Enhancing photosynthetic CO2 fixation by assembling metal-organic frameworks on Chlorella pyrenoidosa

The CO2 concentration at ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is crucial to improve photosynthetic efficiency for biomass yield. However, how to concentrate and transport atmospheric CO2 towards the Rubisco carboxylation is a big challenge. Herein, we report the self-assembly of metal-organic frameworks (MOFs) on the surface of the green alga Chlorella pyrenoidosa that can greatly enhance the photosynthetic carbon fixation. The chemical CO2 concentrating approach improves the apparent photo conversion efficiency to about 1.9 folds, which is up to 9.8% in ambient air from an intrinsic 5.1%. We find that the efficient carbon fixation lies in the conversion of the captured CO2 to the transportable HCO3− species at bio-organic interface. This work demonstrates a chemical approach of concentrating atmospheric CO2 for enhancing biomass yield of photosynthesis.

The effect of MOFs on algal growth is then investigated by following OD change at 750 nm. Based on the calibration curve shown on Fig. S9, it is concluded that MOFs induce a two-fold increase in biomass production (Fig. 3a). Clearly, measuring OD change is not sufficient to conclude to changes in biomass production. Indeed, from Fig. 2 b,c it seems highly probable that the diffusive properties of algal cells are changed upon absorption of MOFs. However, the calibration curve shown on Fig. S9 was only made on algal cells without MOFs, and absorption spectra shown on Fig. S5 show similar absorbance at 750 nm, which is rather surprising. What were cell densities in the experiments shown on Fig. S5 and how were the spectra normalized? Clearly, more direct determination of algal biomass must be performed in order to validate the effect of MOFs. Further, it is concluded that MOFs enhance the photosynthetic activity of algae. Again, in order to conclude that photosynthesis is enhanced in the presence of MOFs it is necessary to perform more direct measurements of photosynthesis by measuring either CO2 uptake or O2 production, the latter being generally used to characterize the photosynthetic activity of algal suspensions.
At a 50 ppm MOF concentration how many MOFs particles or molecules are present in the solution and how many are available for each single algal cell? Are the MOFs present in large excess? How many MOFs molecule/particle are available per cell when the biomass increases 8-fold as shown on Fig. 3 (at pH7). Again, the determination of the cell concentration is needed here, particularly at different time points during the growth of cells in the presence of MOFs.
The authors observe small particles at the surface of MOFs by TEM and propose that free enzymes diffusing out of the cell are adsorbed on MOFs. Then, based on the effect of the carbonic anhydrase inhibitor acetazolamide and on the measurement of carbonic anhydrase activity in the supernatant, it is concluded that the external carbonic anhydrase adsorbed on MOFs would supply bicarbonate to algal, the conversion of CO2 into HCO3-being catalyzed by the adsorbed carbonic anhydrase. Such a scenario should be supported by more direct evidence of the presence of carbonic anhydrase on the surface of MOFs (e.g. western blot, immunolocalization…). Further, the fact that the MOF-induced increase in biomass is similar at pH6 and pH8 (Fig. 3b), conditions in which HCO3-concentration at Discussion This part is only data analysis. It should be included more previous results for discussion of the other possibilities or related mechanism. I think that it is important that authors should check whether there is similar effect of MOF treatment on other species of microalgae.

Point-by-point response to the reviewers' comments
We thank the three reviewers for your efforts in reviewing our manuscript and providing constructive comments. All comments and suggested changes were carefully considered and addressed. All changes have been highlighted with yellow color in the revised version of the manuscript.

Reviewer #1
In this paper, Li and co-authors report on the enhancement of photosynthesis in the microalga C. pyrenoidosa by using a metal organic framework (MOF) which selfassembles at the surface of algae. This paper reports interesting observations related to a stimulating effect of MOFs on biomass production, but would need much better characterization of the stimulating effect to conclude to a stimulating effect on algal photosynthesis.

Reply:
Thank you for your positive comments and professional suggestions on our work.

Comment 1
The absorbance spectra of the Chlorella suspension are rather strange, the chlorophyll peak is not well-resolved and the general trend is an increase of absorbance at higher wavelengths, which is not in line with what is generally observed with algal suspensions where a decrease of absorbance at higher wavelength is observed. This may indicate the presence of diffusive materials (like algal debris) in algal samples.

Reply：
Thank you for your kind advices. We have carefully checked the absorbance spectra of the C. pyrenoidosa suspension and found that the unnormal results were caused by an instrumental error and cuvatte polution. After eliminating the instrumental faults, we obtained reliable results of the UV-Vis spectra. Figure R1 shows that UV-Vis absorption spectra of the MOF/Algae hybrid system. The chlorophyll peak is characterized at the wavelength of about 680 nm. After self-assembly, the absorption spectrum of the hybrid is composed of the spectra both MOFs and the C. pyrenoidosa. We have updated the figure in the revised manuscript.

Comment 2
A better characterization of algal suspensions in the presence of MOFs should be performed, for instance by using flow cytometry, in order to determine the number and size of cells and free particles present in the suspensions.

