Increasing Temperature Alters the Effects of Extracellular Copper on Thalassiosira pseudonana Physiology and Transcription

Copper (Cu) is essential for many physiological processes in phytoplankton, including electron-transfer reactions and high-affinity Fe transport systems. However, at high concentrations, Cu can have a toxic effect on phytoplankton. Phytoplankton’s tolerance to certain toxicants in marine ecosystems cannot always be solely attributed to the presence of compounds, as various environmental factors including temperature can also indirectly influence their effects on organisms. In this study, we investigated the effects of different Cu concentrations (0, 19.6, 160, 800, and 8000 nM) on the growth and physiological changes in the centric diatom Thalassiosira pseudonana (CCMP 1335) when simultaneously applied with temperature cultivation of 20 °C and 25 °C. At low (0 nM) and high (8000 nM) Cu concentrations, the growth rate of T. pseudonana was inhibited, though an increase in temperature lessened this inhibition. There were no significant changes in the POC:PON ratio during all of the treatments. However, increasing the temperature significantly decreased the POC: POP, PON: POP and BSi: POP ratios of T. pseudonana. The intracellular Cu content of T. pseudonana varied from 0.13 to 13.28 fg cell−1 in response to increases in ambient Cu concentrations. Lastly, an increase in the Cu concentration decreased the transcriptional expression of CTR (copper transporter), 3Hfcp (photosynthetic protein), and Sit1 (silica shell formation) encoding genes. In conclusion, our results indicated that T. pseudonana can adapt to physiological processes and molecular mechanisms in response to varying Cu concentrations and ambient temperatures.


Introduction
Trace metals are required for numerous physiological processes in phytoplankton [1][2][3]. In particular, copper (Cu) is critical for photosynthesis [4], respiration [5] and the quenching of reactive oxygen species via superoxide dismutase [6]. Its availability can limit or inhibit the growth of phytoplankton in natural environments [1,2,[7][8][9]. Cu concentrations in open oceans range from 0.3-6.0 nM, and it is often a factor in limiting the growth of phytoplankton [10]. However, Cu concentration can reach up to 300 nM in coastal waters, and such high Cu concentrations have been observed to reduce growth rate of marine phytoplankton [10,11], inhibit photosynthesis [12], decrease nitrate uptake [13], hinder silicic acid uptake [14] and alter the community composition of phytoplankton [8]. Most organisms have evolved mechanisms to survive under both optimal and suboptimal conditions in order to acclimatize to dynamic environments [3]. Plasticity is particularly evident in the use of trace metals, wherein elemental substitutions can occur under limiting conditions and excess intracellular accumulation of metals can occur when external concentrations are high [15,16]. Phytoplankton can modulate their growth, elemental stoichiometry and photosynthetic activity to respond to fluctuating levels of Cu concentration [17,18]. Moreover, varying Cu concentrations can alter the community compositions of phytoplankton, because diatoms have a higher tolerance than cyanobacteria and dinoflagellates [8].
Diatoms contribute roughly 20% of the total global primary production [19,20]. Consequently, diatoms were used as optimal-biological models to investigate ecotoxicological and environmental effects arising from stressors on marine ecosystems [21][22][23]. Cu influences phytoplankton growth [24,25] and cellular-elemental stoichiometry, as well as a range of biochemical process such as pigmentation, photosynthesis and biochemical composition [26]. Furthermore, variation in physicochemical factors including temperature, salinity and pH, can impose additional stress on marine organisms, while also influencing metal contaminants, thereby affecting their availability and toxicity to organisms. In particular, seawater temperature is a key factor that influences the toxicity of chemicals on phytoplankton [17]. Chemical toxicity generally varies with temperature [27] and a rise in temperature can promote a higher Cu tolerance in periphytic bacterial communities. These early studies assumed that an acceleration in pollution induced the acquisition of community tolerance via temperatureinduced physiological enhancements [28]. In contrast, Morin et al. [28] demonstrated that increased Cu concentrations induced tolerance in diatom communities, but that tolerance decreases with increasing temperatures. Further, other studies have indicated that the toxicity of cadmium (Cd) generally increases with increasing temperature in the marine diatom Thalassiosira nordenskioeldii [29]. Thus, both the specific metal element and the diatom species should be considered when designing experiments to investigate metal toxicity in diatoms. The Cu transport system is well-understood at the molecular level in Chlamydomonas reinhardtii [3], Saccharomyces cerevisiae [30] and plants [31]. However, the mechanisms of Cu transport and homeostasis in diatoms are not well understood. Cu uptake is biphasic, wherein two types of transporters may be involved: high-affinity systems (CTR) and lowaffinity systems (ZIPs and NRAMP) [32,33]. Previous studies have suggested that the two Cu-transport systems are differentially controlled by low Fe and/or Cu. Other studies have demonstrated that exposure to Cu can induce the expression of photosynthetic genes and silica-shell formation genes. Furthermore, temperature can alter the metabolism of marine organisms, altering their response to heavy metal. Thalassiosira pseudonana is one of the most widely distributed marine diatoms worldwide and a genome is available for this species. In this study, integrative analyses were used to explore the possible response mechanisms of the T. pseudonana physiology and transcription toward the combined effects of Cu exposure and temperature variation.

