The Biotechnological Potential of the Marine Diatom Skeletonema dohrnii to the Elevated Temperature and pCO2

Marine diatoms are promising candidates for biotechnological applications, since they contain high-value compounds, naturally. To facilitate the production of these compounds, stress conditions are often preferable; however, challenges remain with respect to maximizing a metabolic potential for the large-scale cultivation. Here, we sequenced the transcriptome of diatom Skeletonema dohrnii under the actual (21 °C, 400 ppm) and elevated (25 °C, 1000 ppm) temperature and pCO2 condition. Results indicated that cells grown at higher temperature and pCO2 showed increasing growth rate, pigment composition, and biochemical productivity as did the expression of chlorophyll, carotenoid and bioactive compound related genes or transcripts. Furthermore, performing de novo transcriptome, we identified 32,884 transcript clusters and found 10,974 of them were differentially expressed between these two conditions. Analyzing the functions of differentially expressed transcripts, we found many of them involved in core metabolic and biosynthesis pathways, including chlorophyll metabolism, carotenoid, phenylpropanoid, phenylalanine and tyrosine, and flavonoid biosynthesis was upregulated. Moreover, we here demonstrated that utilizing a unique bio-fixation ability, S. dohrnii is capable of suppressing central carbon metabolism to promote lipid productivity, fatty acid contents and other bioactive compounds under high temperature and pCO2 treatment. Our study suggests that this S. dohrnii species could be a potential candidate for wide-scale biotechnological applications under elevated temperature and CO2 conditions.