Reply:
Thank you for your advice. Size distribution of algal and MOF/Algae cells are measured by flow cytometry. Figure R2 shows flow cytometric scatter plot and FSC-A (forward scattering-area) distribution of the MOF/Algae system. The size distribution of algal cell in the suspension is uniform ( Figure R2a). After self-assembly of algae with MOFs, we find that the main size distribution and number of MOF/Algae cells are the same as that of the bare lagal cells ( Figure R2b). It's noted that the size of single MOF particle is about 300 nm ( Figure 2a). Therefore, self-assembly of MOFs on the algal surface did not change the size of the cell and make them aggragated obviously in the suspension. The results indicate that MOFs in algal suspension efficiently bind on the surface of C. pyrenoidosa cell.
We have revised the related contents as follows: "Algal suspensions in the presence of MOFs are characterized by using flow cytometry to determine the number and size of cells ( Supplementary Fig. 5). The size distribution of algal cell in the suspension is uniform (Supplementary Fig. 5a). After self-assembly of algae with MOFs, we find that the main size distribution and number of MOF/Algae cells are the same as that of the bare algal cells (Supplementary Fig. 5b). Therefore, self-assembly of MOFs on the algal surface can not change the size of the cell and make them aggregated in the suspension. The results indicate that MOFs in algal suspension bind on the surface of C. pyrenoidosa cell."

Comment 3
Based on absorbance spectra, the authors argue that MOFs protect PSII from photodamage by UV and do not impair harvesting properties (lines 84-85). This sole experiment is clearly not sufficient to support such a statement. Additional characterization of photosynthetic properties should be carried out to determine changes in light harvesting properties, such as measuring light saturation curves of photosynthetic O2 production of algal samples and algal-MOF samples of similar cell densities. Moreover, a possible UV protection by MOFs should be demonstrated by using an experimental protocol in which algal samples are subjected to UV stress.

Reply:
We have carried out the experiments of photosynthetic O2 production of the algal samples under saturation light irradiation. As shown in Figure R3, the O2 evolution rate of the bare algal sample decreases gradually, while the rate of the MOF/Algae sample keeps constant. About of the effect on UV stress, We measured the intrinsic electron transfer rate (ETR) of photosystem II (PSII) by a chlorophyll fluorimeter to characterize the photoinhibition or photodamage of PSII. Figure R4 shows that the effect of UV stress on the ETR of PSII in the algae and MOF/Algae samples versus light intensities.
Before irradiation with UV light, the ETR of PSII in the algae is the same as that the MOF/Algae sample. But after 30 min 10 μE m -2 s -1 UV light irradiation, they show the difference with light intensity from 750 μE m -2 s -1 to 2500 μE m -2 s -1 , and the ETR of PSII in the MOF/Algae sample performs higher than that in the bare algae. The results indicate the MOFs self-assembled on the surface of algal cell can block the UV light to a certain extent.
Thus, summarize comments 1and 3, we have revised the related contents as follows: "UV-Vis absorption spectra of the MOF/Algae hybrid system is characterized ( Supplementary Fig. 6). After self-assembly, the absorption spectrum of the hybrid is composed of both MOFs and bare alga. The light shade of MOFs on the cell surface at the wavelength of less than 400 nm is in favor of protecting photosystem II against UV light. To investigate the photoprotection of photosystem II, we carry out the experiments of photosynthetic O2 production of the algal samples under saturation light irradiation ( Supplementary Fig. 7). The O2 evolution rate of the bare algal sample decreases gradually, while the rate of the MOF/Algae sample keeps constant. And the intrinsic electron transfer rate (ETR) of photosystem II (PSII) by a chlorophyll fluorimeter to characterize the photoinhibition or photodamage of PSII ( Supplementary   Fig. 8). Before irradiation with UV light, the ETR of PSII in the algae is the same as that the MOF/Algae sample. But after 30 min 10 μE m -2 s -1 UV light irradiation, they show the difference with light intensity from 750 μE m -2 s -1 to 2500 μE m -2 s -1 , and the ETR of PSII in the MOF/Algae sample performs higher than that in the bare algae.
These results indicate the MOFs self-assembled on the surface of algal cell can protect PSII against photodamage by the UV light to a certain extent." The experimental methods have added in the revised supporting information file. Supplementary Fig. 7). Kinetics of photosynthetic oxygen evolution and average oxygen evolution rates under 1000 μE m -2 s -1 irradiation (λ > 600 nm) of C. pyrenoidosa in the absence and in the presence of MOF. Error bars represent the standard deviation of three experimental results. Supplementary Fig. 8). The plots of electron transfer rates of PSII versus light intensities of C. pyrenoidosa in the absence and in the presence of MOF before and after 30 min 10 μE m -2 s -1 UV irradiation ( λ = 254 nm). Error bars represent the standard deviation of three experimental results.

Comment 3
The effect of MOFs on algal growth is then investigated by following OD change at 750 nm. Based on the calibration curve shown on Fig. S9, it is concluded that MOFs induce a two-fold increase in biomass production (Fig. 3a). Clearly, measuring OD change is not sufficient to conclude to changes in biomass production. Indeed, from Fig.   2 b,c it seems highly probable that the diffusive properties of algal cells are changed upon absorption of MOFs. However, the calibration curve shown on Fig. S9 was only made on algal cells without MOFs, and absorption spectra shown on Fig. S5 show similar absorbance at 750 nm, which is rather surprising. What were cell densities in the experiments shown on Fig. S5 and how were the spectra normalized? Clearly, more direct determination of algal biomass must be performed in order to validate the effect of MOFs.

Reply:
It's conformed that the MOFs boost the biomass prudction. Indeed, using OD change to quantify the amount of biomass is inaccurate or imprecise in the presence of MOFs.
Due to the difference of absorbance spectra at 750 nm, there is an effect of MOFs on the amount of dry cell weight with the OD change. According your suggestion, we have updated the calibration curve of MOF/Algae samples. Figure R5 shows the calibration curve of the dry cell weight and algal cell number versus the optical density at 750 nm of C. pyrenoidosa in the absence and in the presence of MOF. The cell numbers per OD750 that is equal to 35.5 in the presence of MOFs are less than that ratio of 40.9 in the absence of MOF in 0.04 mm 2 square. Therefore, based on the accurate relationship, we calculate that the average biomass growth rate of MOF/Algae is 0.25 g L -1 day -1 , which is slightly less than the quantity of the algae alone. Therefore, the biomass growth rate of MOF/Algae is about 1.9 times as that of the bare algae.
We have added the calibration curves as Figure S12 in the supproting information and reivsed all the data related the biomass amount and the photosynthetic efficiency in the revised manuscript. Figure R5 (Supplementary Fig.12). The calibration curves of (a) algal cell number

Comment 4
Further, it is concluded that MOFs enhance the photosynthetic activity of algae. Again, in order to conclude that photosynthesis is enhanced in the presence of MOFs it is necessary to perform more direct measurements of photosynthesis by measuring either CO2 uptake or O2 production, the latter being generally used to characterize the photosynthetic activity of algal suspensions.

