Involvement of thylakoid membranes in ... - Science Direct

1 downloads 0 Views 2MB Size Report
ever, in spite of these difficulties, multienzyme complexes among Calvin cycle enzymes have been demonstrated by several groups (Sainis and Harris 1986, ...
J. Plant Physiol. 160. 23 – 32 (2003)  Urban & Fischer Verlag http://www.urbanfischer.de/journals/jpp

Involvement of thylakoid membranes in supramolecular organisation of Calvin cycle enzymes in Anacystis nidulans Jayashree Krishna Sainis1 *, Diksha Narhar Dani1, Gautam Kumar Dey2 1

Molecular Biology and Agriculture Division, Bhabha Atomic Research Centre, Trombay, Mumbai, 400 085, India

2

Material Science Division, Bhabha Atomic Research Centre, Trombay, Mumbai, 400 085, India

Received January 10, 2002 · Accepted August 16, 2002

Summary The cells of unicellular photosynthetic cyanobacterium Anacystis nidulans were permeated with lysozyme, toluene, toluene-triton, toluene-triton-lysozyme. Transmission electron microscopy of semi-thin sections (500 nm) using TEM at 160 kV showed that cells permeated with only lysozyme or toluene showed the typical concentric arrangement of thylakoid membranes. However, when toluene-treated cells were further treated with triton and lysozyme the thylakoid membranes were disrupted. Sequential reactions of Calvin cycle were studied in the differentially permeated cells in vivo, using various intermediates such as 3-PGA, GA-3-P, FDP, SDP, R-5-P, RuBP and cofactors like ATP, NADPH depending on the requirement. RuBP and R-5-P + ATP dependent activities could be observed in all types of permeated cells. Sequential reactions of the entire Calvin cycle using 3-PGA could be detected in the cells that had retained the internal organisation of the thylakoid membranes after permeation and were lost on disruption of this organisation. Light dependent CO2 fixation could be detected only in the cells permeated with lysozyme. This activity was abolished in the cells after treatment with toluene. The results suggested that the integrity of thylakoid membranes may be essential for the organisation of sequential enzymes of the Calvin cycle in vivo and facilitate their functioning. Key words: Calvin cycle enzymes – organisation – thylakoid membranes Abbreviations: FDP = fructose-1,6-diphosphate. – GA-3-P = glyceraldehyde-3-phosphate. – 3-PGA = 3-phosphoglyceric acid. – R-5-P = ribose-5-phosphate. – RuBP = ribulose-1,5-bisphosphate. – SDP = sedoheptulose-1,7-diphosphate

Introduction Calvin cycle enzymes are present in the stroma of chloroplasts and can be extracted easily in soluble form, in aque* E-mail corresponding author: [email protected]

ous fraction after disruption of chloroplasts by hypotonic buffers. Typically these enzymes are further purified and characterised in isolation, in dilute buffers at carefully chosen concentrations of salts, substrates and pH. In contrast, these enzymes function in a protein-crowded, water-limited environment in vivo. Such environments would necessitate associa0176-1617/03/160/01-23 $ 15.00/0

