Abstract
Nitrogen (N) deficiency is a frequently encountered situation that constrains global biomass productivity. In response to N deficiency, cell division stops and photosynthetic electron transfer are downregulated, while carbon storage is enhanced. However, the molecular mechanism downregulating photosynthesis during N deficiency and its relationship with carbon storage are not fully understood. The Proton Gradient Regulator-like 1 (PGRL1)-involved in cyclic electron flow (CEF) and Flavodiiron proteins involved in pseudo-(CEF) are major players in the acclimation of photosynthesis. To determine the role of PGRL1 or FLV in photosynthesis under N deficiency, we measured photosynthetic electron transfer, oxygen gas exchange and carbon storage in the knockout of Chlamydomonas pgrl1 and flvB mutants. Under N deficiency, pgrl1 maintains higher net photosynthesis and O2 photoreduction rates, while flvB shows similar responses compared to control strains. The amount of cytochrome b6f was maintained at a higher level in pgrl1. The photosynthetic activity of pgrl1 flvB double mutants decreases in response to N deficiency similar to the control strains. Furthermore, the triacylglycerol content of pgrl1 was twice higher than the controls under N deficiency. Taken together, our results suggest that in the absence of PGRL1, FLV-mediated O2 photoreduction through PCEF maintains net photosynthesis at a high level, resulting in increased triacylglycerol biosynthesis. This study reveals that PGRL1 and FLV play antagonistic roles during N deficiency. It further illustrates how nutrient status can affect the regulation of photosynthetic energy production in relation to carbon storage and provides new strategies for improving lipid productivity in algae.
Significance statement Nitrogen (N) deficiency, an often-encountered phenomenon in nature, triggers growth arrest and massive lipid accumulation in microalgae. The downregulation of photosynthesis is necessary to ensure cell viability. We demonstrate that a well-conserved protein in chlorophytes, the Proton Gradient Regulator-like 1 (PGRL1) is a key (down) regulator of photosynthesis. In its absence, cells exhibited sustained photosynthesis and over-accumulated lipids thanks to the Flavodiiron (FLV) protein. We propose that both PGRL1 and FLV, by having antagonistic roles in N deficiency, manage the redox landscape, carbon storage and biomass production. Our work revolves around the current paradigm of photosynthesis regulation during N deficiency and provides a new framework for improving lipid accumulation in microalgae for biotechnological purposes.
Introduction
Nitrogen (N) deficiency is one of the most harsh environmental situation that constrains global primary biomass productivity in all ecosystems (1–3). Under N deficiency, cell division and photosynthetic CO2 assimilation are downregulated, the carbon and energy are used to synthesize starch and triacylglycerols (TAGs) (4–7). The downregulation of electron transfer reactions, together with a re-routing of the excess of reducing power towards carbon storage, prevents over-production of reactive oxygen species, thus ensuring cell fitness (8–12).
Due to the variability in their natural habitat, microalgae must constantly adjust the production of energy by photosynthesis to match the metabolic demand and have therefore developed a set of regulatory mechanisms to fine-tune electron transfer reactions. During photosynthesis, energy in the form of ATP and NADPH is mostly produced by the linear electron flow (LEF) (13). The balance of ATP and NADPH is essential for optimal CO2 capture and metabolism to which cyclic electron flow (CEF) and pseudo-cyclic electron flow (PCEF) play a critical role (14–17). Two pathways of CEF around PSI have been described in the green microalga Chlamydomonas reinhardtii (Chlamydomonas hereafter), one involving the type II NADPH dehydrogenase (NDA2) (18), and the other the proton gradient regulator 5 (PGR5)/PGR-like 1 (PGRL1) proteins (19). Both CEF pathways reduce the plastoquinone (PQ) pool by using either NADPH or ferredoxin (Fd) as electron donor respectively (18, 20). CEF has been shown to take part in the acidification of thylakoid lumen through generating an extra proton motive force (pmf) in addition to the one produced by LEF (18, 19, 21, 22). PCEF mediated by Flavodiiron proteins (FLVs), by transferring electrons from Fd toward O2, also contributes to the establishment of the pmf and therefore to the lumen acidification (23–25). The pmf is used to either (1) produce ATP (26), (2) trigger light energy quenching via a low lumenal pH (27–29) or (3) repress electron transfer at the level the cytochrome b6f complex through the photosynthetic control triggered by the low lumenal pH (21, 30–32). Both CEF and PCEF have been shown to be critical under various conditions of light, CO2 availability or sulfur deficiency, where both mechanisms have been found to play a synergistic role (14, 24, 25, 33, 34).