Reply:
We agree that O2 evolution activity is often used for researching photosynthesis. O2 production of the MOF/Algae system is measured by Clark-type oxyxgen electrode.

Comment 4
At a 50 ppm MOF concentration how many MOFs particles or molecules are present in the solution and how many are available for each single algal cell? Are the MOFs present in large excess? How many MOFs molecule/particle are available per cell when the biomass increases 8-fold as shown on Fig. 3 (at pH 7). Again, the determination of the cell concentration is needed here, particularly at different time points during the growth of cells in the presence of MOFs.

Reply:
There are about 4.  Figure 3 and main text of the revised manuscript.

Comment 5
The authors observe small particles at the surface of MOFs by TEM and propose that free enzymes diffusing out of the cell are adsorbed on MOFs. Then, based on the effect of the carbonic anhydrase inhibitor acetazolamide and on the measurement of carbonic anhydrase activity in the supernatant, it is concluded that the external carbonic anhydrase adsorbed on MOFs would supply bicarbonate to algal, the conversion of CO2 into HCO3being catalyzed by the adsorbed carbonic anhydrase. Such a scenario should be supported by more direct evidence of the presence of carbonic anhydrase on the surface of MOFs (e.g. western blot, immunolocalization…).

Reply:
Acocroding your suggestion, we have tried an antibody recognizing Carbonic anhydrase II rat monoclonal antibody (Catalog No. AG1271, Beyotime, China), but no band was detected. It's difficult to find the available antibody for algae-derived CAs.
Carbonic anhydrase is a kind of important enzymes that catalyze reversible conversion between CO2 and HCO3though hydration and dehydration. There are various carbonic anhydrase enzymes in the photosynthetic cells, which play the critical roles in the overall processes from CO2 capturing, transporting and concentrating to finall fixation by Rubisco enzyme. Determination of enzymatic activity is a simple and valid method to confirm the carbonic anhydrase in the photosynthesis research ( The Plant Journal (1983)). In our work, we verify external carbonic anhydrase (eCA) and its activity by conducting inhibition experiments (Fig. 4b) and testing CO2 hydration with water into HCO3and protons ( Fig. 4c and d). An important result is that the CO2 hydration rate of eCA in the presence of MOF is higher than that of free eCA in the supernatant. Based on the results and TEM imagings, we depict the proposed scenario of CO2 capture and concentrating.

Comments 6
Further, the fact that the MOF-induced increase in biomass is similar at pH 6 and pH 8 ( Fig. 3b), conditions in which HCO3concentration at the equilibrium with air are extremely different, is difficult to explain in the framework of the proposed scenario.

Reply:
Normarlly, we agree with you on that the HCO3concentrations at the equilibrium are different at pH 6 and pH 8 in aqueous solution. Since pKa1 of carbonic acid is about 6.4, CO2 is the major form at pH 6 and HCO3at pH 8 in algal suspension respectively.
However, in our case that MOFs self-assembled on the surface of cell, the proposed scenario is two sequential processes of CO2 concentraing at the bio-inorganic interface: (1) the capture of CO2 by MOF, and (2) the conversion of CO2 to HCO3by extracellular carbonic anhydrase and OH -. At pH 6, the former is fast while the latter is slow due to acidic environment. While at pH 8, the former is slow due to the lack of CO2 to be captured by MOF while the latter is fast. Therefore, their biomass increases are similar.
More minor concerns:

Comment 6
I do not quite well understand what information is expected from the pH drift measurements (Fig. 3c) since the authors use a BG11 medium buffered with 20mM

Reply:
The value of pH drift indicates the nitrogen assimilation rate of C. pyrenoidosa which produces hydroxide when using nitrate as inorganic nitrogen source (Annu. Rev. Plant Biol. 56 (1): 99-131 (2005)). Since the inorganic nitrogen source is NO3 -, OHis inevitably accumulated during the assimilation of NO3to amino acids, which can lead to pH value up to 10 and hampers algal growth. The chemical equation which reflects the conversion of major elements during the cultivation of C. pyrenoidosa in BG11 medium is listed below. Typical cultivation of C. pyrenoidosa in photobioreacter solves the problem by elevating CO2 concentration (e.g. 2%) in airflow to resist the OHaccumulation. In our system, atmospheric CO2 can't stop the pH drift to high pH, so the addition of pH buffer solution is necessary to maintain the normal cultivation.

Comment 7
Many experimental details are missing, particularly in supplemental Figures.

Reply:
Thank you for your kind advice. We have carefully added the details in the revised manuscript and supporting information files with yellow highlight.

Comment 8
Rubisco determination performed on Fig. S14 is not satisfactory. A western blot should be performed to ensure the shown band is due to Rubisco. Also, the total amount of proteins loaded on the gel should be indicated.