24

Jayashree Krishna Sainis, Diksha Narhar Dani, Gautam Kumar Dey

tion and channelling of intermediates among sequential enzymes of all the metabolic pathways. Though very important, the organisation among sequential enzymes has been very difficult to prove by the conventional methods used for isolation and purification of proteins. However, in spite of these difficulties, multienzyme complexes among Calvin cycle enzymes have been demonstrated by several groups (Sainis and Harris 1986, Sainis et al. 1989, Gontero et al. 1988, Suss et al. 1993, Mouche et al. 2002). The Calvin cycle enzymes that showed a tendency to associate were phosphoriboisomerase, phosphoribulokinase, RuBP carboxylase, glyceraldehyde-3-phosphate dehydrogenase, sedoheptulose-1,7-diphosphatase. Suss et al. (1993) have shown association between Calvin cycle enzymes and ferredoxin NADP reductase. Metabolic significance of such organisation has been demonstrated in vitro in many cases (reviewed in Suss and Sainis 1997). However, the main dilemma has been to investigate the location of soluble enzymes and enzyme systems in vivo and to understand the significance of this location to their function. Cryo-immuno electron microscopy was used to localise Calvin cycle enzymes in the chloroplasts. Immuno gold labelling with antibodies suggested that, in vivo, the soluble non-abundant enzymes of the Calvin cycle are located along the surface of thylakoid membranes in the cases of spinach and Chlamydomonas chloroplasts (Suss et al. 1993, 1995). Cryoimmuno electron microscopy also revealed preferential location of soluble Calvin cycle enzymes to the thylakoid membranes in bundle sheath chloroplasts in the leaves of Zea mays. Several soluble enzymes of the other metabolic pathways also could be detected along the thylakoid membranes in both bundle sheath and mesophyll chloroplasts in leaves of maize (Sainis JK, Melzer M and Suss K-H unpublished observations). These observations have added a new dimension to research on supramolecular organisation. The soluble Calvin cycle enzymes not only have a tendency to associate with each other, but some of them may be organised on the surface of thylakoid membranes. Evidently, this will facilitate channelling of ATP and NADPH generated by light reactions, besides offering a platform for organisation. Our experiments were initiated, therefore, to understand the role of thylakoid membranes in the organisation and function of the soluble Calvin cycle enzymes in situ. To address this question, we selected the unicellular photosynthetic organism Anacystis nidulans as a model system, which was easier to handle, compared to the isolated chloroplasts. Anacystis nidulans cells were permeated with various agents and the ultrastructure, especially that of thylakoid membranes was examined with TEM. The activities of complete or partial sequential reactions of the Calvin cycle were measured in the permeated cells. The results indicate that the disruption of thylakoid membranes affected the functioning of the Calvin cycle in vivo.

Material and Methods Cell culture Anacystis nidulans (BD1) cells were grown in BG-11 medium with continuous white light, 21 W/m2, without aeration. Cells, which were 7–10 day-old, were harvested and used for experiments.

Cell permeation Toluene treatment was done according to Tabita (1978) wherein cells were suspended in HEPES 100 mmol/L pH 8.0 (buffer A) and were treated with half volume of toluene for 10 min on ice. Cells were centrifuged and washed with buffer A and resuspended in the same buffer. For triton treatment, cells were treated with 0.2 % triton in buffer A for 10 min on ice. Cells were washed and resuspended as mentioned previously. For lysozyme treatment, cells were incubated for 30 min, at 37 ˚C in buffer containing HEPES 100 mmol/L pH 8.0, EDTA 1 mmol/L (buffer B) and lysozyme 1 mg/ml. Cells were washed and resuspended in buffer A. Differential permeation of cells of Anacystis nidulans was achieved using the following treatments i) lysozyme alone, ii) toluene alone, iii) toluene treatment followed by triton and iv) toluenetriton treatment followed by lysozyme.

Transmission electron microscopy Control and differentially permeated cells were processed for electron microscopy according to Stanier (1988) except that fixation was done with 3 % glutaraldehyde in phosphate buffer (100 mmol/L, pH 7.2) in the dark for two hours. The cells were suspended in molten agarose, dehydrated by graded series of ethanol and embedded in standard araldite, DDSA [(2-Dodecen-1-YL) succinic anhydride], Dibutyl phthalate and DMP 30 [2,4,6-tris(Dimethylamino-methyl)phenol] embedding mixture. Blocks were polymerised for 72 hrs at 60 ˚C. Semi-thin sections (500 nm) and ultra-thin sections (70 nm) were cut using Leica Ultracut UCT microtome and stained with a 10 % alcoholic solution of uranyl acetate. Semi-thin sections were observed with Jeol (JEM 2000 FX) TEM at 160 kV and ultra-thin sections with Jeol (JEM 1010) TEM at 80 kV.

