Effects of Iron Valence on the Growth, Photosynthesis, and Fatty Acid Composition of Phaeodactylum tricornutum

: Iron is a limiting factor that controls the phytoplankton biomass of the ocean and plays an important role in the lipid production of microalgae. Elucidating the effects of different iron valences on microalgae is helpful for their commercial production. We investigated the growth, photosynthesis, and fatty acid proﬁle of the model diatom Phaeodactylum tricornutum cultured with depleted Fe, Fe 2+ , Fe 2+ /Fe 3+ , and Fe 3+ . Samples were taken every 24 h for 8 days, and their cell density, photosynthetic pigment content, chlorophyll ﬂuorescence, total fatty acid content, and fatty acid composition were analyzed. The cell densities of the Fe 2+ and Fe 2+ /Fe 3+ groups were signiﬁcantly higher than those of the control and Fe 3+ groups ( p < 0.05). They were 1.26 times and 1.23 times higher than those in the Fe-depleted group. The contents of chlorophyll a and c in the Fe 2+ group were signiﬁcantly higher than those in the Fe-depleted group ( p < 0.05). The chlorophyll ﬂuorescence results show that Fe 2+ enhanced the photosynthesis of P. tricornutum to a greater extent than Fe 3+ . On the eighth day of harvest, Fv/Fm and Y(II) in the Fe 2+ group were 0.672 and 0.476, respectively, being 1.10 and 1.19 times greater than those in the Fe 3+ group and 1.15 and 1.33 times greater than those in the Fe-depleted group, respectively. Compared with the control group, the levels of saturated fatty acids of the Fe 2+ /Fe 3+ and Fe 3+ groups were signiﬁcantly higher ( p < 0.05) at 21.36 ± 1.24% and 21.20 ± 0.13%, respectively. The levels of polyunsaturated fatty acids of the Fe 2+ /Fe 3+ group were signiﬁcantly lower ( p < 0.05) at 29.82 ± 2.75%. Our results show that P. tricornutum exhibited physiological plasticity, including changes in photosynthetic activities and shifts in fatty acid composition, in response to different iron valences and that Fe 2+ was more beneﬁcial to the biomass production of this species than Fe 3+ . These ﬁndings are applicable to the production of biomass and polyunsaturated fatty acids