Materials & Methods
In order to address the combined effects of temperature and Cu concentrations on Thalassiosira pseudonana, cultures of T. pseudonana (CCMP 1335) were exposed to two levels of different temperatures (20 and 25 °C) and five levels of Cu concentrations. The second factor was Cu concentration (0, 19.6, 160, 800 and 8000 nM). Thus, a two by five factorial design was chosen, with triplicate values of each response variable at each treatment. Details of this overview are given in the following sections.

Cell Culture and Experimental Design
Thalassiosira pseudonana (CCMP 1335) were obtained from the Xiamen University Center for Marine Phytoplankton and semi-continuously cultured at 20 °C in constant light condition in the medium Aquil (N: 300 mM) that was filter-sterilized using 0.22-μm of polycarbonate filters (Supplemental Table S1). Cultures were acclimated to a range of Cu concentrations (0, 19.6, 160, 800 and 8000 nM) at temperatures of 20 °C and 25 °C and a light intensity of 90 μmol photons m −2 s −1 for more than 10 generations prior to experimentation. All the glassware used in the experiments were cleaned with 10% of nitric acid, rinsed with ultrapure water and autoclaved before use. A stock solution of 2 g Cu L −1 was prepared in ultrapure water using copper sulfate pentahydrate (sigma). Cultures from the high temperature (25 °C) and the low temperature (20 °C) were in turn exposed to five Cu-concentration treatments. The algal cells were cultured in a 500 mL conical flask in a light incubator for 3 days, and the algal was mixed every 2 h. Then, cultures were collected to analyze the response variables from each of the triplicates in every treatment.

Growth Rates
Cell densities were obtained using a Z2 Coulter Counter. Growth rates were calculated from observed cell numbers using the equation: μ = (LnNt2 − LnNt1)/(t2 − t1), where Nt2 and Nt1 are the cell concentrations of the culture at t2 and t1. Growth rates were calculated from triplicate measurements of cell abundances for each culture.

Pigments and Chlorophyll Fluorescence Parameters
Absorbance spectra of cell extracts were determined at day 3 by filtering 15 mL of cultures through a Whatman GF/F glass-fiber filter (25 mm diameter). The filters were placed in 10 mL centrifuge tubes and 8 mL of 90% acetone was immediately added while avoiding light. The samples were then placed at a temperature of 4 °C overnight to extract the pigments, and then centrifuged at 5000× g for 10 min. The supernatant was removed and measured using a UV-Vis spectrophotometer (SP-752, Shanghai spectrum instruments) at absorbances of 480, 510, 630, 664 and 750 nm. Pigment values were then calculated using the Ritchie method [34].
Chlorophyll fluorescence was determined using a pulse-amplitude modulated fluorometer (water-PAM, Walz, Germany). Two samples were taken from each flask. One sample was incubated in the dark, adapted for 15 min first to relax any photosynthetic activity. The maximal efficiency of PSⅡ photochemistry was then determined as the ratio of variable to maximal chlorophyll fluorescence (Fv/Fm), where Fv = (Fm − F0), and Fm and F0 represent the maximal and minimal fluorescence yield of a dark-adapted suspension, respectively. The chlorophyll fluorescence of the other sample was determined at an actinic light of 96 μmol photons m −2 s −1 . The effective efficiency of PSⅡ photochemistry was then determined as the ratio of variable to effective chlorophyll fluorescence (Fv′/Fm′), where Fv′ = (Fm′ − F0′), and Fm′ and F0′ are the maximal and minimal fluorescence yield of an actinic light suspension, respectively, and the nonphotochemical quenching-NPQ = (Fm − Fm')/Fm'.

