P l a n t a 9 Springer-Verlag 1990
In-vivo and in-vitro synthesis of photosynthetic fructose-l,6-bisphosphatase from pea (Pisum sntivum L.) M. Sahrawy, A. Chueca, R. Hermoso, J.J. Lfizaro, and J. L6pez Gorge* Department of Plant Biochemistry,Estaci6n Experimental del Zaidin (CSIC), Profesor Albareda 1, E-18008 Granada, Spain Received 28 September 1989; accepted 26 March 1990
Abstract. Etiolated pea (Pisum sativum L. cv. Lincoln) seedlings do not show any capability for the biosynthesis of chloroplast fructose-l,6-bisphosphatase (FBPase), but the rate of biosynthesis of the increases with the pre-illumination time. This light-induced FBPase synthesis appears to be regulated at the transcriptional level, the response of young leaves being greater than that of mature ones. In-vivo labelling experiments demonstrated by immunoprecipitation, followed by sodium dodecyl sulfate electrophoresis and fluorography, the presence of a 49-kilodalton (kDa) band which corresponds to the mature FBPase subunit. In-vitro translation experiments with a wheat-germ synthesizing system and polyadenylated m R N A isolated from illuminated young pea seedlings have demonstrated the appearance of a 59-kDa labelled band corresponding to the precursor of the FBPase basic subunit. When intact pea chloroplasts were added to the above in-vitro incubation mixture, a labelled 49-kDa subunit similar to that of the in-vivo experiments appeared in the organelle under illumination. From these results we can conclude that a 10-kDa transit peptide bound to the translated pea FBPase subunit exists in the cytosol; this transit peptide is lost during passage through the chloroplast envelope, leaving the mature subunit inside the organelle. Key words: Fructose-l,6-bisphosphatase (biosynthesis) Photosynthesis - Pisum (photosynthesis)
Introduction Fructose-l,6-bisphosphatase (FBPase; EC 22.214.171.124) catalyzes the hydrolysis of fructose-l,6-bisphosphate to fructose-6-phosphate and orthophosphate. Two enzyme * To whom correspondence should be addressed Abbreviations: ELISA= enzyme-linked immunosorbent assay; FBPase=fructose-l,6-bisphosphatase; kDa=kilodalton; MW= molecular weight; PBS=phosphate-buffered saline; PMSF= phenylmethylsulfonyl fluoride; poly(A)mRNA= polyadenylated mRNA; SDS = sodium dodecyl sulfate
isoforms of FBPase are present in the photosynthetic cell, both with outstanding regulatory roles. The cytosolic FBPase is a key enzyme of gluconeogenesis and of the sucrose biosynthetic pathways, and shows the regulatory features of FBPases from non-photosynthetic organisms (Zimmermann et al. 1978; Herzog et al. 1984). By contrast, the chloroplast FBPase catalyzes a crucial step of the photosynthetic carbon-reduction cycle (Bassham and Krause 1969), and has a regulatory mechanism different from that of the cytosolic enzyme. Besides the modulation of its photosynthetic activity by some metabolites, this FBPase appears to be activated by the increase in pH and Mg 2 + concentration which take place in the stroma during the dark-light transition (Portis and Heldt 1976). In addition, the enzyme is light-activated by the reducion of essential -SH groups of the enzyme molecule, a process which occurs through the photosynthetic electron-transport chain via the ferredoxin-thioredoxin system (Buchanan 1980; Anderson 1986). In-vivo experiments with protein-synthesis inhibitors have demonstrated that, as for some other chloroplast proteins, the photosynthetic FBPase is nuclear-coded (Chueca et al. 1984). Preliminary assays with a specific antiserum have also shown a rise in the enzyme level when etiolated pea seedlings are illuminated, the FBPase increase being proportional to the light intensity (Sahrawy et al. 1988). This light induction of enzyme synthesis had been found earlier for, among others, NADP +malate dehydrogenase (Vidal and Gadal 1981), and for the assembly as a whole of ribulose-l,5-bisphosphate carboxylase (Bloom et al. 1983). Moreover, Raines et al. (1988) had shown a light-induced increase of the FBPase-related m R N A level in wheat. However, it has not been possible so far to demonstrate the synthesis of photosynthetic FPBase in an in-vitro cell-free mRNAsynthesizing system (Grossman et al. 1982). In this work we have analyzed the role of light in the rate of biosynthesis of chloroplast FBPase when pea seedlings are illuminated under different light conditions. In addition, we have studied the in-vitro synthesis of the pea enzyme in cell-free preparations, as well as the import of the precursor protein into the organelle.