Introduction
In recent years, an increasing consideration has been devoted to the possibility of growing microalgal culture for commercial purposes. These interests are due to microalgal natural sources for wide-scale biotechnological applications including human food [1], animal feed [2], cosmetics [3], drugs [4], health products [5], biodiesel [6], fertilizers [7] and wastewater treatment [8]. On the other hand, increasing anthropogenic CO 2 is causing major changes on our earth. To mitigate the anthropogenic CO 2 concentrations with alternative renewable energy and other valuable products [9], scientists are focusing on the choice of cultivating microalgal strain, which could tolerate higher CO 2 and simultaneously produce bioactive compounds and natural products [10,11].
Marine diatoms are highly diverse and significant contributors to global primary productivity; they are also considered as attractive organisms for the combined CO 2 biofixation and natural Table 1. Physiological parameters of diatom S. dohrnii grown at LC and HC environmental conditions LC (21 • C, 400 ppm) and HC (25 • C, 1000 ppm).
We annotate the assembled unigenes into NCBI-non-redundant proteins (NR), nucleotide (NT), Swiss-Prot, Kyoto Encyclopedia of Genes and Genomes (KEGG), Eukaryotic Orthologous Groups (KOG), Pfam protein data base, and Gene Ontology (GO). Among the achieved 32,884 unigenes, 22,261 (67.70%), 2960 (9.00%) and 11,583 (35.19%)   The GO annotation showed 38,149 unigenes categorizations involved with 36 GO classes, in which biological process (11), cellular component (12) and molecular function (13) are the major classes, with 10,157, 11,711 and 16,281 annotated unigenes respectively (Figure 1a). To analyze the unigenes' intracellular metabolic process, the KOG database was used, indicating that 13,525 unigenes assigned into 25 functional processes (Figure 1b). Among all, amino acid metabolism (753), lipid transport and metabolism (465) and secondary metabolites biosynthesis (276) were the notable functional terms. The distribution of overall fragments mapped (FPKM) for S. dohrnii between HC and LC conditions with assembled transcript clusters (unigenes length) is also shown in ( Figure S1, and Table S1). The FPKM values were used for the evaluation of the expressed value and quantification of transcripts because it reflects the expression of genes, to which the fragments correspond. We chose a significant threshold of FPKM ≥ 0.1 to prevent analysis with expression levels close to zero for transcripts that had FPKM values < 0.05. Results demonstrated that different unigenes could be supported by different numbers of reads in cells grown at HC and LC conditions. Especially, the ranged FPKM unigenes distribution were from 0.004 to 3.918 in HC conditions, while it was considerably different from 0.004 to 3.718 at LC conditions. Moreover, we perform unsupervised principal component analysis (PCA) based on the expressed genes, to understand if metabolites differ between these two treatments ( Figure S2). The matrix contained 10,974 DETs (active variables) and six samples. The Pearson correlation between high and low carbon triplicate samples ranged from 0.19 to 0.47, shows significant variation between these two conditions. Factor 1 and factor 2 account for 62.83% and 22.79% of the variation, respectively. Factor 1 (cells grown at HC) clearly discriminate metabolic response from cells treated with LC. Both factors consistently cluster together, indicating an influence of different environmental conditions on the cellular development processes. The combining of temperature and pCO2 variation in this study exhibits 10,974 differentially expressed transcripts (DETs) comparing between HC and LC treatments. Based on log-2-fold with a false discovery rate (FDR), 0.001, a total of 8414 transcripts were upregulated and 2560 were downregulated in the HC treatment in comparison with the LC treatment ( Figure S3). Out of these, all DETs, 5635 unigenes were successfully annotated into GO enrichment analysis to understand the S. dohrnii functional response, when treated with HC and LC conditions ( Figure 2). The result shows that of the annotated DETs 2460 were involved in the molecular functions, followed by cellular component 1944 and biological process 1261. The membrane transport, lipid and fatty acid biosynthesis were the notably enriched term in the biological process, whereas, cell, membrane and organelle parts are the most enriched term in the cellular process; lipid transport and metabolic process in molecular function were the top enriched terms, respectively.
To further understand the DETs and its involvement in the specific metabolic pathway, we annotated GO enriched genes in KEGG pathway analysis, which represent 3294 transcripts mapped The combining of temperature and pCO 2 variation in this study exhibits 10,974 differentially expressed transcripts (DETs) comparing between HC and LC treatments. Based on log-2-fold with a false discovery rate (FDR), 0.001, a total of 8414 transcripts were upregulated and 2560 were downregulated in the HC treatment in comparison with the LC treatment ( Figure S3). Out of these, all DETs, 5635 unigenes were successfully annotated into GO enrichment analysis to understand the S. dohrnii functional response, when treated with HC and LC conditions ( Figure 2). The result shows that of the annotated DETs 2460 were involved in the molecular functions, followed by cellular component 1944 and biological process 1261. The membrane transport, lipid and fatty acid biosynthesis were the notably enriched term in the biological process, whereas, cell, membrane and organelle parts are the most enriched term in the cellular process; lipid transport and metabolic process in molecular function were the top enriched terms, respectively. metabolic functions with other databases are shown in Table S2. The DETs showing that metabolism associated ABC transports (p < 0.003), chlorophyll metabolism (p < 0.001), carotenoid biosynthesis (p < 0.001), phenylalanine, tyrosine and tryptophan biosynthesis (p < 0.005), phenylpropanoid biosynthesis (p < 0.005), lipoic acid (p < 0.005), fatty acid biosynthesis (p < 0.005) and carbon metabolism (p < 0.004) were significant pathways altered by combining temperature and pCO2 changes in this study. We, therefore, further investigate DETs associated with these metabolisms.   Table S2 for the comprehensive GO enrichment analysis of each sample. (a) differentially expressed genes in biological process, (b) differentially expressed genes in cellular component, (c) differentially expressed genes in molecular function.  Table S2 for the comprehensive GO enrichment analysis of each sample. (a) differentially expressed genes in biological process, (b) differentially expressed genes in cellular component, (c) differentially expressed genes in molecular function.
To further understand the DETs and its involvement in the specific metabolic pathway, we annotated GO enriched genes in KEGG pathway analysis, which represent 3294 transcripts mapped with 127 metabolic pathways (Table S2). The annotated transcripts' regulation and associated metabolic functions with other databases are shown in Table S2. The DETs showing that metabolism associated ABC transports (p < 0.003), chlorophyll metabolism (p < 0.001), carotenoid biosynthesis (p < 0.001), phenylalanine, tyrosine and tryptophan biosynthesis (p < 0.005), phenylpropanoid biosynthesis (p < 0.005), lipoic acid (p < 0.005), fatty acid biosynthesis (p < 0.005) and carbon metabolism (p < 0.004) were significant pathways altered by combining temperature and pCO 2 changes in this study. We, therefore, further investigate DETs associated with these metabolisms.