Despite the importance of CEF and PCEF in response to dynamic environments, little is known about their role during N deficiency. The role of NDA2-involved CEF have been recently addressed during CO2-limiting photoautotrophic N deficiency and the lack of NDA2 was shown to impair the establishment of non-photochemical quenching (22). The role of PGRL1-PGR5 pathway has been investigated under mixotrophic N deficiency and the lack of PGRL1 was shown to decrease the rate of CEF and consequently TAG production (35). However, the role of PGRL1 during photoautotrophic N deficiency and the possible existence of interactions between carbon/energy sinks (TAG and starch) and (P)CEF pathways has not been explored so far. The latter appears particularly important considering that massive carbon reallocation occur under N deficiency (4). Fully understanding the role of these regulatory pathways and how they affect reserve formation is also needed towards engineering photosynthesis and improving lipid production in conditions of nutrient deficiency.
Here, we evaluated the contribution of PGRL1-mediated CEF in the regulation of photosynthesis during N deficiency and its interaction with carbon reserve accumulation in Chlamydomonas cells grown in photoautotrophic conditions in non-limiting CO2 concentrations (using CO2-enriched air), conditions which favor the accumulation of carbon reserves (7, 36, 37). By monitoring photosynthetic activity based on chlorophyll fluorescence and O2 exchange rates measurements, we observed a continuous and high net and gross photosynthetic activity, elevated O2 uptake and hyper-accumulation of TAGs in a PGRL1-deficient strain under N deficiency compared to the control strains. By characterizing mutants deficient in either PGRL1, FLVs or both we showed that FLVs, by maintaining a strong PCEF in the pgrl1 mutant, allows maintaining high photosynthetic rates and high TAG accumulation. Finally, we discuss how modulating photosynthetic electron flow could constitute an efficient strategy to boost TAG production under nutrient deficiency.
Results
A PGRL1-deficient mutant showed sustained photosynthetic activity under N deficiency
To first assess the role of PGRL1 in the regulation of photosynthesis during photoautotrophic N deficiency, we simultaneously monitored chlorophyll fluorescence and O2 exchange following a dark-light-dark transition in N-replete and N-deprived cells. We compared the photosynthesis efficiency of the PGRL1-deficient strain (pgrl1 hereafter) with its control wild-type strain (137AH) and two complemented lines (19) (Fig. 1 and Supplemental Fig. S1 and S2). Under N replete condition, both chlorophyll fluorescence and net O2 evolution patterns were mostly similar in pgrl1 and control cells (Fig. 1 A, B, E and F). pgrl1 exhibited higher O2 uptake and gross O2 production rates in the light as compared to control lines (Fig. 1B and F), as previously reported (14, 25). After 2 days of N deficiency, the PSII operating yield measured in the light decreased by about 75% in the WT and only by 35% in pgrl1 (Fig. 1 C, D, E and F; Supplemental Fig. S1C). As previously reported (5, 38) the dark respiration rate was stimulated in the WT under N deficiency, this effect being reduced in pgrl1 (Supplemental Fig. S2 and S4). Conversely, pgrl1 showed twice higher net O2 evolution and light-dependent O2 uptake as compared to the WT (Fig. 1 D and F). By using pgrl1 complemented lines (19) we observed full recovery in the pgrl1::PGRL1-2 and a partial recovery in pgrl1::PGRL1-1 (Supplemental Fig. S1 and S2). The preservation of photosynthetic activity in pgrl1 suggests that CEF contributes to repressing the activity of photosynthetic electron transfer under photoautotrophic N-deficiency.