Introduction
Iron is one of the most abundant elements on Earth, and it is present in multiple forms in aquatic environments, including in solutions and colloids [1].Like the three biogenic elements of nitrogen, phosphorus, and carbon, iron plays an important role in phytoplankton physiology [2].Iron is both an essential element for phytoplankton growth and a major factor controlling marine phytoplankton biomass.It is essential for multiple metabolic pathways, including electron transport, oxygen metabolism, nitrate utilization, and respiration.It is especially important in several photosynthetic activities [3], as the photosynthetic electron transfer of phytoplankton requires the participation of iron [4].Iron is a key component of the cytochrome b 6 -f complex, which is an enzyme present in the thylakoid membrane of chloroplasts that plays an important role in photosynthesis through electron transfer between photosystems I and II [5].Therefore, iron deficiency adversely affects the growth, photosynthesis, and lipid accumulation of phytoplankton [4,6].
Different iron sources have different effects on phytoplankton growth [7].Polat et al. showed that FeCl 3 inhibited the growth of Auxenochlorella protothecoides [8], whereas FeSO 4 helped it obtain the highest biomass, and cells cultured with Fe-EDTA had the highest specific growth rate.They also reported that the fatty acid yield of A. protothecoides cultured with FeCl 3 was higher than that of cells cultured with FeSO 4 and Fe-EDTA.In another study, Rizwan et al. showed that Dunaliella tertiolecta cultured with FeH 8 N 2 O 8 S 2 exhibited good growth but a lower lipid content, whereas cells cultured with Fe-EDTA had a higher lipid content than cells cultured with other iron sources [9].Iron sources also affected the photosynthesis of microalgae.Ferric EDTA was reported to enhance the mean intensity of chlorophyll fluorescence, compared with ferric citrate in Phaeodactylum tricornutum [10].The results were different for Anabaena flosaquae.When the concentration of iron is in the range of 0.1-0.8mg L −1 , iron species have more significant influences on Anabaena bloom compared with iron concentration.The promoting effect was in the order of ferric ammonium citrate >EDTA-Fe> iron ions > iron oxalate, and among the four iron sources, A. flosaquae had the strongest adsorption capacity for iron ions [11].
The diatom P. tricornutum has become a model species in the biotechnology field because it is a biofuel precursor and a recombinant protein expression host due to its biosynthetic capacity and high growth rates [12].This species can grow photoautotrophically by using CO 2 and light as carbon and energy sources, respectively.It can store carbon and energy in the form of neutral lipids, especially triacylglycerols.This diatom has potential as an energy source for biodiesel production.Moreover, it may also be a source of omega-3, long-chain polyunsaturated fatty acids.In addition, as much as 35-40% of the total fatty acid content (TFAC) of P. tricornutum comes from natural accumulation of eicosapentaenoic acid (EPA) [13,14], which makes it a potential candidate for EPA production.Like other phytoplankton species, the growth performance and lipid accumulation of P. tricornutum are usually affected by iron concentrations [15].Hayward reported that iron deficiency led to growth restriction and reduced pigmentation and photochemical efficiency in this species [16].More recently, iron limitation (0.04 µM iron) has been shown to decrease photosystem II content in P. tricornutum, and the decline in the content of cytochrome c550 under iron-limiting conditions is accompanied by a decrease in the binding of this protein to photosystem II, and also the extrinsic PsbO subunit [17].Other experiments have shown that adding an appropriate amount of iron (FeCl 3 •6H 2 O, 12 and 60 mg L −1 ) can promote the growth of Chlorella sorokiniana [18].Chen et al. found that both iron deficiency and iron excess enhanced lipid accumulation in P. tricornutum [19].Zhao et al. reported that the addition of iron at concentrations in the range of 1.2 × 10 −8 to 1.2 × 10 −5 mol L −1 resulted in growth rates of P. tricornutum that were similar between adjacent iron concentration gradients and that the trend between FeCl 3 and FeSO 4 was similar [20].Additionally, Kosakowska et al. found that Fe 3+ affected the qualitative and quantitative composition of P. tricornutum pigments [21].Liang et al. used different iron sources (FeCl 3 , C 6 H 5 FeO 7 , FeSO 4 ) to culture P. tricornutum mutant strains and found that FeSO 4 at 0.5 mg L −1 was optimal for growth.Higher FeSO 4 concentrations (1 mg L −1 ) promoted EPA accumulation [22].
To date, most studies on the effect of iron on P. tricornutum have focused on the iron source type (chelated or unchelated, FeCl 3 or FeSO 4 ) and concentration, especially under iron depletion.The effects of different iron valences on the growth and physiology are rarely reported.P. tricornutum is a promising algae species for biofuel production and usually cultivated with enriched iron.It is important to know which valence form of iron is the best source for the bulk culture of this species.We hypothesize that Fe 2+ is more effective in promoting the growth, photosynthetic yield, and fatty acid accumulation of P. tricornutum.Therefore, we tested the effects of different iron valences (Fe-depleted, Fe 2+ , Fe 2+ /Fe 3+ , and Fe 3+ ) under replete concentrations on the growth, photosynthesis, and fatty acid composition of P. tricornutum.The purpose of this study was to identify the valence of iron that would result in the highest fatty acid accumulation and photosynthetic yield.The results are applicable to the production of biomass and PUFAs.

Algal Strain and Medium Composition
P. tricornutum was obtained from the Microalgae Culture Center, Ocean University of China (Qingdao, China).The alga was maintained in f/2 medium [23] with laboratoryprepared artificial sea water to avoid the potential influence of iron in natural seawater.The artificial seawater [24] was filtered and sterilized to prepare f/2 culture medium without the addition of silicon.Iron was added to the medium in the form of FeSO 4 •7H 2 O or FeCl 3 •6H 2 O at a final concentration of 10 −6 mol L −1 .The experiments were divided into four groups: control (Fe-depleted, Ctrl), ferrous (Fe 2+ ), ferrous/ferric (each half, Fe 2+ /Fe 3+ ), and ferric (Fe 3+ ).Each treatment was replicated four times.