Membrane Transporters
In this study, 53 ABC transporters were differentially expressed consisting of 45 upregulated and 8 downregulated transcripts (Table S3). Of these DETs, membrane transport associated with multidrug/pheromone exporter, lipid exporter (ABC1), peptide exporter, pleiotropic drug resistance proteins, ABC superfamily (breast cancer resistant) and mitoxantrone resistance were the notable upregulated transcripts, whereas, metal transporters and mitochondrial Fe/S exporter were the important downregulated transcripts.

Phenylpropanoid and Flavonoid Biosynthesis
A total of 102 transcripts associated with phenylpropanoid biosynthesis were significantly regulated (Table S6), by elevated temperature and pCO 2 , which represent 79 upregulated and 23 downregulated transcripts. The vital upregulated enzymes of these pathways are alcohol dehydrogenase (ADH), pectinesterase (PE), gelatinase, gibberellin-related protein (GAs), collagens, and β-glucosidase. Furthermore, seven transcripts were significantly altered in flavonoid biosynthesis, and all of them were increased in abundance (Table S6), including anthocyanidin synthase (ANS) and caffeoyl-CoA-O-methyltransferase.

Lipid Metabolism and Fatty Acid Biosynthesis
Six transcripts encompassing α-lipoic metabolism were regulated in this study, which consists upregulation of lipoate synthase and lipoyl transferase enzymes (Table S7), whereas seven DETs in ether lipid metabolism have shown upregulation (Table S7), including phospholipase A2 (PLA2s), heterochromatin (HP1) and alpha-beta hydrolase superfamily. Further to this, a total of 90 transcripts associated with fatty acid biosynthesis were significantly regulated in this study, consisting of 74 upregulated and 16 downregulated transcripts (Table S7). The upregulated transcripts of glutaryl-CoA dehydrogenase (GCDH), β-ketoacyl-ACP reductase, long-chain-fatty-acid-CoA ligase, enoyl-CoA hydratase (ECH), acetyl-CoA acetyltransferase, very-long-chain enoyl-CoA reductase, ACP transacylase, fatty acyl-CoA elongase and 3-oxoacyl-acyl-carrier-protein are the important enzymes involved in fatty acid production, while fabA-like domain and AMP-binding enzyme were the notable downregulated enzymes in this metabolism.

