Pavlova gyrans as a potential source of essential fatty acids, sterols and pigments: culture under low temperature

Haptophytes are emerging as sustainable sources of high-value metabolites such as polyunsaturated fatty acids (PUFAs). The goal of this work was to isolate a native haptophyte from the southwestern Atlantic coast and to evaluate the effect of low-temperature stress on the concentration of PUFAs, sterols, and pigments during its cultivation. The strain was identified as Pavlova gyrans. Cultures were carried out in a photobioreactor for 10 days at 20 °C (Control), lowering the culture temperature to 10 °C at the beginning of the stationary growth phase (LTS). The lipid content of the biomass represented 13% of the dry weight, neutral lipids being the main lipid fraction. Under LTS, biomass increased significantly, reaching a value of 305 mg L−1. The same effect was observed with PUFAs content, which represented 41.3% of total fatty acids. The most abundant omega-3 (ω3) and omega-6 (ω6) fatty acids were eicosapentaenoic (EPA) and docosapentaenoic (DPA), respectively. PUFAs concentration under LTS reached 6.65 mg L−1 of which 3.59 mg L−1 corresponded to ω3 and 3.06 mg L−1 to ω6. EPA reached a value of 2.82 mg L−1 while that for DPA was of 1.59 mg L−1. The maximum content of phytosterols was obtained during the exponential growth phase. The most abundant phytosterol was 24-ethylcholesta-5,22-dien-3β-ol, which represented ≈41–49% of the free sterol fraction, with a volumetric concentration of ≈320 μg L−1. Finally, under LTS pigments reached a value of ≈700 μg L−1. These results open the way for further progress towards the commercial and profitable cultivation of P. gyrans for food and aquaculture applications.


Introduction
Microalgae have received widespread attention due to their ability to synthesize metabolites (Maltsev and Maltseva 2021). Some marine microalgae are natural sources of polyunsaturated fatty acids (PUFAs), in particular omega-3 (ω3) and omega-6 (ω6) fatty acids, essential nutrients in the diet of vertebrates, including fish and humans (Tocher 2015). Certain ω3 PUFAs, such as eicosapentaenoic (EPA,20:5) and docosahexaenoic acids (DHA,22:6), have beneficial effects on the development of the neural system and therapeutic properties for cardiovascular diseases, hypertension and autoimmune disorders (Tocher et al. 2019;Remize et al. 2021). ω6 PUFAs, including arachidonic (ARA, 20:4) and docosapentaenoic (DPA,22:5) acids, are required by marine fish larvae for physiological functions such as survival and growth (Basford et al. 2020). At present, the main sources of PUFAs are marine fish oils; 1 3 however, their availability is currently limited due to overfishing and global warming. The high demand for PUFAs is due to their potential applications in the aquaculture, nutraceutical and pharmaceutical industries (Oliver et al. 2020; Barta et al. 2021). In this context, alternative sources of PUFAs, such as microalgae, are a promising option as they are the main producers of these fatty acids in natural marine food webs (Oliver et al. 2020).
Microalgal biomass is a sustainable source of high value-added metabolites. However, its chemical composition varies among different phylogenetic groups, making the concept of phylogenetic species particularly useful (Bongiovani et al. 2014;Jónasdóttir 2019;Taipale et al. 2020;Ansari et al. 2021). The synthesis of metabolites can be modified by manipulating the growth conditions Remize et al. 2021). Under stress conditions such as nutrient deprivation, certain microalgal species accumulate neutral lipids, mainly in the form of triacylglycerides (TAGs), which can serve as a reservoir for PUFAs (Cohen et al. 2000;Zienkiewicz et al. 2016). Temperature plays a key role in the synthesis of PUFAs. Low temperatures reduce membrane fluidity, leading to increased synthesis of unsaturated fatty acids as an adaptive response of cells to environmental changes. In this regard, Balakrishnan and Shanmugam (2021) reported that although low temperature greatly increased ω3 fatty acids in Isochrysis sp., biomass yield decreased. It is therefore of interest to search for culture strategies that allow increased PUFA synthesis at low temperatures without affecting the growth of the culture and biomass production.
Haptophyta, also known as Prymnesiophyta, are unicellular microalgae characterized by the presence of a filamentous appendage, called haptonema, between two flagella (Edvardsen and Medlin 2007). Some of these species, such as Isochrysis galbana and Pavlova viridis, have been proposed as a source of PUFAs for food and aquafeed. Unlike green algae and diatoms, these species do not have recalcitrant cell walls, making them easier for oysters and farmed fish to digest and for processing biomass for commercial purposes (Martínez-Fernández et al. 2004;Tibbetts et al. 2020). In addition, some Pavlova species have been reported to be a source of phytosterols and pigments, with potential applications in the food, aquafeed, and pharmaceutical industries (Ahmed et al. 2015;Robertson et al. 2015). It is therefore important to explore the potential of haptophytes as a sustainable alternative source of PUFAs and other high-value metabolites. In light of the above, the aim of this study was to isolate a native haptophyte from the southwestern Atlantic coast and study the effect of low-temperature stress on biomass and PUFAs, sterol and pigment content.

