period the synthesis of LS and of the 32 kDa protein could hardly be stimulated by light even though their mRNAs were present. Furthermore, individual chloroplast proteins showed different patterns of synthesis within the cell cycle [17,24,25,51]. In the late dark period some synthesis of EF-Tu might occur, explaining the increased concentration of EF-Tu observed at the beginning of the light period. In synchronized C. reinhardii the highest rate of EF-Tu synthesis preceded the peak of maximal chloroplast protein synthesis and also the highest rate of synthesis of the LS and of the 32 kDa protein [17]. Therefore, the abundance of EF-Tu during the cell cycle could influence chloroplast protein synthesis. In C. reinhardii the transcript levels of t u f a increased when chloroplast protein synthesis was inhibited with chloramphenicol or dark treatment [49]. The authors speculated that t u f a expression may be an important regulatory mechanism in the chloroplast. The lack of EF-Tu synthesis in the second half of the light period could limit this factor for chloroplast translation and therefore be responsible for the decreasing chloroplast translational activity at this time (Ref. 43; and Leu et al., unpublished data). Therefore we determined the abundance of EF-Tu per cell by quantification on Western blots. We found an almost constant amount of EF-Tu per cell throughout the cell cycle. However, increasing chloroplast volume of growing cells led to a significant reduction of the concentration of EF-Tu during the light period. We have no data about the exact volume of chloroplasts during the cell cycle. Since cell volume and chlorophyll content per cell increase in parallel during the light period, it seems reasonable to assume that the chloroplast volume increases also in parallel. Therefore, the decrease in the relative concentration of EF-Tu with increasing cell volume also reflects the decrease of the concentration in the chloroplast (Fig. 3). The concentration of EF-Tu and the activity of chloroplast protein synthesis [17] were not correlated in the first half of the light period. A further indication against a limiting function of EF-Tu comes from reports that in E. coli the EF-Tu exceeds in concentration the other elongation factors and the ribosomes by a factor of about 10 [52]. At this time we have no direct evidence for a regulatory role of EF-Tu on chloroplast protein synthesis.
The m R N A for EF-Tu and its distribution between thylakoids and stroma The complex hybridization pattern of the tufA specific gene probe with chloroplast RNA points to a complicated m R N A processing or turnover mechanism. It is yet unknown whether the tufA gene in chloroplasts of C. reinhardii is transcribed polycistronically as the tufB operon or the str operon containing the tufa gene in E. coli [53]. Although we have no proof that the high molecular transcript of 3 kb is the primary transcript from which the 2.1 kb m R N A is derived, it is worth to note that this presumptive primary transcript was exclusively found in the stroma (Fig. 4b). The main band of 2.1 kb was mainly associated with thylakoids and the molecular size of our main band corresponds well to published data [49,54]. Similar to our results, Silk and Wu [49] also found extensive tuf A degradation products in their hybridizations. Furthermore, our Northern blots indicated a pronounced degradation of the tufA transcript towards the middle of the light period, when EF-Tu synthesis rapidly declined. This correlates with the finding that transcription of t u f a occurs immediately at the beginning of the light period but that the abundance of m R N A decreases around the middle of the light period (Ref. 43; and Leu et al. unpublished data). The m R N A for EF-Tu was distributed in an almost constant ratio between thylakoids and stroma during the light period of the cell cycle. This differed from the mRNAs for the LS and the 32 kDa protein, which move to the membrane fraction during the light period, but were predominantly in the soluble fraction during the dark [17]. In contrast, the m R N A for psaB was bound to the membrane during the whole cell cycle. This m R N A was also exclusively found in membrane-bound form in etioplasts and chloroplasts of barley [19] and of Vicia faba [18]. At L2, when protein synthesis was maximal, the content of mRNA for EF-Tu was highest. About 80% of the present mRNA was associated with thylakoid-bound polysomes. Therefore, it seems reasonable to postulate that the actual site of chloroplast translation is the thylakoid membrane, independently of the nature of the protein. Indeed all chloroplast m R N A species investigated so far were found in membranebound form during the time of intensive protein synthesis [15,17-19].