Discussion
Diatoms are promising microorganisms for commercial purposes, as they grow and proliferate rapidly in a short time with highly enriched valuable products naturally [22]. These features make them lead unique candidates as a source of valuable bioproducts, such as pigments, lipids, and bioactive compounds. However, there are still gaps between altering the diatom metabolism for wide-scale commercial applications [23]. Using global transcriptome with modern bioinformatics, here, we investigate the combining effect of increased temperature and pCO 2 concentration on diatom metabolism and their associated biotechnological applications. Despite earlier studies demonstrated elevated temperature and pCO 2 impact on diatoms [6,14], there is no information on S. dohrnii for commercial applications, which, if confirmed, could be a significant advancement for wide-scale commercial use of this algae.
Recent reviews demonstrated the influence of elevating temperature and CO 2 on diatom's physiology and cellular development [24,25]. Some investigation noted that high temperature and CO 2 has positive effects on diatom's physiology [26][27][28] while others noted negative effects [29,30], or unaffected [31,32]. These contradictory findings could be the reason for varied experimental conditions, i.e., different pH [28] light intensities [27] temperature [30] and physiological complexity of diatoms [26,32]. In this study, the combining effect of elevated temperature and pCO 2 stimulated S. dohrnii biomass, and pigment contents, which is consistent with other investigations, showed increased growth and biomass of diatoms by elevated temperature and pCO 2 concentration [26,27,32]. The high pigment contents under HC conditions could be the reason for photo-acclimation or a consequence of increasing cell volume rather than an increase in pCO 2 availability [27,31]. Furthermore, consistent with our present study, recent investigation on the diatom Navicula directa showed the effect of increasing temperature and pCO 2 promoted the photosynthesis process, pigment composition and growth rate by approximately 43% [30].
Among the DETs, we found significant modulation of ABC transporters, including upregulation of multidrug exporter by 8.08-fold, pleiotropic drug resistance (5.58-fold), breast cancer resistance (9.0-fold) and lipid exporter proteins (7.23-fold). Multidrug proteins are designed by plants to confer resistance to various drugs [33] to protect cells from stress conditions and toxic environments. These multidrug proteins observed earlier in the marine organisms including diatoms, sponges, fish, and mussels, as potential anticancer drugs for human cancer [34][35][36]; therefore, over expression of these transporters in this study by combining temperature and pCO 2 elevation could be a valuable source for the biotechnological application.
Pigments composition is highly sensitive to environmental stress, including temperature [14] pH [16], and even combining temperature and pCO 2 changes [30]. Consistently, upregulated phytoene desaturase (10.36-fold) in this study by elevated temperature and pCO 2, increased ketocarotenoids accumulation during changes of temperature and pH together. A similar finding was seen earlier in Synechococcus sp [36] and Haematococous sp [37] response to higher temperature. Further to this, it was reported that pH variation increased the violaxanthin de-epoxidase enzyme to accumulate zeaxanthin on model diatom P. tricornutum [38]. Likewise, pH variation in this study increased the abundance of violaxanthin de-epoxidase (2.76-fold), and elevated zeaxanthin epoxidase by (3.71-fold), which can be extracted for industrial application. Beta-carotene upregulation (2.31-fold) in this study plays an important role by protecting cells from free radicals involved in light-harvesting functions [39]. However, the earlier investigation noted β-carotene as a potential antitumor agent in humans [40]. In agreement with these results, Nishino et al. [41] reported that carotenoid contents, including β-carotene, zeaxanthin and violaxanthin de-epoxidase exhibit greater anti-carcinogenic activity in the market. Besides, the upregulation of methyltransferase (2.05-fold) in chlorophyll metabolism would also promote betaine lipid biosynthesis and those were earlier found to be upregulated in a diatom under phosphate stress [42]. While other upregulated transcripts in chlorophyll metabolism by combining temperature and pCO 2 variation in this study are involved in heme-binding larger chloroplast, those were earlier found to be regulated under phosphate [43] and nitrate deprivation on diatoms [44].
In green algae and higher plants, three aromatic amino acids (Phe), (Tyr) and (Try) are derived from the common precursors of the Shikimate pathway; those were similar in diatoms based on the gene homology [45]. These amino acids are known to be essential products for human food and pharmacological industries [46]. In diatoms, amino acid biosynthesis metabolism could be strictly regulating by stress conditions [45]; as a consequence, the commitment enzymes to relieve or eliminate the end products are crucial to increasing target compounds [45]. In this study, wide regulation of enzymes associated with phenylalanine (Phe) metabolism was upregulated, suggesting increased (Phe) end products for commercial purposes from S. dohrnii. A similar investigation was carried out earlier on eukaryotes by stress conditions, resulting in increased end product of (Phe) [47]. A previous study on P. tricornutum upon nitrate assimilation [15] revealed an upregulated aspartate aminotransferase enzyme that remodels the amino acid metabolism and its concentration. Consistently, the upregulation of this aspartate aminotransferase (7.76-fold) in this study would be expected to do the same functions under combining temperature and CO 2 elevation, whereas other upregulated transcripts, are important for the precursor of variant amino acids and derived metabolites for the cellular growth, and defense mechanism during stress conditions [48]. Moreover, it was reported that increased concentrations of these three aromatic amino acids resulting in elevation of plant secondary metabolites, could be used as valuable products [49,50]. Similarly, we predict the same results by triggering these three amino acid biosynthesis alterations in S. dohrnii by elevated temperature and pCO 2 concentration.
Our results showed that elevation of temperature along with pCO 2 influenced on amino acids composition. This finding agrees with previous results on the diatom Cylindrotheca fusiformis [51], green algae Chlorella strain MFD-1 and Nannochloropsis strain MFD-2, which showed increasing amino acids accumulation up to 36% and 38% respectively, when temperature alone shifted from 15 to 25 • C [52]. The results of these studies indicate that temperature elevation alone can increase AA composition in algae because temperature has greater influence on algal physiological process than CO 2 variation [51,52]. This could be due to temperature involvement in a high number of metabolic process, i.e., enzymatic reactions and photosynthesis process, that subsequently lead to high growth and metabolic rates [53] of the cell than CO 2 . Consequently, the amount of synthesized protein and AA composition by temperature influenced metabolic pathways higher than what is produced by CO 2 influenced mechanisms [51]. Nevertheless, these findings do not describe why proportionally, AA associated genes are more increased in this study by both elevating temperature and CO 2 changes.