A-C, Chlorophyll fluorescence was measured using a Dual-PAM during the dark-light-dark transition from N-replete (A) and N-deficient (B) conditions.
D-F, O2 exchange rates were measured using a MIMS in the presence of [18O]-enriched O2. Net O2 evolution (green) was calculated as gross O2 production (blue) – gross O2 uptake (red).
E, PSII quantum yield before and after 2 d of N deficiency. PSII (Y) was measured using green actinic light (1250 µmol photon m-2 s -1, green LEDs) and calculated as ΦPSII= (FM’-Fs)/ FM’ with FM’ the fluorescence value after saturating pulse, Fs the stationary fluorescence during actinic light exposure. Asterisks represent statistically significant differences compared to the control 137AH (* p<0.05, ** p<0.01 and *** p<0.001) using one-way ANOVA.
F, The Net O2 production before and after 2 d of N deficiency in the wild-type 137AH, pgrl1 and the complemented lines pgrl1::PGRL1-1 and pgrl1::PGRL1-2.
G, Immunoblot analysis of PGRL1 accumulation in the wild-type 137AH, pgrl1 and the complemented lines. L30 antibody was used as a positive control. Coomassie blue staining was used as the loading control. Protein samples were obtained from 2 d N-starved cells.
Wild type 137AH, pgrl1, pgrl1::PGRL1-1 and pgrl1::PGRL1-2 N-replete cells cultivated photoautotrophically with 1% CO2 in air under continuous light (50 μmol photons m−2 s−1) were transferred into N-free media for 2 d prior to measurements. Shown are an average of at least three biological replicates ± SD.
Cyt b6f, PsaD and FLVs accumulate to higher levels in pgrl1 as compared to its control under N deficiency
The amount of photosynthetic proteins drops when Chlamydomonas cells undergo N deficiency under mixotrophic conditions (5, 39). However, this effect strongly depends on growth conditions, as the amount of photosynthetic proteins is maintained at high level when cells are grown under CO2-limiting photoautotrophic conditions (22). By loading similar protein amounts, we observed detectable amount of PGRL1, NDA2, FLVA and FLVB proteins up to 48h of N deficiency although FLVs were found at lower amount compared to N-replete conditions (Fig. 2 A and B). A strong reduction in the amounts of Cyt f, PsbD, PsaC and PsaD subunits were observed when N deficiency was performed in non-limiting CO2 photoautotrophic, thus indicating a reduction in the amounts of Cyt b6f, PSII and PSI complexes (Fig. 2 A and B). We conclude from these analyses that N-limitation induces a strong decrease in most proteins involved in photosynthetic electron transport with the exception of the CEF proteins PGRL1 and NDA2.
A, Immunodetection of photosynthetic proteins in the wild-type 137AH during N deficiency. Cells were harvested at 0, 16, 24, 40 and 48 of N deficiency. Total proteins were extracted from the same cell number (20 million of cells) to quantitatively follow changes in specific protein amounts. Shown are representative images of three biological replicates. We validated our quantitative analysis by showing unchanged accumulation of the housekeeping α-tubulin protein and the timely induction of ATG8 autophagy protein despite the gradual degradation of total cellular proteins as shown by Coomassie staining during N deficiency.
B, Heatmap showing the abundance of proteins during N deficiency in the WT cells. The time 0 (N replete) was used for normalization. The color scale is shown in the right. Data are average of at least two biological replicates.
C, Immunodetection of photosynthetic proteins in the wild-type 137AH and pgrl1 during N deficiency. Total proteins were extracted from cells harvested at 0, 8, 22, 30 and 50 h of N deficiency. Samples were loaded at equal total protein amounts as shown by Coomassie blue staining. Shown are representative images of three biological replicates.
D, Heatmap showing the proteins abundance during N deficiency in the pgrl1 cells. The data from different time points were normalized to the WT protein amount. The color scale is shown in the right. Data are average of at least two biological replicates.