Reply:
Thank you for your kind advices. We identified the protein by immunoblot with an anti-Rubisco monoclonal antibody (Catalog No. AG5359, Beyotime, China). A specific band was detected at the expected molecular weight corresponding to the large subunit of Rubisco ( Figure R7). Therefore, we confirmed that the band dectected in Fig

Reviewer #2 (Remarks to the Author):
This is a very novel study on the environmental effects of MOF materials. It reported for the first time that MOF could adsorb CO2 and transport it to microalgae, which resulted in higher photosynthetic efficiency. The phenomenon is interesting and provides a new approach to concentrate and transport atmospheric CO2 towards to the Rubisco carboxylation. However, there are several issues to address.
Thank you for your postive comments.

Comment 1
The chlorophyll contents of C. pyrenoidosa in the presence/absence of MOFs should be measured.

Reply:
The chlorophyll contents of C. pyrenoidosa are measured. Figure R8 shows the contents of photosynthetic pigments of C. pyrenoidosa in the presence/absence of MOFs. The contents of chlorophyll a, chlorophyll b and carotenoids of C. pyrenoidosa are the same whether in the presence of MOFs or not. It indicates that biomass increasing has no effect on the amount of photosystem II and photosystem I proteins.
We have added the reuslts in the revised manusctipt as follow: "The contents of chlorophyll a, chlorophyll b and carotenoids of C. pyrenoidosa are the same whether in the presence of MOFs or not. Biomass increasing has no effect on the amount of photosystem II and photosystem I proteins ( Supplementary Fig. 14

Comment 2
Is there any possibility to measure the HCO3contents in solution and in C. pyrenoidosa?
HCO3content change is the key issue of this study.

Reply:
We agree with you that HCO3as the main form of CO2 is crucial to understand the mechanism of CO2 fixation in the work. The HCO3contents in solution and in C.
pyrenoidosa is possible to be measured by Raman spectroscopy. And we tried to use the method to meassure the content change of the HCO3 -, but it's difficut to obtain viable data under the rather low concentration in solution and especially in C.
pyrenoidosa. More sensitive detection method need to be developed for further use. Therefore, We have measured the contents of dissolved inorganic carbon (DIC, including CO2, HCO3and CO3 2-). Figure R9 shows the DIC in the media of MOF/Algae and bare algae after two-day cultivation. DIC in medium of the MOF/Algae system is slight higher than that of the bare with the initial pH from 6 to 8.
The DIC increase is derived from the MOFs capturing CO2.
We have added the reuslts in the revised manusctipt as follow: "Meanwhile, it's noted that DIC in medium of the MOF/Algae system is slightly higher than that of bare algae with the initial pH from 6 to 8. The DIC increase is derived from the MOFs capturing CO2 (Supplementary Fig. 19)" Figure R9 (Supplementary Fig. 19). The contents of dissolved inorganic carbon (DIC, including CO2, HCO3and CO3 2-) in medium of C. pyrenoidosa in the absence and presence of MOF under different initial pH value before (Blank, only medium without algal cells) and after two-day cultivation (Algae and MOF/Algae). Error bars represent the standard deviation of three experimental results.

Comment 3
If MOF enhanced the CO2 adsorption and generate more HCO3 -, the pH should be lower.
Why Alga/MOF groups have higher pH than Alga groups in Fig. 3.

Reply:
The value of pH drift in Fig.3c is caused by the nitrogen assimilation of C. pyrenoidosa which produces hydroxide when using nitrate as inorganic nitrogen source. The chemical equation which reflects the conversion of major elements during the cultivation of C. pyrenoidosa in BG11 medium is listed below. Since the inorganic nitrogen source is NO3 -, OHis inevitably accumulated during the assimilation of NO3to synthesize amino acids. The pH drift rate can also indicate the biomass accumulation rate. Because MOF/Algae grows faster than bare algae lead the more OHaccumulation, a higher pH of MOF/Algae groups than that of algae groups. 8CO2+8H2O+2NaNO3 = 2C4H7O1.5N+2NaOH+12.5O2

Comment 4
So many MOF particles were found on the alga surface in Fig. 2. Would they damage the cell wall/membrane? Why MOF particles were not found in TEM observation in

Reply:
We observes that MOF particles are assembled on the surface of C. pyrenoidosa ( Figure. 2c and 2d,). The opposite electrostatic potentials between MOF and C. pyrenoidosa tend to drive them self-assembly by van der Waals interaction (Figure 1d). Due to abundant carboxyl group on the surface of MOFs, it is biocompatible for binding with the cell wall though weak interaction without damage. In addition, algal suspensions in the presence of MOFs are characterized by using flow cytometry and we find that selfassembly of MOFs on the algal surface can not change the size of the cell and make them aggregated in the suspension (Revised version of Supplementary Fig.5).
The sample is the bare C. pyrenoidosa in Fig. S16, which is used for observing the intrisic cell structure of C. pyrenoidosa.

Comment 5
Fig. S10 does not have growth curves. Please check the figure or the caption.

Reply:
We must apologize for the mistake. We have correct the figure and the caption of "ligand concentration" in Fig. S10a, and "biomass growth" in Fig. S10b.

Comment 6
The text should be checked and improved. For example, "Here in" should be "Herein" and "shoule" should be "should".

Reply:
Thank you for your kind advice. We have corrected these spelling mistakes and examined the sentences in our manuscript carefully.

Reviewer #3 (Remarks to the Author):
Authors found that self-assembly of metal-organic frameworks (MOFs) on one of microalgae can enhance its photosynthetic efficiency by about 2 folds by the conversion of the captured CO2 to the transportable HCO3species at bio-organic interface. This work provided evidence for an artificial approach of concentrating atmospheric CO2 for enhancing biomass yield of photosynthesis in microalgae. The work is novel and is of significance to the photosynthesis research for improvement of photosynthesis in microalgae although the evidence to support the conclusion is a little weak at present stage. Suggestion and comments are as follows.
Thank you for your positive comments and professional suggestions on our work.

Title
The title is too broad because the work focused on a green alga.

Reply:
Thank you for your kind advice. We have changes the title as "Enhancing photosynthetic efficiency by assembling metal-organic frameworks on Chlorella

Introduction
Given that the research focused on a green alga, it is better to introduce the CCM background of green alga.

Reply:
Thank you for your kind advice. We have extended the introduction of the intrinsic CCM in green alga. (see page 2, lines 10-26 in the revised manuscript).

Comment 4
Last paragraph: Please provide the full name of C. pyrenoidosa when first used the name.

Reply:
Thank you for your kind advice. We have added the full name of C. pyrenoidosa as Chlorella pyrenoidosa.

Results
Line 94: "It indicates that the free enzymes are diffused out of the cell and adsorbed on the MOF". What kinds of enzymes are? The sentence is better to be removed to the explanation of Figure 4.

Reply:
The enyzme absorbed on the surface of MOFs is one of kind carbonate anhyderase. We observe the samll particles on the MOFs from the SEM data of Fig. 2d and infer that the particles may be a kind of free enzyme. Subsequently, we verify external carbonic anhydrase, eCA and its activity by conducting inhibition experiments (Fig. 4b) and testing CO2 hydration with water into HCO3and protons ( Fig. 4c and d (1983)). Therefore, these free enzymes are extracellular carbonic anhydrase.

Comment 6
Fig.S14. I suggest that authors identify the protein by immunoblot analysis included the loading control.

Reply:
We identified the protein by immunoblot with an anti-Rubisco monoclonal antibody (Catalog No. AG5359, Beyotime, China). A specific band was detected at the expected molecular weight corresponding to the large subunit of Rubisco ( Figure R11). Therefore, we confirmed that the band dectected in Fig. S14 is Rubisco.

Discussion
This part is only data analysis. It should be included more previous results for discussion of the other possibilities or related mechanism.

Reply:
We have rewritten this part as you suggested in the new version of the manuscript as

Reply：
Thank you for your kind advice. We carry out similar experiments on a model green alga, Chlamydomonas reinhardtii. Figure R12 shows the growth curves of

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): In this modified version the authors have correctly taken into account some of my concerns and significantly improved the manuscript. However, I still think that the characterization of the photosynthetic activity of MOFS is too scarce and is not sufficient to support the proposed mechanisms for improved CO2 assimilation based on a MOF-dependent CO2 concentration.
The novel Fig. S7 (and Figure R6) shows a slight increase in the O2 production rate in algae when in the presence of MOF, which is clearly not sufficient to conclude to an improved photosynthetic activity by surface-bound MOF which would play a role of CO2 concentrator. Indeed, if the system acts as a CO2 concentrator, as claimed by the authors, the improvement of photosynthetic activity by MOF should depend on the CO2 availability, a much stronger effect should be observed under limiting CO2 than under non-limiting CO2 concentration. In other words, the apparent affinity of the photosynthetic O2 production for Ci should be affected as it is the case when algae harboring or not a CCM are compared (for instance when comparing high CO2 and low CO2 grown cells). Clearly, the authors should perform experiments at different Ci levels (and different pH see below), as it is generally done in CCM studies.
The authors claim that the function of MOF and this is coupled to that of the intrinsic CCM of C. pyrenoidosa. However, the interaction of MOF with the CCM has not been investigated. To conclude so, the authors should compare the effect of MOF on high-CO2 and low-CO2 grown C. pyrenoidosa or C.reinhardtii cells. If MOF really acts as a CO2 concentrator, they should observe a much stronger effect on algae not harboring a CCM (high CO2 grown) as compared to algae harboring a CCM (low CO2 grown cells).
The proposed scenario assumes the eCA when absorbed on MOF would allow a faster conversion of CO2 absorbed on MOF into HCO3-and then a faster carbon assimilation. However, the evidence for such a scenario is rather scarce. First the CA activity absorbed on MOF was not experimentally determined, it is only based on the fact that eCA activity decrease in the external medium in the presence of MOF. To explain that the increase in biomass is similar at pH 6 and pH 8 (reply to my comment #6) the authors argue that in the first case MOF would only concentrate CO2 and that at pH 8 MOF would have a dual functon(increase CO2 and convert it to HCO3-). In order to support such a scenario, the apparent affinity of net O2 evolution for Ci should be measured by performing measurements at different pH and different Ci concentrations. The effect of the eCA inhibitor AZA would be also probably needed to support the proposed scenario which would predict a stronger effect of AZA on MOF stimulation at pH 8 than at pH6.

Reviewer #2 (Remarks to the Author):
My concerns are well addressed. It could be published as it is.

Reviewer #3 (Remarks to the Author):
The revised version is much improved. The evidence to support authors' conclusion is more solid. I am satisfied with this revision and have no further suggestions and comments.

Point-by-point response to the reviewers' comments
Reviewer #1 (Remarks to the Author): In this modified version the authors have correctly taken into account some of my concerns and significantly improved the manuscript. However, I still think that the characterization of the photosynthetic activity of MOFs is too scarce and is not sufficient to support the proposed mechanisms for improved CO2 assimilation based on a MOF-dependent CO2 concentration.

Reply：
Thank you for your efforts in reviewing our manuscript and conferring on us professional comments. We have taken your comments seriously and performed all the experiments carefully to provide more evidences for supporting the proposed scenario.
All changes have been highlighted with yellow color in the revised version of the manuscript.

Comment 1
The novel Fig. S7 (and Figure R6) shows a slight increase in the O2 production rate in algae when in the presence of MOF, which is clearly not sufficient to conclude to an improved photosynthetic activity by surface-bound MOF which would play a role of CO2 concentrator. Indeed, if the system acts as a CO2 concentrator, as claimed by the authors, the improvement of photosynthetic activity by MOF should depend on the CO2 availability, a much stronger effect should be observed under limiting CO2 than under non-limiting CO2 concentration. In other words, the apparent affinity of the photosynthetic O2 production for Ci should be affected as it is the case when algae harboring or not a CCM are compared (for instance when comparing high CO2 and low CO2 grown cells). Clearly, the authors should perform experiments at different Ci levels (and different pH see below), as it is generally done in CCM studies.

Reply：
To systematic investigation of CCM, we measured the net photosynthetic oxygen evolution rates at various Ci (in the form of CO2 and HCO3 -, respectively) of C.

Comment 2
The authors claim that the function of MOF and this is coupled to that of the intrinsic CCM of C. pyrenoidosa. However, the interaction of MOF with the CCM has not been investigated. To conclude so, the authors should compare the effect of MOF on high-CO2 and low-CO2 grown C. pyrenoidosa or C.reinhardtii cells. If MOF really acts as a CO2 concentrator, they should observe a much stronger effect on algae not harboring a CCM (high CO2 grown) as compared to algae harboring a CCM (low CO2 grown cells).

Reply：
From the results of Figure R1a and c, we observe that the MOF as a CO2 concentrator coupled with the intrinsic CCM of C. pyrenoidosa enable the CO2 affinity increased by 68%, which is close to the HCO3affinities of K1/2 around 10 μM. Furthermore, as shown in Figure

Comment 3
The proposed scenario assumes the eCA when absorbed on MOF would allow a faster conversion of CO2 absorbed on MOF into HCO3and then a faster carbon assimilation.
However, the evidence for such a scenario is rather scarce. First the CA activity absorbed on MOF was not experimentally determined, it is only based on the fact that eCA activity decrease in the external medium in the presence of MOF.

Reply：
The activities of CA enzyme adsorbed on MOF particles and in medium are assayed by a colorimetric assay kit (Solarbio, China). It is analyzed according to the hydrolysis of p-nitrophenyl acetate (p-NPA) to p-nitrophenol (p-NP) catalyzed by CA. The method has been added in the revised supporting information. The activity of CA in medium is 0.6 mM h -1 and the activity of CA enzyme adsorbed on MOF is 0.2 mM h -1 ( Figure R3).
The adsorption amount of CA proteins on the surface of MOFs is 37% of that in medium.
Therefore, after the activities normalized by protein amount, the specific activity of CA on MOFs is similar with that in medium, indicating that eCA adsorbed on MOF enables efficient conversion of MOF-captured CO2 into HCO3 -( Figure R3b).

Comment 4
To explain that the increase in biomass is similar at pH 6 and pH 8 (reply to my comment #6) the authors argue that in the first case MOF would only concentrate CO2 and that at pH 8 MOF would have a dual functon (increase CO2 and convert it to HCO3 -). In order to support such a scenario, the apparent affinity of net O2 evolution for Ci should be measured by performing measurements at different pH and different Ci concentrations.

Reply：
We have measured the apparent affinity of net O2 evolution for CO2 concentration grown under low CO2 at different pH ( Figure R4). We observe that the K1/2 for CO2 of

Comment 5
The effect of the eCA inhibitor AZA would be also probably needed to support the proposed scenario which would predict a stronger effect of AZA on MOF stimulation at pH 8 than at pH 6.

Reply：
We have measured the biomass growth of C. pyrenoidosa (Algae) and MOF/C.

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): In this new version the authors have addressed my concerns by performing additional experiments to better characterize the photosynthetic activity of algae associated or not with MOFs. The manuscript has been significantly improved, but there are still major improvements to be made.
It is shown (Fig. 4C) that the apparent affinity of photosynthesis for Ci is somehow increased by the presence of MOFs when algae are grown in air but not when they are grown in high CO2. It is however difficult to evaluate the effect on the figure since the scale of the X-axis of the graph of Fig. S21 does not allow to distinguish lower CO2 values (the scale should be extended between 0 and 0.5 to better evaluate the effect visually). The way errors bars shown on Fig. 4C and Fig. R1e,f is not explained and in the absence of statistical analysis, it is difficult to be convinced that the difference observed in K1/2 is significant. Moreover, these affinity determinations were performed using a light intensity of 1200 µmol photons m-2 s-1 which is quite far from the light intensity used for the algal cultivation (50 µmol photons m-2 s-1). Altogether this makes the evidence quite weak to explain the observed effect on biomass productivity in terms of affinity. I would suggest the authors to discuss these points both in the result and in the discussion sections.
The authors also observe an increase intracellular accumulation of Rubisco in MOFs-treated algae, but do not really explain what could be the origin of such an effect and what could be its consequence in term of biomass productivity. Is there any evidence in the literature that Rubisco can limit photosynthesis under low CO2 conditions? How is this additional Rubisco organized within the cell, for instance does it contribute to form larger pyrenoids? It is not clear how the kinetic properties of Rubisco could be modified. Please discuss this point in relation to the current knowledge of Rubisco activase for instance.
The first paragraph of the discussion has no link with the current work. In connection with my concerns above, I would suggest that authors replace this section with a critical discussion of their own data. How can be the increase in biomass explained? What part is due to an increased affinity for CO2 and what part could be due to other effects line an increase in the Rubisco content, or a change in kinetics properties of Rubisco? How this relates to limitations previously reported in the literature?