Intracellular Copper
Cells were filtered with 1.2 μm of polycarbonate filters, and the cells retained by the filters were washed with 15 mL of a 1 mM EDTA·2Na solution to remove surfaceadsorbed metals and then rinsed with ultrapure water [35]. Samples were then digested for 6 h at 65 °C on a hot plate with 2 mL of ultrapure HNO3. Total and intercellular metal extracts were then filtered before diluting to 10 mL of ultrapure water in a 15 mL volumetric flask, and then analyzed using an Agilen ICP-MS (7700).

Intracellular POC, PON, POP and BSi
To measure particulate organic carbon (POC), nitrogen (PON), Particular organic phosphorus (POP) and biogenic silicon (BSi), each sample was taken at 25 mL to filter through pre-weighed, pre-combusted (450 °C for 6 h) GF/F filters. The filter membranes were then placed in a sealed sample box with concentrated hydrochloric acid and then dried at 65 °C. The result samples were then analyzed with an elemental analyzer (PerkinElmer 2400 Ⅱ). Biogenic silicon (BSi) and particular organic phosphorus (POP) were analyzed following procedures by Passche [36] and Solórzano et al. [37], respectively.

RNA Extraction and Quantitative Real-Time PCR
Cells from the cultures were harvested onto 25 mm polycarbonate filters with 1.2 μm porosity and then flash-frozen in liquid nitrogen and stored at −80 °C until further RNA extraction. Cell lysis and total-cellular RNA isolation were performed using an RNAiso Plus kit (Takara). In the first treatment (hereafter designated as the RNAiso Plus + liquid N2 method), the homogenization step was replaced by a simultaneous flash-freezing of the cells using liquid nitrogen in a pre-chilled mortar, followed by continuous grinding with a pestle while adding 1 mL of the RNAiso Plus reagent to the lyse cells and efficiently extracting RNA. Care was taken in order to prevent the cells from thawing before the RNAiso Plus regent was added into the Sample. As the frozen RNAiso Plus-cell mixture started to thaw, homogenization was continued until the RNAiso Plus was also pulverized, as previously described [38]. The remaining steps were as recommended in the manufacturer RNAiso Plus protocol. The purity of extracted RNA was verified by RT-qPCR to ensure the lack of genomic DNA contamination. Purified total-RNA was quantified spectrophotometrically, and 1000 ng were reverse-transcribed into first-strand cDNAs in 20 μL reaction using the 2×RealStar Power SYBR Mixture (GenStar). Genespecific primers for RT-qPCR, with amplicon sizes in the range of 90-200 bp, were designed based on the available genomic sequence data for T. pseudonana or otherwise based on previous reports [17,39,40]. cDNA was then amplified with RT-qPCR using those gene-specific primers (Table 1) and SYBR Green Supermix. Each 20 μL PCR reaction contained 0.2 μmol L −1 each of the forward and reverse primers, and 2 ng transcribed RNA. RT-qPCR was performed on the iQ5 real-time system with thermal cycling conditions as follows: 95 °C for 10 min followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. Melt curve analysis began at 95 °C for 1 s, followed by 65 °C for 15 s. Transcript abundance was then determined using the 2 (−ΔCt) method [41] from technical duplicates of each biological triplicate. The ΔCt values were calculated after normalization to the template numbers of the endogenous reference gene TpActin_25772.