Strain isolation and growth under laboratory conditions
The marine haptophyte Pavlova gyrans Butcher was isolated in summer from samples collected with a plankton net from the Bahía Blanca Estuary (38° 45' S, 62° 22' W), Argentina. The isolation was performed under optical microscope by the single cell method using centrifugal micropipettes (Andersen and Kawachi 2005). Non-axenic stock cultures were established in f/2 medium and maintained at the Laboratorio de Estudios Básicos y Biotecnológicos en Algas (LEBBA), CERZOS-CONICET, Bahía Blanca, at 20 °C under 70 µmol photons m −2 s −1 light intensity provided by fluorescent lamps in a 12:12-h light/dark cycle. In order to obtain preliminary information about the strain growth under laboratory conditions, small-scale cultures were performed both at 20 °C and at 10 °C. Cultures were carried out in triplicate (n = 3) in Erlenmeyer flasks of 250 mL containing 100 mL of f/2 medium and an initial inoculum of 50,000 cells mL −1 for 10 days. Except for temperature, culture conditions were the same as those detailed above. Triplicate daily samples were collected to determine growth parameters. Final biomass was harvested by centrifugation at 3000 rpm for 10 min and stored at − 80 °C until use.

Molecular identification and phylogenetic analysis
Approximately 250 mL of culture (n = 3) were concentrated by centrifugation and kept at -80 °C. DNA was extracted using the DNeasy Plant Mini Kit (Qiagen). PCR amplifications of the nuclear 18S rDNA were performed with the forward primers: 18S-F53 (TTG TCT CAA AGA TTA AGC CATG) reported by Olivares-Rubio et al. (2017), SSUNS5a (AAC TTA AAG AAA TTG ACG GAAG), and SSUNS3a (GCA AGT CTG GTG CCA GCA GCC) reported by Fawley et al. (2005), and the reverse primer 1528R (TGA TCC TTC TGC AGG TTC ACC TAC ) reported by Medlin et al. (1988). PCR conditions included an initial denaturing step at 95 °C for 2 min, followed by 35 cycles at 94 °C for 40 s, 52 °C for 50 s, 72 °C for 1 min and a final extension step at 72 °C for 7 min. Amplified products were sequenced by Sanger Sequencing at Macrogen. Homologous sequences from other species of the Pavlovophyceae were obtained from GenBank databases using BLASTn searches. Sequences were aligned with AliView v.1.26 (Larsson 2014) and IQtree2 (Nguyen et al. 2015) was used to run Maximum Likelihood phylogenetic analyses with 1,000 ultrafast bootstrap replicates. The 18S rDNA sequence described here was deposited in GenBank (GenBank accession number OQ582087).

Cultures in photobioreactor
A cylindrical photobioreactor (Figmay 15L, FIGMAY S.R.L, Córdoba, Argentina) 0.25 m in diameter and 0.70 m in height, equipped with a thermostatic bath (Alpha RA 8, Lauda, Germany) for temperature regulation and a gas mixing system for air and CO 2 injection (SMG-01, Electronics service of the Centro Científico Tecnológico Bahía Blanca, Argentina) were used to obtain biomass. Cells at the exponential growth phase, acclimated at 20 °C, obtained through progressive scaling-up of stock cultures (from 10 mL to 1L) were used as inoculum. Autotrophic batch cultures were carried out during 10 days in f/2 medium to a final volume of 5 L using an initial inoculum of 300,000 cells mL −1 . Two temperature conditions were assayed: maintaining culture temperature at 20 ± 1 °C throughout the entire experiment (Control); and lowering the culture temperature to 10 ± 1 °C at the beginning of the stationary growth phase (on day 7) (low-temperature stress; LTS). Cultures were carried out in triplicate (n = 3) under continuously air bubbled with 1% CO 2 and stirred at 9 rpm. LED lamps provided 60 μmol photons m −2 s −1 of light intensity under a cycle of 12:12 h light/darkness. Finally, biomass was harvested by centrifugation at 3500 rpm for 15 min, washed with phosphate saline buffer, lyophilized and stored at − 80 °C.

Determination of growth parameters and dry weight
Cell density (CD) (cells mL −1 ) was determined daily by counting triplicate samples in a Sedgwick-Rafter chamber. Specific maximum growth rate (μ max ; div day −1 ) was estimated during the exponential growth phase by a least squares fit to a straight line of logarithmically transformed data (Guillard 1973). This occurred between days 2 and 4 (at 20 °C) and days 4 and 6 (at 10 °C) in the case of laboratory scale experiments; for photobioreactor experiments values were obtained considering values from day 0 to day 2. Doubling time (DT) (hours) was calculated as DT: (1/μ max ) × 24. Biomass, expressed as dry weight (DW) (mg L −1 ), was estimated gravimetrically. Daily duplicate samples of 10 mL were filtered through pre-conditioned and pre-weighed Whatman GF/C filters (1.2 µm), washed with 10 mL of distilled water, dried for 12 h at 70 °C in an oven, cooled in a desiccator and weighed until constant weight (DW).

Chlorophyll a autofluorescence
In vivo detection of chlorophyll a autofluorescence (Chl a) was carried out according to Martín et al. (2018). Triplicate samples of 5 mL were taken on day 10. Chl a fluorescence intensity (Chl a-FI), expressed as arbitrary units (au), was measured in a spectrofluorometer (Schimadzu RF-5301PC) using an excitation/emission wavelength of 430/600-750 nm read at 680 nm.