Culture and Growth
Algal cultures in the log phase were harvested via centrifugation at 4000× g for 10 min and washed twice with artificial sea water, and the pellets were then resuspended in 500 mL of medium to generate the inoculum.Cells were grown at 20 • C under cool white fluorescent tubes (100 µmol m −2 s −1 ) with a photoperiod of 12 h light:12 h dark for 8 days, and they were manually shaken periodically.The initial inoculation density of all experimental groups was 10 6 cells mL −1 .Daily sampling for analysis was conducted 1 h after the end of the light cycle.Cell numbers were counted under a microscope using the Neubauer-improved chamber (Marienfeld, Lauda-Konigshofen, Germany), and cell growth was recorded.At the end of the growth experiment, cells were harvested via centrifugation at 4000× g for 10 min and washed twice with artificial seawater.Cell pellets were then prepared for the analysis of photosynthetic pigments and fatty acid composition.

Photosynthesis
Each day, the photosynthesis of 2 mL cultures from each group was measured using a pulse-amplitude-modulated fluorometer (Water-PAM; Walz, Effeltrich, Germany).First, each sample was dark-adapted for 15 min, and then the minimum (F o ) and maximum (F m ) fluorescence were immediately determined as the maximal photochemical efficiency of photosystem II (PSII) (F v /F m ).The actinic light was then turned on to promote photosynthesis, and the following parameters were determined: the effective photochemical efficiency of PSII in the light [Y(II) = yield = F v '/F m ' = (F m ' − F o ')/F m '], the electron transport rate (ETR = Y(II)•PAR•0.84•0.5),and nonphotochemical quenching (NPQ).All the determinations were carried out in darkness.Based on the results of Ralph [25], we derived cardinal points of a rapid light curve (α, E k and ETR max ) to describe the photosynthetic capacity of P. tricornutum, its light adaptation state, and its capacity to tolerate short-term changes in light.

Fatty Acid Analysis
The fatty acid composition of each sample was determined according to our previous report [29].A 0.1 g wet pellet with 0.1 mg glyceryl triheptadecanoate as the internal standard was transesterified using a sequential combination of base and acid catalysts.Fatty acid methyl esters were then analyzed using a gas chromatograph (GC-2010; Shimadzu, Tokyo, Japan) equipped with a fused silica capillary column (Supelco SP-2560: 100 m × 0.25 mm, film thickness 0.20 µm; Bellefonte, PA, USA).Injector and flame ionization detector temperatures were controlled at 260 • C with nitrogen as the carrier gas.The column temperature was programmed from 140 to 240 • C at a rate of 4 • C min −1 .Fatty acids were identified by comparing the relative retention time with the reference standards (Merck, Rahway, NJ, USA).

Statistical Analysis
Experimental results are expressed as the mean ± standard deviation (SD) of the four replicates.The Kolmogorov-Smirnov test and Levene's F-test were used to check normality and heterogeneity of variances.Data with parametric distribution were subjected to one-way analysis of variance.The Duncan multiple-range test [30] or least-significant difference test [31] was performed to identify significant differences between any two means that differed at p < 0.05.All statistical analyses were performed using IBM SPSS statistics version 20.0 (IBM, Armonk, NY, USA).

Cell Density
The cell density clearly increased from 24 to 48 h, during which the specific growth rate of the Fe 2+ group changed the most, reaching 0.278 d −1 .The cell density of the Fe 3+ group was lower than that of the control group until day 5, but it showed a slight increase compared with the control group from day 6 to day 8. On the eight day, the cell densities among the treatments were significantly different.As shown in Figure 1b, the Fe 2+ group had the highest density, followed by the Fe 2+ /Fe 3+ group.Both groups had significantly higher values than the control and Fe 3+ groups (p < 0.05).However, the cell density under the Fe 3+ treatment was not significantly different from that of the control group (p > 0.05).