Data analysis
All data are expressed as mean ± 1 standard deviation (SD). Each treatment was performed in triplicate. t-test and One-way ANOVA (LSD test) were used to compare the different parameters (e.g., growth rate, Chla, Chlc, Carotenoid, Fv/Fm, Yield, NPQ, Cu content, POC: Chla) among the 10 different treatments. We checked the normality, equal variances and independence were met before we conducted the one-way ANOVA. Multiple-way analysis of variance (MANOVA) was used to determine whether there were significant differences among the two factors (temperature and Cu content). All statistical analyses were performed using SPSS 16.0 (SPSS, Inc., Chicago, IL, USA). The significant level was set at p < 0.05.

Growth Rates
Copper addition significantly increased the growth rates of Thalassiosira pseudonana at 20 °C and 25 °C when Cu concentrations were less than 160 nM (Figure 1). However, increasing copper concentration significantly reduced the growth rates of T. pseudonana at 20 °C when Cu concentrations were more than 160 nM (Figure 1). In contrast, increased Cu did not significantly alter the growth rates at 25 °C (Figure 1). In addition, the 5 °C temperature increase also influenced the growth rates of T. pseudonana (Figure 1). The growth rates of T. pseudonana at 25 °C significantly increased relative to T. pseudonana at 20 °C by 41.8%, 32.1%, 26.1%, 39.7% and 49.3% at the five different Cu concentrations (p < 0.05) (Figure 1, Supplemental Table S2).

Pigments and Chlorophyll Fluorescence Parameters
At the higher temperature (25 °C), the concentrations of pigments including Chla, Chlc and carotenoid increased in T. pseudonana (Figure 2A). In addition, Cu concentration and temperature had no effect on Chla content, and Chlc had significantly increased under high temperature at 160 nM and nM Cu concentrations (Figure 2A). After 25 °C, carotenoid content of five copper-concentration groups increased, and there were significant differences, except for 800 nM (Figure 2A). For T. pseudonana grown at 20 °C and 25 °C, a significant difference was observed between Fv/Fm and Fv′/Fm′ with increased Cu concentration. The results indicate that elevated temperature can improve the Fv/Fm and Fv′/Fm′ of T. pseudonana ( Figure 2B). Additionally, the NPQ also increased significantly in the treatment of 8000 nM Cu concentration at 20 °C and 25 °C.

Intracellular Elemental Contents
The intracellular Cu content in T. pseudonana ranged from 0.13 to 13.28 fg cell −1 and increased with increasing Cu-concentration treatments (Figure 3). Increased growth temperature significantly reduced the intracellular Cu content of T. pseudonana, with the exception of the highest Cu concentration. There were no significant changes in the POC:PON ratio throughout all of the treatments ( Figure 4A). The POC:POP, PON:POP and BSi:POP ratios of T. pseudonana grown at 25 °C decreased relative to that grown at 20 °C in all of the treatments (Figure 4). There were no significant changes in POC:POP and PON:POP ratios when Cu was increased at 20 °C. However, increasing Cu significantly decreased the POC:POP and PON:POP ratios at 25 °C ( Figure 4B,C). Increasing Cu decreased the BSi:POP ratio of T. pseudonana in all of the treatments, except that grown at 0 nM Cu ( Figure 4D). The POC:Chla ratio of T. pseudonana grown at 25 °C decreased, relative to that grown at 20 °C, by 8.7%, 12.7%, 5.1%, 15.4% and 1.9% when grown with the five different Cu concentration treatments, respectively ( Figure 5). The POC:Chla ratio decreased when Cu concentrations increased from 0 to 19.6 nM, but the ratio increased when Cu concentrations increased from 19.6 to 8000 nM ( Figure 5).

