Skip to main content
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 1998 May 12;95(10):5830–5835. doi: 10.1073/pnas.95.10.5830

Chlorophyll a availability affects psbA translation and D1 precursor processing in vivo in Synechocystis sp. PCC 6803

Qingfang He 1, Wim Vermaas 1,*
PMCID: PMC20465  PMID: 9576970

Abstract

Transcript accumulation and translation of psbA as well as processing of the D1 precursor protein were investigated in relation to chlorophyll availability in vivo in cyanobacterial strains lacking photosystem I (PS I). The psbA transcript level was almost independent of chlorophyll availability and was ≈3-fold lower in darkness than in continuous light (5 μE m−2 s−1). Upon illumination, it reached a steady–state level within several hours. Upon growth under light-activated heterotrophic growth conditions (LAHG) in the PS I-less strain, D1 synthesis occurred immediately upon illumination. However, in PS I-less/chlL cells, which lacked the light-independent chlorophyll biosynthesis pathway and had very low chlorophyll levels after LAHG growth, very little D1 synthesis occurred upon illumination, and the synthesis rate increased with time. This result suggests a translational control of D1 biosynthesis related to chlorophyll availability. Upon illumination, initially a high level of the nonprocessed D1 precursor was observed by pulse labeling and immunodetection in LAHG-grown PS I-less/chlL cells but not in PS I-less cells. A significant amount of the D1 precursor eventually was processed to mature D1, and the half-life of the D1 precursor decreased as the chlorophyll content of the cells increased. The D1 processing enzyme CtpA was found to be present at similar levels regardless illumination or chlorophyll levels. We conclude that, directly or indirectly, chlorophyll availability is needed for D1 translation as well as for efficient processing of the D1 precursor.


The D1 protein, encoded by psbA, binds chlorophyll and constitutes part of the core of the photosystem II (PS II) reaction center. In contrast to other photosynthetic reaction center proteins, D1 turns over rapidly in light. Photodamaged D1 is replaced by newly synthesized D1 that is thought to be reassembled into existing PS II complexes. Chlorophyll binding to D1 is thought to occur during or after D1 synthesis, and this binding stabilizes the protein, whereas at least under in vitro conditions some unassembled D1 protein can be detected (1, 2).

In higher plants, synthesis of D1 is tightly regulated: translation of psbA mRNA is facilitated by binding of nuclear-encoded factors to the 5′-untranslated region that may be regulated by phosphorylation and by the redox potential of the cells (37); other nuclear-encoded factors as well as the stromal ATP content also play a role in the accumulation of mature D1 protein (8, 9). Transcription of the psbA isogenes in cyanobacteria responds to changes in light intensity (10, 11), and it has been suggested that in cyanobacteria psbA expression is regulated mainly at the transcriptional level by light (10, 12). However, at least in certain conditions, psbA transcripts accumulate in Synechocystis 6803 cells grown in the dark (13), even though rapid D1 synthesis is not required in such conditions.

The regulation of D1 synthesis also may depend on chlorophyll availability. In vitro evidence has suggested that chlorophyll regulates the accumulation of D1 by stabilizing the apoprotein (14, 15) without affecting psbA mRNA translation initiation or elongation (15); results obtained in this in vitro system were interpreted to be indicative of ribosomal pausing during the translation of D1, which was suggested to facilitate cotranslational chlorophyll binding (16, 17). Little or no in vivo evidence addressing the relationship between chlorophyll availability and D1 translation and accumulation is presently available.

The D1 protein is translated as a precursor in all of the PS II-containing organisms except Euglena (1821). The D1 precursor is posttranslationally processed by the C-terminal processing protease CtpA (22) at amino acid 344 near the C terminus to generate the mature D1 protein (2327). However, the role of D1 processing is unknown and genetic deletion of the C-terminal extension of the D1 protein in Synechocystis 6803 (27) and Chlamydomonas reinhardtii (28, 29) does not have a significant phenotypic effect.

