Gracilariopsis lemaneiformis 981 was originally collected in March 2020 with great growth, more lateral whiskers, purple-red color, no disease, and less miscellaneous algae from aquaculture base of Xiapu (26°65′N, 119°66′E), Fujian Province, China. Then algae were sampled into a bubble chamber containing ice mass and delivered to the laboratory quickly. The tender thalli (approximately 8 cm) with good growth status were washed repeatedly in filtered seawater to remove sediment and epiphytes. Materials were cultivated in an illumination incubator (GXZ-280B, Dongnan Instrument Factory, Ningbo, China) at 23°C with a photoperiod of 12 h light and 12 h darkness, and 40 μmol photons m^–2 s^–1 light intensity as pre-treatment, sterile seawater with a salinity of 25 was conducted as the medium and was renewed every 3 days during this period. After pretreatment for 2 weeks, the thalli were transferred into sterile seawater enriched with Provasoli medium (Provasoli, 1968).

The concentration of L-arginine (Arg; 0.5 mM) was chosen based on a preliminary test with a range of concentrations (0, 0.1, 0.5, and 1 mM) for the most effective and economic concentration on phenotypic changes. Then algae thalli were subjected to 30°C conditions (day/night = 12:12) with or without 0.5 mM Arg treatment for variable time periods. Each concentration was performed in biological triplicate. After treatment, G. lemaneiformis samples were frozen with liquid nitrogen and stored at − 80°C for subsequent experiments.

The effects of Arg on the growth of G. lemaneiformis at 30°C were based on the change of fresh weight (FW) after 7 days. The calculation formula for Relative Growth Rate (RGR) was conducted as follows:

RGR (% d^–1) = ln (W_t/W₀)/t × 100, where W_t represents the final weight after t days, W₀ is the weight at the initial, and t refers to the treatment time (d).

For the determination of Malondialdehyde (MDA), tissues (0.1 g) were homogenized with 1 ml 10% tricarboxylic acid (TCA) solution for each sample, the supernatant was measured using the MDA test kit (Jianchengbio, Nanjing, China) after centrifuged at 8,000 g for 10 min at 4°C.

Measurement of H₂O₂ content and SOD, POD, and CAT enzyme activity determination can be referred to the method described by Sun et al. (2016) and Zhang et al. (2019). For the antioxidant enzyme activity, briefly, approximately 0.1 g fresh seaweeds were powdered with liquid nitrogen, then extracted separately with 1 ml extracting solution from the detection kit. After centrifugation at 8,000 g for 10 min, the absorbance at 560, 470, and 240 nm was recorded and used to calculate, respectively, according to the instructions of the manufacturer (Cominbio, Suzhou, China).

Phycoerythrin (PE) and phycocyanin (PC) were measured according to the method reported with slight modification (Wang et al., 2015). Approximately 0.1 g powder was weighed and extracted with 10 ml phosphate buffer (10 mM, pH 5.5) at 4°C. After centrifugation at 15,000 g for 30 min, the supernatant was used and determined by UV-Vis spectrophotometer (UV-6100, METASH, Shanghai, China).

The contents of PE and PC were calculated using the following formulas:

where V represents the volume of extract solution (ml), and g refers to the weight of algae powder (g).

Proline (Pro) content was determined with the reagents in assay kits (Cominbio, Suzhou, China). Approximate 0.1 g fresh seaweeds were powdered in liquid nitrogen and extracted with 1 ml extracting solution from the detection kit. After centrifugation at 8,000 g for 10 min, the absorbance of the supernatant was measured at 520 nm and used to calculate Pro content according to the instructions of the manufacturer.

An autoanalyzer based on the ninhydrin method was used to quantify the 17 free AA contents according to the Chinese National Standard (GB 5009.124-2016).

Approximately 0.2 g of fresh G. lemaneiformis was weighed into the beaker with 25 ml 0.02 mol/L HCl, the solution was passed through the C18 column after ultrasonic extraction, then the mixture was filtrated through a 0.22 μm ultrafiltration membrane (Millipore, Temecula, CA, United States), and analyzed using AA automatic analyzer (LA8080, Hitachi, Japan). The content of free AA was calculated according to the following equations: X = A × D × V₁/V₂/10,000/F_w. In this study, X, A, D, V₁, V₂, and F_w refer to AA content (g/100 g), quality of AAs in the sample solution (ng), dilution ratio, the volume of extract solution (ml), sample liquid volume (μl), and FW of samples, respectively.