A detailed investigation exclusively on AA metabolism of these individual stress condition will be required to clarify these effects.
Recent investigations on S. marinoi revealed upregulation of ALDH enzyme with associate production of ethanol, drug biotransformation and oxylipin production [4]. Consistently, this enzyme was upregulated (5.48-fold) in this study, by combining elevated temperature and pCO 2 , suggesting possible enhancement of these above-mentioned by-products from S. dohrnii. Yamauchi et al. found that the upregulation of (GAs) in A. thaliana response to a lower temperature was involved in higher ethylene production [50], which is widely used for the chemical industry. Consistently, varied temperatures in this study increased this gibberellin-regulated protein (1.46-fold), suggesting possible development of ethylene production. Likewise, other upregulated transcripts in phenylpropanoid metabolism, including beta-glucosidase (4.99-fold), pectinesterase (4.87-fold), and gelatinase (2.07-fold) would promote the efficiency of glucose production, cell wall modification, and other amino acids compounds of S. dohrnii.
In recent years, the evaluation of genomic approaches targeted crop plants for the high efficiency of flavonoid biosynthesis and associated valuable products. These genetically modified flavonoid pathways in higher plants can be achieved as follows: (i) upregulating transcripts to increase the carbon flow direction towards the target products and (ii) upregulating transcripts involved in competing pathways and identifying exogenous genes towards the biosynthesis of new products [54]. Such edited transcripts evidenced increasing health-promoting pigments [55] and pharmacological products in higher plants [56]. Similarly, upregulating flavonoid biosynthesis transcripts in this study by stress condition expected to favor commercial purposes.
The effect of temperature and CO 2 variation on lipid content and fatty acid composition of diatoms [6] and other microalgae have been previously reported [57] and showed increased accumulation of these contents under these stress conditions. Accordingly, in this study, the total lipid content and productivity increased in cells grown under combining high temperature and CO 2 . Some studies showed that the ratio of these accumulated bioproducts significantly enhanced by the stress conditions, i.e., nutrients [58], temperature [14] and pH variation [16] to direct carbon sources for commercial products. Indeed, being photoautotrophic to accumulate the lipid and fatty acids during stress, diatoms required carbon sources. Because diatoms use polysaccharide, chrysolaminaran (β-1,3-gluvan) during normal conditions as their carbon sink [59], while during stress, the carbon flux reprogrammed towards lipid synthesis and stored as TAGs [60]. Consistently in this study, many upregulated acyl-CoAs by temperature and CO 2 variation could be used for TAG synthesis, resulting in increased lipid content under high temperature and CO 2 conditions. These findings suggest that under these combining temperature and CO 2 conditions , S. dohrnii's main direction of carbon flow is towards the lipid and fatty acid accumulation. Such energy exchange of different diatoms was previously tested by Wu et al. [61] under elevated pCO 2 concentrations. This study observed that diatoms that grew with high CO 2 levels decreased carbon metabolism to save energy to increase growth rates and lipid productivity. This finding suggests that elevated CO 2 in this study can also induce the carbon flux towards lipid biosynthesis, which could be a useful source for wide-scale industrial applications.
The fatty acid profile diatoms are highly rich in monounsaturated acids (MUFAs) to produce biodiesel; nevertheless, properties of biodiesel are determined by carbon chain length and unsaturation extent presence in the fuel [6,13]. This tendency was also identified in this study by upregulating many long-chain fatty acids proteins (which have 20 or more carbon atoms) and downregulating saturating enzymes in fatty acid metabolism. This increasing unsaturated fatty acid and decreasing saturated fatty acids have been observed recently on T. pseudonana, with respect to elevated CO 2 for biodiesel production [6]. However, this trend was not identified in other species [62], showing a unique feature of S. dohrnii for biotechnological application under the combining of elevated temperature and CO 2 concentration. Similarly, the combining effects of changing temperature and CO 2 were also reported on the diatom Nitzschia lecointei, where cellular PUFA content regulated up to 40% [30].
Despite the unclear mechanisms of elevated CO 2 influence on diatom FA metabolism, it was predicted that regulated saturated fatty acids (SFA) in cells grown at high CO 2 environment is due to FA synthesis and accumulation [63]. It has been noted that high CO 2 levels with low pH in algae can severely regulate the cellular pH, thereby distributing tightly regulated homeostasis [64]. Subsequently, regulating SFA synthesis under high CO 2 environment can be a less fluid cell membrane produce mechanism, which is built of short-chained FA, making them less permeable and helping in variation of cellular homeostasis [65]. Besides, temperature can also affect the FA profile of algae [30,66]. Our investigation showed significant temperature related effect on FA metabolic genes of S. dohrnii with CO 2 changes, with a higher amount of upregulated transcripts in HC conditions compared with LC conditions. These results agree with the FA profile of the diatom Chaetoceros muelleri, which reveled variance of PUFA content when temperature changed 5 • C at culture condition [66]. Nevertheless, our results are contradictory to previous investigations, in which PUFA concentrations are commonly inversely proportional to temperature variation [67]. This may be the reason of the length of experiment and allocation of carbon sources within the cells as mentioned above.
In addition, recent years of genetic engineering on diatoms demonstrated increasing lipid content and fatty acid composition using overexpression, and silencing transcripts modification [68]. Especially, stimulating fatty acid-elongase, D5-desaturase, Acyl-ACP, and Acetyl-CoA carboxylase were demonstrated to increase the lipid and fatty acids contents [69,70], while transcripts associated with multi-functional lipase/acyltransferase, pyruvate metabolism were silenced to the increased yield of lipid contents [71,72]. Remarkably, all these transcripts were up and downregulated respectively in this study by combining elevated temperature and pCO 2 , promising the significant potential of S. dohrnii for the biodiesel production without genetic modification. We show here, S. dohrnii grown well with biomass, bioactive compounds, and lipid productivity under combining elevated temperature and pCO 2 condition, showed the unique features of CO 2 biofixation capability; collectively, suggesting the significant potential of this species to be cultivated in both in the laboratory and open bonds using elevating temperature and CO 2 for wide-scale biotechnological applications.