To gain insights into the mechanisms behind the high photosynthetic activity in pgrl1 during N deficiency, we compared the relative abundance of major proteins involved in photosynthetic complexes and in CEF and PCEF in the pgrl1 and the WT during N deficiency (Fig. 2 C and D). Interestingly, we observed a higher accumulation of the PSI subunit PsaD and reduced degradation of Cyt f in pgrl1 compared to the WT (Fig. 2 C and D). FLVA amount was found more abundant after 2 d of N deficiency despite a reduced level in N-replete in pgrl1 compared to the WT (Fig. 2 C and D). It is worth noting that the hallmark of autophagy i.e. ATG8 was barely detectable in the mutant whereas it was highly induced in the WT under N deficiency (Fig. 2 C and D). All the other proteins tested accumulated to a similar amount in pgrl1 and WT (Fig. 2 C and D). We conclude from this experiment that pgrl1 mutant keeps a higher photosynthetic activity during N deficiency than the WT thanks to a lower decrease of the Cyt b6f amount and higher accumulation of PSI and FLVs proteins.
FLV-mediated O2 photoreduction drives photosynthesis in pgrl1 under N deficiency
The above data (Fig. 1 and 2) suggest that PGRL1 is involved in the downregulation of PET, and that a light-dependent O2 uptake mechanism is activated in pgrl1 during N deficiency. We then questioned the nature of molecular actors involved in the increased O2 uptake. As FLVs have been reported to be the main O2 uptake mechanism in the light (24, 25, 40), we evaluated steady-state O2 exchange rate in pgrl1, flvB as well as in the pgrl1 flvB double mutants (Fig. 3). In N-deficiency, the light-dependent O2 uptake, was highly increased in pgrl1 but strongly impaired in flvB and pgrl1 flvB mutants (supplemental Fig. S3 and S4B) which mirrors what has been reported for N-replete conditions (24, 25). Interestingly, while both O2 uptake in the light and net O2 production measured during steady-state photosynthesis remained high in pgrl1, no marked difference in gross or net O2 exchange were detected between flvB, pgrl1 flvB and their respective control lines (Figure 3 A-C and supplemental Fig. S2 and S3). Immunoblot analyses showed similar accumulation of NDA2, Cyt f, PsaD, and ATG8 proteins in the prgl1 flvB compared to controls under N replete and deficiency (Fig. 3D). Taken together, these data show that the absence of FLV-mediated PCEF in the pgrl1 background restores the effect of N limitation observed in control strains. We conclude that FLV-mediated PCEF by maintaining photosynthetic electron transfer reactions in pgrl1 allows net photosynthesis to be maintained at a high level during N deficiency.
A, Gross O2 uptake measured between 5-7min as shown by the supplemental figure 3B.
B, Gross O2 production measured between 5-7min as shown by the supplemental figure 3B
C, Net O2 production measured between 5-7min as shown by the supplemental figure 3B
D, Immunodetection of photosynthetic protein in pgrl1 flvB and their control lines.
O2 exchange rates were measured using a MIMS in the presence of [18O]-enriched O2. Net O2 evolution (green) was calculated as gross O2 production (blue) – gross O2 uptake (red). Total proteins were extracted from N-replete and 2 days N-deprived cells and samples were loaded at equal total proteins amounts as shown on Coomassie blue staining. Cells cultivated photoautotrophically with 1% CO2 in air under continuous light of 50 μmol photons m−2 s−1 were transferred into N-free media for 2 days prior to measurements or sampling for immunoblot. Shown are an average of at least three biological replicates ± SD. Asterisks represent statistically significant difference compared to the WTs (* p<0.05, ** p<0.01, *** p<0.001 and **** p<0.0001) using one-way ANOVA.