Total RNA was extracted from G. lemaneiformis thalli after 30°C treatment for 24 and 48 h using the RNeasy Plant Mini Kit (Qiagen, Shanghai, China) according to the instructions of the manufacturer. Briefly, the quality and quantification of RNA were separately determined by agarose gel electrophoresis and NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, United States). More accurate RNA quantification was performed by an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, United States) for 28S/18S and RNA integrity number (RIN).

A total of 12 samples were performed by mRNA library preparation kit (MGIEasy, MGI, Shenzhen, China) for the establishment of the cDNA library. The mRNAs were enriched with oligo magnetic adsorption, fragmented, then reversed into double-strand cDNAs, which were linked with sequencing adaptors after purification. The single-stranded circular DNA libraries were conducted by polymerase chain reaction (PCR) amplification and then sequenced according to 50 bp single-end sequencing method by using BGISEQ-500 platform performed by Wuhan Genomic Institution (BGI, Shenzhen, China).

RNA-seq, reads filtering, and unigene annotations were performed by using SOAPnuke software (BGI, Shenzhen, China). Adaptor sequences and reads with unknown bases > 10% and low-quality reads (reads with a mass value below 15 which accounted for more than 50% of the total base number) (Cock et al., 2010) were removed to obtain clean reads. Clean reads were then mapped to a mixed reference transcriptome assembled by using HISAT (Kim et al., 2015) and Bowtie2 (Langmead and Salzberg, 2012), respectively. The expression levels of the genes and transcripts were represented by fragments per kilobase per million fragments (FPKM) mapped values and were calculated by RSEM software (Li and Dewey, 2011). The DEGs were identified using the DESeq analysis (Wang et al., 2010). The criteria of adjusted p-value ≤ 0.001 and log₂ (fold change) ≥ 1 were set as the threshold for DEGs. Q value was obtained from p-value and was adjusted using KOBAS 2.0 according to the Beyer-Hardwick (BH) method (Mao et al., 2005). The significance enrichment levels of gene oncology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were set at Q value < 0.05.

Ten genes were chosen randomly from the transcriptomic results for validation by qRT-PCR. The total RNA of each sample was prepared as described in Section “RNA Isolation and cDNA Library Construction.” First-strand cDNA was synthesized from 1 μg of total RNA by reverse transcription using the HiScript II Q RT SuperMix for qPCR (Vazyme, Nanjing, China) according to the protocol of the manufacturer. qPCR was performed on an ABI7500 qRT-PCR System (Applied Biosystems, Foster City, CA, United States) and each reaction contained 10 μl ChamQ Universal SYBR qPCR Master Mix (Vazyme, Nanjing, China), 0.1 μg cDNA, and 7.5 pmol of each gene-specific primer in a final volume of 20 μl. Three independent biological replicates were used for analyses and the relative expression level of genes was determined by the 2^–ΔΔCT method using 18S rRNA as the respective reference gene. Primers for qPCR are listed in Supplementary Table 1.

The data are expressed as the mean and SD. Statistical difference of data was evaluated with one-way ANOVA and Duncan’s multiple range tests at the p < 0.05 level of significance using SPSS (version 19.0, SPSS Institute, Chicago, IL, United States).

To gain insights into the possible function of Arg on G. lemaneiformis under HT stress, we analyzed the effects of different Arg concentrations (0, 0.1, 0.5, and 1 mM) on morphological changes of algal thalli under HT stress. Under prolonged stress, it showed that the algae thalli in the control treatment (CK; 0 mM Arg) group exhibited an obvious accelerated yellowing phenotype as compared with those of 0.5 and 1 mM Arg treatment, however, no obvious morphological change in 0.1 mM Arg treatment as compared with the CK group (Supplementary Figure 1A) was observed. Then, we measured the content of PBP present in the red algae, such as PE and PC. Results showed that G. lemaneiformis with 0.5 and 1 mM Arg treatment contained higher content of PBP as compared with the CK group at 7 d under HT stress treatment, while 0.1 mM Arg treatment was comparable and remained unchanged to the CK group. Both 0.5 and 1 mM Arg treatments were efficient in alleviating HT stress-triggered decrease of algal PBP, but there was no significant difference between the two concentrations (Supplementary Figures 1B,C). Thus, 0.5 mM arginine was chosen in the following experiments for economic consideration.