Chlorophyll and carotenoid quantification
Pigment kinetics was determined through the spectrophotometric method according to Scodelaro Bilbao et al. (2016). For each growth condition, duplicate samples of 40 mL were taken at days 0, 3, 6 and 9 and harvested by centrifugation at 3500 rpm for 10 min. The pellet was resuspended in methanol and kept in darkness. Chlorophyll a (Chl a) chlorophyll c (Chl c) and carotenoids were estimated by measuring the absorbance of methanolic extracts at different wavelengths and considering the equations proposed by Jeffrey and Humphrey (1975) and Strickland and Parsons (1968) for chlorophylls and carotenoid, respectively. Analyses were carried out in duplicate.

Nile Red staining and measurement
To determine the presence of neutral lipids, the lipophilic fluorescent dye Nile Red (NR) (9-diethylamino-5H-benzo[a] phenoxazine-5-one, Sigma) was used. To that end, triplicated samples were collected and analyzed every day through the addition of the fluorescent dye (1 μL mL −1 , stock solution 1 mg mL −1 , prepared in acetone) (Martin et al. 2016). After a five minutes incubation, NR fluorescence was detected by a RF-5301 PC Schimatzu spectrofluorometer (excitation wavelength = 480 nm, emission wavelength = 580 ± 10 nm). The resulting relative fluorescence intensity (RFI) was expressed in arbitrary units (a.u.f.) after considering microalgal cells autofluorescence and self-fluorescence of NR.

Total lipids and lipid fractionation
Total lipid (TL) determination was carried out from 200 mg of lyophilized and macerated biomass obtained at the end of the experiment (day 10). Lipid extraction was performed as previously described (Folch et al. 1957). In addition, 30 mg of total lipid extracts (day 10) were fractionated into neutral lipids (NL), phospholipids (PL), and glycolipids (GL) using Sep-Pak silica cartridges, according to Popovich et al. (2012). The cartridges were activated with chloroform and each lipid fraction was consecutively eluted with chloroform:acetic acid (9:1 v/v) (NL), acetone:methanol (9:1 v/v) (GL), and methanol (PL). Each fraction was recovered and dried under nitrogen atmosphere. Finally, total lipid content and each lipid fraction were determined gravimetrically Lipid analyses were performed in duplicate.

Chromatographic analysis of sterols and fatty acids
The fatty acid methyl ester (FAME) profiles of duplicated total lipid (TL) and NL, GL and PL fractions from P. gyrans (obtained by duplicate) were analyzed at the end of the experiment (day 10) by means of an HP Agilent 4890D gas chromatograph (Hewlett Packard, USA) equipped with an SP2560 capillary column (100 m, 0.25 mm, and 0.2 μm) (Supelco Inc., US) using a Supelco FAME 10 mix 37 standard (CRM47885, Supelco Inc., USA) as an external standard. Samples were injected twice. Data were analyzed using HP3398A GC Chemstation Software (Hewlett Packard, USA) . FAME quantification (μg L −1 or mg mg −1 biomass) was performed employing customized external calibration solutions coupled to gas chromatography.

Statistical analysis
The results are expressed as mean values ± standard deviation (SD) (n = 3). Statistical analyses were performed using the INFOSTAT software (Di Rienzo et al. 2018). The T-Student test was applied to detect significance differences between conditions (p < 0.05).

Strain identification and growth under laboratory conditions
The isolated strain was identified by means of morphological characters as Pavlova gyrans Butcher (1952) (Fig. 1). Cells were observed under the optical microscope to be solitary and very motile ( Fig. 1a-c). Their shape was variable, being mostly oval or elongated (3-5 μm wide and 4-8 μm long), with two flagella (Fig. 1a) two yellow-green chloroplasts (Fig. 1b) and a conspicuous stigma in the apical region (Fig. 1c). Hyaline bodies were usually present in the posterior region (Fig. 1b). Part of the hyaline content of the cells was extruded, forming pseudopodia-like structures. (Fig. 1c). The morphological characterization was consistent with the molecular identification. The resulting phylogenetic tree based on the 18S rDNA gene sequence is shown in Fig. 2. The gene sequences obtained were 100% identical to those of other P. gyrans strains. Pavlova gyrans sequences formed a monophyletic group, which was sister to sequences from P. pinguis and P. granifera, described as Clade 3 by Bendif et al. (2011). Other species of Pavlova were associated with those of Diacronema (Fig. 2). The other two genera in the Pavlovaceae family are represented by the single species in the genus Exanthemachrysis, E. gayraliae, and the type species of the genus Rebecca, R. salina (Fig. 2). Figure 3a shows the growth of the isolated microalgal strain in laboratory cultures performed at 20 °C and 10 °C, under laboratory conditions. The graph clearly shows that, at 20 °C, P. gyrans reached the stationary growth phase on day 7, in which aggregates of cells embedded in mucilage were observed (Fig. 1d). The maximum specific growth rate (μ max ) and doubling time values obtained at 20 °C were 1.01 ± 0.17 div day −1 and 24.31 ± 4.1 h, respectively. In contrast, P. gyrans (3) HH = (18 ∶ 1 + ΣPUFA)∕(12 ∶ 0 + 14 ∶ 0 + 16 ∶ 0) growth at 10 °C was slower than that observed at 20 °C, showing μ max and DT values of 0.58 ± 0.1 div day −1 and 41.8 ± 6.4 h.