Protein estimation Total soluble proteins were estimated in the cell-pellet and supernatant after treatment of cells with various permeating agents. Proteins were extracted in TRIS-HCl (60 mmol/L, pH 6.75) containing 2 % SDS and 5 % mercapto-ethanol and estimated by dot blots using Ponceau S according to Bannur et al. (1999). Absorbance at 660 nm was measured to estimate approximate cell density. One OD at 660 nm corresponds to 2 × 105 cells per ml in case of 7–10 day-old untreated cells, which were used in all these experiments. It was difficult to measure viable cell count after permeation, therefore, OD at 660 was routinely used as the basis to express enzyme activities.

Light dependent CO2 fixation Cells were suspended in HEPES buffer (100 mmol/L, pH 8.0) containing MgCl2 20 mmol/L, KH2PO4 10 mmol/L and incubated in light for

Thylakoid membranes and organisation of Calvin cycle enzymes 15 min. The optical density of cell suspension was around 2.5 at 660 nm. The reaction was started by addition of radioactive NaH14CO3 20 mmol/L (specific activity 0.5 mCi/mmole). Light intensity during CO2 fixation was 21W/m2. The reaction was terminated after 10 min by adding an equal volume of 6 N acetic acid. The acid stable reaction product was counted in a scintillation counter.

Enzyme assays in vivo In vivo activities of several sequential Calvin cycle enzymes and RuBP carboxylase were estimated in permeated cells as described below.

In vivo activity of RuBP carboxylase Permeated cells were incubated in assay buffer comprising of HEPES 50 mmol/L pH 8.0; MgCl2 20 mmol/L; and NaH14CO3 20 mmol/L (specific activity 0.5 mCi/mmole). The reaction was started after 15 min with RuBP. Final concentration of RuBP was 1mmol/L and total volume of reaction mixture was 200 µl. The reaction was terminated after 2 min by transferring 100 µl of reaction mixture to a scintillation vial containing 200 µl 6 N acetic acid. The acid stable reaction product was counted in a liquid scintillation counter.

In vivo activities of sequential enzymes of the Calvin cycle Permeated cells were incubated with assay buffer, as described for RuBP carboxylase assay, containing DTT 10 mmol/L. The reaction was started by adding either of the following substrate combinations i) R-5-P + ATP, ii) SDP + GA-3-P + ATP, iii) FDP + GA-3-P + ATP, iv) GA-3P + ATP, v) 3-PGA + NADPH + ATP. Total volume of reaction mixture was 200 µl and final concentration of ATP and substrates was 2 mmol/L, whereas that of NADPH was 1 mmol/L. The reaction was terminated after 10 min in the case of R-5-P and after 30 min for the rest of the substrates by transferring 100 µl of reaction mixture to a scintillation vial containing 200 µl of 6 N acetic acid. The acid stable reaction product was counted as described previously. The blank values were obtained by omission of substrates from reaction mixtures.

25

have previously been used by Nierzwicki-Bauer et al. (1984) to observe overall thylakoid arrangement in cyanobacterium Agmenellum quadruplicatum. Figure 2 a and b show the ultrastructure of cells treated with lysozyme. In this case, the concentric layers of thylakoid membranes and centroplasm were seen clearly. Figure 3 a and b show the intracellular structure of cells treated with toluene. In these cells, the outer cell envelope appeared to be detached from the cell mass though the cell shape was maintained. The concentric layers of thylakoid membranes were visible, but the thylakoid membranes appeared swollen in transverse section. When the toluenised cells were further treated with triton, the cells were turgid and the concentric layers of thylakoid membranes were seen in transverse section, but appeared to be distorted in the longitudinal sections (Figs. 4 a, b). The swelling of thylakoid membranes seen in toluenised cells was reduced after triton treatment. However when cells permeated with toluene-triton were further treated with lysozyme, the internal structure of thylakoid membranes appeared to be completely destroyed in most of the cells (Figs. 5 a, b and c). Figures 5 a and b are the images of ultrathin sections and Figure 5 c is the image of semi-thin section of Anacystis nidulans cells treated with toluene-triton-lysozyme.