Fatty Acid Analysis
The fatty acid composition of each sample was determined according to our previous report [29].A 0.1 g wet pellet with 0.1 mg glyceryl triheptadecanoate as the internal standard was transesterified using a sequential combination of base and acid catalysts.Fatty acid methyl esters were then analyzed using a gas chromatograph (GC-2010; Shimadzu, Tokyo, Japan) equipped with a fused silica capillary column (Supelco SP-2560: 100 m × 0.25 mm, film thickness 0.20 µm; Bellefonte, PA, USA).Injector and flame ionization detector temperatures were controlled at 260 °C with nitrogen as the carrier gas.The column temperature was programmed from 140 to 240 °C at a rate of 4 °C min -1 .Fatty acids were identified by comparing the relative retention time with the reference standards (Merck, Rahway, NJ, USA).

Statistical Analysis
Experimental results are expressed as the mean ± standard deviation (SD) of the four replicates.The Kolmogorov-Smirnov test and Levene's F-test were used to check normality and heterogeneity of variances.Data with parametric distribution were subjected to one-way analysis of variance.The Duncan multiple-range test [30] or least-significant difference test [31] was performed to identify significant differences between any two means that differed at p < 0.05.All statistical analyses were performed using IBM SPSS statistics version 20.0 (IBM, Armonk, NY, USA).

Cell Density
The cell density clearly increased from 24 to 48 h, during which the specific growth rate of the Fe 2+ group changed the most, reaching 0.278 d -1 .The cell density of the Fe 3+ group was lower than that of the control group until day 5, but it showed a slight increase compared with the control group from day 6 to day 8. On the eight day, the cell densities among the treatments were significantly different.As shown in Figure 1b, the Fe 2+ group had the highest density, followed by the Fe 2+ /Fe 3+ group.Both groups had significantly higher values than the control and Fe 3+ groups (p < 0.05).However, the cell density under the Fe 3+ treatment was not significantly different from that of the control group (p > 0.05).

Photosynthetic Pigments
Chl-a content was significantly higher in the Fe 2+ and Fe 2+ /Fe 3+ groups than in the control group (p < 0.05), with increases of 26.4% and 19.8% by the end of day 8, respectively (Figure 2).Chl-c content was significantly higher in Fe 2+ -treated cells than in the control group (p < 0.05), but there was no significant difference in Chl-c concentration between the control and Fe 2+ /Fe 3+ or Fe 3+ groups (p > 0.05).The highest carotenoid concentrations were found in the Fe 3+ group and then the Fe 2+ /Fe 3+ group; however, the differences among all groups were not statistically significant (p > 0.05).

Photosynthetic Pigments
Chl-a content was significantly higher in the Fe 2+ and Fe 2+ /Fe 3+ groups than in control group (p < 0.05), with increases of 26.4% and 19.8% by the end of day respectively (Figure 2).Chl-c content was significantly higher in Fe 2+ -treated cells than the control group (p < 0.05), but there was no significant difference in Ch concentration between the control and Fe 2+ /Fe 3+ or Fe 3+ groups (p > 0.05).The high carotenoid concentrations were found in the Fe 3+ group and then the Fe 2+ /Fe 3+ gro however, the differences among all groups were not statistically significant (p > 0.05).

Chl Fluorescence
Both F v /F m and Y(II) increased immediately after iron treatment, and the rate increase was highest in the first 24 h (Figure 3).F v /F m peaked at 48 h in all iron treatm groups (Figure 3a), followed by a downward trend until it leveled off.The F v /F m incre rate of the Fe 2+ group was always higher than that of the other groups during experimental period, and this difference was statistically significant from day 4 until end of the experiment (p < 0.05).The difference between the Fe 2+ /Fe 3+ and Fe 3+ grou began to appear on day 1, peaked on day 3, and then gradually decreased until the e of the experiment.Y(II) showed a similar trend of increasing first and then decreasi with the maximum value appearing on day 3 (Figure 3b).The results clearly show t Fe 2+ and Fe 2+ /Fe 3+ had a significant promoting effect on F v /F m and Y(II), and the effec Fe 2+ was better than that of Fe 2+ /Fe 3+ .Additionally, the NPQ values of the Fe 2+ a Fe 2+ /Fe 3+ groups were significantly lower than that of the control group on day 4 ( 0.05), and the values had decreased by 60.4% and 64.1%, respectively, by the end of experiment (Figure 4).