Exogenous application of 0.5 mM Arg was significantly decreased MDA content by 56.01% as compared with the CK group after 7 days of HT treatment (Figure 1A). Importantly, the RGR furtherly indicated that 0.5 mM Arg promoted the growth of G. lemaneiformis as compared with those without Arg treatment under HT stress (Figure 1B) since the RGR of G. lemaneiformis under HT stress (30°C) was significantly decreased by 20.45% compared with those cultivated under optimal temperature (23°C; Supplementary Figure 2). Collectively, these results indicated that the application of Arg could remarkably alleviate the membrane lipid peroxidation and the growth decline of G. lemaneiformis under HT stress.

To explore the involvement of Arg in ROS scavenging against HT stress, we firstly measured the content of H₂O₂ in thalli under HT with or without Arg treatments after 1, 3, and 7 days of incubation. H₂O₂ content was continued to accumulate in algal thalli without Arg treatment, showing a significant increase at day 7, whereas the H₂O₂ content in the Arg group was declining in 7 days (Figure 2A).

We then measured the antioxidant enzyme (CAT, POD, and SOD) activities. CAT activity was remarkably activated in the Arg-treated group, reached a maximum at 3 days, with an obviously 59.11% increase in contrast to control (Figure 2C). Superoxide dismutase activity was slightly activated in the Arg-treated group as compared with the CK group at 3 days under HT stress (Figure 2B). But Arg treatment had no significant promotion in POD activity relative to CK under HT stress (Figure 2D). These results indicated that Arg could upregulate the activities of antioxidant enzymes, especially activities of CAT, thus effectively protecting the algal cells from heat damage.

To reveal the mechanism of the effect of Arg in enhancing G. lemaneiformis tolerance to HT stress, the algal thalli from both CK and Arg groups after HT treatment at 24 and 48 h were subjected to RNA-seq analysis. RNA samples were tested using the BGISEQ platform, generated 21.58 M clean reads average in each sample, and obtained a total of 23,479 unigenes, the detailed information of each sample was listed in Supplementary Table 2. The total mapping genome and gene ratio were average at 94.30 and 96.42%, respectively. Then all the unigenes were blasted against public databases (such as TFs, PRG, GO, KEGG, and NR) for identification of gene function, of which approximately 73.34% of unigenes annotated at least in one database except TFs and PRG (Supplementary Table 3).

Principal Component Analysis (PCA) was performed to describe transcriptome differences between Arg and CK groups under HT stress at 24 and 48 h. It suggested that Arg changed the transcriptome of G. lemaneiformis under HT stress, especially at 48 h (Figures 3A,B). The volcano plots of DEGs between Arg-treated and CK groups revealed a total of 1,414 unigenes (629 upregulated and 785 downregulated) at 24 h, and a total of 3,825 unigenes (2,219 upregulated and 1,606 downregulated) at 48 h (Supplementary Figure 3). To validate the reliability of the RNA-Seq results, 10 unigenes were randomly selected for RT-PCR analysis. The expression patterns of the 10 unigenes were generally in agreement with the RNA-Seq results (Supplementary Figure 4).

To analyze the functions of the DEGs, GO and KEGG enrichment analyses were performed. GO analysis showed that DEGs at 24 h were enriched in membrane part, membrane, POD activity, oxidoreductase activity, an integral component of membrane, an intrinsic component of membrane, and antioxidant activity, etc. However, DEGs at 48 h were associated with the structural constituent of the cytoskeleton, generation of precursor metabolites and energy, thylakoid, and photosynthesis (Supplementary Figure 5). Kyoto Encyclopedia of Genes and Genomes enrichment analysis showed that there was only one remarkably enriched pathway at 24 h, the basal transcription factors (TFs; Figure 3C). While the top three significantly enriched KEGG pathways at 48 h were related to metabolisms and were mostly involved in AA metabolism, energy metabolism, global and overview maps, and carbohydrate metabolism (glycolysis, pyruvate metabolism, ascorbate, aldarate metabolism, and alactose metabolism; Figure 3D).