Protein efflux and extraction of pigment-protein complexes from permeated cells We observed that treatment with toluene did not extract the intracellular proteins and pigments. However, when the toluenised cells were further treated with triton, nearly 50 % of soluble proteins were extracted and the extracts showed peak absorbance at 663 nm indicating release of chlorophyll containing pigment complexes. Further treatment of cells with lysozyme resulted in solubilisation of phycobiliproteins, showing a peak absorbance at 620 nm along with chlorophyll containing pigment complexes (data not given).

Results Light dependent CO2 fixation Ultrastructure of normal and permeated cells of Anacystis nidulans Ultrastructure of Anacystis nidulans cells was visualised using semi-thin and ultra-thin sections. Figures 1 a and 1 b show TEM images of semi-thin (500 nm) transverse and longitudinal sections of the untreated cells of Anacystis nidulans. Figure 1c is the image of transverse ultra-thin section (70 nm). In both semi-thin and ultra-thin sections, 3 – 4 concentric layers of thylakoid membranes as well as centroplasm containing nucleoplasm, carboxysomes and polyphosphate granules were clearly visible. Regular globular bodies were observed in the electron opaque region along the thylakoid membranes. Semi-thick sections ranging from 250 nm–1 µm

Light dependent CO2 fixation activity was observed only in cells permeated with lysozyme. This activity was comparable to that of untreated cells (Table 1). The results indicate that the sequential photochemical and biochemical reactions for assimilation of CO2 were not affected by lysozyme treatment though the cells showed slight changes in TEM (Figs. 2 a, b). The cells permeated with toluene, toluene-triton and toluenetriton-lysozyme did not show any light dependent CO2 fixation activity suggesting that these permeation treatments had affected functioning of photo-assimilatory processes. Photochemical activities of PSI and PSII were measured after permeation according to the procedures described in Setlikova et al. (1999). The control and lysozyme treated cells showed

26

Jayashree Krishna Sainis, Diksha Narhar Dani, Gautam Kumar Dey

Figure 1. TEM images of Control cells. a) Semi-thin (500 nm) transverse section, magnification × 72000, bar represents 100 nm, b) Semi-thin (500 nm) longitudinal section, magnification × 327800, bar represents 200 nm, c) Ultra-thin (70 nm) transverse section, magnification × 39900 bar represents 200 nm.

both these activities, whereas cells treated with toluene and toluene-triton did not show any of these light reactions (data not given). Thus permeation with toluene had mainly affected the light reactions and hence cells involving permeation with toluene alone or in combination did not show any light dependent CO2 fixation activity.

In vivo activities of RuBP carboxylase and other sequential Calvin cycle enzymes In vivo activity of RuBP carboxylase was measured in permeated cells (Table 1). The control cells did not show appreciable increase in CO2 fixation when RuBP was included in

Thylakoid membranes and organisation of Calvin cycle enzymes

27

Figure 2. TEM images of cells treated with lysozyme. a) Semi-thin (500 nm) transverse section, magnification × 52800, bar represents 200 nm, b) Semi-thin longitudinal (500 nm) section magnification × 36900, bar represents 200 nm.

Figure 3. TEM images of cells treated with toluene. a) Semi-thin (500 nm) transverse section, magnification × 90000, bar represents 100 nm, b) Semi-thin longitudinal (500 nm) section magnification × 50400, bar represents 200 nm.