Growth and biomass parameters in the photobioreactor
Once the robustness of P. gyrans growth under laboratory conditions was proven, cultures were carried out in a cylindrical photobioreactor. Figure 3b presents the cell density and dry weight (DW) kinetics of P. gyrans throughout 10 days of culture under control and LTS conditions. When P. gyrans was grown under control conditions, the stationary phase occurred at day 7. However, under low temperature stress the stationary phase was reached at day 9, resulting in a higher cell density value compared to the control condition (p < 0.05) ( Fig. 3b and Supplementary Table S1). Moreover, chlorophyll a autofluorescence at day 10 was significantly higher under LTS conditions than under the control condition (p < 0.05) (Supplementary Table S1). No lag phase or decline phase was observed. Growth parameters such as μ max and DT were similar to those obtained under laboratory scale culture and no statistically significant differences were observed between the control and LTS conditions (p > 0.05), (Supplementary Table S1). Biomass increased in line with the lower culture temperature (Fig. 3b). Under the LTS condition, final biomass was significantly higher than that observed under the control condition (p < 0.05), reaching a maximum value of 305 mg L −1 ( Fig. 3b and Supplementary Table S1).

Neutral lipid accumulation
The accumulation kinetics of neutral lipids was evaluated using Nile Red fluorescence. As can be seen in Fig. 4a, spectrofluorometric measurements showed that neutral lipids increased from day 5 under both control and LTS conditions. However, no significant differences were detected between the two experimental conditions (p > 0.05). Maximum values were reached by the end of the experiment (day 10) and represented an 85-fold increase over the values obtained for day 0. Figure 4b shows oil bodies in P. gyrans at the stationary growth phase under low-temperature stress. TAGs at the end of culture were similar under LTS conditions as in the control, with a value ≈6 mg L −1 (p < 0.05) (Supplementary Table S1).

Total lipid and lipid fractions
Total lipid (TL) content of the final biomass represented ≈12-13% of the dry weight. Neutral lipids (NL) was the main lipid fraction, representing ≈63% of TL for both temperature conditions, followed by glycolipids (GL) (≈32% of TL) and phospholipids (PL) (≈5% of TL) (Fig. 4c) (p > 0.05). However, when the results were expressed in terms of volumetric concentration, TL increased significantly up to 38.31 mg L −1 (p < 0.05) at low temperatures, representing a ≈ 40% increase over the value obtained for the control (Fig. 4d). In addition, NL, GL and PL fractions significantly increased up to 24.41, 11.64 and 2.26 mg L −1 , respectively under low-temperature stress (p < 0.05) (Fig. 4d).  Table 1 shows the fatty acid (FAs) profiles obtained from the TL fraction of P. gyrans via gas chromatography. Results are expressed as % of total FAs. The main saturated fatty acids (SFAs) were 14:0 and 16:0 for both experimental conditions. Lowering culture temperature did not affect the relative percentage of 16:0 FA (p > 0.05), whereas a marked decrease was detected in the 14:0 FA (p < 0.05), which decreased from 17.53% (control) to 14.79% (LTS condition) ( Table 1). The relative amount of SFAs (in % total FAs) significantly decreased under the LTS condition (p < 0.05), whereas monounsaturated fatty acids (MUFAs) remained at ≈28% for both experimental conditions assayed (Table 1), 16:1 being the most abundant FA of this fraction (≈25% of the total FAs) ( Table 1). Polyunsaturated fatty acids (PUFAs) were the dominant lipid class in P. gyrans both under control and LTS conditions. In addition, these FAs showed a marked increase from 35.71% (± 4.6) (control) to 41.6% (± 1.5) of total FAs under low-temperature stress (p < 0.05) ( Table 1). The relative content of ω3 FAs was not significant (p > 0.05) (Fig. 5 a) whereas the variation detected for ω6 FA (% of total FAs) increased from 16.09% (± 1.45) to 18.65% (± 0.93) in response to low temperature (p < 0.05) (Fig. 5b). The main PUFAs detected after low temperature exposure were EPA and DPA, which accounted for 80% and 50% of total ω3 and ω6, respectively (Fig. 5 c, d).
The FA profile of NL, GL and PL was also analyzed (Table 1). For the NL fraction, low temperature significantly changed the content of 14:0 and 15:0 SFAs (p < 0.05) ( Table 1). The main MUFA (16:1 FA) was not affected by low temperature stress (p > 0.05). Except for DHA, significant variations were observed in PUFA content in response to low-temperature stress ( Table 1). The ω6 FAs increased significantly under the LTS condition (p < 0.05) (Fig. 5b). On the other hand, the GL fraction showed a slight decrease in the percentage of PUFAs triggered by low temperature, at the expense of an increase in SFAs and MUFAs (p < 0.05) (Table 1). Finally, MUFAs significantly increased while PUFAs and SFAs decreased in the PL fraction when P. gyrans was exposed to LTS (p < 0.05) ( Table 1). In terms of whole lipid fractions, the most abundant ω3 PUFAs detected under LTS was EPA, accounting for more than 80% of the total ω3 FAs (Fig. 5c). The main ω6 PUFAs were docosapentaenoic acid (DPA) for the NL and PL fractions and linoleic acid (LA) for the GL fraction, accounting in each case for more than 40% of the total ω6 (Fig. 5d). The analysis of TL showed a significant increase in PUFAs under low-temperature stress from 3711 (± 474) μg L −1 to 6653 (± 253 μg L −1 ) (p < 0.05) (Fig. 6). Furthermore, both ω3 and ω6 FAs increased markedly under the LTS condition (p < 0.05) (Fig. 6, inserted). This effect was mainly due to the increase in the concentration of EPA and DPA. The NL fraction was shown to be the main contributor to the effect observed in the concentration of PUFAs ( Supplementary  Fig. S1). Table 2 shows values corresponding to the hypocholesterolemic/hypercholesterolemic ratio (HH), atherogenic index (AI) and thrombogenic index (TI) obtained from P. gyrans grown under the two experimental conditions. AI and TI decreased under low-temperature stress in the TL fraction while HH increased (p < 0.05), as also observed for the NL fraction. Furthermore, the HH of the LTS condition was lower than that of the control in the case of GL and PL fractions (p < 0.05). AI and TI showed an increase over the lipid composition of the GL fraction under LTS (p < 0.05). Growth is expressed as cell density (CD) (cells mL −1 ) and biomass as dry weight (DW) (mg L −1 ). Values are expressed as mean ± standard deviation (n = 3). * Stands for statistically significant differences (p < 0.05)