Synechocystis 6803 contains both a light-dependent and light-independent protochlorophyllide reductase. By deletion of chlL (encoding one of the components of the light-independent protochlorophyllide reductase) from a photosystem I (PS I)-less Synechocystis 6803 strain, a mutant has been constructed that, upon continued growth under light-activated heterotrophic growth (LAHG) conditions (i.e., growth in darkness except for several minutes of light every day) (30), can be depleted almost entirely of chlorophyll and of PS II (31). Deletion of the por gene (encoding the light-dependent NADPH: protochlorophyllide oxidoreductase) from this strain generated the PS I-less/chlL/por (del) strain, which synthesized reduced amounts of chlorophyll in a light-dependent manner and with apparently poor quantum efficiency. This chlorophyll was associated mostly with components other than PS II (32). This system is suitable to study the effects of light as compared with those of chlorophyll availability on synthesis of the D1 protein under in vivo conditions. In this paper, we demonstrate that neither light nor chlorophyll greatly modifies psbA transcript levels but that the availability of chlorophyll is required for psbA translation. Interestingly, also the rate of processing of the D1 precursor was found to be influenced, directly or indirectly, by the availability of chlorophyll.

MATERIALS AND METHODS

RNA Isolation and Northern Blots.

Total cellular RNA was isolated from 50–100 ml of Synechocystis 6803 cultures at OD730 ≈ 0.5 as described (10). The RNA preparation obtained by this method was treated with 5 units of RNase-free DNase at 37°C for 20 min to remove any trace of DNA and then reprecipitated in 70% ethanol in the presence of 0.25 M sodium acetate (pH 5.3) overnight after extraction with chloroform. The pellet was washed with 80% ethanol, resuspended in H2O, and stored in aliquots at −70°C.

RNAs were separated on 1.2% agarose-formaldehyde gels and transferred to GeneScreen Plus membrane (NEN/Dupont) according to the manufacturer’s recommendation. The blots were hybridized with 32P-labeled probes (33) that were prepared by PCR (32). For rehybridization, the blots were stripped by incubating with boiling 2X standard saline citrate containing 1% (wt/vol) SDS twice for 10 min each and were reprobed after the blots had been verified to be nonradioactive.

Protein Pulse Labeling and Pulse–Chase with [35S]Protein Labeling Mix.

To probe the synthesis of cyanobacterial membrane proteins, pulse-labeling with protein-labeling mix (EXPRE35S35S, containing 73% l-[35S]methionine and 22% l-[35S]cysteine) (DuPont/NEN) was used. To follow protein synthesis upon initiation of chlorophyll synthesis in PS I-less/chlL strains, the appropriate Synechocystis 6803 strains were propagated in liquid under LAHG conditions (15 min of light at 10 μE m−2 s−1/day; dark otherwise) for 2–3 wk. At the time cultures were transferred from LAHG conditions to continuous light at 5 μE m−2 s−1 (time 0), they were in logarithmic growth phase (OD730 ≈ 0.3–0.5). Upon continuous illumination, protochlorophyllide reduction and hence, chlorophyll synthesis resumed in chlLstrains. At different time points after the start of continuous illumination, 50–100 ml of cultures were harvested by centrifugation at 1,600 × g for 5 min at room temperature. The cell pellet was resuspended in sulfur-less BG 11 (33) with 10 mM glucose. Protein labeling mix was added to the medium to a final concentration of 1 μCi/ml (1 Ci = 37 GBq). The culture was incubated in a 300-ml transparent centrifuge tube at 10 μE m−2 s−1 for 15 min, after which it was chilled rapidly by adding ice to the culture and spun down by centrifugation. Cells were used immediately for thylakoid isolation in the presence of protease inhibitors or they were frozen in liquid nitrogen for thylakoid preparation on the same day.

Pulse–chase experiments were performed as described (33). To measure the half-life of the D1 precursor, appropriate Synechocystis 6803 strains were grown under conditions as specified in text and figure legends. Cells (OD730 ≈ 0.5) were first labeled in vivo; the labeled cells were then harvested by centrifugation at room temperature and resuspended in one-half the original volume of sulfur-less BG 11 medium supplemented with 25 μM each of unlabeled methionine and cysteine. Immediately, a 20-ml fraction of the cell suspension was aliquoted into a centrifuge tube half-filled with ice and harvested by a 3-min centrifugation at 4°C (0 min chase). The cell pellets were frozen in liquid nitrogen until thylakoid isolation. The remainder of the cells was incubated for different lengths of time under normal growth conditions either in light or in darkness, after which 20-ml aliquots of the cell suspension were chilled, harvested, and frozen as stated above until thylakoid isolation on the same day.

