one marine microalgal strain D. salina (GY-H13) was purchased from Shanghai Guangyu Biological Technology Co., LTD, and maintained in petri dishes using BG11 solid medium containing 3% NaCl. D. salina cells were successively transferred from petri dishes to 250 ml flasks, and then cultivated in 400 ml bubble column photobioreactors (PBR, working volume 300 ml) with 1% CO₂ under 110 μmol m⁻² s⁻¹ and 25°C conditions. The cells of D. salina were harvested during their logarithmic growth phase by centrifugation, and then the harvested cells were resuspended into BG11 medium containing 3% NaCl with the required biomass density and used in the following experiments.

In order to investigate the effect of CO₂ concentration on the growth of D. salina, the microalgal cells were cultured at five CO₂ levels, air (0.03%CO₂), 1%CO₂, 10%CO₂, 20%CO₂, and 30%CO₂. The culture was incubated in 400 ml PBR (working volume 300 ml) under 110 μmol m⁻² s⁻¹ and 25°C conditions. The concentration of CO₂ (v/v) was manipulated by adjusting the flow rates of pure CO₂ and air with gas mass flow controllers, respectively. The mixed gases were filtered (0.22 µm) and then transferred into the bottom of PBR through one slender glass tube (inner diameter 0.25 cm) with an aeration rate of 0.2 L min⁻¹. The initial cell density of the cultures was maintained at 0.1 g L⁻¹. The temperature was maintained using constant temperature water bath.

The cells of D. salina were cultivated for 8 days at 0.03%–30% CO₂. Samples were taken daily from the PBR to estimate microalgal growth and the pH of the culture medium. Generally, 12 ml of culture medium was collected in a clean glass tube, and the pH of the sample was immediately measured using a Five Easy pH meter (METTLER TOLEDO). The maximum quantum yield of photosystem Ⅱ was determined using 2 ml of the sample. The F_v/F_m ratio was measured using a fluorescence monitoring system (FMS2, Lufthansa Scientific Instruments Co., Ltd., United Kingdom) after the sample has been stored in dark conditions for 30 min (García-Cañedo et al., 2016). Of the remaining sample 10 ml was filtered using a pre-dried and pre-weighed cellulose membrane (0.45 µm pore size), washed with deionized water, dried for 24 h at 105°C, cooled in a desiccator and then weighed again to determine uncorrected dry algae biomass. The dry weight of the blank filter was subtracted from that of the loaded filter to obtain the corrected algae dry cell weight. At the end of the microalgal cultivation period (8 days), microalgal samples were harvested to test the contents of Chlorophyll and Carotenoid content. The pigments were extracted and calculated as described in Pruvost et al. (2011). The absorbances at 480, 652, 665, and 750 nm were measured by HACH DR2800 in a glass cell with a path length of 1 cm.

Transcriptomic analysis was used to investigate and test potential CO₂ tolerance mechanisms at a genetic level. D. salina cells cultivated under air (0.03%), 1% and 30% CO₂ were collected on day 1 and day 2, and denoted as C (0.03%) D1, C (1%) D1 and C (30%) D1; C (0.03%) D2, C (1%) D2 and C (30%) D2, respectively. Each sample was determined for two replicates. The sample treatment, transcriptomic determination, gene annotation and the related bioinformatics analysis have been described in previous studies (Zhou W et al., 2017; Cheng et al., 2019). NCBI non-redundant protein (NR) database classification of transcriptome sequences using DIAMOND software. Swiss-Prot database classification of transcriptome sequences using DIAMOND software. Protein family (Pfam) database classification of transcriptome sequences using HMMER3 software. Protein direct homology cluster (COG) database classification of transcriptome sequences using DIAMOND software. Gene ontology (GO) classification of transcriptome sequences using BLAST2GO software. Kyoto Encyclopedia of Genes and Genomes (KEGG) classification of transcriptome sequences using KOBAS software. Expression levels of genes and transcripts were quantified separately using the expression quantification software RESM, quantified as FPKM (fragments per kilobase exon model per million mapped reads), FPKM takes into account the effect of gene length and sequencing volume differences on the calculation of gene expression by first homogenizing the sequencing volume and then the gene length to allow visual comparison of expression. In order to control the probability or frequency of errors in the overall inferred results, the p-values obtained from the statistical tests are corrected using BH (FDR correction with Benjamini & Hochberg) for multiple testing, and the corrected p-values are known as p-adjust. Differential folds FC (Fold change) for different CO₂ concentration treatments were calculated based on the FPKM values of gene expression using RSEM software for differential analysis. Genes meeting the screening criteria of |log₂ FC|≥1 & p-adjust <0.05 were considered as differentially expressed genes (DEGs). GO and KEGG enrichment analyses were performed on genes in the gene set using Goatools. Raw data was deposited at the NCBI Sequence Read Archive (SRA) with BioProject record PRJNA901516.