Gene Expression
Temperature alters the transcription and expression of many genes, including Cu transport, ZIP, photosynthetic, and silica shell formation-related genes. Under low Cu concentrations, CTR_24275 and CTR_9391 expression increased at both 20 °C and 25 °C ( Figure 6A). However, the expression of these two genes differed with respect to temperature, wherein the higher-growth temperature induced the expression of CTR_9391 ( Figure 6A). Increased Cu concentrations inhibited the expression of CTP_263051 and CTP_264357 ( Figure 6A). In particular, the expression of CTP_264357 was significantly inhibited at 25 °C ( Figure 6A). At high-Cu concentrations, the expression of COX_37139 and COX17_264096 were reduced at both 20 °C and 25 °C ( Figure 6A). Likewise, the expression of SCO1_30986 was reduced at both 20 °C and 25 °C under high-Cu concentration treatments ( Figure 6A). Increased temperature will promote the expression of these two genes. Higher-growth temperature induced the expression of FET3_5574, FRE1_11375, and FTR1_268009 during most of the Cu concentration treatments ( Figure 6B). In addition, increased-growth temperature and Cu concentrations induced the expression of NRAMP_9840 ( Figure 6B). The expression of five ZIP genes differed based on temperature and Cu concentrations. For example, the expression levels of ZIP_11826 and ZIP_268980 were inhibited at 25 °C ( Figure 6B). Under high-Cu concentrations, the expression of ZIP_26890 was inhibited when cultures were grown at 20 °C, but induced with growth at 25 °C ( Figure 6B). 3HfcpA and 3HfcpB also exhibited lower expression levels at 25 °C, but the transcription expression of 3HfcpA_1778744 and 3HfcpB_1778746 decreased with increasing Cu concentrations ( Figure 6C). Lastly, higher Cu concentrations inhibited the expression of Sil1_52355302 and Sit1_224002055 ( Figure  6C).  (3HscpA_1778744, 3HscpB_1778746), silicon syntheScheme 1. Sit1_224002055). Data are mean ± standard deviation (n = 3). Different letters at the top of each bar indicate significant differences between different Cu concentration treatments and * indicates significant differences between different temperatures (p < 0.05).

Growth Rates
Here, we investigate the role of Cu and ambient temperatures on the growth and physiological responses of the diatom T. pseudonana. Cu is essential for many physiological processes in phytoplankton, including electron-transfer reactions and highaffinity Fe transport systems, therefore low-Cu concentrations can also limit phytoplankton growth, as observed in the coccolithophore and T. oceanic [7]. Therefore, slight increases in ambient Cu concentrations have been shown to trigger the growth of P. tricornutum [25]. We found that increased Cu concentrations appeared to stimulate the growth of T. pseudonana at lower concentrations, which is consistent with previous studies [43,44]. Taken together, the results of this study and others suggest that Cu is an essential element for phytoplankton. However, Cu can also be potentially toxic at high concentrations. Increased copper concentration significantly reduced the growth rates of T. pseudonana at 20 °C (Figure 1). Jordi, Basterretxea et al. have shown that Cu aerosols can inhibit phytoplankton growth in the Mediterranean sea [45]. Further, high-Cu concentrations can inhibit Phaeocystis cordata growth, both in batch and semi-continuous cultures [46]. T. pseudonana exhibited a reduced growth at 20 °C under high-Cu concentrations, but little inhibition at 25 °C ( Figure 1). Overall, the high-Cu concentration treatments decreased the growth of T. pseudonana.
Earth-system models predict that global warming will result in significant a decline in net-primary production by marine phytoplankton throughout the 21st century; this will be driven by rising temperatures which exceed the limits of thermal tolerances of phytoplankton, in addition to increasing grazing, nutrient limitations and more stratified ocean waters [47]. However, some species grow faster under increased cultivation temperatures [48,49]. Accordingly, our results also show that increased cultivation temperature significantly increased the growth rate of T. pseudonana (Figure 1). In addition to the direct effects of temperature on physiology, it can indirectly alter the effects of Cu concentrations on phytoplankton growth. Our results indicated that declines in growth rates were a result of higher Cu concentrations that were affected by increased temperatures. Consistent with this result, spring lacustrine phototrophic periphyton growing at 18 °C were more sensitive to a herbicide mixture than those growing at higher temperatures [50]. Further, a recent study indicated a decrease in Cu tolerance of winter phototrophic periphyton coincided with temperature changes, where higher temperatures decreased the diversity of diatom species and also exacerbated the community sensitivity towards Cu [28].