Preparation of Cell Extracts and Thylakoid Membranes.

Cyanobacterial cell pellets were resuspended in thylakoid buffer (1/100 of the original culture volume) containing 20 mM Mes/NaOH (pH 6.4), 5 mM MgCl2, 5 mM CaCl2, 20% glycerol (vol/vol), 1 mM freshly made phenylmethylsulfonyl fluoride, and 5 mM benzamidine. Cell suspensions (0.4–0.6 ml) were transferred to an ice-chilled microcentrifuge tube filled with ≈0.5 ml of glass beads prewetted by thylakoid buffer. The cells were broken in a MiniBeadBeater (BioSpec Products, Bartlesville, OK) by six-breaking cycles (30 sec of shaking, each followed by a 3–5 min of cooling on ice-water). After centrifugation at 1,600 × g for 10 min to remove unbroken cells and cell debris, the supernatant was used as total cell extract for Western blot analysis of CtpA or was diluted 30- to 50-fold in thylakoid buffer and centrifuged again to sediment the thylakoids at 4°C (20 min at 40,000 rpm in a Beckman Ti 50.2 rotor). The thylakoids were washed once and resuspended in thylakoid buffer (1/200 of the original culture volume).

SDS/PAGE and Western Blotting.

SDS/PAGE and Western blotting procedures were described elsewhere (32). A total of 25 μg of protein was loaded per lane unless indicated otherwise. To estimate the protein content of a sample, the cell extract or thylakoid sample (9 μl) was incubated in 2% SDS at 37°C for 15 min. After centrifugation to remove insoluble fragments, the supernatant was diluted 10-fold and the protein content was measured using bicinchoninic acid-protein assay reagents (Pierce) according to the manufacturer’s instructions.

RESULTS

psbA Transcript Levels.

A first aim of this study was to investigate the effects of chlorophyll availability and illumination on psbA transcript accumulation and at the same time, to clarify discrepancies reported in the literature regarding psbA transcript accumulation in darkness in cyanobacterial systems: no psbA transcript was found in darkness by Mohamed and Jansson (10), whereas Smart and McIntosh (13) observed a high level of psbA transcript in dark-incubated cultures. To investigate the role of chlorophyll and light in psbA transcript accumulation, the PS I-less/chlL and PS I-less strains were grown under LAHG conditions for 2 wk and transferred to continuous illumination at 5 μE m−2 s−1 immediately before the daily 15-min light period of the LAHG incubator. Samples were taken at different time points after the onset of continuous illumination. After isolation of RNA from these samples, Northern blot analysis was performed to investigate psbA transcript levels. The results are shown in Fig. 1. The labeled gene-internal fragment of psbAII hybridized to a 1.2-kilobase (kb) transcript corresponding in size to that of psbAII and psbAIII (10, 13).

Figure 1.

Figure 1

Transcript analysis of psbA and rps1. LAHG-propagated PS I-less/chlL (A) and PS I-less (B) cells were transferred to continuous illumination at 5 μE m−2 s−1 for the length of time indicated, and RNA was isolated. RNA of the light-grown PS I-less/chlL/por (del) strain was included in A, lane C* and that of the light-grown PS I-less strain was in B, lane C. (C) Transcript from wild-type Synechocystis 6803 grown photoautotrophically (light, −G) and photomixotrophically (light, +G) and after 2 days of darkness in the absence of glucose (dark, −G) and in presence of 5 mM glucose (dark, +G). The wild-type cells used for these experiments were grown photoautotrophically in continuous light at 50 μE m−2 s−1 to OD730 ≈ 0.4 before dark incubation. The blots were first probed with psbAII and then with rps1 after the psbA probe was stripped.