pgrl1 over accumulates TAGs under N deficiency
Emerging literature suggests that alterations of energy management pathways affect not only biomass production but also its composition (16, 17, 36, 41). Starch and TAGs are major forms of carbon storage in Chlamydomonas during N deficiency. Since pgrl1 mutant shows stronger net photosynthetic rate, we measured its ability to accumulate storage compounds. Starch accumulation was similar in pgrl1 and control lines under both N replete and deficiency (Fig. 4A; Supplemental Fig. S5, S7 A and B). In contrast, thin-layer chromatography (TLC) revealed twice more TAG accumulation in pgrl1 compared to the control lines under N deficiency whereas no difference was observed under N-replete condition (Fig. 4B; Supplemental Fig. S5 and S7 C and D). Lipid droplet imaging confirmed the observation of TAG quantification by TLC (Fig. 4C). To determine whether the increase in TAG is due to membrane lipid remodeling or to an improved de novo fatty acid biosynthesis in the chloroplast, we quantified total fatty acids by gas chromatography coupled to a mass spectrometer (GC-MS). We observed a net increase in total fatty acids in pgrl1, suggesting an increase in de novo fatty acid biosynthesis (Supplemental Fig. S6).
A, Starch amount.
B, TAG amount.
C, representative images of confocal microscopy observation of lipid droplet stained with Bodipy dye. Pseudo-colors were used: BODIPY-stained LDs in yellow, chlorophyll autofluorescence in red. DIC: differential interference contrast. Scale bar = 5 µm.
Cells cultivated photoautotrophically with 1% CO2 in the air under continuous light of 50 μmol photons m−2 s−1 were transferred into N-free media for 2 d prior to sampling for starch and TAG or confocal imaging. Asterisks represent statistically significant difference compared to the control 137AH (* p<0.05, ** p<0.01 and *** p<0.001) using one-way ANOVA.
We then evaluated whether the changes in photosynthesis and TAG amount affect biomass accumulation, we followed changes in cell volume for 6 d during N deficiency. Cell volume is used here as a proxy for biomass accumulation. The cell volume increased twice faster in the pgrl1 compared to the control lines (Supplemental Fig. 5A). We conclude from above experiments that the higher photosynthetic activity in pgrl1 results in increased biomass and TAG productivity under N deficiency.
We then analyzed starch and TAG amount in flvB as well as pgrl1 flvB double mutants before and under N deficiency (Supplemental Fig. S7). Similar to the observations on photosynthetic activity, the pgrl1 flvB double mutant accumulated similar amounts of TAG and starch as compared to control strains either in N replete or N limited conditions (Supplemental Fig. S7I-L). Intriguingly, under optimal growth but not under N deficiency, the flvB single mutants accumulated much higher amount of starch than WT (Supplemental Fig. S7E) and accumulated lower amounts of TAG under N deficiency (Supplemental Fig. S7H). These data indicate that modifying regulatory mechanisms of photosynthetic electron transfer reactions strongly affects carbon storage in both quality and quantity (Fig. 1 and 2).
We conclude that during phototrophic N-deficiency, photosynthetic electron flow mediated by FLVs allows to drive photosynthetic TAG accumulation while PGRL1 tends to constrain it, likely via an activation of photosynthetic control. We propose that both PGRL1 and FLVs, by having an antagonistic role in N-deprivation, manage the redox landscape, carbon storage and biomass production.
Discussion
The ability of microalgae to coordinate their energy production to meet the metabolic demand is crucial for their survival in a constantly fluctuating environment. Mechanisms involved in photosynthesis regulations have been abundantly studied in response to light or CO2 levels, but not much is known during nutrient deficiency when a massive re-orientation of metabolic pathways occur. In the absence of N, the major cellular energy sinks (cell division, protein biosynthesis and photosynthetic CO2 fixation) are restricted whereas the carbon and energy are stored as TAGs or starch. Moreover, the photosynthetic electron transfer (PET) is downregulated to match the demand and maintain cell viability. In this study, we demonstrated that PGRL1 is involved in the downregulation of PET during N deficiency. This down-regulation likely operates via the photosynthetic control mechanism (see below for details). Lack of PGRL1 resulted in sustained efficient photosynthetic activity after 2 d of N deficiency. We have further shown that the release of photosynthetic control in pgrl1 mutants resulted in higher FLVs-mediated PCEF. By channeling excess electrons toward O2, PCEF prevents PSI acceptor side-limitation, allows maintaining PET at high level and favors high TAG accumulation. A schematic diagram explaining the role of PGRL1 in the downregulation of photosynthesis and metabolic consequences of its absence during N deficiency is presented in Fig. 5.