Acknowledgements We thank J. Palmer and S. Baldauf for the cloned DNA probe, and E. Vijgenboom for the antisera. This work was partly supported by the Swiss National Foundation for Scientific Research to A.B. and E.B. (grant 3.506-0.83) and by the National Science Foundation to A.M. and S.L. (grant DCB-8607745).
216 References 1 Falk, H. (1969) J. Cell Biol. 42, 582-587. 2 Chua, N.-H., Blobel, G., Siekevitz, P. and Palade, G.E. (1973) Proc. Natl. Acad. Sci. USA 70, 1554-1558. 3 Chua, N.-H., Blobel, G., Siekevitz, P. and Palade, G.E. (1976) J. Cell Biol. 71,497-514. 4 Chen, J.L. and Wildman, S.G. (1970) Biochim. Biophys. Acta 209, 207-219. 5 Margulies, M.M. and Michaels, A. (1975) Biochim. Biophys. Acta 402, 297-308. 6 Margulies, M.M. (1986) in Regulation of Chloroplast Differentiation, pp. 171-190, Alan R. Liss, New York. 7 Herrin, D., Michaels, A. and Hickey, E. (1981) Biochim. Biophys. Acta 655, 136-145. 8 Herrin, D. and Michaels, A. (1985) FEBS Lett. 184, 90-95. 9 Minami, E.-I., Shinohara, K., Kuwabara, T. and Watanabe, A. (1986) Arch. Biochem. Biophys. 244, 517-527. 10 Minami, E.-I. and Watanabe, A. (1984) Arch. Biochem. Biophys. 235, 562-570. 11 Herrin, D. and Michaels, A. (1985) Arch. Biochem. Biophys. 237, 225-236. 12 Bhaya, D. and Jagendorf, A.T. (1984) Plant Mol. Biol. 3, 277-280. 13 Margulies, M.M., Tiffany, H.L. and Hattori, T. (1987) Arch. Biochem. Biophys. 254, 454-461. 14 Leu, S., Bolli, R., Mendiola-Morgenthaler, L. and Boschetti, A. (1984) Planta 160, 204-211. 15 Minami, E.-I., Shinohara, K., Kwakami, N. and Watanabe, A. (1988) Plant Cell Physiol. 29, 1303-1309. 16 Breidenbach, E., Jenni, E., Leu, S. and Boschetti, A. (1988) Plant Cell Physiol. 29, 1-7. 17 Breidenbach, E., Jenni, E. and Boschetti, A. (1988) Eur. J. Biochem. 177, 225-232. 18 Friemann, A. and Hachtel, W. (1988) Planta 175, 50-59. 19 Klein, R.R., Mason, H.S. and Mullet, J.E. (1988) J. Cell Biol. 106, 289-301. 20 Hattori, T. and Margulies, M.M. (1986) Arch. Biochem. Biophys. 244, 630-640. 21 Bhaya, D. and Jagendorf, A.T. (1985) Arch. Biochem. Biophys. 237, 217-223. 22 Bolli, R., Mendiola-Morgenthaler, L. and Boschetti, A. (1981) Biochim. Biophys. Acta 653, 276-287. 23 Fish, L.E. and Jagendorf, A.T. (1982) Plant Physiol. 69, 814-825. 24 Howell, S.H., Posakony, W.J. and Hill, K.R. (1977) J. Cell Biol. 72, 223-241. 25 Iwanij, V., Chua, N.-H. and Siekevitz, P. (1975) J. Cell Biol. 64, 572-585. 26 Herrin, D.L., Michaels, A.S. and Paul, A.-L. (1986) J. Cell Biol. 103, 1837-1845. 27 Boschetti, A., Breidenbach, E., Leu, S. and Clemetson-Nussbaum, J. (1987) in Algal Development (Molecular and Cellular Aspects), (Wiessner, W., Robinson, D. and Starr, R.C., eds.), pp. 114-122, Springer-Verlag, Berlin.