Experimental Setup, Species Collection and Culture Condition
To reduce contamination, accurate methods were used in this study. Briefly, a culture experiment was conducted in 5 L polycarbonate bottles (3 + 3 = 6 nos), which had been soaked for 1 week in 1% Citranox ® and 2 weeks in 1 mol L −1 hydrochloric acid (HPLC) grade. The bottles were rinsed seven times with ultra-pure water (Merck Millipore Corporation) between each soaking step. Finally, the cleaned bottles were air-dried under a clean desk and packed with three polyethylene bags for storage prior to beginning the experiment. This investigation was conducted with globally abundant, and ecologically important bloom-forming marine diatom Skeletonema dohrnii, which was isolated from the Yellow sea [20]. The isolated strain was maintained at Artificial Sea Water medium according to the Aquil recipe [73] at 21 • C with (~400 ppm) 12 h:12 h light: dark cycle, with a light intensity of 100 µmol m −2 s −1 using cool white fluorescent tubes [20]. For the experiment, two different levels of temperature (21 and 25 • C) and pCO 2 culture conditions (400 and 1000 ppm) were set, referred to as LC (400 ppm, 21 • C normal) and HC (1000 ppm, 25 • C stress condition) respectively, were grown in the above-mentioned conditions, and were gently bubbled pCO 2 through filtration (0.22 µm) using CO 2 incubators [26]. The LC and HC conditions represent the present and future (2100 years) atmospheric CO 2 concentrations [74]. The duration of the overall experiment was eight days; nevertheless, in each treatment (LC and HC), triplicate samples were collected during the mid-exponential stage (day 4) for all the physiological analyses.