We proposed that PGRL1-mediated CEF is involved in the downregulation of photosynthesis during N deficiency through the Cyt b6f-dependent photosynthetic control (black arrow). In the absence of PGRL1, FLVs-mediated PCEF is strongly induced to drive the photosynthetic electron transfer. As a result, excess photosynthetic electrons are diverted toward fatty acid and triacylglycerol biosynthesis (* out of the chloroplast). The red arrows show the pathways upregulated in the PGRL1-deficient strain compared to the WT (gray arrows).
PGRL1-involved CEF contributes to photosynthetic control during N deficiency
Unlike in higher plants (21, 42–44), the photosynthetic control mechanism has been poorly studied in microalgae. The term ‘photosynthetic control’ refers to mechanisms that restrict the reactions of PET during the steady-state photosynthesis mostly occur in response to environmental fluctuations (31, 44). The Cyt b6f is considered as the central hub for controlling PET between PSII to PSI and prevents the over reduction of P700 (44, 45). This is achieved either by slowing down the turnover rate (conductance) of the Cyt b6f triggered by the acidification of the thylakoid lumen (44) followed (or not) by the selective autophagy of Cyt b6f (39). In addition to Cyt b6f, the PGRL1-PGR5-dependent CEF was shown to contribute to the photosynthetic control in higher plants (21, 43, 46) and lately in microalgae (19, 47–50). However, a recent work in Chlamydomonas reported that PGRL1 only slightly contribute to the lumenal proton generation and that the pgrl1 displayed similar proton gradient to the WT during photoautotrophic N deficiency under CO2 limiting conditions (22). Our work shows that impairing the accumulation of the PGRL1 removes a bottleneck of photosynthetic electron flow during N deficiency under non-limiting CO2 conditions (Fig. 1 and 3; Supplemental Fig. S1 and S2). Similar observations were reported under sulphur deprivation where deletion of PGRL1 or PGR5 resulted in sustained linear electron flow toward H2 production in Chlamydomonas (19, 49). The contribution of PGRL1 in the photosynthetic control is likely through the acidification of lumen that will exert a backpressure on Cyt b6f conductance, slowing down the electron transfer toward plastocyanin. However, further works will be required to shed light on the exact participation of PGR5/PGRL1 in the photosynthetic control. Recently, PGR5 (which interacts with PGRL1 during the CEF (43) was shown to be required for efficient Q cycle through the reduction of the heme-ci in the b6f using reduced ferredoxin (50). The operation of the Q cycle is crucial for lumen acidification to trigger the onset of photosynthetic control (51–53). By doing so, PGRL1-PGR5 connects the stromal redox poise to the b6f-dependent photosynthetic control.
FLVs become predominant in the absence of PGRL1-mediated cyclic electron flow
So far, a few studies reported a compensation mechanism between PGRL1 and FLVs during algal adaptation to high light or low-CO2 conditions (14, 25). Under high light or low-CO2 conditions, the increased activity of FLVs in the PGRL1-deficient strain had no significant improvement on the net photosynthesis but rather FLVs serve either as a safety valve of excess electrons or to balance ATP/NAD(P)H by generating more pmf (14, 25). Similar compensation has also been observed in higher plants where orthologous expression of FLVs rescues the pgr5 mutants phenotype without further improvement photosynthesis (46, 54) although some level of increased biomass where measured in WT Arabidopsis expressing FLVs under light fluctuations thanks the protective role of FLVs in these conditions (55). In contrast, our results suggest that under N deficiency, FLVs and PGRL1 have an antagonistic role. Indeed, the increased activity of FLVs in the PGRL1-deficient strain resulted in higher net photosynthesis and TAG accumulation than in the WT (Fig. 3D and 4B), which shows that rather than compensating each other, PGRL1 is indirectly controlling the activity of FLVs by limiting the electron flow. Our observation of the sustained accumulation of