Chlorophyll Fluorescence Parameters
Quantum yield analysis indicated that T. pseudonana were characterized by relatively high Fv/Fm, suggesting that T. pseudonana possessed relatively high activity PSⅡ complexes. A slight decrease in Fv/Fm values was observed in T. pseudonana grown with increased Cu concentration and at 20 °C ( Figure 2B). In contrast, no significant differences in Fv/Fm values were observed when grown under different Cu concentrations and at 25 °C, although the Fv/Fm values were generally higher than those for T. pseudonana grown at 20 °C ( Figure 2B). Taken together, these results suggest that the T. pseudonana PS II complex was relatively sensitive to Cu stress, but increased temperatures mitigated these deleterious effects. Thus, increased growth temperatures may help improve PS II recovery of T. pseudonana treat with high-Cu concentration stress.

Intracellular Elemental Contents
Diatom (e.g., T. pseudonana, P. tricornutum and T. weissflogii) exhibit a higher tolerance to increased Cu levels compared to chrysophyceae (e.g., Gephyrocapsa oceanica) dinoflagellates (e.g., Peridinium sp.) and Cyanobacteria (e.g., Synechococcus spp.) [8]. It is possible that copper binds more strongly to the mucopolysaccharide cell wall of the cyanobacteria than it does to the fundamentally different types of cell walls of eukaryotic algae (predominantly cellulose) [8]. The Cu content in P. tricornutum has been shown to range from 1.3 to 12.8 fg cell −1 [25], which coincides with the range reported here for T. psuedonana. T. pseudonana appeared to particularly adapt to higher Cu concentrations and grew well in the presence of 8000 nM Cu ( Figure 1). Concomitantly, their intracellular Cu was far lower than that observed in other species that exhibit Cu toxicity symptoms [51,52].
Only a few studies have explored the phenotypic flexibility in C:N:P of microalgae during exponential growth; therefore, more studies are needed to better understand the effect of micro/macro-nutrient and temperature [53]. The POC:PON ratio is a key parameter in algal growth and is dependent on the variability in the availability of nutrients as well as growth temperature. Reay et al. [54] found that phytoplanktonspecific affinity for nitrate is temperature-dependent below optimal temperatures, and diatom nitrate reductase activity declines rapidly as temperatures exceed optimal growth levels [55]. The higher POC:PON ratios of T. australis at low temperature, compared to high temperature (optimal temperature), suggest that suboptimal temperatures could have limited nitrate uptake relative to carbon fixation [56]. No significant differences in the POC:PON ratio were observed among the cultures in this study, although the POC:PON ratio was decreased with an increasing growth temperature ( Figure 4). However, previous studies have suggested that warming can reduce intracellular POC:PON, POC:POP, PON:POP and BSi: POP ratios [57][58][59] due to higher temperatures promoting cell metabolism and increased carbon consumption [47,60]. Hillebrand et al. [61] observed a trend of decreasing N:P ratio and variability with increasing growth rate in phytoplankton, suggesting that fast-growing phytoplankton in general require more P and also have a relatively low PON:POP ratio compared with slow-growing phytoplankton. Another key parameter that characterizes algal growth is the POC: Chla ratio. The POC: Chla ratios increased at low and high Cu levels or low temperature, which suggested that the cell needed more energy to maintain Cu homeostasis when they are in extreme environments. Geider [62] demonstrated that the relationship between the POC: Chla ratio and growth can be described by a linear model. He also parameterized the dependence of the POC: Chla ratio on temperature and demonstrated that the POC: Chla ratio decreases with increasing growth temperature, similar to the results observed in this study ( Figure 5). POC: Chla could be a useful indicator for ecological and physiological conditions of phytoplankton communities and, additionally, it is a photoacclimation index, which can help us to understand primary productivity, and is highly variable over space and time and among different species, ranging from about 20 to over 200 g g −1 in the ocean. Variation in POC: Chla ratios have been established as a sensitive indicator that quantifies the adjustment of phytoplankton cellular pigment levels to the new demands for photosynthesis, caused by changes in ambient temperature, nutrients, and light conditions [63].