Sterols
The effect of low temperature on P. gyrans sterols was studied (Fig. 7 c, Table 3). From day 0 to day 7, growth conditions were the same for both control and LTS conditions and values are therefore shown as an average between the two (ACS; average of control and stress conditions) (Fig. 7). The maximum value was observed on day 4, when cells exhibited 3.500-4000 μg L −1 sterols under both experimental conditions. No significant differences were detected on day 9 (p > 0.05) (Fig. 7a). Figure 7c shows the relative percentage of free and esterified sterols throughout the days of culture. The highest percentage of esterified sterols was found early on, at day 2 of culture. Towards the end of the culture (day 9), a significant increase in the percentage of free sterols due to low-temperature stress was observed (p < 0.05) (Fig. 7c). Sterol profiling via gas chromatography for P. gyrans extracts at day 9 revealed that Values represent means ± standard deviation (n = 3) for C and LTS growth conditions. * Indicates statistically significant differences between conditions (p < 0.05) 24-ethylcholesta-5,22-dien-3β-ol and 24-ethylcholesta-5,24dien-3β-ol were the most abundant molecular species identified (Table 3). Low-temperature stress did not significantly affect 24-ethylcholesta-5,22-dien-3β-ol (p > 0.05), which represented ≈41-49% of free sterols with a volumetric concentration of ≈320 μg L −1 . Furthermore, 24-ethylcholesta-5,24-dien-3β-ol showed a significant reduction due to low temperature, from 77 to 73 μg L −1 (p < 0.05) under control and LTS conditions, respectively.

Pigments
Finally, the effect of low-temperature stress on chlorophylls (Chl) a, c and carotenoids in P. gyrans was studied. As with sterols, the results were analyzed from day 0 to day 7 (ACS), and on day 9 as control and LTS conditions (Fig. 7). Figure 7b shows that Chl a and carotenoids reached their maximum value (in µg L −1 ) on day 6, increasing 21-fold and 13-fold, respectively, with respect to the starting condition (day 0). Low-temperature stress (day 9) induced a marked decrease in all the primary and secondary pigments compared to the control (1.43, 1.93 and 1.48-fold for Chl a, c and carotenoids, respectively). Relative contents of Chls a, c and carotenoids, expressed as a % of total pigment (Fig. 7 d), was similar at the beginning of the experiment (day 0). However, before low-temperature stress (days 0-6) the relative content of Chl a increased at the expense of a decrease in Chl c with the days of culture, whereas the percentage of carotenoids did not show significant changes. Finally, the relative content of the analyzed pigments did not show significant changes associated with low temperature stress (p > 0.05) (Fig. 7d).