As shown in Figure 1B, the growth of D. salina was significantly influenced by the CO₂ level. The biomass of microalgal cells was 0.61 g L⁻¹ after 8 days of incubation in air (0.03%CO₂). The cell biomass of the 1% CO₂ group grew rapidly, reaching 1.7 g·L⁻¹ on day 8, which was 2.79 times higher than that of incubation in air. Further increasing CO₂ concentrations resulted in the lower biomass concentrations under 10%–30% CO₂. However, the cell biomass of 30% CO₂ group also reached 1.13 g·L⁻¹, which was 1.85 times higher than that of incubation in air. The effect of different CO₂ concentration treatments on the growth of D. salina is further evidenced by the picture of D. salina growth. (Figure 1A). Under the similar experimental conditions, the growth parameters of D. salina used in this study are higher than previous marine microalgal strain. (Ishida et al., 2000; Chiu et al., 2009), Chaetoceros muelleri (Wang et al., 2014), P. glacialis and A. longicornis (Artamonova et al., 2017) and Phormidium valderianum (Dineshbabu et al., 2020). These results suggest that D. salina can tolerate a high CO₂ concentration environment.

Chlorophyll fluorescence parameters have been gradually applied in the study of the effects of environmental stress on cell photosynthesis (Bhagooli et al., 2021). Among the many parameters, the parameter F_v/F_m has little change under non-stress conditions and decreases significantly under stress conditions. It is suitable parameter to reflect the influence of environment on the growth of microalgae. Under 0.03% CO₂ conditions, the F_v/F_m ratio decreased to 0.48 on day 1, then gradually increased to 0.65 on day 5, and finally remained around this value until day 8. Under 1% CO₂ conditions, the F_v/F_m ratio increased to 0.54 on day 1, reached a maximum of 0.67 on day 6, and then decreased to 0.53 on day 8. Under 30% CO₂ conditions, the F_v/F_m ratio decreased to 0.47 on day 1, then gradually increased to 0.58 on day 3, and finally gradually decreased to 0.35 on day 8 (Figure 1C). The variation of F_v/F_m ratios under different CO₂ conditions indicated that the 1% CO₂ treatment condition was more suitable for D. salina growth. In contrast, the changes of F_v/F_m ratios at 30% CO₂ showed a reduced photochemical efficiency through damage to PSII under higher CO₂ stress. These results show consistency with the growth of D. salina.

The pH of the culture medium affects the permeability of the cell membrane of microalgae cells, the uptake and utilization of ions in the culture medium, the utilization of CO₂, and may alter the nutrient metabolism of microalgal cells (Barati et al., 2021). The starting pH of the culture medium was 7.8. Under 0.03% CO₂ conditions, the pH increased to 9.2 on day 1, then continued to increase and remained around 9.6. Under 1% CO₂ conditions, the pH dropped to 7.06 on day 1, then increased to 7.94 on day 2, and finally remained around 8.05. Under 30% CO₂ conditions, the pH dropped to 4.89 on day 1, then increased to 5.5 on day 3, and finally remained around 5.9 (Figure 1D). It can be seen that a pH-shift occurred for all cultures resulting in more alkaline conditions over time, and the pH level decreased with increasing CO₂ concentration. Generally, the pH of the culture medium was dictated by processes of acidification resulting from CO₂ dissolution and alkalization due to microalgal photosynthesis removing dissolved CO₂. The pH of the culture medium increased with time because of photosynthesis which elevated pH after algae acclimation in the culture medium. Alkalization of the culture medium is thought to compensate for the acidification effect of high CO₂ concentration (Solovchenko and Khozin-Goldberg, 2013). At higher CO₂ concentrations (10–30%), the combination of greater acidification and diminished photosynthesis resulted in the lower pH range.

Chloroplasts are important organelles in most photosynthetic microalgae, and chloroplasts also serve as sensors of the external environment (Zhang et al., 2020). Chlorophyll content is an important indicator of the photosynthetic intensity of microalgal cells, and the adaptation of microalgal cells to different CO₂ concentration conditions can be detected by chlorophyll content (Kumari et al., 2021). We determined the chlorophyll contents of D. salina cells after 8 days of incubation under different CO₂ concentrations conditions. As shown in Figure 1E, the maximum chlorophyll contents of D. salina cells was obtained at 1% CO₂. The chlorophyll contents of D. salina cells under 30% CO₂ was much higher than these of incubation in air. The above results were consistent with cell biomass. During photosynthesis, oxygen and light generate reactive oxygen species (ROS). As a defense mechanism, organisms produce many endogenous antioxidants to eliminate harmful ROS, thus maintaining normal cellular function and organismal health (Ascherio and Schwarzschild, 2017). Carotenoidscan protect chlorophyll from photooxidation by absorbing thermal energy from singlet oxygen and quenching it by releasing energy through polyene vibrations (Abreu et al., 2020). As shown in Figure 1 F, the contents of carotenoid in D. salina cells cultivated with 20%–30% CO₂ were higher than that in D. salina cells cultivated with 0.30%–10% CO₂. This indicates that D. salina cells were stressed by ROS under 20%–30% CO₂ conditions and produced carotenoids as antioxidants to eliminate the damage caused by ROS.