M. Sahrawy et al. : Biosynthesis of photosynthetic fructose-l,6-bisphosphatase in pea
Material and methods Plant material and in-vivo FBPase labelling. Pea (Pisum sativum L. cv. Lincoln; Fit6 Seeds, Spain) seeds were soaked in running water, germinated in moistened vermiculite contained in plastic trays, and grown in darkness in a growth chamber for 8 d at 20~ C. For in-vivo labelling experiments, leaves of 8-d dark-grown seedlings were gently brushed with [35S]methionine (>37 TBq" mmol 1; Amersham Int., Amersham, UK) solution in 2% Tween 80, about 370 kBq for each seedling, and then illuminated before harvesting with white light (222 gmol- m - a. s - ~) for 4, 15 and 20 h. In other experiments a set of 8-d dark-grown seedlings were preilluminated for 4, 10, 15 and 24 h before labelling, and harvested just 1.5 h after the [35S]methionine application. In all cases, leaves of 5-10 seedlings were thoroughly washed with distilled water, and homogenised by hand in a chilled mortar with (1 : 3, w/v) 25 mM 2-amino-2-(hydroxymethyl)-l,3-propanediol (Tris)-HC1 pH7.5, 5 mM MgClz, 5 mM 2-mercaptoethanol, 0.5 mM Naz-ethylenediaminetetraacetate and 1 M phenylmethylsulfonylfluoride (PMSF). The homogenate was centrifuged at 12000.g for 30 min, and the supernatant collected for analytical determinations. Radiochemical and biochemic'al determinations. The 35S incorporation to the soluble components of leaves was determined by soaking up 3-gl aliquots of supernatants in small pieces (0.5'0.5 cm 2) of Whatman No I (Springfield Mill, UK) chromatographic paper, and counting the radioactivity in a liquid scintillator. The 35S labelling of the soluble total protein was determined as above, but for 20 min before counting, the chromatographic paper was washed, with a hot solution of 0.1% methionine in 10% trichloroacetic acid, and successively rinsed with cold 10% trichloroacetic acid, absolute ethanol, and diethyl ether. Finally, the 35S incorporation in de-novo-synthesized FBPase was determined by adding to an aliquot of supernatant 4 vol. of 1 mM phosphate buffer pH 7.4, 0.15 M NaC1, 2.7 M KC1 (PBS buffer), containing 50 mM methionine and 10% Triton X-100. A sufficient quantity of an antibody solution containing 5 rag- ml- 1 of anti-FBPase IgG immunoglobulins in PBS buffer was added to the above mixture to link the FBPase (Bard et al. 1985). After overnight incubation at 4~ C, the FBPase-IgG complex was sequestered by 1 h incubation at room temperature with 0.5 mg of protein A-Sepharose per microliter of the anti-FBPase IgG solution used earlier. The pellet obtained after a short centrifugation was washed three times with the PBS buffer containing 10mM methionine and 1% Triton X-100, and then given two additional washes with the same buffer made up with 10 mM methionine and 0.1% sodium dodecyl sulfate (SDS). The pellet was then dissolved in a few microliters of 125 mM Tris-HC1 buffer pH 6.8, 4% SDS, 5% 2-mercaptoethanol, by heating at 100~ C for 5 rain, and the FBPase subunit isolated by SDS-polyacrylamide gel electrophoresis according to Laemmli (1970). The gel was stained for proteins with Coomassie Brilliant Blue 250R and, after drying, the radioactive bands were visualized by fluorography at - 7 0 ~ C with a high-sensitivity photographic plate (Hyperfilm; Amersham, UK). The labelled band corresponding to the FBPase subunit was solubilized by 4 h digestion at 80~ with 30% hydrogen peroxide, and counted in a liquid scintillator. Phosphorylase b (94 kDa MW), serum-albumin (67 kDa MW), ovalburnin (43 kDa MW), carbonic anhydrase (30 kDa MW), soybean trypsin inhibitor (20./kDa MW) and lactoalbumin (14.4 kDa MW) were used as MW (molecular weight) markers (all from Pharmacia, Uppsala, Sweden). The anti-FBPase serum was prepared by immunization of rabbits with homogeneous pea photosynthetic FBPase, whereas the enriched IgG fraction was obtained from the antiserum by precipitation with 40% saturation (NH4)z~04, DEAE-cellulose chromatography, and affinity chromatography on an FBPase-Sepharose colunm; the IgG concentration was determined according to Hum and Chantler (1980). Homogeneous photosynthetic FBPase from pea leaves was obtained as described by Plfi and L6pez Gorg6 (1981). Proteins were measured according to Lowry et al. (1951), and chlorophyll (Chl) by the method of Arnon (1949). Fructose-
1,6-bisphosphatase activity was assayed as described by Lfizaro et al. (1974), and FBPase protein by the competitive enzyme-linked immunosorbent assay (ELISA) of Hermoso et al. (1987).