Discussion
In order to ensure the production of high value-added metabolites of commercial interest, the selection of robust microalga strains with favorable growth and biomass Table 1 Fatty acid profiles (% total FAs) of Pavlova gyrans total lipid and lipid fraction extracts. The profiles obtained under control (C) and stress (SLT) conditions at the stationary growth phase are shown.
Lipid fractions: NL (neutral lipids); GL (glycolipids); PL (phospholipids). Summation (Ʃ) of saturated (SFAs); monounsaturated (MUFAs); and polyunsaturated (PUFAs) fatty acids Values are mean ± standard deviation (n = 3). * Indicates statistically significant differences (p < 0.05) between conditions production characteristics is of great importance. While specific abiotic factors such as low temperature increase PUFA content, this condition negatively affects microalgal growth (Aussant et al. 2018). One of the main goals of this study was to assess the effect of lowering the culture temperature at the stationary growth phase to maintain high biomass levels. This strategy did not affect the maximum rate growth (μ ≈ 1 div day −1 ). Furthermore, the growth rates obtained for P. gyrans were similar to those reported for other Pavlova species (0.7-0.9 div day −1 ) and for commercially used species, such as the haptophyte I. galbana and the cryptophyte Rhodomonas sp. (μ ≈ 1 div day −1 ), under optimal growth conditions (Ponis et al. 2006;Oostlander et al. 2020;Zarrinmehr et al. 2020;. Low-temperature stress induced an increase in biomass in P. gyrans. This effect might be a consequence of low temperature, as it was shown to be a crucial factor to prevent biomass loss caused by respiration during dark periods (Cheng and He 2014). This finding is in consonance with that reported for Cylindrotheca closterium, a diatom proposed as an alternative PUFAs source and exposed to similar stress conditions , although the final biomass was higher for P. gyrans (305 mg L −1 ) than that obtained for C. closterium (226 mg L −1 ). Thus, P. gyrans represents a highly promising strain in terms of growth and biomass. Analysis of the kinetics of neutral lipid accumulation showed the expected trend, with higher accumulation towards the stationary growth phase (Tan and Lee 2016;Schüler et al. 2017;Ananthi et al. 2021). According to Brindhadevi et al. (2021), changes in culture temperature can positively or negatively affect the total lipid content in microalgae. However, in this study, low-temperature stress did not affect total lipid content, reaching similar yields under both control and stress conditions (≈13% of dry weight). This latter value is lower than that reported for other Pavlova, Diacronema and Isochrysis species (17.9-30% DW) (Martínez-Fernández et al. 2006;Fradique et al. 2013;Syazwina et al. 2022).
Neutral lipids were quantitatively the main fraction found in P. gyrans, followed by glycolipid and phospholipid fractions. This is in line with the data reported for I. galbana, where neutral lipids accounted for 50-60% of total lipids (Zhu et al.1997). In addition, neither neutral lipid nor glycolipid or phospholipid fraction distribution was affected by low-temperature stress in P. gyrans. However, due to the increase in biomass obtained under the stress condition, an increase of around 40% in total lipids and lipid fractions (mg L −1 ) was obtained.
Omega 3 and omega 6 PUFAs from microalgae are highly valuable to the aquaculture and nutraceutical industries, as they can replace those obtained from fishmeal or fish oil (Barta et al. 2021;Maltsev et al. 2021). The biosynthesis of PUFAs by microalgae is species-specific and depends on the culture conditions (Barta el al. 2021), temperature being a crucial factor for biomass and PUFA synthesis. In agreement with previous studies reporting Pavlova species as good sources of PUFAs (Milke et al. 2008;Guihéneuf et al. 2015;Haas et al. 2016), P. gyrans showed high levels of PUFAs, representing 41% of total FAs under low-temperature exposure. EPA and DPA were the most abundant ω3 and ω6 fatty acids, respectively. Similar results were reported by Tibbets et al. (2020) in Pavlova sp. 459 cultured under 20 °C for 10 days, where PUFAs accounted for 57% of total FAs, and EPA y DPA were also the main FAs. The relative content of PUFA found in P. gyrans was comparable to that reported for the diatoms Phaeodactylum tricornutum and C. closterium (≈21-26% of total FAs) grown under similar temperature conditions (Jian and Gao 2004;Almeyda et al. 2020). In addition, total PUFAs and EPA and ARA contents in P. gyrans were higher than those reported for anchovy fish oils (≈30%, 9% and 1% of total FAs, respectively) (Kaya and Turan 2008). In terms of lipid fractions, PUFAs formed part of both polar lipids and neutral lipids in P. gyrans. EPA was the main ω3 PUFA found in all lipid fractions, accounting for more than 80% of total ω3 under low temperature. DPA was the major ω6 FA, except for the GL fraction, reaching gyrans grown under both conditions tested. Values are presented as mean ± standard deviation (n = 3). * Indicates significant differences (p < 0.05) between growth conditions Table 2 Health lipid indices of Pavlova gyrans. The table shows the values of atherogenic index (AI), thrombogenic index (TI) and hypocholesterolemic/hypercholesterolemic ratio (HH) for total lipid (TL), neutral lipid (NL), glycolipid (GL) and phospholypid (PL) frac-tions calculated from fatty acid composition under Control (C) and Low temperature stress conditions (LTS). Values are presented as mean ± standard deviation (n = 3). *Indicates significant differences (p < 0.05) between conditions for each lipid class PUFAs. An interesting fact emerging from this work is that low temperatures lead to higher relative PUFA content since they are also increased as part of the neutral lipid fraction. On the other hand, the polar lipid fractions showed a decrease in the relative percentage of PUFA at expense of that of MUFA. These results differ from those obtained in C. closterium, where low-temperature stress led to an increase in PUFAs in the polar lipid fractions and to a decrease in the neutral fraction . The increase in PUFAs observed in the neutral lipid fraction may be due to an increase in their synthesis and subsequent storage in TAGs. Alternatively, these PUFAs could be remodeled from membrane lipids into TAGs (Schüler et al. 2017). The molecular mechanism underlying PUFA synthesis in P. gyrans should be considered in future research. Low-temperature stress increased not only PUFAs content in P. gyrans but also biomass. This effective culture strategy for increasing PUFAs could possibly be applied to other Pavlova strains. PUFAs obtained from P. gyrans under low-temperature stress was compared with that reported for C. closterium grown under similar stress conditions and with other haptophyte species (Table 4). Although the relative amount of PUFAs found in P. gyrans (% of total FAs) was twice that of C. closterium, PUFAs content was higher in C. closterium (5630 mg 100 g −1 biomass) . Moreover, in this latter strain, EPA was the main ω3 found, as in P. gyrans, whereas the most abundant ω6 fatty acids were ARA in C. closterium and DPA in P. gyrans. Comparison with other haptophytes shows that Pavlova species, including P. gyrans, are a good source of EPA and DPA ω6 (Xu et al. 2008;Tibbetts et al. 2020), while I. galbana has a high DHA content (Bonfanti et al. 2018;Matos et al. 2021). The ω3/ω6 ratio obtained for P. gyrans (1.17) was lower than that reported for the other haptophytes species detailed before and similar to that obtained for C. closterium.