The response of psbA transcript levels to illumination was similar for the PS I-less/chlL and PS I-less strains, with ≈35% of the psbA transcript level present in LAHG-propagated cells as compared with cells exposed to light for 24 h or more (Fig. 1 A and B). The transcripts increased to 80% of the normal level within 3 h of illumination and had reached a maximum level after 24 h in the light (Fig. 1B). The light-grown PS I-less/chlL/por (del) cells, which accumulated only 15% of the amount of chlorophyll present in a PS I-less strain with normal chlorophyll synthesis (32), contained ≈70% of the psbA transcript amount present in light-grown PS I-less/chlL or PS I-less strains (Fig. 1A).

For an estimation of transcript levels by Northern blots, an appropriate loading control is important. Because rRNA and many mRNAs are expected to have different stability, we chose to probe for the rps1 (encoding the ribosomal protein S1) transcript as a constitutive and moderately abundant loading control. The rps1 probe hybridized to a 1.2-kb transcript band, consistent with what was reported in the Synechococcus sp. strain PCC 6301 (34). As reported by Mohamed and Jansson (10), no significant amount of psbA transcript was observed in cultures of wild-type Synechocystis 6803 that were left in darkness for 48 h in the absence of glucose (Fig. 1C; a similar amount of total RNA was loaded as compared with other lanes). However, the rps1 transcript also was reduced significantly and became barely detectable at 3 days of incubation in total darkness (not shown) even though cells were still viable and the yield of total RNA did not change much. This result indicates that rps1 of Synechocystis 6803 responds to carbon (glucose) starvation in a similar way as was observed in Escherichia coli, yeast, and other organisms (3537). Therefore, we propose that the lack of psbA transcript in darkness in the absence of glucose (10) reflects starvation; however, an unspecific negative regulation cannot be excluded. The rps1 level present in the PS I-less/chlL/por (del) cells was found to be reduced, which is consistent with the notion that the level of ribosomal protein transcript is growth-rate related (36, 37).

In Vivo Synthesis of D1 as a Function of Chlorophyll Biosynthesis.

Mostly based on in vitro and in organello data, D1 synthesis and accumulation has been shown to be affected by chlorophyll availability (38). The current interpretation of the data suggests a posttranslational stabilization of D1 by chlorophyll (15). Our cyanobacterial system allows an investigation of the relationship between chlorophyll availability and biosynthesis of D1 in vivo. The PS I-less and PS I-less/chlL strains were grown under LAHG conditions for 2 wk before these cells were returned to continuous light at 5 μE m−2 s−1. After a period of continuous illumination (0–24 h), pulse labeling (15 min) was carried out. In the PS I-less strain, which was able to synthesize chlorophyll in darkness, the mature D1 protein was labeled prominently regardless the time of illumination (Fig. 2B). The increase in the amount of labeling with illumination time roughly correlates with transcript availability. However, in the PS I-less/chlL strain (Fig. 2A), which was depleted of chlorophyll during LAHG growth because this strain does not synthesize chlorophyll in darkness, the mature form of the D1 protein barely was detectable during the first 2 h in the light and became increasingly prominent with increasing illumination time. Even in very short pulses (3 min + 3 min centrifugation), very little D1 was labeled in early stages of greening (data not shown), essentially excluding the possibility that D1 is synthesized but rapidly degraded. Because the psbA transcript level at the start of illumination is a third of the level present upon continuous illumination in both strains, the virtual absence of D1 translation shortly after the start of illumination in the PS I-less/chlLstrain but not in the PS I-less strain suggests a translational control of D1 synthesis that may be related (directly or indirectly) to chlorophyll availability.

Figure 2.

Figure 2

Pulse labeling of thylakoid proteins. LAHG-propagated PS I-less/chlL (A) or PS I-less (B) cells were transferred to continuous illumination at 5 μE m−2 s−1 for the length of time indicated. At this time point, cells were taken, and newly synthesized proteins were labeled in the light in vivo for 15 min. Thylakoid membranes were isolated, and thylakoid proteins were separated on SDS/PAGE; labeled proteins were detected by autoradiography.