During stress, ROS, such as superoxide radicals (O₂^–), singlet oxygen (¹O₂), hydrogen peroxide (H₂O₂), and hydroxyl radicals (OH^∙), are produced, which exert a detrimental impact on plant cells and result in cellular damage and death (Thorpe et al., 2013). However, plants have evolved corresponding mechanisms to eliminate excessive ROS via ROS scavenging enzymes, such as SOD, CAT, POD, APX, MDAR, TRX, and PRXR (Anderson, 2002; Pérez-Pérez et al., 2009; Smirnoff, 2010; Das and Roychoudhury, 2014). In our study, DEG analysis showed that Arg treatment enhanced the transcript levels of most genes encoding those protective enzymes, such as SOD, APX, MDAR, PRXR, and TRX (Figure 4). Moreover, exogenous Arg reduced the content of H₂O₂ and elevated the activities of antioxidant enzymes, such as SOD and CAT in G. lemaneiformis under HT stress (Figure 2). However, it was an interesting phenomenon that the H₂O₂ content of the Arg-treated group was increased at 1 day. SPD could be catalyzed into 4-amino-butanal to produce H₂O₂ by PAO, and the expression of genes encoding PAO could be upregulated by Arg, which may be the reason for the H₂O₂ increase (Figure 5 and Supplementary Table 4). These results indicated the positive effects of Arg in regulating thermotolerance via upregulation of the antioxidant enzymes under HT stress.

Heat shock proteins, as vital cellular chaperones, play a pivotal role in conferring abiotic stress tolerance. Moreover, HSP can detoxify ROS by positively regulating the antioxidant enzymes system and improving the stability of cell membranes (ul Haq et al., 2019). At least six types of HSPs, i.e., HSP100, HSP90, HSP70, HSP60, HSP40, and small HSPs (sHSPs), have been identified in higher plants. HSP90 and HSP70 can interact with some co-chaperone proteins, such as HSP20 which regulates its activities and aid in the folding of specific substrate proteins to ameliorate heat damage and maintain cell homeostasis (Wang et al., 2004; Guo et al., 2020). HSP40 is an obligatory co-chaperone partner of HSP70 families, which provides HSP70s with the activities and functional specificity to ensure their chaperone function effectively (Craig et al., 2006). Many pieces of evidence have indicated the importance of HSPs and sHSPs in regulating high-temperature tolerance in algae, such as in red algae G. lemaneiformis, Cyanidioschyzon merolae, and green algae Chlamydomonas reinhardtii (Gu et al., 2012; Kobayashi et al., 2014). Our previous study showed that HSPs were involved in the regulation of phytohormone-mediated heat tolerance in response to heat stress (Wang et al., 2017). In the present study, most of the genes encoding HSP90, HSP70, HSP40, and sHSPs were induced by Arg application, particularly at 48 h, while many of those genes were not significantly expressed at 24 h. We speculate that these HSPs respond to the Arg-induced HT tolerance mainly at 48 h (Table 1), which is generally consistent with the expression pattern of other possibly HT-responsive DEGs analyzed in our study. Therefore, it indicated the crucial role of Arg in regulating thermotolerance via upregulation of HSPs when G. lemaneiformis responds to heat stress.

In the signaling network from the stress signals perception to the downstream expression of stress-responsive genes, TFs play an essential role (Hussain et al., 2011). TFs have also emerged as powerful tools for the manipulation of complex metabolic pathways (Hussain et al., 2011). Members of the HSF, MYB, ERF, bZIP, and WRKY TF families have already been implicated in the regulation of stress responses in algae (Strenkert et al., 2011; Ritter et al., 2014; Liang and Jiang, 2017; Ji et al., 2018). Notably, the MYB group of TFs renders tolerance to elevated temperatures, and this high tolerance is associated with enhanced AA metabolism (El-kereamy et al., 2012; Xing et al., 2021). In the present study, Arg remarkably enhanced several TF encoding genes, covering 12 families, such as HSF, C2H2, C2C2-GATA, bHLH, MADS, MYB, and bZIP (Table 2). Particularly, the transcript level of genes encoding MYB showed considerable enhancement both at 24 and 48 h in response to HT stress. It indicated that the upregulation of TFs may contribute to Arg-induced heat tolerance in G. lemaneiformis.