Gene Expression
Gene transcription represents a rapid physiological response to environmental change. Previous research has demonstrated that Cu addition provokes a decreased relative expression of two CTR-like genes (CTR_24275 and CTR_9391), two putative Cutransporting P1B-type ATPases (CTP_263051 and CTP_264357) and two Cu-chaperones (COX11_37139 and COX17_264096) over short time scales [40]. In the aforementioned study, CTR_24275 and COX_264096 mRNA had rebounded to almost to the initial expression levels after 12 h of incubation with Cu addition. These results suggest that the expression of the two genes CTR_24275 and COX_264096 can acclimate to environmental changes. In this study, the relative expression of COX_264096 was only inhibited by increasing copper additions ( Figure 6A). This difference may be explained by coastal diatoms exhibiting a lower sensitivity to environmental conditions than oceanic diatoms. In addition, the genes ToCTR1 and ToCTR2 were down-regulated in T. oceanica in response to increasing Cu concentrations [10]. ZIPs and NARMP are the low affinity Cu transport system that also exhibited altered expression with varying Cu concentrations and temperature. In addition, the expression of respiration-related genes can be altered when acclimating to varying Cu concentrations and temperatures. COX11 is described as an inner mitochondrial membrane integral protein that is essential for cytochrome c oxidase assembly [64,65]. Likewise, COX17 is involved in the assembly of the cytochrome c oxidase protein and is involved in copper donation to both SCO1 and COX11 [66], in the intermembrane space [16,67,68]. Our results indicated that T. pseudonana reduced the absorption of Cu by reducing the expression of COX17_264096, COX11_37139 and SCO1_30986 in order to protect against Cu toxicity to the respiration system. Numerous studies have indicated that warming accelerates the respiratory activity of organisms [47,60]. Consistent with these observations, our results indicated that the expression of COX17_264096, COX11_37139 and SCO1_30986 increased with increased growth temperatures ( Figure 6A). The increased expression of photosynthetic proteins 3HfcpA_1778744 and 3HfcpB_1778746 in T. pseudonana likely occurred to maintain photosynthetic performance in response to environmental stress, which is consistent with other studies [17,67]. Our results indicate that T. pseudonana can grow fast by inducing the expression of 3HfcpA_1778744 and 3HfcpB_1778746 under low Cu concentration conditions ( Figure 6C).

Conclusions
Cu is a micronutrient that is involved in many of the metabolic processes in diatoms including photosynthesis, respiration, antioxidant activities, cell well metabolism, and hormone perception. Consequently, the growth of diatoms is inhibited when they experience Cu-limitation in open oceans. However, excessive Cu concentrations can also physiologically damage diatoms growing in coastal waters. In this study, increased growth temperature altered the effect of excess Cu concentrations on T. pseudonana growth. There is no interactive effects of temperature and Cu concentrations on T. pseudonana pigment content, POC, PON, POP, BSi (or POC: PON, POC: POP, PON: POP and BSi: POP ratios), although each was responsive to individual factors. Nevertheless, the combination of increased temperature and Cu concentrations altered the transcription of genes, including photosynthetic and silicon synthesis genes under different Cu concentration and temperature conditions. Heavy metal accumulation is an intensive contemporary problem for marine environments [69]. Consequently, understanding metal homeostasis is critical to ascertain the effects of heavy metals on marine ecosystem taxa and processes. Cu is a particularly integral trace metal for phytoplankton, and thus, exploring Cu homeostasis, including g intracellular location [70] and Cu-sensing and regulation-processing [71], is necessary to evaluate the combined effect of Cu and environmental stressors on marine taxa and ecosystems. The results reported here provide a framework for better understanding the adaptability of important marine diatoms to increase metal stress and ambient temperatures.