A band above the mature D1 band was relatively prominent in early stages of illumination and chlorophyll synthesis in the PS I-less/chlL strain but not in the PS I-less strain with normal chlL (Fig. 2A). This band also was present in the light-grown PS I-less/chlL/por (del) strain, which synthesizes only a fraction of chlorophyll as compared with the PS I-less or the PS I-less/chlL strains grown in light (data not shown). Because the mobility of the band above D1 was similar to that of the D1 precursor, we probed the immunological relationship between this band and D1. Western blots (Fig. 3A) showed that the band above D1 was immunoreactive with D1 antiserum whereas no crossreaction was observed at this molecular mass in the PS I-less strain that can synthesize chlorophyll in darkness (Fig. 3B). Therefore, we assigned the band that, upon pulse labeling, accumulated in early stages of chlorophyll synthesis to be the D1 precursor. The D1 precursor was not detectable after PS I-less/chlL cells had been in continuous light for 24 h. A small amount of mature D1 was present in samples of the PS I-less/chlL strain even before they were exposed to continuous light.

Figure 3.

Figure 3

Western blot analysis of D1 and CtpA proteins. LAHG-propagated PS I-less/chlL (A, C) or PS I-less (B) cells were transferred to continuous illumination at 5 μE m−2 s−1 for the length of time indicated, and thylakoid membranes were isolated. Thylakoid proteins were separated on SDS/PAGE, transferred to poly(vinylidene difluoride) membranes, and probed with D1 antiserum (A, B) or with CtpA antiserum (C). In lanes labeled 24* and 24**, 10% and 25%, respectively, of the amount of sample in the 24–h lane were loaded.

The apparent relative stability of the D1 precursor at early stages of chlorophyll synthesis is linked to chlorophyll synthesis rather than to light. In the anti-D1-probed Western blot, only mature D1 was detected in the PS I-less strain that contained chlL (Fig. 3B) regardless whether it was grown in continuous light or under LAHG conditions. LAHG-grown PS I-less cells contained ≈30% of the level of the D1 protein that was present in PS I-less or PS I-less/chlL cells that were grown in continuous light.

Fig. 3A indicates that, even at early times of illumination, no smaller polypeptides are seen that crossreact with D1 antisera. This is in contrast to results of in organello studies carried out with higher plants (14, 16, 17) in which smaller bands were observed on Western blots under conditions of insufficient chlorophyll abundance; these bands were attributed to D1 “pausing” products resulting from translational arrest.

The accumulation of detectable levels of the D1 precursor during early stages of illumination in PS I-less/chlL cells that had been propagated under LAHG conditions suggests that processing of the D1 precursor is not efficient if insufficient chlorophyll is available.

Expression of CtpA upon Greening.

Because the processing of the D1 precursor seemed to be attenuated upon insufficient availability of chlorophyll, we investigated whether this inefficient processing was caused by differences in the amounts of CtpA, the protease responsible for D1 processing. To this aim, we probed the presence of CtpA by Western blot; the CtpA antiserum was a kind gift of Dr. Himadri Pakrasi, Washington University. In contrast to D1, which showed at least a 10-fold induction upon illumination for 24 h, the level of CtpA protein (Fig. 3C) remained almost constant during this time period. This result indicates that CtpA availability is not a major factor causing the attenuation of D1 processing.

Pulse–Chase of D1 and its Precursor.

As shown in Figs. 2A and 3A in PS I-less/chlL cells propagated under LAHG conditions, the D1 precursor accumulated in early stages of greening, and the amount of mature D1 increased at later stages. The question is whether the D1 precursor is processed and may give rise to mature D1 or whether the precursor is degraded and mature D1 results from protein that was translated later. To address this, the PS I-less/chlL strain was grown under LAHG conditions for 2 wk and transferred to continuous illumination for 1 h (5 μE m−2 s−1). Cells were then pulse-labeled for 15 min in the light. Labeled cells were divided into two aliquots: in one, the labeled proteins were chased in light; in the other, the chase was carried out in darkness. Approximately 65% of the D1 precursor disappeared within 15 min, and the D1 precursor became invisible after 1.5 h of chase regardless of the presence of light, indicating that the decay kinetics of the D1 precursor are independent of illumination (Fig. 4A). Meanwhile, the amount of mature D1 (normalized to the intensity of label in the entire lane) increased slightly after a 15-min chase but less than would be expected on the basis of the decrease in the amount of D1 precursor (Fig. 4B). Therefore, part of the D1 precursor is processed to mature and relatively stable D1, whereas the remainder appears to be degraded, with or without processing, without accumulation of sizeable breakdown products.