Chl Fluorescence
Both F v /F m and Y(II) increased immediately after iron treatment, and the rate of increase was highest in the first 24 h (Figure 3).F v /F m peaked at 48 h in all iron treatment groups (Figure 3a), followed by a downward trend until it leveled off.The F v /F m increase rate of the Fe 2+ group was always higher than that of the other groups during the experimental period, and this difference was statistically significant from day 4 until the end of the experiment (p < 0.05).The difference between the Fe 2+ /Fe 3+ and Fe 3+ groups began to appear on day 1, peaked on day 3, and then gradually decreased until the end of the experiment.Y(II) showed a similar trend of increasing first and then decreasing, with the maximum value appearing on day 3 (Figure 3b).The results clearly show that Fe 2+ and Fe 2+ /Fe 3+ had a significant promoting effect on F v /F m and Y(II), and the effect of Fe 2+ was better than that of Fe 2+ /Fe 3+ .Additionally, the NPQ values of the Fe 2+ and Fe 2+ /Fe 3+ groups were significantly lower than that of the control group on day 4 (p < 0.05), and the values had decreased by 60.4% and 64.1%, respectively, by the end of the experiment (Figure 4).The light-response curves of all treatments showed a trend of rising first and th gradually becoming stable (Figure 5).During this process, the advantage of the F group gradually became significant.The saturation irradiance (E k ) of the Fe 2+ /Fe 3+ a Fe 3+ groups was lower than that of the control group (Table 1).The smallest E k occurr in the Fe 3+ group (144.38 ± 11.68 µmol photons m -2 s -1 ), and the largest E k was about times higher in the Fe 2+ group.In addition, the α values of all Fe-treated groups we significantly higher than those of the control group (p < 0.05), and the value of the F group (α = 0.23) was significantly higher than those of the other two iron groups (α 0.21, p < 0.05).The trend for ETR max was similar to that for the α value (Table 1).The light-response curves of all treatments showed a trend of rising first and then gradually becoming stable (Figure 5).During this process, the advantage of the Fe 2+ group gradually became significant.The saturation irradiance (E k ) of the Fe 2+ /Fe 3+ and Fe 3+ groups was lower than that of the control group (Table 1).The smallest E k occurred in the Fe 3+ group (144.38 ± 11.68 µmol photons m −2 s −1 ), and the largest E k was about 1.3 times higher in the Fe 2+ group.In addition, the α values of all Fe-treated groups were significantly higher than those of the control group (p < 0.05), and the value of the Fe 2+ group (α = 0.23) was significantly higher than those of the other two iron groups (α = 0.21, p < 0.05).The trend for ETR max was similar to that for the α value (Table 1
Table 2. Fatty acid composition (% of total fatty acid) and total fatty acid content (TFAC, mg g −1 ) of

Total Fatty Acid Content (TFAC) and Fatty Acid Composition
No significant differences in the TFAC were found among the experimental groups (p < 0.05, Table 2).The fatty acid compositions of all treatments on day 8 were analyzed, and the main constituents were C 14:0 (6.03-6.24%),C 16:0 (13.03-14.42%),C 16:1n-7 (19.80-21.78%),C 18:1n-9 (7.34-7.81%),and EPA (23.86-25.84%) in each treatment (Table 2).There were no significant differences in the C 14:0 , C 16:1n-7 , or C 18:1n-9 percentage contents among the four groups.The C 16:0 percentage content of Fe 2+ /Fe 3+ was significantly higher than that of the control (p < 0.05).The EPA percentage contents of Fe 2+ /Fe 3+ and Fe 3+ groups were found to be significantly lower than those of the control (p < 0.05), whereas there were no significant differences between the Fe 2+ group and the control (p > 0.05).A gradual upward trend was found in the levels of saturated fatty acids (SFAs) with increasing iron valences, and those of the Fe 2+ /Fe 3+ and Fe 2+ groups were significantly higher than those of the control (p < 0.05).The levels of monounsaturated fatty acids (MUFAs) showed a similar trend, with a significantly higher value in the Fe 2+ /Fe 3+ group than in the Fe 2+ group (p < 0.05).However, the levels of polyunsaturated fatty acids (PUFAs) showed the opposite trend, and those of the Fe 2+ /Fe 3+ groups were significantly lower than those of the control (p < 0.05).The ratio of docosahexaenoic acid (DHA) to EPA remained steady and did not differ significantly between the control and experimental groups (Table 2).