28

Jayashree Krishna Sainis, Diksha Narhar Dani, Gautam Kumar Dey

Figure 4. TEM images of cells treated with toluene-triton. a) Semi-thin (500 nm) transverse section, magnification × 76800, bar represents 100 nm, b) Semi-thin longitudinal (500 nm) section magnification × 50000, bar represents 200 nm.

the reaction mixture, showing that cells were not permeable to RuBP. The in vivo activity of RuBP carboxylase increased when the cells were permeated and reached its maximum in cells treated with toluene-triton-lysozyme. We were interested in studying the function of the Calvin cycle in differentially permeated cells. Calvin cycle involves sequential reactions of 13 enzymes. These enzymes are RuBP carboxylase, 3-PGA kinase, NADP-glycelardehyde-3phospahte dehydrogenase, triose-phosphate isomerase, fructose-1,6-diphosphate aldolase, fructose-1,6-diphosphatase, transketolase, aldolase, sedoheptulose-1,7-diphosphatase, transketolase, phosphopentose isomerase, phoshoriboisomerase and phosphoribulokinase (Table 2). During in vivo assays, activities of the sequential enzymes were interpreted in terms of the amount of RuBP generated and used by RuBP carboxylase for CO2 fixation. For example when 3-PGA, NADPH and ATP are given as substrates, all the enzymes starting from 3-PGA kinase have to work sequentially to finally synthesise RuBP, which will be used by RuBP carboxylase for CO2 fixation reaction. Thus, each enzyme in the sequence will use the product generated by the previous enzyme as substrate. The Calvin cycle intermediates and the corresponding enzymes involved in regeneration of RuBP in the partial reactions are given in Table 2. Table 3 shows the rates of CO2 fixation using various intermediates of the Calvin cycle in differentially permeated cells.

Among all the intermediates used to test functioning of partial sequential reactions of the Calvin cycle, the highest rate of CO2 fixation was observed with R-5-P (Table 3). The next best substrate was FDP followed by 3-PGA in most cases. The rates of CO2 fixation with GA-3-P and SDP were low for all treatments. Primarily, this may be due to non accessibility of the corresponding enzymes to these externally added intermediates since 3-PGA dependent CO2 fixation activity will eventually depend on internally generated GA-3-P through sequential reactions. Similarly FDP dependent CO2 fixation activity will ultimately depend on internally generated SDP from sequential reactions using FDP. Interestingly the extent of permeation differentially affected rates of CO2 fixation using various intermediate substrates. R-5-P dependent activity was low in cells treated with lysozyme and increased when cells were treated with toluene, toluene-triton, toluenetriton-lysozyme. The activities obtained with other intermediates of Calvin cycle such as 3-PGA, GA-3-P, FDP and SDP were highest in cells permeated only with toluene. Toluene-triton treated cells showed reduction in GA-3-P and 3-PGA dependent CO2 fixation activities while the rates of other sequential reactions were comparable to those in toluene treated cells (Table 3). In cells permeated with toluene-tritonlysozyme, rates of CO2 fixation using 3-PGA, GA-3-P, FDP, SDP were considerably reduced. In Tables 1 and 3 all the activities are expressed on the basis of OD at 660 nm. Triton treatment extracted pigment-protein complexes, hence there

Thylakoid membranes and organisation of Calvin cycle enzymes

29

Figure 5. TEM images of cells treated with toluene-triton-lysozyme. a) Ultra-thin (70 nm) transverse section, magnification × 61500, bar represents 100 nm, b) Ultra-thin (70 nm) longitudinal section, magnification × 35700, bar represents 200 nm, c) Semi-thin (500 nm) longitudinal section, magnification × 34800, bar represents 200 nm.

was an apparent increase in activity on this basis. We had estimated activity of RuBP carboxylase in the cells and in the supernatants after treatment with various permeating agents. We observed that more that 98 % of RuBP carboxylase activity was retained in the cells even after treatment with toluenetriton and toluene-triton-lysozyme. Therefore, we also ex-

pressed activities of other enzymes as a percent of RuBP dependent activity and these values are given in Table 3. Also in this case, the sequential activities with SDP, FDP, GA-3-P and 3-PGA were highest in toluenised cells, reduced after triton treatment and were negligible when cells were further treated with lysozyme.