Figure 4.

Figure 4

(A) Pulse–chase labeling of thylakoid proteins. LAHG-propagated PS I-less/chlL cells were transferred to continuous illumination at 5 μE m−2 s−1 for 1 h and after this time, cells were pulse-labeled in light for 15 min. The radioactivity was chased subsequently with a 15,000-fold excess of unlabeled methionine and cysteine for the length of time indicated in darkness or in light. (B) Relative intensity of labeled D1 and its precursor. The intensity of radiolabeled D1 polypeptide (circles) and its precursor (squares) upon chasing in darkness (closed symbols) and in light (open symbols) was determined by densitometry. The integrated intensity of each labeled band was normalized to the total intensity of the corresponding lane. The intensity of D1 label at chase time zero was normalized to one.

These results suggest that the lifetime of the D1 precursor was inversely correlated with the amount of chlorophyll available. To test this possibility, pulse–chase experiments were carried out on light-grown PS I-less/chlL/por (del) cells, which have ≈15% of the amount of chlorophyll in light-grown PS I-less strains with or without chlL, and on LAHG-propagated PS I-less/chlL cells that were illuminated for different lengths of time (Table 1). The results indeed showed a correlation between the rate of processing of the D1 precursor and the chlorophyll content of the cells.

Table 1.

Half-life of the D1 precursor at low chlorophyll availability

Strain illumination (h) PS I-less/chlL
PS I-less/chlL/ por (del) continuous
1 6 24
Half-life (min) 10 2 n.d. 7
Chlorophyll (μM/OD730) 0.09 0.38 0.71 0.12

PS I-less/chlL cells were propagated under LAHG conditions for 2 wk before they were transferred to continuous illumination at 5 μE m−2 s−1. Cell aliquots were taken at specified time points after the start of illumination, labeled with [35S]methionine and cysteine for 15 min, and chased in light after addition of 25 μM each of unlabeled methionine and cysteine. The intensity of label associated with D1 precursor on autoradiograms was measured by densitometry, and the half-life of the D1 precursor was calculated. n.d., precursor not detectable. 

DISCUSSION

Accumulation of psbA Transcript.

Upon propagation of Synechocystis 6803 strains under LAHG conditions for 1–2 wk, steady–state psbA transcript levels dropped to 35% of those detected in light-grown cultures regardless the presence of chlorophyll. This result suggests that psbA is transcribed in darkness because even if the psbA transcript is very stable in darkness with a half-life of 7 h (39), the transcript level will decrease to 13% of the original level in the dark period before the next 15-min daily illumination, excluding the effect of dilution of psbA transcripts by cell division (the PS I-less/chlL and PS I-less strains have a doubling time of ≈26 h).

The LAHG-grown PS I-less/chlL cells contained a similar amount of psbA transcript as the LAHG-propagated PS I-less cells; upon illumination, the psbA transcript level followed the same kinetics to recover to a continuous light level. This indicates that psbA transcript accumulation is independent of chlorophyll availability. This result is supported by a comparison with psbA transcript levels in the light-grown PS I-less/chlL/por (del) cells, which accumulate only ≈15% of chlorophyll as compared with the PS I-less or the PS I-less/chlL strains grown in continuous light: the psbA transcript level in the PS I-less/chlL/por (del) mutant is ≈70% of that in strains with normal chlorophyll levels.

In Vivo Synthesis of D1 Is Controlled by the Overall Availability of Chlorophyll.

D1 protein synthesis occurs on ribosomes bound to thylakoid membranes (40, 41). Dark-grown plants lack D1 (42), even though polysome-associated psbA mRNA is present (14, 43). Therefore, it is believed that, in higher plant systems, no D1 synthesis occurs in darkness (44), but the factor(s) involved in activation of D1 synthesis are as yet unknown.