Seawater Carbonate System and Parameters
During experiment to ensure stable seawater carbonate chemistry (pH variation < 0.05), culture was diluted every 24 h using pre-CO 2 equilibrated medium and the cell density was maintained at no more than 10 µg chlorophyll-a L −1 . The pH variation of culture was monitored every day prior to dilution using pH probe (Mettler Toledo DL 15 Titrator, Stockholm, Sweden), calibrated with National Bureau of Standard (NBS) buffer solutions of pH 7.0 and 10 (Sigma-Aldrich, St Louis, MO, USA). DIC was measured using Shimadzu Total Organic Carbon Analyzer (TOC-5000A, Kyoto, Japan) and total alkalinity (TA) was measured by potentiometric titration. The other parameters of the carbonate system were calculated using CO 2 SYS software based on the known values of DIC, pH, salinity, alkalinity, and temperature [75].

Determination of Growth Rate and Cell Density
The initial cell density was 5 × 10 5 (cells mL −1 ) at the beginning of experiment. In this study, cell density was determined daily using a Beckman Multisizer ™ 3 Coulter Counter ® with a 100 µm aperture. Growth rate µ(d −1 ) was calculated as follows: where, N 0 and N t are at the end and start of the exponential phase of growth, respectively, and t is the duration of the exponential growth phase.

Determination of Pigments
Fifty mL of culture samples, grown under LC and HC conditions, were filtered onto GF/F membrane (25mm, Millipore, MA, USA) and placed overnight in a refrigerator (4 • C) after adding 10 mL of methanol. Consequently, the samples were centrifuged (5000× g, 10 min), and the supernatant was measured using a microplate reader (Thermo, Multiskan, GO, USA). The final concentrations of carotenoid and chlorophyll were calculated following Gao et al. [76].

Determination of Biochemical Analysis
The protein content of S. dohrnii was determined by the Bradford method [77] and the total carbohydrate was measured by the Anthrone-Sulphuric Acid Colorimetric method [78]. Further to this, a modified Folch method [79] was conducted to determine the lipid contents of cells as follows. Approximately, 4 L of algal cells at each sample was harvested by centrifugation (7500× g, 5 min) and dried in an oven at 60 • C to a constant weight. Fifty (mg) of grinded algae powder was transferred to a 10 mL centrifuge tube, with added solutions of 5 mL chloroform-methanol mixture and (V:V = 2:1) and 1 mL NaCl (0.88%) respectively. After being shaken by a multi-tube vortex mixer (DMT-2500, Zhengzhou, China) for 20 min (2000 rpm), the solution was centrifuged (5000× g, 5 min) and the supernatant was removed; subsequently, 5 mL of methanol-water solution (V:V) was added to at 25 • C trifuge tube. After centrifuging and removing the supernatant one more time, the bottom phase was dried using a steady stream of nitrogen supplied by the nitrogen blowing device (Autoscience, MTN-5800, Tianjin, China). Lipid content (pg cell −1 ) was calculated based on the lipid weight and cell number/weight. Lipid productivity (mg L −1 ) was determined by the changes in lipid production for two days of comparison periods.