30

Jayashree Krishna Sainis, Diksha Narhar Dani, Gautam Kumar Dey the enzymes of the Calvin cycle in control as well as lysozyme treated cells. Tabita et al. (1978) demonstrated that the cells of the photosynthetic prokaryotes permeated with toluene show activities of two Calvin cycle enzymes viz RuBP carboxylase and phosphoribulokinase. We investigated the functioning of the entire Calvin cycle in differentially permeated cells of Anacystis nidulans in order to find out whether the ultra-structural changes occurring in cells affect the in vivo functioning of sequential reactions of this cycle. Though the lysozyme treated

Table 1. Light dependent and RuBP dependent CO2 fixation activity in control and differentially permeated cells of Anacystis nidulans. Treatment

nmoles of CO2 fixed/O.D. 660 nm, hr Light dependent CO2 fixation Mean ± SE

Control 590.36 ± 13.32 Lysozyme 547.38 ± 13.73 Toluene 0 Toluene-triton 0 Toluene-triton-lysozyme 0

RuBP dependent CO2 fixation Without RuBP With RuBP Mean ± SE Mean ± SE 55.95 ± 18.36 17.22 ± 2.62 5.59 ± 0.81 3.01 ± 0.39 7.53 ± 5.20

77.84 ± 25.96 162.9 ± 94.15 182.02 ± 2.44 169.53 ± 48.55 498.50 ± 34.22

Table 2. Calvin cycle intermediates used in assay mixtures and the corresponding sequential enzymes involved in the partial reactions for CO2 fixation activity in differentially permeated cells of Anacystis nidulans.

The mean values represent the average of three or more independent experiments along with values of standard error.

Discussion Light dependent CO2 fixation as well as the functioning of sequential reactions in the Calvin cycle differed remarkably in differentially permeated cells. Electron microscopy also revealed considerable differences in ultrastructure. Cells permeated with lysozyme retained the ability to photoassimilate CO2, which was lost when the cells were treated with toluene. In lysozyme treated cells the thylakoid membrane system was least disturbed and probably the composition was well conserved also, hence the photo assimilatory machinery functioned as efficiently as that of the untreated cells. NADPH and ATP produced by light reactions could be used efficiently by

Calvin cycle intermediates and cofactors used in assay mixtures

Enzymes involved

R-5-P + ATP SDP + ATP + GA-3-P FDP + ATP + GA-3-P GA-3-P + ATP 3-PGA + ATP + NADPH

E12, E13 and E1 E9 to E13 and E1 E6 to E13 and E1 E4 to E13 and E1 E2 to E13 and E1

The enzymes are designated as RuBP carboxylase (E1), 3-PGA kinase (E2), NADP-glycelardehyde-3-phospahte dehydrogenase (E3), triose-phosphate isomerase (E4), fructose-1, 6-diphosphate aldolase (E5), fructose-1, 6-diphosphatase (E6), transketolase (E7), aldolase (E8), sedoheptulose-1,7-diphosphatase (E9), transketolase (E10), phosphopentose isomerase (E11) phoshoriboisomerase (E12) and phosphoribulokinase (E13).

Table 3. Calvin cycle intermediate dependent CO2 fixation activity in differentially permeated cells of Anacystis nidulans. Permeabilizing agents used

Lysozyme Toluene Toluene-triton Toluene-triton-lysozyme

Intermediates used for assay of sequential reactions R5P + ATP

SDP + GA-3-P + ATP

FDP + GA-3-P + ATP

mean ± SE

mean ± SE

132.14 ± 64.13 *90.73 % 311.08 ± 25.17 *242.78 % 347.38 ± 35.97 *250.24 % 403.14 ± 163.64 *89.07 %

1.10 ± 0.59 *0.76 % 15.26 ± 2.64 *11.91 % 12.5 ± 1.84 *9.00 % negligible

GA-3-P + ATP

nmoles of CO2 fixed/O.D. 660 nm, hr mean ± SE mean ± SE 1.66 ± 0.83 *1.14 % 67.00 ± 12.74 *52.29 % 53.48 ± 4.21 *38.52 % 6.96 ± 3.52 *1.54 %

negligible 4.30 ± 1.03 *3.36 % 1.91 ± 0.67 *1.38 % negligible

PGA + NADPH + ATP mean ± SE 3.17 ± 1.88 *2.18 % 22.04 ± 3.63 *17.20 % 7.95 ± 4.06 *5.73 % negligible