Preparation of total and polyadenylated mRNAs (poly ( A ) mRNA ). For in-vitro synthesis experiments a crude nucleic-acid preparation was obtained from 10 g of 14-d-old light-grown pea seedlings, according to Wallace (1987). Leaves were frozen at - 7 0 ~ C, and ground (Omni-mixer; Sorvall, Norwalk, USA) with 30ml of 100 mM Tris-HC1 buffer pH 7.5, made up to 0.5% in SDS, and 30 ml of ice-cold liquid phenol. The homogenised mixture was centrifuged at 2000.g for 10 rain at room temperature, and the aqueous layer re-extracted with an additional 30 ml of phenol. The resulting aqueous phase obtained after centrifugation was adjusted with a NaC1 solution to 0.2 M concentration, and the nucleic acids precipitated by addition of 2.5 vol. of ethanol. After standing the mixture overnight at - 2 0 ~ C, nucleic acids were recovered by 10 rain centrifugation at 15000-g. The pellet was redissolved in distilled water, and the RNA precipitated overnight at 4~ by addition of NaC1 to 2.5 M final concentration. This cycle was repeated twice and, after drying by vacuum, the precipitate was resuspended in 0.01 M Tris-HC1 pH 7.5, 1 M in NaC1. A poly(A)mRNA preparation was obtained from the above crude RNA by affinity chromatography on a column of oligo-dT cellulose, according to Maniatis et al. (1982). The column was equilibrated with 0.01 M Tris-HC1 pH 7.5, 0.5 M NaC1, and the fixed poly(A)mRNA eluted with the same buffer but without NaC1. The RNA was precipitated by addition of Na-acetate to a final 2% concentration, and of 2.5 vol. of ethanol. After overnight sedimentation at - 2 0 ~ C, the poly(A)mRNA was recovered by 10 rain centrifugation at 15000"g; the pellet was vacuum dried, and stored at - 7 0 ~ C after suspension in a few microliters of distilled water. Cell-free in-vitro synthesis of FBPase. For the in-vitro translation we have used the isolated poly(A)mRNA coupled to a nucleasetreated amino-acid-depleted wheat-germ preparation (Amersham); this system was complemented with [35S]methionine, and a mixture of non-labelled amino acids minus methionine (Anderson et al. 1983). The assay conditions were optimized in relation to poly(A)mRNA effectivity, incubation time, and concentrations of the reaction-mixture components. A typical experiment contained, in 50 [xl of 5 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonicacid (Hepes) buffer pH 7.6, 3 gg of poly(A)mRNA, 25 gl of wheat-germ preparation, 1665 kBq of [35S]methionine (> 37 TBq.mmol- 1; Amersham), and the 19 remaining amino acids to a final concentration of 50 gM each. In addition, the reaction mixture contained the ATP-generating system creatine-creatine phosphokinase, 58 gM spermine, 2 mM dithiothreitol, 20 gM guanosine 5'-triphosphate, 2.5 mM Mg-acetate, and 120 mM K-acetate. The mixture was incubated at 27~ C for 60 rain, and dialyzed on a 0.025-gm Millipore V (Bedford, Mass., USA) membrane against PBS buffer. The radioactivity incorporated in total translated proteins was determined in an aliquot, as described for the in-vivo procedure. For determination of FBPase labelling the dialyzed mixture was treated first with a rabbit preimmune serum, and then with an anti-FBPase IgG, just as above. The recovery of the FBPase-antienzyme complex, SDS-electrophoresis and fluorography of the plate were performed as described earlier, but with 50 mM methionine in the washing solutions. In-vitro FBPase import into chloroplasts. The analysis of the in-vitro uptake of FBPase into chloroplasts was carried out according to Grossman et al. (1982), by using intact pea chloroplasts isolated by centrifugation in a Percoll gradient (Mills and Joy 1980). The pellet of intact chloroplasts was washed once and then resuspended in the import buffer of 50 mM Hepes-KOH pH 8.0, 0.33 M sorbitol, 10 mM methionine, to a Chl concentration of 1 mg.ml -~. A typical assay for FBPase uptake contained, in 0.3 rnl of reaction mixture, a volume of the in-vitro-synthesis incubation mixture equivalent to 3-4.106 counts per minute of labelled translated protein, and intact chloroplasts corresponding to 0.15-0.20 mg Chl.