Omega 3 and omega 6 PUFAs play an important role in preventing cardiovascular and nervous system disorders, inflammatory reactions, along with certain types of cancer (Kapoor et al. 2021). Consequently, an imbalance in the Σω6/Σω3 ratio is associated with health risks. Oils enriched in ω3 PUFA, mainly DHA and EPA, are considered healthy food supplements especially in those countries correspond to ACS: average between conditions control (C) and low temperature stress (LTS) (ACS) for 2, 4, 7 and 0, 3, 6 days of culture for sterols and pigment, respectively; and the values corresponding to average for day 9 under C and LTS conditions. In all cases, they are represented as mean ± standard deviation (n = 3) where fish consumption is low (Gharajeh et al. 2020;. A good source of ω3 PUFA must contain at least 0.03 g of ALA per 100 g of product or > 0.04 g of EPA + DHA per 100 g of product ). In addition, according to Gladyshev and Sushchik (2019) a daily personal consumption of 0.5-1.0 g of EPA + DHA is recommended by the World Health Organization to obtain health benefits. In this study, P. gyrans was shown to have 0.012 g of ALA per 100 g biomass and 1.16 g of EPA + DHA per 100 g biomass, under the LTS condition. P. gyrans can therefore be considered a good source of ω3 PUFAs. In addition, microalgal PUFAs may have a direct effect on blood cholesterol concentration, reducing the risk associated with atherosclerosis, coronary thrombosis, and blood vessel clots (Aussant et al. 2018). In this regard, antiatherogenic (AI), antithrombogenic (TI) and hypercholesterolemic (HH) indices are good indicators of lipid nutritional quality (Gharajeh et al. 2020;. When AI and TI values are lower than 1 there is a reduced risk of coronary heart disease, whereas a relatively high HH is desirable for a healthy diet (Pleadin et al. 2017;Chen et al. 2020). The reduced AI and IT indices found in the present study show that low-temperature stress improved the nutritional quality of P. gyrans FAs at the same time as increasing the HH ratio compared to the control condition. Based on human health indices, P. gyrans can be considered a potential candidate for use as a dietary supplement. There is evidence of the use of haptophyte biomass as a source of PUFAs in food. For example, I. galbana has been added as a functional ingredient to sweet biscuit, pasta, tomato puree and yogurt (Gouveia et al. 2008;Fradique et al. 2013;Gheysen et al. 2019;Matos el al. 2021).
In addition to nutraceutical applications, PUFAs from microalgae have an important role in aquafeed. At present, the main source of long chain (LC)-PUFAs for aquafeed is marine fish oil. A possible alternative to replace it could be plant oils (Ayisi et al. 2019). However, these oils have large amounts of C 18:2 ω6 (LA) FA while they are deficient in essential ω3 and ω6 LC-PUFAs for fish and other aquatic organisms. Some of these, such as DPA, ARA, EPA and DHA, must be provided through the diet as the cannot be synthesized from shorter-chain and less unsaturated precursors. Moreover, they have a crucial role in promoting growth and survival of marine fish larvae (Milke et al. 2008;Basford et al. 2020). In addition, the dietary availability of EPA and DHA has implications for fish growth and survival (Santigosa et al. 2021). P. gyrans can therefore be considered a good source for aquafeed owing to its FAs composition in terms of its ω3 and ω6 LC-PUFAs. In this way, some reports show the use of Pavlova species for aquaculture feeds. Unlike other microalga species used in cultured fish feed, such as Chlorella or Nannochloropsis, the cell walls of Pavlova species are less recalcitrant and do not depend on cell disruption for proper digestion by fish (Tibbetts et al. 2020). Thus, dried biomass can be added directly to fish feed. Diets combining P. pinguis and Pavlova sp. 459 with the diatom Chaetoceros muelleri resulted in much higher growth rates in scallops than those obtained with C. muelleri and P. lutheri (Milke et al. 2004(Milke et al. , 2008. These authors concluded that the main differences in scallop growth were attributed to the higher level of DPA detected in Pavlova sp. 459 and P. pinguis, compared to that in P. lutheri. Together these results show the potential of P. gyrans as a source of PUFAs for aquaculture feeds. Microalgae are also considered a potential and sustainable source of phytosterols, health-promoting components with cholesterol-lowering effects and neuromodulatory, neuroprotective, antiatherogenic, anti-inflammatory, and anticancer activities (Randhir et al. 2020). The sterol content of P. gyrans under low temperature (5.95 μg mg −1 ) was higher than that reported for P. salina and I. galbana (< 2.5 μg mg −1 ) and lower than that of P. lutheri (26.05 μg mg −1 ) (Ahmed et al. 2015). The sterol composition of microalgae is more stable than that of fatty acid, although it is variable among species. Some species have a single predominant sterol while others have a complex mixture of sterols (Xu et al. 2008;Lu et al. 2014). In this work, eleven phytosterols were identified, with 24-ethylcholesta-5,22-dien-3β-ol being predominant (≈40-50% of total sterols). This finding is in agreement with that reported by Volkman et al. (1990) for Pavlova species, where 24-ethylcholesta-5,22-dien-3β-ol represented between 30-50% of the total sterols. Environmental variations, such as temperature and nutrients, can significantly influence the production of sterols in microalgae (Volkman 2016). In this regard, sterol kinetics in P. gyrans showed a reduction in total sterol volumetric concentration as a function of culture age. Maximum values (≈3.6 mg L −1 ) were obtained at the exponential growth phase while values decreased when P. gyrans reached the stationary growth phase. These findings are the converse of those reported for the diatom C. closterium, where total sterols increased with culture age under similar temperature conditions ). According to Xu et al. (2008), changes in sterol content and composition depend on a particular sterol response. In this sense, these authors reported a decrease in 24-ethylcholesta-5,22E-dien3-β-ol content within the days of culture in P. lutheri, whereas (Lu et al. 2014) found an increase in cholest-5-en-3β-ol in Nannochloropsis oceanica. It can therefore be assumed that the P. gyrans and C. Closterium species have different cell adaptation and survival responses, resulting in a different sterol composition, with 24-ethylcholesta-5,22-dien-3β-ol and cholest-5-en-3β-ol, respectively, being the most abundant ones.
Pigments have biological functions such as antioxidant, antiinflammatory and anti-obesity agents (D'Alessandro and Filho 2016). Van Lenning et al. (2003), reported that chlorophyll a, chlorophylls (Chls) c 1 and c 2 , and carotenoids such as fucoxanthin, diadinoxanthin, diatoxanthin and β carotene are the most common pigments found in Pavlovophyceae species, representing about 85% of the total. The values obtained in this work under optimal growth conditions for Chl a, Chl c and carotenoids are similar to those reported for P. gyrans (51% for Chl a, 5% for Chl c (c1 + c2) and 44% for carotenoids), grown at 20 °C and harvested at exponential growth phase (D'Alessandro and Filho 2016). Stress conditions have been widely used to trigger pigment synthesis and accumulation in microalgae. Previous studies have shown that low-temperature exposure could significantly increase/decrease carotenoid and chlorophyll contents in these microorganisms (Orset and Young 1999;García González et al. 2005;Markou and Nerantzis 2013;Guihéneuf and Stengel 2017;Ferro et al. 2018;Potijun et al. 2021). In the present study, the content of chlorophylls and carotenoids decreased due to low-temperature stress. In general, chlorophylls and primary carotenoids are degraded under stress and therefore their biomass content decreases significantly, while the synthesis of secondary carotenoids is enhanced by reactive oxygen species (ROS), generated by stress conditions (Markou and Nerantzis 2013). However, a more exhaustive study is needed to discriminate the effect of low-temperature stress on the different types of carotenoids in P. gyrans.