The mean values represent the average of three independent experiments along with the values of the standard error. * These values represent percent of RuBP dependent activity with the respective substrates after differential permeation treatments mentioned in column 1. The 100 % values for RuBP dependent CO2 fixation activity expressed as n moles/O.D. 660 nm, hr for each treatment were as follows. 1. Lysozyme – 145.64 ± 91.53; 2. Toluene – 128.13 ± 51.57; 3. Toluene-Triton – 138.82 ± 28.46; 4. Toluene-Triton-Lysozyme – 452.61 ± 58.36

Thylakoid membranes and organisation of Calvin cycle enzymes cells showed light dependent CO2 fixation, activities of the sequential reactions of the Calvin cycle intermediates were low when compared with cells permeated with other reagents. This may be due to lower accessibility of Calvin cycle enzymes to the externally added substrates in these cells. The cells that were permeated with toluene exhibited the highest rates of SDP, FDP, GA-3-P and 3-PGA dependent CO2 fixation using ATP and/or NADPH as cofactors, suggesting that even though light reactions were disturbed by this treatment, operation of the Calvin cycle could be observed in these cells. Interestingly, the rate of 3-PGA + ATP + NADPH dependent CO2 fixation, where ATP and NADPH were externally provided, was observed to be much lower in these permeated cells as compared to light dependent CO2 fixation in non-permeated cells, indicating that the in situ generated ATP and NADPH could be used more efficiently. This supports the prediction of Suss et al. (1993) that there could be networking of Calvin cycle enzymes with the components of thylakoid membranes in vivo, which should aid in channelling of NADPH and ATP produced by electron transport to the respective enzymes. TEM images of toluene and toluene-triton treated cells showed the concentric layers of thylakoid membranes, but the appearance of thylakoid membranes was different than normal (Figs. 3 and 4). Interestingly, even after 50 % loss of proteins, triton treatment did not affect sequential reactions from FDP onwards, but resulted in reduction of activities with GA-3-P and 3-PGA. However, when cells were treated with toluene-triton-lysozyme, concentric arrangement of thylakoid membranes as well as internal ultrastructure was completely destroyed (Figs. 5 a and b). These cells showed R-5-P + ATP dependent CO2 fixation. This activity involves sequential reactions of phosphoriboisomerase, phosphoribulokinase and RuBP carboxylase, which have molecular weights of 56 kDA, 200 kDA and 550 kDA respectively. Anacystis nidulans cells appear to retain enzymes irrespective of their molecular weights, even after extensive permeation with toluenetriton-lysozyme (Figs. 5 a, b, c). This may be due to the formation of a multi-enzyme complex or attachment of these enzymes to intracellular matrix. In the cells treated with toluenetriton-lysozyme, the activities of sequential enzymes using 3-PGA, GA-3-P, FDP and SDP were notably reduced, probably due to further loss of these enzymes after disruption of thylakoid membranes by lysozyme treatment. It cannot be due to inhibition of the sequential enzymes by toluene and triton, as cells treated with toluene and toluene-triton show all these activities (Table 3). Earlier cryo-immuno electron microscopy demonstrated membrane localisation of Calvin cycle enzymes. In order to understand the functional significance of such organisation, we monitored the activities of sequential reactions of the Calvin cycle in situ, by entering the cells of cyanobacterium Anacystis nidulans after differential permeation. Disruption in structural integrity of thylakoid membranes apparently resulted in loss of the organisation and function of the Calvin