Effects of Different Iron Valences on the Growth of P. tricornutum
The growth of P. tricornutum cells treated with Fe 2+ was better than that of cells treated with Fe 3+ , Fe 2+ /Fe 3+ , or no iron.Zhou et al. compared the effects of Fe 2+ and Fe 3+ on the growth of natural mixed algal colonies and single-species colonies of Scenedesmus quadricauda and reported that Fe 2+ exhibited higher biological time-effectiveness than Fe 3+ for algal reproduction [32], which is consistent with our finding.However, our growth results are inconsistent with those of Zhao et al., who reported a similar growth rate between FeCl 3 and FeSO 4 at the same iron concentration [20], most likely because we used artificial seawater instead of natural seawater throughout the experimental period.We also found that, compared to the P. tricornutum cells under the Fe 3+ conditions, the growth of the control group did not show signs of significant iron limitation.A similar result was reported by Liang et al. [22].
On the one hand, P. tricornutum could minimize the need for iron by replacing iron-rich proteins with iron-free proteins to withstand the pressure of iron limitation [20].At the same time, we speculate that certain metal ions in the experiment replaced the function of iron.According to Kothamasi and Kothamasi, cobalt is a possible competitive inhibitor of iron uptake because of its similar size [33].The presence of cobalt in the f/2 medium used in our experiment may have narrowed the advantage of the iron-treated groups over the control group.In addition, Rico et al. studied iron-limited diatom cells and found that copper can replace iron in certain metabolic functions [34], such as some high-affinity iron transport systems.Another response of phytoplankton to iron deprivation is the substitution of Cyt c553 (one Fe atom) with copper-dependent plastocyanin (no Fe atoms) and the partial replacement of ferredoxin (two Fe atoms) with flavodoxin (no Fe atoms) [35].In another study, Peers and Price suggested that selection pressure imposed by Fe limitation resulted in the use of a copper protein for photosynthesis in an oceanic diatom, and this biochemical switch reduced the need for Fe and increased the requirement for copper [36].
The P. tricornutum genome contains at least three clusters of Fe-responsive genes.One of the Fe-regulated clusters contains the gene encoding a copper tyrosinase (TYR1) and a protein of unknown function (ISIP1) [37].Because the role of copper nutrition in iron-limited diatoms has become increasingly apparent [38], we infer that the addition of Cu 2+ in f/2 medium may play a role in replacing iron.