Twelve samples were used for transcriptome assembly, and a total of 97.67 Gb Clean Data were obtained. The Clean Data of each sample was more than 7.28 Gb, and the percentage of Q30 base was more than 95.06%. Statistical analysis of raw counts was performed by DESeq2 software based on negative binomial distribution, and genes with expression differences between comparison groups were obtained basing on certain screening conditions (p-adjust<0.05 and |log₂FC|≥ 1). Compared with C (0.03%)D1, there were 807 differential genes in C (1%)D1 group, including 354 up-regulated genes and 453 down-regulated genes. Compared with C (0.03%)D1, C (30%)D1 group has 13408 differential genes, including 6342 up-regulated genes and 7066 down-regulated genes (Figure 2A).

To further elucidate the response of D. salina under high CO₂ conditions, DEGs between C(0.03%)D1 and C(30%)D1 were subjected to GO analysis using the software Goatools (P-adjust<0.05). A total of 6039 DEGs were annotated in GO and divided into 225 functional subcategories, containing 137 groups of biological processes, 28 groups of cellular components and 60 groups of molecular functions. Within the category of biological processes, transmembrane transport, regulation of primary metabolic process, cell surface receptor signaling pathway, organic cyclic compound biosynthetic process, and cGMP biosynthetic process were prominent (Figure 2B). Within the category of cellular components, intrinsic component of membrane, integral component of membrane, endoplasmic reticulum, cytosol, and proteasome core complex were significantly enriched (Figure 2C). For the molecular functional categories, active transmembrane transporter activity, ATPase-coupled transmembrane transporter activity, primary active transmembrane transporter activity, transcription coregulator activity, peptide receptor activity were significantly enriched (Figure 2D).

In addition, KEGG enrichment analysis showed that 4056 DEGs were allocated to 125 KEGG pathways. The up-regulated genes (Figure 2E) are mainly enriched in ABC transporters, ribosome, galactose metabolism, pentose phosphate pathway, pentose and gluconate interconversions. The down-regulated genes (Figure 2F) were mainly enriched in homologous recombination, DNA replication, mismatch repair, other types of O-glycan biosynthesis, spliceosome, etc.

As mentioned above the low pH is associated with high CO₂ concentrations, this could induce serious oxidative stress and lead to ROS production in microalgal cells (Thompson et al., 2017). As the main executor of ROS scavenging system, the antioxidant system of algae consists of enzymatic antioxidant system and non-enzymatic antioxidant system. Their main function is to scavenge active oxygen (Liao et al., 2018). Enzymatic antioxidant system includes various antioxidant enzymes, such as superoxide dismutase (SOD), catalase (CAT), thioredoxin peroxidase (TrxR), ascorbic acid peroxidase (APX), glutathione peroxidase (GPX). Non-enzymatic antioxidant system includes various antioxidants, such as ascorbic acid (vitamin C), glutathione, vitamin E, carotenoids and proline (Serrano et al., 2021). SOD is the first line of defense in the enzymatic active oxygen scavenging pathway of algae. It is a metal enzyme family that can catalyze the transformation of superoxide into oxygen and hydrogen peroxide (H₂O₂) through disproportionation reaction (Zelko et al., 2002; Ben Hamouda et al., 2022). CAT is a tetramer enzyme containing heme, which is found in all aerobic organisms. CAT mainly exists in peroxides. Because its activity does not require reductive substrates and its high maximum reaction rate value and low Michaelis constant value to H₂O₂, it is essential to eliminate H₂O₂ produced in large quantities when cells are under stress (Gauthier et al., 2020). APX is found in higher plants, eukaryotic algae and some cyanobacteria (Roy et al., 2021). It exists in a variety of organelles, such as chloroplasts, mitochondria, peroxides and cytoplasm, and is the key enzyme to remove H₂O₂ in chloroplasts. Many studies have shown that stress has a significant impact on the activity of the antioxidant enzyme system of algae, but there are differences in the response of various enzymes to different stresses in different algal strains (Gauthier et al., 2020). Besides, the HSP/chaperone network includes Hsp70 family genes (DnaK, DnaJ, DnaJC7, DnaJC11, DnaJC13, Hsp70), Hsp family D gene (HspD1) and small Hsp genes (Hsp20, Hsp33) (Jacob et al., 2017). Compared with C(0.03%)D1, we found that the expression levels of genes encoding SOD, CAT and APX were significantly up-regulated in C(30%)D1. Compared to C(0.03%) D1, we found that the expression levels of genes encoding SOD, catalase CAT and APX were significantly up-regulated in C (30%) D1. The expression levels of genes encoding DnaK, DnaJ, Hsp20, HspA5 and HspA4 were significantly up-regulated in C (30%) D1(Table 3). The high expression of heat shock proteins indicated that D. salina was subjected to oxidative stress. On the other hand, D. salina can respond to high CO₂ stress through the antioxidant enzyme system under high CO₂ stress.