Conclusions
The native haptophyte Pavlova gyrans is shown to be sufficiently robust for cultivation in a photobioreactor under controlled growth conditions. The strain has proven to be a good source of PUFAs and sterols with potential applications in functional food and aquafeed industries. Low-temperature stress increased the concentration of PUFAs, including the essential fatty acids EPA and DPA, but without affecting total lipid content. PUFAs are mainly located in the neutral lipid fraction, thus providing a starting point in the search for other culture strategies that promote TAG accumulation. Therefore, the commercial cultivation of P. gyrans as an alternative and sustainable source of PUFAs, sterols and pigments, is a future challenge to explore considering its economic feasibility.

Acknowledgements
The technical assistance of Dra. M. Cecilia Damiani, Dr. Federico Delucchi and Jorge Oyola is acknowledged. MDA has a CONICET Fellowship. PIL, PGSB and MVSP are Research Members of CONICET. DC is a member of the Professional Support Personnel of CONICET. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Author's contributions All authors participated in the conception and design of the experiments. MDA and PSB performed the isolation of the strain and temperature experiments in the photobioreactor. MVSP performed the molecular identification. MDA, PSB performed the biochemical analysis of the resulting biomass. All authors interpreted the data and discussed the results. MDA drafted the article. PIL, PSB, DC and MVSP critically reviewed the article and contributed with intellectual input. All authors approved submission of the article.
Data availability Data are available from the authors upon reasonable request.

Declarations
Competing interests There are no competing interests.