Effects of Different Iron Valences on Photosynthesis of P. tricornutum
Chlorophyll is an important pigment involved in photosynthesis, and its content indicates the stress resistance of plants to a certain extent [39,40].Although iron is not directly involved in the composition of chlorophyll, it is the precursor of chlorophyll synthesis and plays a key role in the activation of various enzymes in the process of chlorophyll synthesis [41][42][43].Therefore, iron deficiency may be sufficient to retard chlorophyll synthesis [44].Carotenoids are important pigments that play a major role in the protection of photosynthetic eukaryotes against photooxidative processes because of their antioxidant activity [45,46].Although the synthesis pathway of carotenoids does not depend on iron, iron reduction may allow their formation to actively compete with chlorophyll synthesis for glycine, thereby reducing the amount of glycine incorporated into carotenes and xanthophylls [47].In the present study, iron significantly increased the production of Chl-a in P. tricornutum cells, and it also affected the content of Chl-c and carotenoids, which is consistent with the above conclusions.At the same time, statistical analysis showed that the enhancing effect of Fe 2+ treatment on Chl-a content was better than that of Fe 3+ .This may be because the pyrrole ring is a chlorophyll precursor [48], and in the dark, ferrous stimulate the formation of porphyrin rings [49][50][51].
Photosynthetic parameters are commonly used to evaluate the effects of environmental stress on microalgae, and they can reflect algal growth status [52].In our study, the values of F v /F m , ETR, and Y(II) of P. tricornutum cells treated with iron were all higher than those of the control group (Figures 3 and 5), indicating that iron has important effects on the photosynthesis of P. tricornutum cells.Similar results were reported for Anacystis nidulans [53], D. tertiolecta [54], and Beta vulgaris leaves [55].NPQ is fluorescence quenching that is caused by heat dissipation, and it reflects the ability of plants to dissipate excess light energy into heat; therefore, it is a photoprotective ability [56].Decreased NPQ was detected in iron-treated P. tricornutum (Figure 4), suggesting an increased ability of cells to dissipate excess energy in the form of heat and fluorescence.This would have a positive protective effect on plant photosynthetic machinery.We also found that the ETR max was higher in the presence of Fe 2+ than in the no-iron or Fe 3+ groups (Figure 5).These results show that Fe 2+ had a certain positive effect on improving the potential maximum relative electron transfer rate of P. tricornutum.Fe 2+ treatment also resulted in greater α and E k values of P. tricornutum cells (Figure 5), which suggests a higher light intensity tolerance.
Virtually all light reactions take place in the thylakoid membranes in oxygenic organisms.These membranes contain two photochemical systems (PSII and PSI), their associated membrane-intrinsic or membrane-bound light-harvesting protein complexes, the cytochrome b 6 -f complex, and ATP synthase [57].The cytochrome b 6 -f complex is the photosynthetic electron transport mediator between PSII and PSI, transferring electrons from water to NADP + [58].Iron supply plays an important role in the structure and function of thylakoid membranes and the photochemical energy conversion process [59,60].Zhao et al. experimentally demonstrated that iron affects the synthesis of the cytochrome b 6 -f complex in P. tricornutum [20].This complex contains heme iron and ironululin protein [61], and iron accomplishes the transfer of electrons in the process of photosynthesis via the mutual conversion of Fe 2+ and Fe 3+ [62].The electron transfer process in photosynthesis can thus be affected by many iron-containing compounds [63,64].In our study, the electron transfer rate of P. tricornutum treated with iron was higher than that not treated with iron, and the enhancing effect of Fe 2+ treatment was better than that seen for the Fe 2+ /Fe 3+and Fe 3+ -treated groups.Therefore, we concluded that Fe 2+ has a better promoting effect on electron transport between the PS II and PS I of P. tricornutum than Fe 3+ , most likely because Fe 3+ interrupted the electron transport between PS II and PS I [65].

Effects of Different Iron Valences on Lipid Accumulation and Fatty Acid Composition of P. tricornutum
The rate of lipid production is an essential feature that must be considered in the large-scale microalgal lipid production process because it takes into account both biomass production and lipid content [5].For sustainable biodiesel production, the process should be able to produce lipids at the highest productivity level with the highest cellular lipid content [66].Liu et al. found that adding Fe 3+ directly to Chlorella vulgaris in the late growth stage did not induce lipid accumulation in cells, but increased cell density; however, reinoculation into Fe 3+ -containing medium after centrifugation induced lipid accumulation in the cells [67].In our study, the method of reseeding after centrifugation was used, and no obvious increase in TFAC content was found (Table 2).This may be because the cultivation conditions necessary for maximizing biomass productivity are not always consistent with those leading to the highest lipid production [5].Savvidou et al. previously showed that the TFAC of Nannochloropsis oceanica increased under iron deficiency compared with the control group containing Fe 3+ [68].In other studies, severe iron stress caused Chlamydomonas reinhardtii cells to accumulate a significant amount of lipids, specifically triacylglycerols, by compromising the growth [69,70].However, in the present study, we clearly showed that iron deficiency did not enhance the TFAC content and even obtained the same growth rate as Fe 3+ , and the Fe 2+ and Fe 2+ /Fe 3+ -containing medium promote growth but not lipid accumulation.We inferred that the response of P. tricornutum to iron deficiency is different from that of the green algae.P. tricornutum has been reported to be highly tolerant to Fe limitation and to be able to maintain steady-state growth at an Fe ion concentration in the range of 10-30 pmol L −1 [30], which suggest P. tricornutum should have the ability to utilize and recycle endocellular ferrous iron.This possibility of ferrous recycling via maintaining a steady lipid content in P. tricornutum should be further elucidated.
Studies have shown that the production of PUFAs in certain microalgae could be qualitatively and quantitatively improved under iron-deficient or low-iron conditions [71].For example, N. oceanica cells cultured under iron-limiting conditions had increased PUFA content [68].However, iron deficiency reduced the production of PUFAs in Chlamydomonas but increased the production of SFAs [69].The results of our study are consistent with the report of [68] (Table 2), and further studies are needed to elucidate the various responses of different species.We also found that, compared with the control group, the protection of EPA and PUFAs by Fe 2+ was better than that provided by Fe 2+ /Fe 3+ and Fe 3+ (Table 2).This was most likely because Fe 2+ can combine with a conserved histidine cluster in fatty acid desaturases to form an enzyme active center, thereby catalyzing the formation of fatty acid double bonds [72,73].