To investigate the relationship between chlorophyll availability and D1 biosynthesis, pulse labeling was carried out on the PS I-less/chlL mutant that had been grown under LAHG conditions and subsequently was illuminated. Our results show that D1 synthesis barely is detectable within the first hour after the start of continuous illumination, during which period the chlorophyll biosynthesis rate is high, but only ≈15% of the amount of chlorophyll present in continuous illumination-grown cells is generated (31). These results clearly show that also in cyanobacterial systems transcript levels by no means constitute the sole mechanism regulating D1 synthesis. The psbA transcript levels showed only a 2- to 3-fold induction during the first day of continuous illumination whereas D1 synthesis showed an at least 15-fold induction during the same period of illumination in LAHG-grown PS I-less/chlL cells. However, in strains that synthesize chlorophyll in darkness, D1 synthesis rates are very significant in the first hour of illumination. This result indicates that D1 synthesis is controlled by the overall chlorophyll availability rather than by chlorophyll synthesis per se.

It is interesting to note that no D1 synthesis intermediates (pausing products) were observed by immunodetection or pulse labeling in any of the samples, in contrast to interpretations based on in vitro studies of higher plant chloroplasts (16, 17). Our results obtained on this in vivo cyanobacterial system suggest that the lack of sufficient chlorophyll availability affects the initiation of psbA translation but does not lead to premature termination or pausing of D1 translation in vivo in Synechocystis 6803. The existence of short pausing products cannot be totally excluded if the polyclonal D1 antiserum used for this work recognizes mostly C-terminal epitopes, but none of the pausing products, if any, accumulate sufficiently to be detectable by pulse labeling (Fig. 2).

In C. reinhardtii, at least four proteins have been shown to specifically bind to the psbA 5′-untranslated region (5, 6). These proteins are expressed regardless of illumination but were shown to have a higher affinity to the psbA 5′-untranslated region in light-grown cells. Therefore, it has been suggested that modulation of the binding activity rather than the quantity of these proteins regulates the psbA mRNA translation (5, 45). Based on our data, it is tempting to hypothesize that one of these proteins and/or other unknown psbA translation activators may be a chlorophyll-binding protein whose affinity to psbA mRNA may change depending on the number of chlorophylls it has bound.

Chlorophyll Availability Increases the D1 Processing Rate.

In wild-type or PS I-less Synechocystis 6803 strains that can synthesize chlorophyll both in light and in darkness, no D1 precursor can be observed by pulse labeling or immunodetection (Figs. 2B and 3B). However, in LAHG-propagated PS I-less/chlL cells as well as in the same cells that were illuminated subsequently for several hours, the D1 precursor was detectable both upon pulse labeling and by immunodetection. In addition, pulse–chase experiments demonstrated that the D1 precursor was processed slowly under conditions in which little chlorophyll was available and that the D1 precursor processing rate increased with increasing chlorophyll levels.

Because CtpA is continuously present in the PS I-less/chlL strain and because there is no evidence thus far that CtpA activity is regulated in any way, we interpret our data by suggesting that the D1 precursor changes its conformation when the protein or other proteins of the PS II complex bind sufficient chlorophyll and/or become properly assembled as a function of chlorophyll availability and that this in turn affects the availability of the processing site in the D1 precursor to CtpA. Having the D1 precursor become available to CtpA may involve either a major structural rearrangement or something much more subtle: both in vivo and in vitro studies of enzymatic processing of pD1 by site-directed mutagenesis have indicated that the secondary structure around the cleavage site rather than the presence of specific amino acids is important for the CtpA-catalyzed proteolytic reaction (46, 47).

The experimental system described in this paper has opened up an opportunity to study the biogenesis of chlorophyll-binding proteins under in vivo conditions. Many notions obtained from in vitro and in organello studies regarding the biosynthesis and regulation of PS II proteins have now been tested in vivo, which has provided distinctly different insights. The finding that D1 processing is correlated with chlorophyll availability also has added another dimension to the regulation of D1 synthesis.

Acknowledgments

We thank Dr. Himadri Pakrasi (Washington University, St. Louis, MO) for providing antibodies against CtpA from Synechocystis sp. PCC 6803. This work has been supported by a grant from the U.S. Department of Energy (DE-FG03-95ER20180).

Footnotes

This paper was submitted directly (Track II) to the Proceedings Office.

Abbreviations: PS I, photosystem I; PS II, photosystem II; LAHG, light-activated heterotrophic growth; chlL, gene encoding subunit of the light-independent protochlorophyllide reductase; CtpA, C-terminal protease processing the D1 protein.

References


Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES