In-silico molecular investigation of Nannochloropsis microalgae cellulose synthase under salinity conditions and in-vitro evaluation of the proportionate effects on cellulose production.

In-silico molecular investigation of Nannochloropsis microalgae cellulose synthase under salinity conditions and in-vitro evaluation of the proportionate effects on cellulose production - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice 3 Biotech . 2025 May 21;15(6):180. doi: 10.1007/s13205-025-04329-y In-silico molecular investigation of Nannochloropsis microalgae cellulose synthase under salinity conditions and in-vitro evaluation of the proportionate effects on cellulose production Seyed Mohammad Hasan Haghayeghi 1 , Maryam Azimzadeh Irani 1, ✉ , Hossein Askari 1, ✉ , Madihe Sadat Rasa 1 , Zeinab Shariatmadari 1 Author information Article notes Copyright and License information 1 Faculty of Life Sciences and Biotechnology, Shahid Beheshti University, Tehran, Iran ✉ Corresponding author. Received 2024 Oct 22; Accepted 2025 Apr 22; Issue date 2025 Jun. © King Abdulaziz City for Science and Technology 2025. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. PMC Copyright notice PMCID: PMC12095769  PMID: 40417659 Abstract Nannochloropsis is a microalgae with more than substantially 60–70% cellulose in its cell wall, making it a potential candidate for nanocellulose sustainable production. This study examined the effects of salts in seawater and their role on Nannochloropsis gaditana and Nannochloropsis oculata cellulose synthase activity using In-silico and In-vitro approaches for the first time. Deep-learning-based AlphaFold2 predicted model was selected as the most reliable 3D structure. Molecular docking results revealed that none of the selected ligands occupied the binding site predicted for the native substrate of the enzyme, uridine-diphosphate. To validate the In-silico results, experiments were conducted to investigate the impact of salinity stress (NaCl, NaNO 3 and NaHCO 3 ) on the cell growth and cellulose production. The assessment tools included a UV–visible spectrophotometer and a hemocytometer, with a modified Jayme–Wise method used for cellulose extraction. The results indicated that the following concentrations of 0.443 mol/L, 0.457 mol/L, and 0.469 mol/L of NaCl, 0.072 mol/L, 0.077 mol/L, and 0.082 mol/L of NaNO3, 0.0021 mol/L, 0.0022 mol/L, and 0.0023 mol/L of NaHCO 3 did not lower the growth rate nor the cellulose yield of N. oculata and notable enhancement in growth was observed in cultures supplemented with 0.0023 mol/L NaHCO 3 . Furthermore, when NaCl (0.457 mol/L and 0.469 mol/L), NaNO 3 (0.082 mol/L) and NaHCO 3 (0.0022 mol/L and 0.0023 mol/L) were individually introduced to the culture, cellulose yield increased up to five times compared to the control group. Supplementary Information The online version contains supplementary material available at 10.1007/s13205-025-04329-y. Keywords: Cellulose synthase, Nannochloropsis , Molecular modeling, AlphaFold2, Cellulose extraction Introduction Microalgae play a significant role in bioeconomy due to their diverse applications in industry. These microorganisms offer a promising solution to food supply concerns and can serve as a sustainable source of protein as they grow in harsh saline environments(Tang, et al. 2020 ; Wang et al. 2021 ). The favorable properties which make microalgae a good substitution for nanocellulose production include lack of lignin, high alignment, high crystallinity and surface area(Ross et al. 2021 ). Microalgal nanocellulose is obtained from microalgae after the production of valuable materials, such as lipids and proteins. The cellulose present in microalgae cell wall could be brought back to the cycle and used to produce valuable products, such as nanocellulose (Lee et al. 2018 ). Cellulose-based nanomaterials are eco-friendly and are becoming familiar as potential substitutes for common porous materials such as silica and carbon which mostly present challenges including brittleness, high toxicity, and high production costs(Abraham et al. 2017 ). Cellulosic nanomaterials have also numerous applications as porous agents for biomedical and biophysical purposes thanks to possessing properties, such as (1) low density, (2) high surface area, and (3) fatigue resistance(Abraham et al. 2017 ). The global nanocellulose market was measured to be 0.4 billion USD in 2022 with a compound annual growth rate (CAGR) of 21.9% (Market et al. xxxx). Since the purpose of choosing cellulose synthase in this study was to enhance the production of nanocellulose in Nannochloropsis , the economic importance of nanocellulose should be noted. Nanocellulose has several medical and industrial applications, such as smart packaging agents, contamination sensors, and vehicle compartments due to its impermeability against moisture, antibacterial properties as well as low weight and cost (Anne 2014 ; Dhali et al. 2021 ; Cowie et al. 2014 ). Nannochloropsis gaditana is a microalga related to Eustigmatophyceae. N. gaditana has been reported to have 75% cellulose in the cell wall composition and 25% cellulose content in dry weight (Baba et al. 2016 ). There are two types of nanocellulose produced in microalgae: (1) cellulose nanofibers (CNFs) and (2) cellulose nanocrystals (CNCs) (Zanchetta et al. 2021 ). The CNF obtained from N. gadtiana is suggested to be an achievable candidate as the process of its retrieval from microalgal extracted cellulose needs fewer mechanical and chemical treatment steps. Nanocellulose is obtained from a few microalgae species, one of them being Nannochloropsis. As Nannochloropsis cell wall is not comprised of lignin and hemicellulose, an easy-to-use extraction method (TEMPO-mediated oxidation) was operated which consists of fewer steps in comparison with nanocellulose production from terrestrial sources such as wood. This can be helpful in reduction of environmental consequences of nanocellulose mass production processes 4 . Cellulose is synthesized in cellulose synthase complexes (CSCs) which are located in cell membrane. CSCs directly take part in cellulose synthesis in plants. Salt stress has been reported to impact cellulose synthesis in the plants, potentially resulting in the activity of many metabolic pathways (Colin et al. 2023 ). Cellulose is synthesized in cellulose synthase complexes which are located in cell membrane. Cellulose synthase A catalytic subunit 1 (CESA1), which is the catalytic subunit of cellulose synthase (Fig. 1 ), belongs to glycosyltransferase family 2 which is one of the largest enzyme groups and consists of more than 70 families (Farrokhi et al. 2006 ). CESA has two major roles: (A) synthesizing cellulose and (B) exporting the synthesized cellulose over its transmembrane channel (Purushotham et al. 2020 ). Fig. 1. Open in a new tab Cellulose synthase enzyme overview. This enzyme consists of two subunits. The catalytic ( A ) subunit is represented by orange color N. gaditana CESA1 is reported to have great similarity to bacterial cellulose synthase (BcsA) which produces amorphous cellulose (Scholz et al. 2014 ). Algal microfibrils have a high amount of orientation, whereas wood cellulose mostly consists of randomly oriented microfibrils (Mihranyan 2011 ). Terminal complexes (TCs) are plasma membrane proteins which move in plasma membrane and produce cellulose microfibrils along their way. Algae are reported to have terminal complexes that have correlation with their microfibril dimensions (Roberts and Roberts 2004 ). Harsh environmental stresses such as salinity and drought have different effects on plant biomass production (Endler et al. 2015 ). Furthermore, algae have the ability to grow in very diverse situations. Salinity stress influences many biochemical and physiological mechanisms which take place in microalgae growth (Endler et al. 2015 ). Accordingly in this study, the primary salt ions in seawater composition which is the native habitat of Nannochloropsis , were induced as salinity stressors on N. gaditana by computational methods. Salinity stress in Nannochloropsis , Chlamydomonas sp.JSC4, Desmodesmus abundans and Dunaliella tertiolecta has proven to enhance lipid production (Yun et al. 2019 ). In addition, previous studies have reported a boost in carbohydrate content in Chlorella sp. and Scenedesmus sp. upon salinity stress (Zanchetta et al. 2021 ). In addition, previous studies have employed molecular docking in microalgae to explore the molecular interactions in anti-diabetic (Munawaroh, et al. 2020 ), beta-lactamase inhibiting (Pestana-Nobles 2022 ), and carotenoid biosynthesis (Narang et al. 2021 ) processes. Several studies have validated the molecular docking analysis conducted on microalgal molecules by assessing the growth rate of organisms by Optical Density (OD) or cell density. The response of alkyl-OPFRs on Chlorella pyrenoidosa (Chu et al. 2021 ) and interaction of BDE47 with D1 proteins of Chlorella pyrenoidosa (Hu 2023 ) were analyzed by molecular docking analysis, while the growth inhibition was studied by cell counting using hemocytometer. Furthermore, the growth rate of microalgae has been measured by OD in several studies. Lutein interaction with proteins involved in quorum sensing mechanism in Chlorella pyrenoidosa (Sampathkumar et al. 2019 ) and interaction of Chromium (Cr) with membrane proteins in Cyclotella sp. (Li, et al. 2021 ) were revealed by molecular docking analysis, with the cells growth verified by OD. There are various abiotic factors, including changes in pH, temperature, salinity, light stress and nutrients, influencing the cellular conditions of Nannochloropsis . The goal of this study was to generally validate the results of molecular docking involving salt ions on the cellulose synthase enzyme. Given the importance of light as a vital factor in the photosynthetic organisms life and considering a decline in protein and sugar content of Nannochloropsis during high light intensity (Wang and Jia 2020 ), it was crucial to maintain the optimal situation for the microalgal cells. As Normal light intensity for microalgae optimal growth ranges within 26–400 µmol m −2 s −1 (Singh and Singh 2015 ; Maltsev et al. 2021 ), a light supply with white fluorescent light (74 µmol m −2 s −1 ) was illuminated with a 16/8 light/dark cycle for 12 days to support the growth of Nannochloropsis , maintaining cells at a normal growth stage with no impact on cell conditions from light intensity. In this study, the molecular interactions of salt ions and N. gaditana and N. oculata cellulose synthase were investigated to recognize the enzyme functionality and adaptability under salinity stress conditions for the first time. Several molecular modeling methods such as AlphaFold2, Robetta, SWISS-MODEL, and I-TASSER were employed to predict CESA1 structure. Molecular docking was used to scrutinize cellulose synthase binding sites against stressor agents available in seawater. Moreover, hydrophobicity which is a crucial factor in protein folding was used to survey the properties of the active sites. Accordingly, experimental validations were carried out to uncover the relationship between salinity stress, microalgae growth, and cellulose production by cultivating N. oculata in culture media with different salt concentrations over a 12-day period. Throughout this duration, optical density and cell density were regularly measured using a UV–visible spectrophotometer and a hemocytometer to evaluate the growth rate. On the final day, cellulose yield was assessed using a modified Jayme–Wise method. The results of both computational and experimental analyses demonstrated that the overall activity of cellulose synthase is not interrupted during salinity stress conditions. Putting together, computational and experimental investigations of Nannochloropsis cellulose synthase activity in salinity stress condition reveal that Nannochloropsis could be used as a strong biorefinery candidate to produce nanocellulose under harsh salinity conditions. In conclusion, various approaches including genetic and metabolic engineering of Nannochloropsis can be employed to boost the production of bio-products such as essential lipids, pharmaceutical compounds, and nutraceuticals in addition to nanocellulose production, making Nannochloropsis a promising bio-factory candidate. Materials and methods Computational methods Molecular modeling of Nannochloropsis gaditana cellulose synthase The protein sequence of N. gaditana cellulose synthase was obtained from National Center for Biotechnology Information (NCBI) (Sayers et al. 2022 ) with AZIL01001061.1 accession code. Structural modeling of the cellulose synthase was performed by two different approaches including deep-learning-based with AlphaFold2 and Robetta and homology modeling via SWISS-MODEL and Iterative Threading ASSEmbly Refinement (I-TASSER) to achieve the most reasonable results. AlphaFold2 AlphaFold2 is a novel protein 3D structure prediction method which uses machine learning approaches to combine biological and physical information of protein structure. Multiple sequence alignment information is used to design the deep-learning algorithm in this method. pTM (Predicted Template Modeling Score) which estimates the intrinsic accuracy of the model is used to evaluate the confidence of the predicted model (Jumper, et al. 2021 ; Evans, et al. 2021 ). ColabFold which is an online software based on Google Colaboratory was utilized to run AlphaFold2, ColabFold benefits from MMseqs2 (Many-against–Many-sequence searching) which executes homology searching 40–60 times faster than AlphaFold2. Furthermore, ColabFold operates deep neural networks by accessing powerful Graphics Processing Unit (GPUs) from Google Colaboratory (Mirdita, et al. 2022 ). ColabFold was employed by entering the sequence retrieved from NCBI, with the num_relax setting at 0 and no specific template, aiming for faster performance than AlphaFold2. The structure was not relaxed as it was later conducted using Chimera. ‘mmseqs2_uniref-env’ option was selected for multiple sequence alignments, iterating through two databases: UniRef100 and a database of environmental sequences using MMseqs2. The ‘num_recycles’ was set on 3, considering cellulose synthase had known homologs and did not require further recycling. PAE (Predicted Aligned Error) is a scoring system generated by AlphaFold2 which illustrates the predicted positional error at residue x according to the actual structures aligned on residue y. PAEs are calculated in Angstroms. If the PAE values are low, it indicates that AlphaFold2 has predicted the models with high-confidence and accurate orientations (Varadi et al. 2022 ). PAE plots for top five models generated by AlphaFold2 reveal that most regions have high-confidence (more than 80%) and just the regions in red and white have low confidence (less than 60%). plDDT (local distance difference test) scoring standard estimates the per-residue confidence on a scale from 0 to 100. Residues with pLDDT ≥ 90 have very high-confidence, residues between 90 and 70 are confident, residues between 70 and 50 have low confidence, and residues below 50 are the least confident 34 . Swiss-model SWISS-MODEL (Waterhouse et al. 2018 ) is a protein 3D structure prediction server which operates by homology modeling. This server uses data from PDB templates to discover the closest template to the sequence. Hence, N. gaditana cellulose synthase structure was predicted by selecting bacterial cellulose synthase catalytic subunit: A (PDB ID: 4P02) as the template. SWISS-MODEL benefits from ProMod3 (Studer et al. 2021 ) new modeling engine and QMEANDisCo (Benkert et al. 2011 ) scoring function which is a new version of Qualitative Model Energy ANalysis (QMEAN) (Benkert et al. 2011 ). SWISS-MODEL website (was employed for modeling the protein 3D structure. Cellulose synthase sequence retrieved from NCBI was uploaded for creating the model. Robetta Robetta is a server which predicts protein 3D structure using either genomic data or protein sequence. Robetta mainly relies on RoseTTAFold for the prediction of protein 3D structures. Furthermore, comparative modeling and de novo protein prediction methods are employed in this approach to generate the protein 3D structures(Kim et al. 2004 ). Robetta web-server was used for submitting the cellulose synthase sequence. I-TASSER I-TASSER is a server for protein 3D structure modeling which performs the prediction in a four-step process: (1) template identification, (2) structure assembly simulation, (3) model refinement and (4) structure-based function annotation. I-TASSER web-server was used for entering the cellulose synthase sequence. The server generates five models, which the top-ranked model selected for analysis (Yang et al. 2015 ). Molecular modeling of Nannochloropsis oculata cellulose synthase N. oculata genome assembly (GCA_004335455.1) with accession code PRJNA65107 was retrieved from NCBI. A BLASTn search was conducted between the nucleotide sequence of N. Oculata genome and N. gaditana cellulose synthase ( AZIL01001061.1 ). The resulting nucleotide sequence (accession code: AFGM01001913.1 ) was translated using the Expasy Translate tool (Gasteiger et al. 2003 ). The query included six Open Reading Frames (ORFs) in both forward and reverse strands. All of the ORFs then were utilized in a BLASTp search against N. gaditana cellulose synthase, with the sixth ORF showing the highest identity with N. gaditana cellulose synthase. This amino acid sequence was then employed to predict N. oculata cellulose synthase 3D structure using AlphaFold2. Structure refinement and energy minimization were performed using Chimera. Finally, a Ramachandran plot was generated to evaluate the stability of the N. oculata cellulose synthase for further docking analysis. Docking of the salt stressors Molecular docking of the B, Ca, Cl, Fe, HCO 3 , K, Mg, Na, NH 4 , NO 3 , P, PO 4 , SiO 2 , and SO 4 was performed by Hex (Macindoe et al. 2010 ) which applies Fast Fourier-transform (FFT) correlation (Padhorny et al. 2016 ) to dock salt stressors and ions on N. gaditana and N. oculata cellulose synthase. FFT algorithm considers all rotations in 6D space. Hex benefits from the computational power of a GPU developed using the Common Unified Device Architecture (CUDA), resulting in high-speed calculations. Molecular docking analysis was carried out to compare the interaction between salt stressors and cellulose synthase by measuring the interaction free energy (E-total), which is the interaction free energy (kJ/mol). The analysis of protein–ligand interactions was performed by visualizing the complex structure and the amino acids that existed in the 5 Angstroms distance of the ligand using PyMOL (Schrödinger 2015 ) software. Cellulose synthase (receptor) was shown in cartoon mode, the ligands were visualized in sphere mode, and the interactions were visualized by sticks (Figs. 9 , 10 , 11 , 12 and 14 ). The complete structure of cellulose synthase was chosen as receptor for molecular docking analysis. Fig. 9. Open in a new tab Region I is the binding site of eleven ligands: B, Ca, Cl, Fe, HCO 3 , Mg, Na, NH 4 , P, PO 4 and SO 4 ( A ). Around 5 Å amino acids of B ( B ). Around 5 Å amino acids of Cl ( C ). Around 5 Å amino acids of Fe ( D ). Around 5 Å amino acids of Ca ( E ). Around 5 Å amino acids of HCO 3 ( F ). Around 5 Å amino acids of Mg ( G ). Around 5 Å amino acids of Na ( H ). Around 5 Å amino acids of PO 4 ( I ). Around 5 Å amino acids of NH 4 ( J ). Around 5 Å amino acids of P ( K ). Around 5 Å amino acids of SO 4 ( L ) Fig. 10. Open in a new tab Region II is the binding site of two regions( A ): around 5 Å amino acids of SiO 2 ( B ) and around 5 Å amino acids of NO 3 ( C ) Fig. 11. Open in a new tab Region III is the binding site of UDP which is the native substrate of cellulose synthase. Overview of the third region ( A ). The amino acids located in Around 5 Å of UDP ( B ) Fig. 12. Open in a new tab Molecular docking of saltwater ions on both N. gaditana (Dark blue) and N. oculata (Cornflower blue) cellulose synthase. The results demonstrated that six of the ligands (B, Cl, Fe, K, NH4, P) occupied the same binding site (Region I) in N. oculata as they did in N. gaditana . ( A) Exhibits the Region I binding site on the N. gaditana cellulose synthase. ( B–G) Highlights the docked positions of B, Cl, Fe, K, NH4, and P, respectively. The orange balls represent the ligands position on N. gaditana cellulose synthase and the yellow balls present ligands positions on N. oculata cellulose synthase Fig. 14. Open in a new tab Amino acids located in around 5 Å each ligand visualized by hydrophobicity scale. The most hydrophilic amino acids were visualized in blue and the most hydrophobic amino acids were visualized in yellow. Amino acids in around 5 Å of B ( A ). Amino acids in around 5 Å of Ca ( B ). Amino acids in around 5 Å of Cl ( C ). Amino acids in around 5 Å of K ( D ). Amino acids in around 5 Å of Mg ( E ). Amino acids in around 5 Å of Fe ( F ). Amino acids in around 5 Å of Na ( G ). Amino acids in around 5 Å of HCO 3 ( H ). Amino acids in around 5 Å of NH 4 ( I ). Amino acids in around 5 Å of SiO 2 ( J ). Amino acids in around 5 Å of NO 3 ( K ). Amino acids in around 5 Å of PO 4 ( L ). Amino acids in around 5 Å of P ( M ). Amino acids in around 5 Å of SO 4 ( N ). Blue shows the most hydrophilicity and yellow shows the most hydrophobicity 3D structural analyses and active site engineering PyMOL was used for the structural alignment, Root Mean Square Deviation (RMSD) (Reva et al. 1998 ) calculations, and specifying the amino acids of the active site. RMSD is a method used to compare the predicted and experimentally generated structures. RMSD value computes the average deviation between atoms in two structures so the validity of the predicted structure increases as the RMSD value becomes smaller. Studies demonstrate that alignment of two structures is reliable if the RMSD value is lower than 5 Angstroms (Munawaroh, et al. 2020 ). UCSF ChimeraX (Pettersen et al. 2021 ) was used for analyzing the predicted 3D structure from ColabFold and examining the pLDDT of the top five ranked models, while UCSF Chimera 1.16 (Pettersen et al. 2004 ) was employed for N. gaditana cellulose synthase energy minimization. Energy minimization was carried out by Assisted Model Building with Energy Refinement (AMBER ff14SB) (Maier et al. 2015 ), 2000 Steepest descent steps, and 100 conjugate gradient steps. BIOVIA Discovery Studio (Discovery Studio et al. 2021 ) visualizer was utilized to present the hydrophobicity of amino acids and to show the ligands in the active site. Kyte and Doolittle hydrophobicity scale was employed to evaluate hydrophobicity of active sites; in this scale the most hydrophilic amino acid score number was −4.5 (arginine) and the most hydrophobic was 4.5 (isoleucine), with the other amino acids falling in between (Kyte and Doolittle 1982 ). ProCheck (Laskowski et al. 1993 ) was employed to draw the Ramachandran plot of predicted 3D structures. Ramachandran plot visualizes the general geometry of the molecule and position of amino acids towards each other. Experimental methods Cultivation of Nannochloropsis oculata Due to our limited access to the N. gaditana strain, it was decided to cultivate N. oculata in the experiments . This was because this strain has the highest similarity to N. gaditana in terms of cellulose content in its cell wall (Jeong, et al. 2017 ). N. oculata was supplied by Qeshm Microalgae Refinery Company. F/2 medium which is a slight modification of the F medium (GUILLARD RR, R. JH. 1962 ) was prepared with the following composition: artificial seawater, 8.82 × 10 –4 mol/L of NaNO 3 , 3.62 × 10 –5 mol/L of NaH 2 PO 4 · H 2 O, 1.06 × 10 –4 mol/L of Na 2 SiO 3 · 9H 2 O, trace metals (1.17 × 10 –5 mol/L of FeCl 3 ·6H 2 O, 1.17 × 10 –5 mol/L of Na 2 EDTA·2H 2 O, 3.93 × 10 –8 mol/L of CuSO 4 ·5H 2 O, 2.60 × 10 –8 mol/L of Na 2 MoO 4 ·2H 2 O, 7.65 × 10 –8 mol/L of ZnSO 4 ·7H 2 O, 4.20 × 10 –8 mol/L of CoCl 2 ·6H 2 O and 9.10 × 10 –7 mol/L of MnCl 2 ·4H 2 O) and vitamins (2.96 × 10 –7 mol/L of Thiamine · HCL (vitamin B 1 ), 2.05 × 10 –9 mol/L of Biotin (vitamin H) and 3.69 × 10 –10 mol/L of Cyanocobalamin (vitamin B 12 )) (Andersen 2004 ). Then, each sample was meticulously prepared by combining 200 mL of F/2 medium, 40 mL of N. oculata solution, and 20 mL of varied concentrations of NaCl, NaNO 3 and NaHCO 3 , as specified in Table 1 . The final concentrations of the samples were calculated by combining 20 mL salt solutions, the salt quantities in the artificial seawater used to prepare the medium (including 0.423 mol/L NaCl, 0.00214 mol/L NaHCO 3 ) (Henson et al. 2019 ), and a minimal amount of NaNO 3 (8.82 × 10 –4 mol/L) for creating the F/2 medium (Table 2 ). The concentrations were chosen based on levels typically present in desalination brine (Omerspahic et al. 2022 ). Each sample was assigned a number with the name of the salt for traceability purposes. The control group received 20 mL of distilled water (dH 2 O) instead of the mentioned solutions. Samples were placed in a germinator (Noor Sanat Tajhiz, Karaj, Iran) at 24 °C and were manually agitated daily for 1 h to facilitate aeration. White fluorescent light (74 µmolm −2 s −1 ) was illuminated with a 16/8 h light/dark cycle for 12 days. All tests were carried out in duplicate (Elbarmelgy et al. 2021 ). To monitor the microalgal growth, the optical density and cell density of the samples were measured using a UV–visible spectrophotometer at 680 nm and a hemocytometer under a microscope, respectively, (Reyimu and Özçimen 2017 ) at different timepoints (days 1, 5, 7, 9 and 12). Table 1. Concentrations of NaCl, NaNO 3 , and NaHCO 3 treatment groups. Each group is identified by the name of the salt followed by a number Treatment Concentration (mol/L or M) NaCl-1 0.685 mol/L NaCl-2 0.856 mol/L NaCl-3 1.026 mol/L NaNO 3 −1 0.941 mol/L NaNO 3 −2 1.000 mol/L NaNO 3 −3 1.059 mol/L NaHCO 3 −1 0.002 mol/L NaHCO 3 −2 0.003 mol/L NaHCO 3 −3 0.004 mol/L Open in a new tab Table 2. Final concentrations of NaCl, NaNO 3 , and NaHCO 3 treatment groups Treatment Final concentration (mol/L or M) NaCl-1 0.443 mol/L NaCl-2 0.457 mol/L NaCl-3 0.469 mol/L NaNO 3 −1 0.0724 mol/L NaNO 3 −2 0.0769 mol/L NaNO 3 −3 0.0815 mol/L NaHCO 3 −1 0.0021 mol/L NaHCO 3 −2 0.0022 mol/L NaHCO 3 −3 0.0023 mol/L Open in a new tab Each group is identified by the name of the salt followed by a number Cellulose extraction Cellulose extraction was performed using a modified Jayme–Wise method.(Baba et al. 2016 ) On day 12, 100 ml of each sample was centrifuged and washed several times to separate the microalgae biomass from the remaining culture medium. The pellets were dried using an oven for 24 h at 80 °C. The cellulose extraction process was initiated with the removal of wax from the algae through the treatment of microalgae with a mixture of toluene and ethanol. Microalgae dried powder was mixed with toluene and ethanol at a ratio of 68:32 (v/v) using a magnetic stirrer for 24 h. Subsequently, the resulting mixture was filtered and the obtained residue was treated with 2% or 4% NaOH for 2 h at 80 °C, aimed at eliminating hemicelluloses. The residue was bleached by stirring it with sodium hypochlorite (6% or 10%) at a pH of 4.8 for 2 h at 70 °C. The cellulose pellet obtained through centrifugation was then dried using an oven and weighed (Baba et al. 2016 ). Statistical analysis We conducted repeated measures analysis of variance (ANOVA) ( p < 0.05) with least significance difference (LSD) post hoc analysis and one-way ANOVA with LSD post hoc analysis using IBM SPSS Statistics 27. Data visualization was performed using the matplotlib python package (Hunter 2007 ). A summary of the computational and experimental approaches employed in this study is illustrated in Fig. 2 . Homology modeling approaches were performed by SWISS-MODEL and I-TASSER to predict protein 3D structure of N. gaditana cellulose synthase. AI structure prediction methods were also employed by AlphaFold2 and Robetta, as they represent the most advanced techniques for protein 3D structure. Molecular docking analysis of the ligands on cellulose synthase was conducted using HEX. Biovia Discovery Studio, Pymol, and ChimeraX were used for molecular and atomistic visualizations as well as interpretation of molecular docking results. Ultimately, N. oculata was cultivated in F/2 medium and salinity stress induced. The growth rate of N. oculata was evaluated using optical density and cell density measurements obtained by a UV–visible spectrophotometer plus a hemocytometer. Cellulose extraction was done using the modified Jayme–Wise method. Statistical analysis was conducted using ANOVA and data were visualized using Matplotlib Python package (Fig. 2 ). Fig. 2. Open in a new tab Schematic representation of the computational and experimental methods used in this study for investigation of Nannochloropsis cellulose synthase activity and cellulose production in salinity stress Results Nannochloropsis gaditana cellulose synthase 3D structure modeling by AlphaFold2, SWISS-MODEL, Robetta, and I-TASSER Four servers were employed to obtain the most suitable 3D structure of cellulose synthase: AlphaFold2 and Robetta utilizing deep-learning methods plus SWISS-MODEL and I-TASSER employing homology modeling methods for protein 3D structure prediction. AlphaFold2 Cellulose synthase of N. gaditana is 677 amino acids in length. The AlphaFold2 predicted structure is formed of 22 alpha-helices and 9 beta-strands (Fig. 3 C). Most of the regions including alpha-helices(121–151, 156–182, 192–195, 199–201, 216–227, 233–235, 247–260, 282–293, 314–322, 363–372, 386–392, 402–413, 433–453, 456–458, 464–494, 507–529, 534–546, 548–560, 577–585, 587–611, 617–636 and 638–648 amino acids) and beta-strands (205–211, 236–242, 266–270, 300–303, 323–326, 329–332, 336–340, 382–385 and 416–419 amino acids.) were high in confidence according to PAE scoring standard and only some regions in N-terminal and C-terminal were low in confidence (Fig. 3 A). As plDDT plot shows that these regions are in N-terminal and C-terminal of the protein (Fig. 3 B), it was decided to eliminate these regions for molecular docking simulation study. plDDT score also validates the accuracy of high-confidence regions of the predicted structure. The native active site of the enzyme which occupies a region on the alpha-helices is revealed to be present in blue regions which are high in confidence (Fig. 3 C). Fig. 3. Open in a new tab PAE plots of the first 5-ranked generated models by AlphaFold2. The first ranked model had the best confidence by pTM score of 0.78. This plot represents the alignment prediction error as the red regions show the most errors and the blue regions show the most confident regions ( A ). The plDDT plot shows that most regions of N. gaditana cellulose synthase amino acid residues are in confident regions, but the residues below 60% were not high in confidence ( B ). The AlphaFold2 predicted 3D structure. The blue regions show the most confidence and the red regions demonstrate least confidence ( C ). The Ramachandran plot of the structure reveales that 89.6% of the residues are in the most favored location ( D ) Protein structure evaluation was done by calculating the Ramachandran plot of the structure. Analysis of the plot confirmed that the structure had optimal quality and 90% of the residues were present in the favorable region (Fig. 3 D). Swiss-model The results demonstrated that the structure predicted by the SWISS-MODEL was similar to 4P02.1.A PDB template (Fig. 4 A). The Ramachandran plot made for this structure demonstrated that 88.6% of the residues were placed in favorable position (Fig. 4 B). The predicted model had 0.53 QMEANDisCo local score in which the blue regions were the most similar regions and the orange regions had lower similarity to the template (Fig. 4 C). Fig. 4. Open in a new tab 3D structure predicted by SWISS-MODEL ( A ). The Ramachandran plot shows that 88.6% of the residues are in the most favorable regions ( B ). The Local Quality Estimation plot evaluates the confidence of residues, blue residues have most similarity to the template and the orange regions are less similar regions to the template ( C ) Robetta Robetta generated five models, where the first ranked model was used for comparison of the models (Fig. 5 A). The Ramachandran plot made for this structure showed that 90% of the residues were located in the favorable region (Fig. 5 B). Fig. 5. Open in a new tab 3D structure of Robetta predicted model ( A ). Ramachandran plot for Robetta predicted structure shows that 89.7% of the residues are in the most favorable regions ( B ). 3D structure of I-TASSER predicted model ( C ). Ramachandran plot for I-TASSER predicted structure reveals that 69.4% of the residues are in the most favorable regions ( D ). The plot regions are divided into four groups: most favorable (red), additional allowed (yellow), generously allowed (light yellow), and disallowed (white) I-TASSER The Ramachandran plot calculated for the first ranked model obtained from I-TASSER results demonstrated that 69.4% of the residues were placed in the favorable region (Fig. 5 C, D). Ultimately, all the predicted structures were superimposed with the structure predicted by AlphaFold2 to assess their similarities. The results reveal that all the structures were superimposed with high accuracy and the calculated RMSD values are: AlphaFold2 vs Robetta: 2.6 Å, AlphaFold2 vs SWISS-MODEL: 4.8 Å and AlphaFold2 vs I-TASSER: 5.2 Å (Fig. 6 ). The results prove that AlphaFold2 and Robetta models have the greatest similarity to each other. Fig. 6. Open in a new tab Alignment of predicted structures to AlphaFold2 model evaluated by RMSD. AlphaFold2 vs I-TASSER ( A ). AlphaFold2 vs Robetta ( B ). AlphaFold2 vs SWISS-MODEL ( C ) Nannochloropsis oculata cellulose synthase 3D structure modeling by AlphaFold2 Cellulose synthase of N. oculata is consisted of 333 amino acids after model construction, refinement and removal of the low-confidence regions due to the PAE and plDDT plots (Fig. 7 A–C). The AlphaFold2-predicted 3D structure is composed of 15 alpha-helices (1–9, 19–41, 53–55, 70–81, 87–89, 102–117, 139–150, 171–79, 215–228, 242–248, 258–268, 289–309, and 320–332) and 12 beta-strands (59–65, 90–96, 123–128, 157–160, 165–167, 180–182, 186–188, 192–196, 200–202, 238–241, 272–275 and 281–284). Fig. 7. Open in a new tab PAE plots of the first 5-ranked generated models by AlphaFold2 for the N. oculata cellulose synthase strucutre. The first ranked model had the best confidence by pTM score of 0.43. This plot represents the alignment prediction error as the red regions show the most errors and the blue regions show the most confident regions ( A ). The plDDT plot shows the per-residue score of the predicted N. oculata cellulose synthase, indicating high-confidence regions (closer to 100) in the predicted model ( B ). Energy minimized N. oculata cellulose synthase 3D structure after removal of low-confidence (red) regions ( C ). The Ramachandran plot of the structure reveals that 90.9% of the residues are in the most favored location (D) Energy minimization was performed on N. oculata cellulose synthase to make it ready for docking analysis (Fig. 7 C). Moreover, the protein overall structure was assessed by Ramachandran plot. The plot showed that 90.9% of the residues are located in favorable regions (Fig. 7 D). Molecular docking of salt stressors on Nannochloropsis gadtiana cellulose synthase The ligands selected for cellulose synthase functionality evaluation in salinity stress are the common compositions of seawater which include: B, Ca, Cl, Fe, HCO 3 , K, Mg, NH 4 , NO 3 , Na, NaCl, PO 4 , P, SO 4 and SiO 2 (Elbarmelgy et al. 2021 ; Matos et al. 2017 ), which the Structured Data File (SDF) files were obtained from PubChem database (Kim et al. 2019 ). Table 3 compares the location of active sites, where ligands with different ionic radius bind. Hex was used to dock B, Ca, Cl, Fe, HCO 3 , K, Mg, NH 4 , NO 3 , Na, PO 4 , P, and SO 4 ions as well as SiO 2 molecule on the cellulose synthase (Table 4 ). The ligands were bound to three binding regions on the molecule (I–III, respectively) (Fig. 8 ). Region I exhibited the highest ligand abundance, followed by Region II and Region III, respectively. Region III is the uridine-diphosphate (UDP) binding site which is the native substrate of cellulose synthase. Ranking of the ligands according to their interaction free energy (E-total) is reported in Table 4 which exhibited that SO 4 , PO 4 , Na, and HCO 3 had the most favorable reaction free energy (kJ/mol). Table 3. Amino acids present within a 5 Å radius of the ligands (ion/molecule) Ion/molecule Around 5 Å amino acids of active site B Tyr360–Phe361–Tyr362–Gly363–Val 365–Leu 366 Ca Gln343–Ser344–Phe 345/Gln356–Gln357–Tyr358–Arg359–Tyr 362/Tyr 423 Cl Tyr 360–Phe 361–Tyr 362–Gly 363–Val 365–Leu 366 Fe Tyr 360–Phe 361–Tyr 362–Gly 363–Val 365–Leu 366 HCO 3 Arg 359–Tyr 360–Gly 363–Pro 364/Tyr 529–Val 532–Tyr 536–Leu 537–Ser 540 K Tyr 358–Phe 361–Tyr 362/Trp 443 Mg Gln 343–Ser 344–Phe 345/Gln 356–Gln 357–Tyr 358–Arg 359–Tyr 362//Tyr 423 Na Arg 359–Tyr 360–Gly 363–Pro 364/Tyr 529–Val 532–Tyr 536 NH 4 Tyr 358–Phe 361–Tyr 362 – Trp 443 NO 3 His 434–Lys 438/Lys 567–Val 568–Thr 569–Ser 570–Asp 572 P Tyr 360–Phe 361–Tyr362–Gly 363–Val 365–Leu 366 PO 4 Arg 359–Tyr 360–Phe 361–Gly 363–Pro 364–Val 365/Tyr 529–Val 532–Tyr 536 SO 4 Arg 359–Tyr 360–Phe 361–Gly 363–Pro 364– Val 365/Tyr 529–Val 532–Tyr 536 SiO 2 Met 437/Phe 549–Cys 550–Arg 553//Trp 584–Pro 587–Tyr 588 UDP Gln 138–Tyr 141–Arg 145/Phe 162–Glu 166–Ile 169–Val 170–Leu 171–Gly 172–Gly 173–Ser 174 Open in a new tab Table 4. Molecular docking results were obtained using HEX Ligand E-total (kJ/mol) PO 4 − 90.9 SO 4 − 90.9 Na − 84.4 HCO 3 − 82.3 SiO 2 − 77.8 NO 3 − 76.2 P − 69.7 Cl − 68.6 B − 67.9 Ca − 64 Mg − 64 Fe − 53.6 K − 40.7 NH 4 − 32.1 Open in a new tab These results were based on interaction free energy (E-total). The top-ranked ligands had the lowest free energy. Cellulose synthase was selected as receptor in all of the performed molecular dockings Fig. 8. Open in a new tab Three binding regions were observed after molecular docking Region I is the binding site of eleven ligands and consists of GLN-343, SER-344, PHE-345, GLU-350, ASP-351, ASP-352, PRO-353, LEU-354, GLY-355, GLN-356, GLN-357, TYR-358, ARG-359, TYR-360, PHE-361, TYR-362, GLY-363, VAL-365, LEU-366, and TYR-423 amino acids (Fig. 9 ). Region II is the binding site of SiO 2 and NO 3 and consists HIS-434, MET-437, LYS-438, PHE-549, CYS-550, ARG-553, LYS-567, VAL-568, THR-569, SER-570, ASP-572, TRP-584, PRO-587, and TYR-588 amino acids (Fig. 10 ). Region III is the UDP binding site and is composed of GLN-138، TYR-141، ARG-145، PHE-162، GLU-166، ILE-169، VAL-170، LEU-171، GLY-172، GLY-173، SER-174، MET-483، LEU-486، ILE-487، PRO-490، PHE-505، TYR-510، PHE-514، TYR-517، MET-518، and GLN-521 amino acids (Fig. 11 ). Molecular docking of salt stressors on Nannochloropsis oculata cellulose synthase Molecular docking was performed using all of the selected ligands from seawater on N. oculata cellulose synthase. The results demonstrated that six of them (B, Cl, Fe, K, NH4 and P) occupied the same binding site as they occupied in N. gaditana (Fig. 12 A–G), which was expected due to the incomplete protein sequence of N. oculata that are currently available. It is concluded from Table 5 that P, Cl and B were the most favorable ligands, since they possessed the lowest reaction free energy (kj/mol). In addition, since the results of BLASTp analysis exhibited 84.7% identity between the N. oculata and N. gaditana sequences, the rest of structural analyses were carried out on N. gaditana as the reference species. Table 5. Rection free-energies (E-total) of six ions (P, Cl, B, Fe, K, and NH4) docked on N. oculata cellulose synthase Ligand E-total (kJ/mol) P − 75.2 Cl − 74.2 B − 73.6 Fe − 59.7 K − 57 NH4 − 48.3 Open in a new tab Hydrophobicity scaling of the active site The amino acids that existed in around 5 Å of all the ligands were represented according to their hydrophobicity traits (Figs. 13 and 14 ). The most hydrophilic amino acids were visualized in blue and the most hydrophobic amino acids were visualized in yellow. In this scale, the most hydrophilic amino acid score number was − 4.5 (arginine) and the most hydrophobic was 4.5 (isoleucine), with the other amino acids falling in between (Fig. 13 ). The results of analyzing the active sites hydrophobicity demonstrated that the most hydrophilic active sites belong to Ca, Mg and NO 3 by possessing four hydrophilic amino acids (GLN:343, GLN:356, GLN:357, and ARG:359 for Ca and Mg and HIS:434, LYS:438, LYS:567, and ASP:572 for NO 3 ). Furthermore, the most hydrophobic active sites belong to SO 4 and PO 4 by having two hydrophobic amino acids (VAL:365 and VAL:532) (Table 6 ). Na, HCO 3 , SiO 2 , and NO 3 have also more hydrophilic properties in their active site. P, Cl, B, and Fe had mainly hydrophobic properties in their active sites but also two hydrophilic amino acids were seen in the region. Ca and Mg had the most hydrophilic active sites and finally K and NH 4 active sites were likely neutral (Table 6 ). As displayed in Figs. 13 and 14 , and Table 6 , most of the interacting amino acids in all of the active sites are hydrophilic with the most hydrophile binding sites being seen for Ca, Mg, and NO 3 (Fig. 13 ) and most hydrophobic binding sites for B, Cl, Fe, P, K, PO 4 , SO 4 , and SiO 2 (Fig. 13 ). Meanwhile, the UDP binding site represents both properties (Fig. 13 ). Fig. 13. Open in a new tab Number of amino acids in around 5 Å distance of the active site divided by hydrophobicity scale: B, Cl, Fe and P; Ca and Mg; K; Na; HCO 3 ; NH 4 ; NO 3 ; PO 4 and SO 4 ; SiO 2 And UDP. Blue shows the most hydrophilic (− 4.5 to − 3.5), dark green shows the more hydrophilic (− 3.5 to − 1), green shows mostly neutral regions (− 1 to 1), light green reveals more hydrophobic (1–3.5) and yellow indicates the most hydrophobic (3.5–4.5) Table 6. List of specific amino acids present in each active site, categorized by hydrophobicity scale Number of amino acids in active site divided by hydrophobicity scale Ion/Molecule Blue (most hydrophilic) Dark green Green Light green Yellow (most hydrophobic) B 0 2 (TYR:360–TYR:362) (GLY:363)1 2 (PHE:361–LEU:366) (VAL:365)1 Ca 4 (GLN:343–GLN:356–GLN:357–ARG:359) 3 (TYR:358–TYR:362–TYR:423) (SER:344)1 (PHE:345)1 0 Cl 0 2 (TYR:360–TYR:362) (GLY:363)1 2 (PHE:361–LEU:366) (VAL:365)1 Fe 0 2 (TYR:360–TYR:362) (GLY:363)1 2 (PHE:361–LEU:366) (VAL:365)1 HCO 3 (ARG:359)1 4 (TYR:360–PRO:364–TYR:529–TYR:536) 2 (GLY:363–SER:540) (LEU:537)1 (VAL:532)1 K 0 2 (TYR:358–TYR:362) (TRP:443)1 (PHE:361)1 0 Mg 4 (GLN:343–GLN:356–GLN:357–ARG:359) 3 (TYR:358–TYR:362–TYR:423) (SER:344)1 (PHE:345)1 0 Na (ARG:359)1 4 (TYR:360–PRO:364–TYR:529–TYR:536) (GLY:363)1 0 (VAL:532)1 NH 4 0 2 (TYR:358–TYR:362) (TRP:443)1 (PHE:361)1 0 NO 3 4 (HIS:434–LYS:438–LYS:567–ASP:572) 0 2 (THR:569–SER:570) 0 (VAL:568)1 P 0 2 (TYR:360–TYR:362) (GLY:363)1 2 (PHE:361–LEU:366) (VAL:365)1 PO 4 (ARG:359)1 4 (TYR:360–PRO:364–TYR:529–TYR:536) (GLY:363)1 (PHE:361)1 2 (VAL:365–VAL:532) SO 4 (ARG:359)1 4 (TYR:360–PRO:364–TYR:529–TYR:536) (GLY:363)1 (PHE:361)1 2 (VAL:365–VAL:532) SiO 2 (ARG:553)1 3 (TYR:551–PRO587–TYR:588) (TRP:584)1 3 (MET:437–PHE:549–CYS:550) 0 UDP 3 (GLN:138–ARG:145–GLU:166) (TYR:141)1 3 (GLY:172–GLY:173–SER:174) 2 (PHE:162–LEU:171) 2 (ILE:169–VAL:170) Open in a new tab Blue indicates the highest hydrophilicity and yellow indicates the most hydrophobicity Effect of NaCl, NaNO 3 , and NaHCO 3 treatments on Nannochloropsis oculata cell growth and cellulose production After performing repeated measures ANOVA ( p < 0.05) with LSD post hoc analysis on all groups’ absorbances at 680 nm within 12 days, there was no statistically significant difference between the microalgae growth in the control group and the treatments (Fig. 15 ). This may confirm that all the treatments did not inhibit the microalgae growth under salinity stress conditions and normal light intensity(74 µmol m −2 s −1 ), and therefore, a medium containing these salts such as desalination brine can be a suitable environment for cultivating N. oculata. However, after performing the same ANOVA tests on all groups’ cell densities during 12 days of cultivation, there was a statistically significant difference between the control and one of the treatment groups, NaHCO 3 −3 ( p = 0.022). This implies that treatment with NaHCO 3 (0.0023 mol/L) improved the microalgae growth (Fig. 15 ). While the other treatment groups did not show a significant change from the control. Fig. 15. Open in a new tab Cell density (Cells/mL) × 10 4 of the N. oculata in different NaCl concentrations and control groups ( A ). Absorbance (680 nm) of the N. oculata in different NaCl concentrations and control groups ( B ). Cell density (Cells/mL) × 10 4 of the N. oculata in different NaNO 3 concentrations and control groups ( C ). Absorbance (680 nm) of the N. oculata in different NaNO 3 concentrations and control groups ( D ).Cell density (Cells/mL) × 10 4 of the N. oculata in different NaHCO 3 concentrations and control groups ( E ). Absorbance (680 nm) of the N. oculata in different NaHCO 3 concentrations and control groups ( F ) Figure 15 A and B presents the cell density and optical density of N. oculata in NaCl groups compared to the control, indicating that the growth curves rose gradually until day 9. This trend was followed by a decline until day 12. Neither of the groups was significantly different from the other. Figure 15 C and D belongs to NaNO 3 groups showing fluctuation in the growth curves of N. oculata. Although the groups do not follow a similar trend, they reach their maximum value approximately on day 9 and decline until day 12. The figures are very close to each other and neither is significantly different. However, as depicted in Fig. 15 E, F, which belong to NaHCO 3 groups, N. oculata reaches its highest growth rate in the NaHCO 3 −3 treatment (significantly different cell density compared to the control and NaHCO 3 −1 groups). In the NaHCO 3 −3 group in Fig. 15 E, there was a sharp increase from day 1 until day 7 which peaked with nearly 84 × 10 4 cells per mL and then declined until day 12. Figure 16 illustrates the images of treatment groups and control group on day 9, a crucial point at which nearly all treatments attained their peak cell density. It is evident from the visual representations that NaHCO 3 −3, showing higher growth in comparison with others, displays both higher density and a more pronounced green coloration (Fig. 16 ). Fig. 16. Open in a new tab Images of treatments and control group on day 9, a crucial point at which nearly all treatments attained their peak cell density. The top row corresponds to the NaCl treatment, the middle row to NaNO 3 and the bottom row to NaHCO 3 .The NaHCO 3 -treated samples which exhibited great growth, reveal a vibrant and healthy green appearance. Subsequently, the NaNO 3 -treated samples show intermediate green coloration compared to the top and bottom rows, whereas the NaCl-treated samples display the least pronounced green When comparing each treatment optical density with the control group on a specific day, the NaHCO 3 −3 group was significantly different on days 7 and 12 ( p = 0.048 and p = 0.031, respectively) as determined by one-way ANOVA with LSD post hoc. Performing the same tests on cell densities for each day, it was found that the NaHCO 3 −3 group on day 5, day 7, and day 12 ( p = 0.012, p = 0.002 and p = 0.036, respectively) and NaHCO 3 −2 group on day 12 ( p = 0.024) were significantly different from control (Supplementary Figs. 2 and 3 ). This can demonstrate that concentrations of NaHCO 3 can be appropriate for microalgae growth. Effect of NaCl, NaNO 3 , and NaHCO 3 treatments on cellulose weight of N. oculata Use of one-way ANOVA with LSD post hoc test on cellulose weights, extracted from N. oculata on day 12, illustrated that there was a statistically significant difference between the control group and NaCl-2, NaCl-3, NaNO 3 −3, NaHCO 3 −2, and NaHCO 3 −3 ( p = 0.002, p = 0.048, p = 0.001, p = 0.011 and p = 0.003, respectively). Thus, we can estimate that growing N. oculata in media with NaCl, NaNO 3 , and NaHCO 3 may lead to a greater harvest of cellulose or at least we can conclude that these media do not reduce cellulose production (Fig. 17 ). Fig. 17. Open in a new tab Cellulose weight of the N. oculata in control and different treatment groups on day 12 Discussion In this study, a combination of computational and experimental analyses confirmed that the activity of cellulose synthase is not disrupted in the presence of mineral salts in the habitat of Nannochloropsis . Since no cellulose synthase gene or protein sequence data was available for N. oculata in databases, and the sequence retrieved from N. oculata genome was incomplete, N. oculata was initially modeled using a BLAST search on its genome and AlphaFold2, followed by molecular docking. However, N. gaditana cellulose synthase was used for the rest of the comprehensive computational analysis as it was the closest match (84.7% sequence identity) to the N. oculata cellulose synthase and provided a complete protein structure for the analysis. Using 3D protein modeling and molecular docking, it was specifically shown that none of the salt stressors occupies the enzyme's active site, approving that N. gaditana cellulose synthase can perform effectively in such salt stress conditions. As N. gaditana is a seawater microalga and also is capable of growing in saline environments such as desalination brine (Matos et al. 2017 ), it was expected that only a few elements from seawater composition would inhibit cellulose synthase native active site. The final results supported this idea as none of the ligands occupied the native active site. Different environmental conditions can affect the biomass production yield and growth rate (Shetty et al. 2019 ). Molecular docking analysis of salt ions on N. gaditana cellulose synthase confirmed that PO 4 , SO 4 , Na, HCO 3 , and SiO 2 are the top-ranked ions due to their most favorable interaction free energy. As hydrophobicity is a vital factor in protein folding, this factor was used to investigate the active site characteristics. The results demonstrated that most of the ligands with more favorable binding energy possessed hydrophilic properties in their active site. It was shown from hydrophobicity evaluation that most of interacting hydrophilic amino acids are in Ca, Mg and NO 3 binding sites and most of interacting hydrophobic amino acids are present around B, Cl, Fe, P, K, PO 4 , SO 4 , and SiO 2 binding sites. Furthermore, UDP binding site included both hydrophilic and hydrophobic amino acids. A noticeable point is that as the content of the chosen salts varies in seawater composition, it may influence the ranking of ligands in further studies. Also, as there are some pollutants such as heavy metals (Devriese et al. 2001 ), nitrate, and nitrite present in seawater composition in a lower content, which may be important in cellulose synthase activity, further In-silico analysis based on other possible ligands could be useful to consider the enzyme’s efficiency from a complementary aspect. These findings provide a clear understanding of N. gaditana adaptability in salinity stress by confirming that none of the mineral salts interfere with cellulose synthase active site and by ranking the salt stressors based on the binding free energies as well as the precise binding epitopes on the enzyme. This could result in enhancing this enzyme’s activity in the cellulose production process. In the experimental part of the research, the goal was to investigate whether NaCl, NaNO 3 , and NaHCO 3 affect N. oculata growth and cellulose yield. As the results indicated, these salt stressors did not inhibit the microalgae growth nor did they decrease cellulose yield. In some cases, with NaHCO 3 as the stress-inducing factor, an improvement in growth was observed. Furthermore, there was a moderate increase in the quantity of cellulose in some concentrations of the three salt stresses. Note that the experiments were conducted under optimal light conditions (74 µmolm −2 s −1 ) with a 16/8 light/dark cycle for 12 days, confirming the computational observations showing no interruption of the enzyme’s active site by these stressors. Overall, it was decided to validate the comprehensive results by evaluating N. oculata cell growth using either optical density (OD) or cell density. N. oculata from Iran’s ecosystem was employed for experimentally validation of the computational findings. This choice was made, because N. oculata was found to have similar cellulose content to N. gaditana . In addition, an 84.7% sequence identity was observed between the cellulose synthases between N. oculata and N. gaditana in a BLASTp search, and there was a lack of access to N. gaditana in our local genetic banks. Furthermore, the N. oculata strain is considered the holotype representative of the Nannochloropsis genus. There are also several approaches in molecular docking analysis of microalgae in which the growth level of the selected microalgae has been evaluated by the methods mentioned. Hence, it was decided to employ both these approaches to achieve and validate the main computational results of the study comprehensively. Ultimately, we aimed to ensure that neither cell growth nor cellulose yield was interrupted upon using salt ions in the medium. However, further research is needed to validate the computational ranking of the ligands by evaluating the exact binding energies of the salt stressors and, as well as to compare this microalgae growth rate with other microalgae species in desalination brine. Furthermore, extraction of cellulose from N. oculata in all growth phases (Log, Mid-log and Stationary) could be beneficial for evaluation of the optimal extraction time.. In this study, the aim was not to analyze the cellulose production of individual cells, as an in-depth investigation into multiple underlying factors and the impacts of salts on each enzyme involved in the cellulose synthesis process would be required. The goal was to propose the cultivation of microalgae, such as Nannochloropsis , in disposable materials such as desalination brine could be a promising approach for extracting cellulose for commercial purposes. Therefore, the overall cellulose output of each sample was evaluated, whether it increased as a result of cell growth or other factors. These findings enhance the current knowledge about microalgae cellulose synthase which is the most fundamental cellular machine in the process of nanocellulose production. This paves the path towards nanocellulose production from microalgae in harsh saline environments. It could also be used as a framework to evaluate similar/other stress sources impacts on different species of microalgae. Conclusions In the present study, the molecular interactions of salt ions and N. gaditana cellulose synthase was investigated to recognize the enzyme functionality and adaptability under salinity stress conditions for the first time. Molecular modeling of cellulose synthase and molecular docking analysis of the selected ligands from seawater composition was performed. In addition, cell growth and cellulose production of N. oculata under similar stress conditions were evaluated to validate the in-silico results. Ultimately, molecular docking analysis concluded that none of the salt stressors occupied the native active site of cellulose synthase, meaning the activity of enzyme would not diminish under salinity stress conditions. Consequently, optical density and cell density were regularly measured to evaluate the growth rate. In general, it was shown that the applied salinity stressors—NaCl, NaNO 3 , and NaHCO 3 —did not impede the growth of N. oculata In the case of NaHCO 3 at a specific concentration, it even enhanced microalgae growth. Finally, when it came to cellulose production, certain samples from each salinity stress group exhibited significantly higher yields compared to the control group. Overall, both computational and experimental results proved that cellulose synthase activity during the presence of seawater salt ions will not be interrupted; hence, development of this enzyme in the further studies can be done thanks to its emerging potential in the bio-materials market. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 502 KB) (502.2KB, docx) Author contributions SMHH carried out the computational analyses, provided Figs. 1–12, Tables 3–5 and wrote the corresponding text. MAI co-designed the study, supervised the project, provided the experimental resources, interpreted the results, and edited the entire manuscript. HA co-designed the study and contributed to the interpretation of the results. MSR carried out the experiments, provided Figs. 13–15, Tables 1, 2 and wrote the corresponding text. ZS supervised the experimental assays and edited the corresponding text. All authors have read the manuscript thoroughly. Funding No source of funding was collaborated in this study. Data availability All methods were carried out in accordance with relevant guidelines and regulations. The data sets used and/or analyzed during the current study are available by the corresponding author upon reasonable request. Declarations Conflict of interest All of the authors declare that they have no known conflicts of interests that could have appeared to influence the work reported in this paper. Contributor Information Maryam Azimzadeh Irani, Email: m_azimzadeh@sbu.ac.ir. Hossein Askari, Email: h_askari@sbu.ac.ir. References Abraham E, Weber DE, Sharon S, Lapidot S, Shoseyov O (2017) Multifunctional cellulosic scaffolds from modified cellulose nanocrystals. ACS Appl Mater Interfaces 9:2010–2015 [ DOI ] [ PubMed ] [ Google Scholar ] Andersen RA (2004) Algal culturing techniques Anne J (2014) Market projections of cellulose nanomaterial-enabled products—part 1: applications Baba HS, Baba HMB, Kassouar S, Abi AS (2016) Physicochemical analysis of cellulose from microalgae Nannochloropsis gaditana. Af J Biotechnol 15:1201–1207 [ Google Scholar ] Benkert P, Biasini M, Schwede T (2011) Toward the estimation of the absolute quality of individual protein structure models. Bioinformatics 27:343–350 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Chu Y, Zhang C, Ho SH (2021) Computational simulation associated with biological effects of alkyl organophosphate flame retardants with different carbon chain lengths on Chlorella pyrenoidosa. Chemosphere 263 [ DOI ] [ PubMed ] Colin L et al (2023) The cell biology of primary cell walls during salt stress. Plant Cell 35:201–217 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cowie N, Bilek T, Wegner TH, Shatkin E (2014) Market projections of cellulose nanomaterial-enabled products—part 2: volume estimates Devriese M et al (2001) Effect of heavy metals on nitrate assimilation in the eukaryotic microalga Chlamydomonas reinhardtii. Plant Physiol Biochem 39:443–448 [ Google Scholar ] Dhali K, Ghasemlou M, Daver F, Cass P, Adhikari B (2021) A review of nanocellulose as a new material towards environmental sustainability. Sci Total Environ 775:145871 [ DOI ] [ PubMed ] [ Google Scholar ] BIOVIA discovery studio, dassault systèmes. Preprint at (2021) Elbarmelgy A, Ismail MM, Sewilam H (2021) Biomass productivity of Nannochloropsis sp. grown in desalination brine culture medium. Desalination Water Treat 216:306–314 [ Google Scholar ] Endler A et al (2015) A mechanism for sustained cellulose synthesis during salt stress. Cell 162:1353–1364 [ DOI ] [ PubMed ] [ Google Scholar ] Evans R et al (2021) Protein complex prediction with AlphaFold-Multimer. bioRxiv. 10.1101/2021.10.04.46303434981053 [ Google Scholar ] Farrokhi N et al (2006) Plant cell wall biosynthesis: genetic, biochemical and functional genomics approaches to the identification of key genes. Plant Biotechnol J 4:145–167 [ DOI ] [ PubMed ] [ Google Scholar ] Gasteiger E et al (2003) ExPASy: the proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res 31:3784–3788 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] GUILLARD RR, R. JH. Studies of marine planktonic diatoms. I. Cyclotella nana Hustedt, and Detonula confervacea (cleve) Gran. Can J Microbiol 8, 39–229 (1962). [ DOI ] [ PubMed ] Henson JH, Samasa B, Burg EC (2019) High resolution imaging of the cortex isolated from sea urchin eggs and embryos. Methods Cell Biol 151:419–432 [ DOI ] [ PubMed ] [ Google Scholar ] Hu J et al (2023) Photosynthetic response mechanism to polybrominated diphenyl ether exposure in Chlorella pyrenoidosa. Ecotoxicol Environ Saf 263:115245 [ DOI ] [ PubMed ] [ Google Scholar ] Hunter JD (2007) Matplotlib: a 2D graphics environment. Comput Sci Eng 9:90–95 [ Google Scholar ] Jeong SW et al (2017) Transcriptional regulation of cellulose biosynthesis during the early phase of nitrogen deprivation in nannochloropsis salina. Sci Rep 7:1–11 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Jumper J et al (2021) Highly accurate protein structure prediction with AlphaFold. Nature 596:583–589 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Kim DE, Chivian D, Baker D (2004) Protein structure prediction and analysis using the Robetta server. Nucleic Acids Res 32:W526 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Kim S et al (2019) PubChem 2019 update: improved access to chemical data. Nucleic Acids Res 47:D1102–D1109 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character of a protein. J Mol Biol 157:105–132 [ DOI ] [ PubMed ] [ Google Scholar ] Laskowski RA, MacArthur MW, Moss DS, Thornton JM (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr 26:283–291 [ Google Scholar ] Lee HR, Kim KH, Mun SC, Chang YK, Choi SQ (2018) A new method to produce cellulose nanofibrils from microalgae and the measurement of their mechanical strength. Carbohydr Polym 180:276–285 [ DOI ] [ PubMed ] [ Google Scholar ] Li N et al (2021) Extracellular adsorption, intracellular accumulation and tolerance mechanisms of Cyclotella sp to Cr(VI) stress. Chemosphere 270:128662 [ DOI ] [ PubMed ] [ Google Scholar ] Macindoe G, Mavridis L, Venkatraman V, Devignes MD, Ritchie DW (2010) HexServer: an FFT-based protein docking server powered by graphics processors. Nucleic Acids Res 38:W445–W449 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Maier JA et al (2015) ff14SB: Improving the Accuracy of Protein Side Chain and Backbone Parameters from ff99SB. J Chem Theory Comput 11:3696–3713 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Maltsev Y, Maltseva K, Kulikovskiy M, Maltseva S (2021) Influence of light conditions on microalgae growth and content of lipids, carotenoids, and fatty acid composition. Biology 10:1060 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Matos ÂP, Moecke EHS, Sant’Anna ES (2017) The use of desalination concentrate as a potential substrate for microalgae cultivation in Brazil. Algal Res 24:505–508 [ Google Scholar ] Mihranyan A (2011) Cellulose from cladophorales green algae: from environmental problem to high-tech composite materials. J Appl Polym Sci 119:2449–2460 [ Google Scholar ] Mirdita M et al (2022) ColabFold: making protein folding accessible to all. Nat Methods 19:679–682 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Munawaroh SH et al (2020) In-vitro molecular docking analysis of microalgae extracted phycocyanin as an anti-diabetic candidate. Biochem Eng J 161:107666 [ Google Scholar ] Nanocellulose Market, Latest Global Industry Size Growth Forecast, Trends Report Narang PK et al (2021) Functional annotation and sequence-structure characterization of a hypothetical protein putatively involved in carotenoid biosynthesis in microalgae. S Afr J Bot 141:219–226 [ Google Scholar ] Omerspahic M, Al-Jabri H, Siddiqui SA, Saadaoui I (2022) Characteristics of desalination brine and its impacts on marine chemistry and health, with emphasis on the persian/arabian gulf: a review. Front Mar Sci 9:845113 [ Google Scholar ] Padhorny D et al (2016) Protein-protein docking by fast generalized Fourier transforms on 5D rotational manifolds. Proc Natl Acad Sci U S A 113:E4286–E4293 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Pestana-Nobles R et al (2022) Docking and molecular dynamic of microalgae compounds as potential inhibitors of beta-lactamase. Int J Mol Sci 23:1630 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Pettersen EF et al (2004) UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem 25:1605–1612 [ DOI ] [ PubMed ] [ Google Scholar ] Pettersen EF et al (2021) UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci 30:70–82 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Purushotham P, Ho R, Zimmer J (2020) Architecture of a catalytically active homotrimeric plant cellulose synthase complex. Science 1979(369):1089–1094 [ DOI ] [ PubMed ] [ Google Scholar ] Reva BA, Finkelstein AV, Skolnick J (1998) What is the probability of a chance prediction of a protein structure with an rmsd of 6 å? Fold des 3:141–147 [ DOI ] [ PubMed ] [ Google Scholar ] Reyimu Z, Özçimen D (2017) Batch cultivation of marine microalgae Nannochloropsis oculata and Tetraselmis suecica in treated municipal wastewater toward bioethanol production. J Clean Prod 150:40–46 [ Google Scholar ] Roberts AW, Roberts E (2004) Cellulose synthase (CesA) genes in algae and seedless plants. Cellulose 11:419–435 [ Google Scholar ] Ross IL, Shah S, Hankamer B, Amiralian N (2021) Microalgal nanocellulose—opportunities for a circular bioeconomy. Trends Plant Sci 26:924–939 [ DOI ] [ PubMed ] [ Google Scholar ] Sampathkumar, S. J., Srivastava, P., Ramachandran, S., Sivashanmugam, K. & Gothandam, K. M. Lutein: A potential antibiofilm and antiquorum sensing molecule from green microalga Chlorella pyrenoidosa. Microb Pathog 135, (2019). [ DOI ] [ PubMed ] Sayers EW et al (2022) Database resources of the national center for biotechnology information. Nucleic Acids Res 50:D20 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Scholz MJ et al (2014) Ultrastructure and composition of the Nannochloropsis gaditana cell wall. Eukaryot Cell 13:1450–1464 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schrödinger LLC (2015) The PyMOL molecular graphics system, Version~1.8 Shetty P, Gitau MM, Maróti G (2019) Salinity stress responses and adaptation mechanisms in eukaryotic green Microalgae. Cells 8:1657 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Singh SP, Singh P (2015) Effect of temperature and light on the growth of algae species: a review. Renewable Sustain Energy Rev 50:431–444 [ Google Scholar ] Studer G et al (2021) ProMod3—a versatile homology modelling toolbox. PLoS Comput Biol 17:e1008667 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Tang DYY et al (2020) Green technology for the industrial production of biofuels and bioproducts from microalgae: a review. Environ Chem Lett 18:1967–1985 [ Google Scholar ] Varadi M et al (2022) AlphaFold protein structure database: massively expanding the structural coverage of protein-sequence space with high-accuracy models. Nucleic Acids Res 50:D439–D444 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Wang B, Jia J (2020) Photoprotection mechanisms of Nannochloropsis oceanica in response to light stress. Algal Res 46:101784 [ Google Scholar ] Wang Y, Tibbetts SM, McGinn PJ (2021) Microalgae as sources of high-quality protein for human food and protein supplements. Foods 10:3002 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Waterhouse A et al (2018) SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res 46:W296–W303 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Yang J et al (2015) The I-TASSER Suite: protein structure and function prediction. Nat Methods 12:7 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Yun CJ, Hwang KO, Han SS, Ri HG (2019) The effect of salinity stress on the biofuel production potential of freshwater microalgae Chlorella vulgaris YH703. Biomass Bioenergy 127:105277 [ Google Scholar ] Zanchetta E et al (2021) Algal cellulose, production and potential use in plastics: challenges and opportunities. Algal Res 56:102288 [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary file1 (DOCX 502 KB) (502.2KB, docx) Data Availability Statement All methods were carried out in accordance with relevant guidelines and regulations. The data sets used and/or analyzed during the current study are available by the corresponding author upon reasonable request. Articles from 3 Biotech are provided here courtesy of Springer ACTIONS View on publisher site PDF (4.7 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• All Cellulose Composites Productions From Oil Palm Empty Fruit Bunches#270893
• A Fungal Natural Product that Targets Cellulose Synthase Complex and Inhibits Plant Cellulose Biosynthesis#176491
• 42 U.S.C. § 16251 — Production incentives for cellulosic biofuels#589021
• Enzymes, Proteins, and Peptides as Promising Biosolutions for Crop Production#14425
• Influence of carbohydrate source on xanthone content in root cultures of Gentiana dinarica Beck#280586