RNA Isolation
The complete experimental and carried out bioinformatics analysis of this study are shown in ( Figure S4). Initially, 1 L of cultures grown with high and low pCO 2 cells was harvested (3 + 3 = 6 L) by gentle filtration onto 0.8 µm polycarbonate filters (Millipore, MA, USA) during the mid-exponential stage. Filtered cells were immediately flash-frozen with liquid nitrogen and stored at −80 • C until RNA extraction. Total RNA was extracted from frozen cells using Trizol reagents (Invitrogen, Waltham, MA, USA) following the manufacturer's protocol. The purity and concertation of RNA were checked using the NanoDrop-6000 Assay-Kit and an Agilent 2100 Bioanalyzer system (Agilent Technologies, Santa Clara, CA, USA). An approximation of 5 µg of RNA used as input materials for cDNA library construction. The overall unigenes' FPKM distribution and density variation among six samples are given in Figure S1.

Library Construction and Sequencing
As there is no genome DNA available of these species, the construction and sequencing of the cDNA library in this study were performed as follows. Briefly, the poly (A) messenger RNA was isolated from total RNA samples with oligo (dT) attached magnetic beads (Illumina, San Diego, CA, USA) and using divalent cation fragmentation buffer incubated at 94 • C for five minutes, where the mRNA was fragmented into short fragments. The cleaved RNA fragments were then reverse transcribed to the first strand cDNAs by random hexamer primers. Following this, the second strand cDNAs were synthesized to construct the final cDNA library using the Illumina platform (San Diego, CA, USA). After the final repairing process and ligation of the adapter, RNA was amplified by using PCR and purified by QIAquick Gel Extraction Kit (Qiagen, Venlo, Netherlands). The cDNA library of S. dohrnii tissue was sequenced on the Illumina Hiseq 2000 platform (San Diego, CA, USA) with paired-end reads of 90 nucleotides.

Quality Control and de Novo Assembly
Prior to de novo assembly, we removed low-quality raw reads, and PCR duplicates using PrintSeq v0.20.4 (San Diego, CA, USA); adapters were clipped using Trimmomatics v0.38 in [80]. De novo assembly was performed using Trinity v2.8.4 [81], based on the obtained high-quality clean reads. Further to this, using TransDecoder v5.5.0 (https://transdeooder.github.io/), the raw assembly results were filtered for a minimum transcript length of 300 nucleotides and detectable CDS. The CDS of all the similar alternative splice-form (transcripts) sets was selected as a unigenes; the completeness of all the assembly was assessed using BUSCO v3 [82].

Functional Annotation, DETs Detection and Pathway Enrichment Analysis
The resulting transcripts were annotated using InterProscan v5.33-72.0 [83], to NCBI-NT and NCBI-NR [84], Pfam databases [85] Swiss-Prot, KOG, GO, and KEGG [86]. Furthermore, with a (FDR < 0.01 and fold change > 2), the differentially expressed transcripts (DETs) in this study were obtained based on the methods of empirical Bayesian (fold change > 2 and posterior probability of being equivalent expression < 0.05) [87], noisy distribution (fold change > 2 and probability > 0.8) [88], negative binomial distribution [89], and Poisson distribution (fold change >2 and FDR < 0.001) [90]. The metabolic pathway analysis of differentially expressed transcripts was conducted according to the KEGG pathway database (http://www.genome.jp/kegg/) and the functional enrichment analysis was performed using phyper, a function of R software. The p-value calculating formula in the hypergeometric test is: Then, we calculate the false discovery rate (FDR) for each p-value; in general, the terms where FDR is no larger than 0.01 are defined as significantly enriched.

Statistical Analysis
Statistical analysis of all the physiological data was performed with unpaired t-test analysis using GraphPad Prism 6 software. Triplicate samples of each condition were taken into consideration, and comparison was done between groups of cells grown at high temperature and pCO 2 versus low temperature and pCO 2 .

Conclusions
In this study, utilizing combined elevated temperature and pCO 2, the whole-cell transcriptome profile of S. dohrnii was investigated for the first time to evaluate the metabolic potential associated with biotechnological applications. We showed under elevated temperature and pCO 2 conditions, S. dohrnii grew well with higher biomass, bioactive components synthesis and lipid productivity using the unique CO 2 biofixation ability. Our study also provides evidence of a new potential organism for their bioactive compound and natural products, offering a new exploration of promising algae to microalgal-based derived bioactive compounds for wide-scale biotechnological applications.

Conflicts of Interest:
The authors declare no conflict of interest exists in this study.