M. Sahrawy et al. : Biosynthesis of photosynthetic fructose-l,6-bisphosphatase in pea After 60 min incubation in the light at 27~ C with gently shaking, samples were ice-cooled and mixed with 50 lal of 0.1% trypsin in the chloroplast-suspending medium to break down the non-imported polypeptides. The mixture was maintained at 4~ for 30 min, and layered onto 40% Percoll in the chloroplast suspension buffer. After centrifugation at 2500-g for 4 rain, the green sediment was recovered in the isotonic chloroplast suspension buffer, and re-centrifuged in the same conditions. The pelleted chloroplasts were then lysed by suspension for 30 min in 0.24 M NaC1 containing soybean trypsin inhibitor, precipitated with 10% trichloroacetic acid, washed with cold acetone, and suspended in PBS buffer. After SDS-electrophoresis and protein staining as above, the radioactive bands were identified by fluorography.
mination, but the specific labelling (ratio o f labelled F B P a s e to labelled trichloroacetic-acid-precipitable protein) remained c o n s t a n t (Fig. 2). This means that in m a ture leaves there is a non-selective synthesis o f F B P a s e versus some other light-inducible proteins, which show a noticeable increase with respect to total leaf labelling. However, when F B P a s e was i m m u n o l o g i c a l l y determined by E L I S A , the ratio o f F B P a s e to total soluble
O J v
Light plays an essential role in chloroplast formation and development, which occurs t h r o u g h the biosynthetic induction o f m a n y chloroplast (Palmer 1985; Briar et al. 1986) and nuclear-coded (Ellis 1981; Gallagher and Ellis 1982) organellar proteins. As Fig. 1 shows, 8-d darkg r o w n pea seedlings did n o t c o n t a i n any d e - n o v o FBPase, which could be clearly detected as a 4 9 - k D a subunit only after 4 h pre-illumination before labelling. In spite o f the use o f a protease inhibitor ( P M S F ) , a weakly labelled 3 2 - k D a polypeptide could be visualized as a proteolysis p r o d u c t o f the native subunit. This 32k D a b a n d increased, with a parallel decrease o f the 49k D a subunit, when P M S F was omitted. The a m o u n t o f labelled enzyme slowly increased up to 24 h pre-illu-
Results and discussion
Fig. 1. Fluorography after SDS-electrophoresis of the specific immunoprecipitate obtained from crude extracts of 8-d-etiolated pea seedlings, which were pre-illuminated between (~24 h, then treated with [3SS]methionine, and harvested after 1.5 h of labelling. Incubation mixture includes PMSF as an antiproteolytic agent. Arrows (left) show the position of MW standards visualized after protein staining
Fig. 3. Fluorography after SDS-electrophoresis of the FBPase-specific immunoprecipitate obtained from crude extracts of 8-d-etiolated pea seedlings, treated with [35S]methionine, and then illuminated for 4, 15, and 20 h. In this experiment the antiproteolytic agent (PMSF) was omitted
M. Sahrawy et al. : Biosynthesis of photosynthetic fructose-1,6-bisphosphatase in pea
proteins assayed by the Lowry method increased sharply with the pre-illumination time. So there is evidently a strong light-induced biosynthesis of FBPase in the early stages of leaf development. This steep rise in the level of FBPase synthesis in young leaves was also observed when the [35S]methionine was applied to leaves of 8-d dark-grown seedlings which were then continuously illuminated up to 20 h (Fig. 3); in this experiment the protease inhibitor was omitted as it did not completely suppress the proteolytic breakdown of the FBPase subunit. There are several possible sites for the light-induced control of protein synthesis. Even though many examples of the influence of light on the expression of genetic systems have been described, most of them have been related to a transcriptional control (Gallagher and Ellis 1982; Freyssinet and Buetow 