More minor concern
The authors refer to their approach as a non-genetic concentrating CO2 approach which would be more or less opposed to a genetic approach. I find this somewhat misleading. Indeed, the effect of MOFs is only observed in the presence of a genetic CCM since no improvement is observed in high CO2 grown algae. Also, the effect of MOFs is not to concentrate CO2, but rather to facilitate the functioning of the "genetic" CCM, likely by supplying more substrate to bicarbonate transporters. I would suggest the authors to define their system with a positive rather than negative term, using for instance "chemical approach" rather than as a "non-genetic approach".

Point-by-point response to the reviewers' comments
Reviewer #1 (Remarks to the Author): In this new version the authors have addressed my concerns by performing additional experiments to better characterize the photosynthetic activity of algae associated or not with MOFs. The manuscript has been significantly improved, but there are still major improvements to be made.

Reply：
Thank you for your efforts in reviewing our manuscript and conferring on us professional comments to improve the manuscirpt. We have taken your comments seriously to provide more evidences for supporting the proposed scenario. All changes have been highlighted with yellow color in the revised version of the manuscript.

Comment 1
It is shown (Fig. 4C) that the apparent affinity of photosynthesis for Ci is somehow

Comment 2
The authors also observe an increase intracellular accumulation of Rubisco in MOFs- In the liturature, Akiho Yokota et al [1] reported that the Rubisco content in C.
pyrenoidosa was affected by the CO2 concentration during cultivation. C. pyrenoidosa grown on high CO2 concentration (1% CO2) had higher Rubisco expression level of 1.5 mg (mg Chl) -1 than that of 0.8 mg (mg Chl) -1 in air-grown algae. Meanwhile, by comparing the Rubisco content and the photosynthetic CO2 fixation rate, they found that nearly full activity of Rubisco must be needed during photosynthesis in C.
pyrenoidosa and other green algae, which is different from that in C3 plants. It implies the strong dependence of biomass growth rate on the Rubisco content in C. pyrenoidosa.
Borkhsenious et al [2] applied immunogold particles (with anti-Rubisco antibody) to localize the distribution of Rubisco in green alga Chlamydomonas reinhardtii cells at different CO2 concentrations. When 5% CO2-grown algae were adapted to air-grown condition, the proportion of Rubisco in pyrenoid increased from 50% to 90% within 5 hours (other Rubisco was dissolved in the chloroplast stroma), but the Rubisco content in pyrenoid remained unchanged. The additional Rubisco is possibly distributed in the chloroplast stroma rather than assembled in the pyrenoid.
It indicates that the CO2 concentration mainly affects the stroma-localized Rubisco content as shown in our work. Meanwhile, the starch sheath got thicker with the CO2 concentration decreased, which indicated the induction of CCM to form the diffusive barrier for preventing the leakage of CO2 from the pyrenoid [3] .

Comment 3
It is not clear how the kinetic properties of Rubisco could be modified. Please discuss this point in relation to the current knowledge of Rubisco activase for instance.

Reply:
Thank you for your comments. The kinetic properties of Rubisco vary among different photosynthetic organisms, but remain steady on a certain species [4] . Rubisco can be inactivated by mistaken generation of tight-binding sugar phosphates on the active sites, while Rubisco activase (Rca) regenerates the Rubisco catalytic site by release of the inhibitors after structual remodeling [5] . Mirkko Flecken et al [6] reported that Rca is positioned over the Rubisco catalytic site under repair and pulls the N-terminal tail of the large Rubisco subunit (RbcL) into the hexamer pore. Simultaneous displacement of the C terminus of the adjacent RbcL opens the catalytic site for inhibitor release. For the close-packed Rubisco in the pyrenoid of green alga, it seems not to be accessible for Rca like stromal Rubisco. Freeman Rosenzweig et al [7] observed the liquid-like behavior of the pyrenoid matrix in Chlamydomonas reinhardtii cell, enables the facile transportation of relatively insufficient Rca to enough Rubisco for efficient carbon fixation in the pyrenoid of green algae.
Based on the litureatures, we can reasonably infer that the enhanced Rubisco activity of MOF-treated C. pyrenoidosa is due to the increase of Rubisco content, but not the increase of the kinetic properties of Rubisco. According your suggestion, we have discussed the point in the revised manuscirpt.

Comment 4
The first paragraph of the discussion has no link with the current work. In connection with my concerns above, I would suggest that authors replace this section with a critical discussion of their own data. How can be the increase in biomass explained? What part is due to an increased affinity for CO2 and what part could be due to other effects line an increase in the Rubisco content, or a change in kinetics properties of Rubisco? How this relates to limitations previously reported in the literature?

Reply:
Thank you for your precious suggestion and opinion. We have rewritten the first paragraph of the discussion part in the revised manuscript. We critically discussed our data and cited the opinions in several literatures to explain the proposed scenario.
"In this study, we find that the affinity for CO2 of MOF/C. pyrenoidosa cell is stronger than that of bare C. pyrenoidosa cell ( Supplementary Fig. 27). But the presence of MOF particles doesn't remarkably have influence on the affinity of cells grown under high CO2 (Supplementary Fig. 27). Because C. pyrenoidosa cells grown in high CO2 is not harboring the intrinsic CCM [8] . The complete CCM is vital to transfer CO2 toward Rubisco enzyme for accelerating the rate of CO2 fixation. Moreover, we find that the expression level of Rubisco protein is upregulated in MOF/C. pyrenoidosa cells (Fig.   4e). It was reported that the Rubisco content in C. pyrenoidosa was affected by the CO2 concentration during cultivation and nearly full activity of Rubisco must be needed during photosynthesis in C. pyrenoidosa and other green algae [1] . It was observed that Rubisco protein is more likely to stay in the chloroplast stroma than to be close-packed in the pyrenoid when CO2 concentration is elevated. And the additional Rubisco was distributed in the chloroplast stroma rather than to be organized in the pyrenoid [2] . The kinetic properties of Rubisco vary among different photosynthetic organisms, but remain steady on a certain species [4] . It implies the biomass growth rate strong dependence on the Rubisco content in C. pyrenoidosa. Thus, the synergetic effect of the functional MOF and the intrinsic CCM in C. pyrenoidosa cells enables a strong affinity for CO2 and the Rubisco content in C. pyrenoidosa for accelerating CO2 fixation."