31

cycle indicating a possible structural and functional role for these membranes. The conclusive demonstration of the exact role of thylakoid membranes in the organisation of soluble Calvin cycle enzymes may be possible in the future when the technology of cryo-electron tomography and the use of structural signatures to localise the enzymes in vivo (reviewed in Baumeister and Steven 2000) will be feasible. The functional significance of this location could then be studied by the generation of precise mutations using site-directed mutagenesis, where membrane location of Calvin cycle enzymes will be specifically disrupted. Acknowledgements. We wish to thank Dr. S. K. Mahajan, Head, Molecular Biology and Agriculture Division; Dr. S. Banerjee, Director, Material Sciences Group; Dr. A. S. Bhagwat, Molecular Biology and Agriculture Division, BARC Mumbai and Dr. A. R. Chitale, Head Electron Microscopy Section, Jaslok Hospital Mumbai for encouragement during the course of this work. We also want to thank Shri D. B. Kanaskar and Shri S. S. Bhonsle, Electron Microscopy Section, Jaslok Hospital for help in taking images with 80 kV TEM. Our sincere thanks to Mrs. P. S. Agashe for help in photography. We thank Dr. N. K. Ramaswamy, Biosciences Group, BARC for kind help in the assay of photochemical activities.

References Bannur SV, Kulgod SV, Metkar SS, Mahajan SK, Sainis JK (1999) Protein determination by Ponceau S using digital color image analysis of protein spots on nitrocellulose membranes. Analyt Biochem 267: 382 – 389 Baumeister W, Steven AC (2000) Macromolecular electron microscopy in the era of structural genomics. Trends Biochem Sci 25: 624 – 631 Gontero B, Cardenas ML, Ricard J (1988) A functional five-enzyme complex of chloroplast involved in the Calvin cycle. Eur J Biochem 173: 437– 443 Mouche F, Gontero B, Callebaut I, Mornon JP, Boisset N (2002) Striking conformational changes suspected within the phosphoribulokinase dimer induced by interaction with GAPDH. J Biol Chem 277: 6743 – 6749 Nierzwicki-Bauer SA, Balkwill DL, Stevens SE (1984) The use of highvoltage electron microscopy and semi-thick sections for examination of cyanobacterial thylakoid arrangements. J Microscopy 133: 55 – 60 Sainis JK, Harris GC (1986) The association of Ribulose-1,5-bisphosphate carboxylase with phosphoriboisomerase and phosphoribulokinase. Biochem Biophys Res Commun 139: 947– 954 Sainis JK, Merriam K, Harris GC (1989) The association of Ribulose1,5-bisphosphate carboxylase/oxygenase with phosphoribulokinase. Plant Physiol 89: 368 – 374 Setlikova E, Sofrova D, Prasil O, Budac P, Koblizek M, Setlik I (1999) Integrity and activity of photosystem 2 complexes isolated from the thermophilic cyanobacterium Synechococcus elongatus using various detergents. Photosynthetica 37: 183 – 200 Stanier G (1988) Fine structure of Cyanobacteria. Methods in Enzymolgy 167: 157–172

32

Jayashree Krishna Sainis, Diksha Narhar Dani, Gautam Kumar Dey

Suss KH, Arkona C, Manteuffel R, Adler K (1993) Calvin cycle multienzyme complexes are bound to chloroplast thylakoid membranes of higher plants in situ. Proc Natl Acad Sci USA 90: 5514 – 5518 Suss KH, Prokhorenko I, Adler K (1995) In Situ association of Calvin cycle enzymes Ribulose-1,5-bisphosphate carboxylase/oxygenase, activase, ferredoxin-NADP reductase and nitrite reductase with thylakoid and pyrenoid membranes of Chlamydomonas reinhardtii chloroplasts as revealed by immunoelectron microscopy. Plant Physiol 107: 1387–1397

Suss KH, Sainis JK (1997) Supramolecular organization of water-soluble photosynthetic enzymes in chloroplasts. In: Pessarakli M (ed) Handbook of Photosynthesis. Marcel & Dekker Inc, New York Basel Honkong pp 305 – 314 Tabita FR, Caruso P, Whitman W (1978) Facile assay of enzymes unique to the Calvin cycle in intact cells, with reference to ribulose1,5-bisphosphate carboxylase. Analyt Biochem 84: 462 – 472