Cyt b6f and increased accumulation of PSI in the PGRL1-deficient strain under N deficiency (Fig. 2C) could be explained by the strong activity of FLVs draining electrons from the photosynthetic chain thereby releasing the over-reduction pressure and preventing PSI from damage. However, we were expecting that removal of both PGRL1 and FLVs should have more severe consequences on cell physiology and metabolism. Instead, we observed similar photosynthetic activity and TAG biosynthesis in the pgrl1 flvB double mutants as their control strains (Fig. 3 and supplemental Fig. S7), i.e. the additional removal of FLVs suppressed the phenotype of PGRL1-deficient strain. The WT-like phenotype of pgrl1 flvB double mutants further supports our conclusion that PGRL1 and FLV are antagonist during N deficiency. A third pathway taking over could explain this WT-like phenotype of the double mutant pgrl1 flvB. The NDA2 pathway or chloroplast-mitochondria electron flow (CMEF) could be good candidates. NDA2 protein level was shown to increase during air photoautotrophic N deficiency and also the CEF rate in Chlamydomonas was decreased by 50% in nda2 mutants (22). However, we observed a similar protein level of NDA2 in the pgrl1 flvB as their control lines (Fig. 3C and Supplemental Fig. S4D). Nevertheless, the protein level might not always correlate with the activity and other regulatory mechanisms (e.g., phosphorylation or redox regulation) can modulate enzyme activity. However, the high rates of respiration in the dark in N-deprived WT and pgrl1 flvB double mutants (Fig. 1 and Supplemental Fig. S3 and S4A) is a good indication that mitochondria is strongly active and likely reflects a strong CMEF, which would take over CEF and PCEF in N-deprived conditions. Altogether, we conclude that FLVs act as a major driving force leading to the increased net photosynthesis in pgrl1 under N deficiency (Fig. 3).
Relationship between cellular redox landscape and carbon storage
A major biotechnological challenge in algal domestication for biofuel is the low lipid productivity. In Chlamydomonas and many other microalgae, starch and TAG massively accumulate but mostly under stress conditions in particular N deficiency when cell division stops and productivity is impaired. Considerable efforts have focused on the study of the molecular mechanisms behind the onset of reserve accumulation by monitoring omics responses to a stress (5, 56–59), or focused on specific steps of fatty acid and TAG biosynthesis, which have resulted in some limited improvement in productivity (60, 61). Improving productivity requires a better understanding of the crosstalk between photosynthetic carbon fixation, environmental signals and the redox balance, which all govern reserve accumulation (62–64). Here by studying mutants affected in CEF and PCEF, we explored the relationships between the cellular redox status and carbon storage. The increased accumulation of TAG but not starch observed in the PGRL1-deficient strain under N deficiency (Fig. 4 A and B) is due to an increase in de novo biosynthesis in the chloroplast as evidenced by the observed increase in total fatty acids (Supplemental Fig. S6). The increase in TAG amount is a consequence of the continuous production of NADPH and ATP through PCEF in the PGRL1-deficient strain under nitrogen deficiency. By alleviating the limitation of the acceptor side of PSI, PCEF released the over-reduction pressure of the photosynthetic chain, thereby sustaining the PET reactions. In line with our finding, the Chlamydomonas pgd1 mutants (e.g. Plastid galactoglycerolipid degradation 1) with reduced LEF rate are shown to produce less TAG under N starvation (11, 65). The report that the pgrl1 mutant made less TAG than WT under mixotrophic conditions (66) is not surprising. It is well known that the bioenergetics of Chlamydomonas under photoautotrophic conditions differ from mixotrophic conditions (67–69). The presence of acetate can drastically affect cellular bioenergetics level i.e. its uptake consumes ATP and its metabolism produces NADH, therefore further favoring oil synthesis (67, 68, 70, 71).