Comment 5
Finally

Reply:
Thank you for your kind advice. Our previous photosynthetic efficiency calculation is not accurate and strict. Chemical energy fixed in biomass should not be estimated by the value of glucose molecule, the assessment of incident light energy should be accurate according to the naure of the light source. We carefully learn the literature you recommand to us and revised our calculation as below: Apparent photo conversion efficiency calculations. The apparent photo conversion efficiency (APCE) values in this work were calculated according to Wagner et al [9] . We For the calculation of EB, the calorific value (HB) of 23.4 KJ g −1 for C. pyrenoidosa [10] was taken to calculate the chemical energy stored in biomass. During two days cultivation, EB can be calculated as: Values were averaged over three independent experiments.
As a result, the apparent photo conversion efficiency (APCE) values are 5.1 % for the control group (bare C. pyrenoidosa) and 9.8 % for MOF/Algae hybrids. Supplementary Fig. 15). The light spectrum of the light source for the cultivation of C. pyrenoidosa.

REVIEWERS' COMMENTS
Reviewer #1 (Remarks to the Author): I appreciate that the authors have taken my concerns seriously. Prior to acceptance, I think that the interpretation of data related to the increase in Rubisco levels in MOF-treated algae should be more careful.
The conclusion of the authors (lines 206-207) stating that "the acceleration of biomass growth is mainly due to upregulation of the expression level of the Rubisco…" is quite confusing. I understand from their previous experiments that the increase productivity at limiting CO2 would result from a higher affinity for Ci due to the effect of MOF. The observed increase in the Rubisco content/activity may be a consequence, but in no case do the authors establish a link between this increase and the increased biomass productivity. Both effects may contribute, but the fact that the Vmax values measured in MOF-treated algae are not increased (see Figs S22 & S27) suggests that the increase in Rubisco content does not improve the maximal capacity of CO2 fixation. Therefore, I would suggest the authors to moderate their conclusion by mentioning that Vmax values are not modified in MOFtreated algae, and write something more neutral like "such an increase in Rubisco content may result from an increased internal Ci concentration in MOF-treated algae, and may contribute to reaching high CO2 fixation rates and high biomass productivity. Note however, that no increase in the maximal CO2 fixation rate was observed non-limiting CO2 (Fig S22, S27), indicating that the contribution of increased Rubisco content is probably limited in MOF-treated algae".
Line 62: please write "apparent photo conversion efficiency" Line 64: please replace "on" by "by" the excreted carbonic anhydrase… Line 209: please replace "stronger" by "higher" and correct the English in the whole paragraph.

Point-by-point response to the reviewers' comments
Reviewer #1 (Remarks to the Author): I appreciate that the authors have taken my concerns seriously. Prior to acceptance, I think that the interpretation of data related to the increase in Rubisco levels in MOFtreated algae should be more careful.

Reply：
Thank you for your efforts in reviewing our manuscript and conferring on us professional comments to improve the manuscirpt. We have taken your comments seriously to provide more evidences for supporting the proposed scenario. All changes have been highlighted with yellow color in the revised version of the manuscript.

Comment 1
The conclusion of the authors (lines 206-207) stating that "the acceleration of biomass growth is mainly due to upregulation of the expression level of the Rubisco…" is quite confusing. I understand from their previous experiments that the increase productivity at limiting CO2 would result from a higher affinity for Ci due to the effect of MOF. The observed increase in the Rubisco content/activity may be a consequence, but in no case do the authors establish a link between this increase and the increased biomass productivity. Both effects may contribute, but the fact that the Vmax values measured in MOF-treated algae are not increased (see Figs S22 & S27) suggests that the increase in Rubisco content does not improve the maximal capacity of CO2 fixation. Therefore, I would suggest the authors to moderate their conclusion by mentioning that Vmax values are not modified in MOF-treated algae, and write something more neutral like "such an increase in Rubisco content may result from an increased internal Ci concentration in MOF-treated algae, and may contribute to reaching high CO2 fixation rates and high biomass productivity. Note however, that no increase in the maximal CO2 fixation rate was observed non-limiting CO2 (Fig S22, S27), indicating that the contribution of increased Rubisco content is probably limited in MOF-treated algae".

Reply：
Thank you for your comments. Since the increased Rubisco content in MOF-treated algae results from the increased internal Ci concentration, the latter one should be the basic reason for the increased biomass growth rate in MOF-treated algae. We have replaced the sentence "the acceleration of biomass growth is mainly due to upregulation of the expression level of the Rubisco…" by more neutral explanation like "such an increase in Rubisco content may result from an increased internal Ci concentration in MOF-treated algae, and may contribute to reaching high CO2 fixation rates and high biomass productivity. However, no increase in the maximal O2 evolution rate was observed under non-limiting CO2 ( Supplementary Fig. S22, S27), indicating that the increased Rubisco content in MOF-treated algae mainly contributes to dark reactions".

Minor questions
Line 62: please write "apparent photo conversion efficiency"; Line 64: please replace "on" by "by" the excreted carbonic anhydrase; Line 209: please replace "stronger" by "higher" and correct the English in the whole paragraph.
Reply：Thank you for your kind advice. We have corrected all the mistaken words, and examined the text carefully.