28 Brandt, P., Winter, J., Von Kessel, B., Kohnke, B. (1987) in Algal Development (Molecular and Cellular Aspects), (Wiessner, W., Robinson, D.G. and Starr, R.C., eds.), pp. 134-141, SpringerVerlag, Berlin. 29 Brandt, P., Breidenbach, E., Prestin, B. and Boschetti, A. (1989) J. Plant Physiol. 134, 420-426. 30 lnamine, G., Nash, B., Weissbach, H. and Brot, N. (1985) Proc. Natl. Acad. Sci. USA 82, 5690-5694. 31 Klein, R.R. and Mullet, J.E. (1986) J. Biol. Chem. 261, 11138-11145. 32 Klein, R.R. and Mullet, J.E. (1987) J. Biol. Chem. 262, 4341-4348. 33 Laing, W., Kreuz, K. and Apel, K. (1988) Planta 176, 269-276. 34 Deng, X.-W. and Gruissem, W. (1987) Cell 49, 379-387. 35 Berry, J.O., Carr, J.P. and Klessig, D.F. (1988) Proc. Natl. Acad. Sci. USA 85, 4190-4194. 36 Bouet, C., Schantz, R., Dubertret, G., Pineau, B. and Ledoigt, G. (1986) Planta 167, 511-520. 37 Monroy, A.F., McCarthy, S.A. and Schwartzbach, S.D. (1987) Plant Sci. 61-76. 38 Ellis, R.J. (1977) Biochim. Biophys. Acta 463, 185-215. 39 Watson, J.C. and Surzycki, S.J. (1982) Proc. Natl. Acad. Sci. USA 79, 2264-2267. 40 Montandon, P.E. and Stutz, E. (1983) Nucleic Acids Res. 11, 5877-5892. 41 Sreedharan, S., Beck, C.M. and Spremulli, L.L. (1985) J. Biol. Chem. 260, 3126-3231. 42 Spremulli, G.H. and Spremulli, L.L. (1987) Biochem. Biophys. Res. Commun. 148, 1490-1495. 43 Leu, S., Herrin, D. and Michaels, A. (1989) in Proceedings of the Brauschweig Symposium of Applied Plant Molecular Biology (Galling, G., ed.), Zentralstelle ftir Weiterbildung der Technischen Universit~it Braunschweig, Braunschweig, F.R.G. pp. 22-37. 44 Mendiola-Morgenthaler, L., Leu, S. and Boschetti, A. (1985) Plant Sci. 38, 33-39. 45 Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: a Laboratory Manual, pp. 188-189, Cold Spring Harbor Laboratory, New York, Cold Spring Harbor, NY. 46 Laemmli, U.K. (1970) Nature (Lond.) 227, 680-685. 47 Vernon, L.P. (1960) Anal. Chem. 32, 1144-1150. 48 Ktick, U., Choquet, Y., Schneider, M., Dron, M. and Bennoun, P. (1987) EMBO J. 6, 2185 2195. 49 Silk, G.W. and Wu, M. (1988) Curr. Genet. 14, 119-126. 50 Tiboni, O. and Ciferri, O. (1986) Methods Enzymol. 118, 296-308. 51 Herrin, D. and Michaels, A. (1984) in Recombinant DNA and Cell Proliferation (Stein, G. and Stein, S.L., eds.), pp. 87-106, Academic press, Orlando, FL. 52 Van der Meide, P.H., Vijgenboom, E., Talens, A., Bosch, L. (1983) Eur. J. Biochem. 130, 397-407. 53 Van Delft, J.H.M., Schmidt, D.S. and Bosch, L. (1987) J. Mol. Biol. 197, 647-657. 54 Montandon, P.E. and Stutz, E. (1989) Experientia 45, A-33.