Figure 1 .
Figure 1.(a) Growth curve of Phaeodactylum tricornutum cultured under different iron valence treatments.(b) Growth of P. tricornutum cells under different iron valence treatments on day 8. Column values (mean ± SD of four replicates) with different lowercase letters are significantly different (p < 0.05).

Figure 2 .
Figure 2. Chlorophyll content per cell of Phaeodactylum tricornutum under different iron vale treatments on day 8. Data are means (±SD), n = 4. Significant differences (labeled with aster compared with the Ctrl were determined using the least-significant difference test, where p < is considered statistically significant.

Figure 2 .
Figure 2. Chlorophyll content per cell of Phaeodactylum tricornutum under different iron valence treatments on day 8. Data are means (±SD), n = 4. Significant differences (labeled with asterisk) compared with the Ctrl were determined using the least-significant difference test, where p < 0.05 is considered statistically significant.

Figure 3 .
Figure 3. (a) Day-to-day variation in maximum photochemical efficiency of photosystem II in dark-adapted Phaeodactylum tricornutum.(b) Day-to-day variation in PSII photochemical efficiency (YII) of P. tricornutum under different iron valence treatments.Error bars represent ±SD of four replicates.

Figure 3 .
Figure 3. (a) Day-to-day variation in maximum photochemical efficiency of photosystem II in darkadapted Phaeodactylum tricornutum.(b) Day-to-day variation in PSII photochemical efficiency (YII) of P. tricornutum under different iron valence treatments.Error bars represent ±SD of four replicates.

Figure 3 .
Figure 3. (a) Day-to-day variation in maximum photochemical efficiency of photosystem II in dark-adapted Phaeodactylum tricornutum.(b) Day-to-day variation in PSII photochemical efficien (YII) of P. tricornutum under different iron valence treatments.Error bars represent ±SD of four replicates.

Figure 4 .
Figure 4. Day-to-day variation in nonphotochemical quenching (NPQ) of Phaeodactyl tricornutum under different iron valence treatments.Error bars represent ±SD of four replicates.

Figure 4 .
Figure 4. Day-to-day variation in nonphotochemical quenching (NPQ) of Phaeodactylum tricornutum under different iron valence treatments.Error bars represent ±SD of four replicates.

Table 1 .
The parameters of light-response curve of Phaeodactylum tricornutum under different ir valence treatments on day 8. Values (mean ± SD of four replicate) within the same row w different lowercase letters are significantly different (p < 0.05).

Table 1 .
The parameters of light-response curve of Phaeodactylum tricornutum under different iron valence treatments on day 8. Values (mean ± SD of four replicate) within the same row with different lowercase letters are significantly different (p < 0.05).

Table 2 .
Fatty acid composition (% of total fatty acid) and total fatty acid content (TFAC, mg g −1 ) of Phaeodactylum tricornutum under different iron valence treatments on day 8. Values (mean ± SD of four replicate) within the same row with different lowercase letters are significantly different (p < 0.05).