Moreover, in addition to higher TAG amount, we also observed higher biomass (cell volume) in the PGRL1-deficient strain during N deficiency (Supplemental Fig. 5A). This high biomass can be interpreted by the combined increased of photosynthetic activity (Fig. 1), TAG biosynthesis (Fig. 4) and reduced autophagy as indicated by the amount of ATG8 (Fig. 2). Here we show that removing the photosynthetic control mediated by PGRL1 allows the FLVs-mediated PCEF to massively take part in the energization of the de novo fatty acid biosynthesis pathway resulting in improved TAG biosynthesis and double TAG productivity. Taken together, our data highlight the importance of energetic balance on carbon reserve formation for biomass production and for altering biomass composition. Rewiring the alternative photosynthetic electron flows opens new possibilities in fine tuning algal productivity and the organic compounds they are producing.
Materials and methods
Growth conditions and strains
The pgrl1 from the background 137AH, its two complementing strains, flvB mutants from the background CC-4533 and the double mutants pgrl1 flvB were previously described (19, 24, 25). Cells were routinely cultivated in an incubation shaker (INFORS Multitron pro) maintained at 25°C, with 120 rpm shaking and constant illumination at 50 µmol m−2 s−1. Fluorescent tubes delivering white light enriched in red wavelength supplied lightings in the INFORS. All experiments were performed under photoautotrophic condition with minimum medium with or without nitrogen source (MM and MM-N) buffered with 20 µM MOPS at pH 7.2 in air enriched with 1% CO2. Due to cell aggregation (pgrl1) that prevent accurate cell counting, total cellular volume was measured using a Multisizer 4 Coulter counter (Beckman Coulter) and the different strains were diluted to reach a similar cellular concentration before N deficiency experiments.
Measurement of chlorophyll fluorescence using a Pulse Amplitude Modulation (PAM)
Chlorophyll fluorescence was measured using a PAM fluorimeter (Dual-PAM 100, Walz GmbH, Effeltrich, Germany) on the MIMS chamber as described in (40) using green actinic light (1250 µmol photon m-2 s -1, green LEDs). Red saturating flashes (8,000 µmol photons m−2 s −1, 600 ms) were delivered to measure the maximum fluorescence (FM) every 30 s (before and upon actinic light exposure). The maximum PSII quantum yield was calculated as Fv/Fm= (Fm-Fo)/Fm where F0 is the basal fluorescence obtained with the measuring light and Fm the fluorescence emitted after saturating pulse (72). PSII operating yield (ΦPSII) was calculated as ΦPSII= (FM’-Fs)/ FM’ with FM’ the fluorescence value after saturating pulse, Fs the stationary fluorescence during actinic light exposure.
O2 exchange measurement using Membrane Inlet Mass Spectrometry (MIMS)
O2 exchanges were measured in the presence of [18O]-enriched O2 using a water-jacketed, thermoregulated (25°C) reaction vessel coupled to a mass spectrometer (model Prima ΔB; Thermo Electronics) through a membrane inlet system (73). The cell suspension (1.5 mL) was placed in the reaction vessel and bicarbonate (10 mM final concentration) was added to reach a saturating CO2 concentration. One hundred microliters of [18O]-enriched O2 (99% 18O2 isotope content; Euriso-Top) was bubbled at the top of the suspension just before vessel closure and gas exchange measurements. O2 exchanges were measured during a 3-min period in the dark, then the suspension was illuminated at 1250 μmol photons m−2 s−1 for 5 min using green LEDs followed by 3-min in the dark. Isotopic O2 species [18O18O] (m/e = 36), [18O16O] (m/e = 34), and [16O16O] (m/e = 32) were monitored, and O2 exchange rates were determined (73). Argon gas was used to correct O2 exchange measured by the spectrometer as described in (73).
All others methods are described in SI Materials and Methods
Acknowledgments
O.D. thanks The French Atomic Energy and Alternative Energy Commission (CEA) for a PhD scholarship. G.P. and Y.L.-B. acknowledge the continuous financial support of CEA (LD-power, CO2Storage). A.B. acknowledges the support from the Carnegie Institution for Science. We thank Bertrand Legeret for maintaining the HelioBiotec lipidomics platform and thank Mallaury Cabanel for assistance in performing immunoblot. We also acknowledge the ZoOM microscopy facility.
Footnotes
Competing Interest Statement: There are no conflicts of interest.