Polyamines, such ad PUT, SPM, and SPD, as second messengers, manifest a multifunction defensive role in conferring plant abiotic stresses tolerance (Liu et al., 2007; Hasanuzzaman et al., 2019). PA metabolism may lead to increased GABA accumulation, which also mediates plant responses to several environmental stresses (Al-Quraan, 2015). The synthesis of PA in plants requires Arg and Met metabolism (de Oliveira et al., 2018). The present study showed that Arg treatment enhanced the transcript levels of most PA synthesis-related genes (Figure 5 and Supplementary Table 4). This is similar to the results of several studies that the overexpression of PA biosynthetic genes provides increased stress tolerance in the plant (Liu et al., 2015). Moreover, exogenous application of Arg significantly upregulated the genes involved in the catabolism of PA to produce GABA (Figure 5). Then GABA from this step converts into succinate, which further enters the TCA cycle to generate more energy for stress acclimation. Collectively, these data indicated that exogenous Arg could promote PA and GABA syntheses that effectively improved the adaptive heat stress response in G. lemaneiformis.

Besides PA and GABA, Pro has also been proved that it can improve the resistance of plants under abiotic stress (Sairam et al., 2002; Haghighi et al., 2020; Wang X. et al., 2020). Algae and higher plants tend to accumulate Pro in response to salt and heavy metal stresses (Zhang et al., 2008; Cheng et al., 2014). In the present study, Arg treatments significantly increased Pro content, as compared to the non-treated group under high-temperature stress (Supplementary Figure 6) and upregulated the expression of the Pro synthesis genes (P5CR genes; Figure 5). Besides Pro, some other AAs, such as His, Ser, Met, Gly, alanine (Ala), Asp, and valine (Val), were also reported to play vital roles in the protection of plants against various stresses (Di Martino et al., 2003; Sharma and Dietz, 2006; Liu and Lin, 2020). In the present study, we measured the contents of 17 free AAs between the Arg and CK treatments under HT stress for 3 d. The levels of most AAs were increased, particularly, Asp, Thr, Arg, Ala, and His, which were markedly increased as compared to CK treatments under HT stress. This is similar to the transcriptome data that Arg upregulated the expression of AA synthesis genes encoding Gly, Cys, Ser, Pro, Met, His, Thr, Val, Asp, and Ala in the Arg-treated group, especially at 48 h (Figure 6 and Table 3). Collectively, these results indicated that Arg might promote algae HT tolerance by upregulating the synthesis of Pro and other stress-related AAs in G. lemaneiformis.

Glycolysis, TCA cycle, and oxidative phosphorylation are cellular interconnected energy metabolic pathways, which generate carbon and nitrogen skeletons, the electron transfer chain (ETC), and provide ATP for plant and algae growth or defense (Fernie et al., 2004; Fan et al., 2018). Genes related to glycolysis and the TCA cycle, such as HK, PFK, GAPDH, PK, SDH, and MDH, were reported to be involved in the tolerance to temperature stress (Dong et al., 2020; Wang L. et al., 2020). As reported, HT stress downregulated energy metabolisms, such as glycolysis/gluconeogenesis, TCA cycle, oxidative phosphorylation, and starch and sucrose metabolism (Wang et al., 2018). In the present study, Arg remarkably enhanced the energy metabolism at 48 h by activating the expression of genes encoding numerous supplementary enzymes in carbon and nitrogen metabolism (e.g., pyruvate kinase, PDH, CSY, IDH, and 2-OGDH) and electron transport chain (e.g., NADH dehydrogenase, cytochrome c reductase, cytochrome c oxidase, and ATPase; Figures 7, 8), which were consistent with the enhanced RGR of G. lemaneiformis under HT stress (Figure 1B and Supplementary Figure 1).