1984), and only in a few cases has a translational (Fromm et al. 1985) or posttranslational (Apel and Kloppstech 1978; Cuming and Bennet 1981) modulation been considered. In barley the level of total poly(A)mRNA increases from 4 to 6 gg per g leaf FW, when etiolated seedlings are illuminated for 6 h. Similarly, the m R N A for the small subunit of ribulose-l,5-bisphosphate carboxylase, a nuclear-coded protein, increases many times when etiolated plants are exposed to light (Coruzzi et al. 1984; Zhu et al. 1985); this means that there is a transcriptional control o f protein synthesis since the gene copy number is not substantially affected. By contrast, the transcript of the chloroplast D N A coding for the large subunit of ribulose-l,5bisphosphate carboxylase shows only a small m R N A increase (Coruzzi et al. 1984), which can be related to a parallel increase in the copy number of the chloroplast genome. In this case the light control must occur at the translational level (Berry et al. 1985), but why this m R N A is not expressed in the dark is a matter of speculation; perhaps there is an absence o f some light-induced specific factor necessary for m R N A translation. In the case of pea FBPase, light regulation of the synthesis o f the precursor protein seems to occur at the transcriptional level, since the FBPase precursor appears when cytosolic m R N A is translated in the dark in an in-vitro translation system (see below). Raines et al. (1988) have demonstrated the absence o f any FBPase m R N A in etiolated young wheat seedlings; the m R N A could be detected only after 2 min illumination, followed by a sharp increase after 24 h light exposure. By contrast, the concentration of FBPase m R N A in mature wheat plants subjected to a 40-h dark period rose very quickly in the presence of light, such that after 4 h illumination the FBPase m R N A level was similar to that of 24-h-illuminated young seedlings. Our in-vitro translation system was optimized to obtain an optimum yield of translated protein. Figure 4 shows the influence of incubation time and of the m R N A concentration in the reaction mixture, as well as the role of light on the effectiveness of the cytosolic m R N A . From these results we selected an m R N A concentration of 60 ~tg.ml-~, and an incubation time of 60 min. In addition, to rule out any participation of chloroplast m R N A in the biosynthesis of the precursor FBPase when intact organelles were added to the incuba-
b60E 13u ,2, ~ 4ottl .c_ 020B_
Time(min) Fig. 4. Patterns of 3SS-labelled trichloroacetic-acid-precipitable
protein in in-vitro synthesis experiments with [35S]methionine,and poly(A)mRNA isolated from 8-d dark-grown pea seedlings, preilluminated for 24 h before mRNA isolation. The in-vitro incubation times were between 15 and 60 rain, with a mRNA concentration of 0.5 ( o - - o ) , 1.5 (D--~) or 3.0 ( m - - l ) gg in the incubation mixture. Blanks without mRNA ( o - - e ) and with poly(A)mRNA isolated from non pre-illuminated seedlings ( • - - x ) were also included. Each value is the mean of four determinations +SE
c" s O l_ O
E ~610O_ m tl_ no
m [a. i
Time (rain) Fig. 5. Patterns of FBPase activity ( • • ) and FBPase protein (ELISA) ( e - - o ) of intact pea chloroplasts subjected to a lightdark transition. A blank of FBPase activity in non-illuminated chloroplasts is also included ( o - - o ) . The dark bar on the x-axis indicates the duration of the dark period. Each value is the mean of four determinations _+SE tion mixture, a suspension of pea chloroplasts was subjected to a light-dark cycle, and FBPase activity and the enzyme protein determined in organellar lysates. As Fig. 5 shows, the enzyme activity increases during illumination, but the level of the FBPase protein determined
M. Sahrawy et al. : Biosynthesisof photosynthetic fructose-/,6-bisphosphatase in pea 1