Enhanced coproduction of astaxanthin and lipids by the green microalga Chromochloris zofingiensis: Selected phytohormones as positive stimulators

Journal Pre-proofs Enhanced coproduction of astaxanthin and lipids by the green microalga Chromochloris zofingiensis: Selected phytohormones as positive stimulators Jun-hui Chen, Dong Wei, Phaik-Eem Lim PII: DOI: Reference:

S0960-8524(19)31472-5 https://doi.org/10.1016/j.biortech.2019.122242 BITE 122242

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

21 August 2019 2 October 2019 3 October 2019

Please cite this article as: Chen, J-h., Wei, D., Lim, P-E., Enhanced coproduction of astaxanthin and lipids by the green microalga Chromochloris zofingiensis: Selected phytohormones as positive stimulators, Bioresource Technology (2019), doi: https://doi.org/10.1016/j.biortech.2019.122242

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Ltd.

Enhanced coproduction of astaxanthin and lipids by the green microalga Chromochloris zofingiensis: Selected phytohormones as positive stimulators

Jun-hui Chen a, Dong Wei a, b*, Phaik-Eem Lim c

a

School of Food Science and Engineering, South China University of Technology, Guangzhou

510641, P.R. China b

Research Institute for Food Nutrition and Human Health, Guangzhou, China

c

Institute of Ocean and Earth Sciences (IOES), University of Malaya, 50603 Kuala Lumpur,

Malaysia

* Correspondence: Prof. Dong Wei, School of Food Science and Engineering, South China University of Technology, Guangzhou 510641, P.R. China. Tel. & Fax: +86 20-87113849. E-mail: [email protected]

1

Abstract: Phytohormones comprise a variety of trace bioactive compounds that can stimulate cell growth and promote metabolic shifts. In the present work, a two-stage screening strategy was innovatively established to identify positive phytohormones for enhancement of astaxanthin and lipid coproduction in microplate-based cultures of mixotrophic Chromochloris zofingiensis. The results showed that auxins were the most efficient stimulators for astaxanthin accumulation. The maximum content of 13.1 mg/g and yield of 89.9 mg/L were obtained using indole propionic acid (10 mg/L) and indoleacetic acid (7.8 mg/L), representing the highest levels of astaxanthin in this microalga reported to date. Total lipids with the highest content (64.5 % DW) and productivity (445.7 mg/L/d) were coproduced with astaxanthin using indoleacetic acid. Statistical analysis revealed close relations between phytohormones and astaxanthin and lipid biosynthesis. This study provides a novel original strategy for improving astaxanthin and lipid coproduction in C. zofingiensis using the selected phytohormones as positive stimulators.

Keywords: Chromochloris zofingiensis; Astaxanthin; Lipids; Phytohormones; Induction

2

1. Introduction Astaxanthin (3, 3’-dihydroxy-β, β-carotene-4, 4’-dione) is a high-value ketocarotenoid with a broad range of commercial applications in aquaculture feed, functional food, nutraceuticals, cosmetics and pharmaceuticals due to its superhigh antioxidant activities and pigmentation (Ambati et al., 2014). Compared with synthetic astaxanthin consisting of a mixture of three isomers, natural astaxanthin from various microorganism mainly comprising the (3S, 3`S) stereoisomer has stronger antioxidative activity and meets food safety requirements. Thus, natural astaxanthin commands a high price of thousands US dollars per kg in the global market, which is several-fold higher than synthetic astaxanthin (Shah et al., 2016). Currently, Haematococcus pluvialis as well as Xanthophyllomyces dendrorhous are dominant producers for natural astaxanthin production in industry due to their ability to biosynthesize high levels of astaxanthin. However, H. pluvialis-based astaxanthin production has several limitations, such as a long period of cultivation, greater contamination by other microbes and zooplankton, and the highest costs for astaxanthin production (Shah et al., 2016). Regarding X. dendrorhous, the primary astaxanthin stereoisomer found is 3R, 3`R; thus, it has recently been sold as a pigment in animal feed. To overcome these crucial techno-economic challenges, numerous studies have been conducted either on the selection of ideal alternative species with superior characteristics or the utilization of integrated culture strategies for astaxanthin production. In recent years, the green microalga Chromochloris zofingiensis has been regarded as the most promising candidate for the economical production of astaxanthin (Liu et al., 2014). In comparison to H. pluvialis, C. zofingiensis has the ability to accumulate large amounts of astaxanthin and lipids simultaneously under stress conditions, subsequently storing them in lipid droplets in the form of mono- and diesters with fatty acids. These properties make C. zofingiensis capable of meeting commercial 3

demands for multiple bioproducts (lipids, carbohydrates, etc.) in addition to astaxanthin due to its high growth rate in organic trophic mode to achieve high cell densities and carotenoid accumulation. Several breakthrough findings and remarkable progress have been achieved; however, most of them have been focused on the optimization process by nutrient deficiency, high light irradiation and oxidative stress. Nevertheless, astaxanthin productivity by C. zofingiensis is still far from the requirements necessary for commercial production. Phytohormones are a broad-spectrum collection of plant-derived bioactive compounds that have biofunctions in modulating cell reproduction and regulating the biosynthesis of specific metabolites (e.g., lipids and secondary carotenoids) to resist environmental stresses (Han et al., 2018). For instance, many recent studies have shown that gibberellic acid, abscisic acid, salicylic acid, jasmonic acid (or methyl jasmonate), cytokinins and auxins (e.g., indole-3-acetic acid, indole3-propinic acid, etc.) have the ability to enhance astaxanthin and/or lipid accumulation in H. pluvialis or other microalgal species (Bajguz & Piotrowska-Niczyporuk, 2014; Hunt et al., 2010; Jiang et al., 2015; Lee et al., 2016; Liu et al., 2016b; Yu et al., 2015). However, few studies have examined the type and capability of phytohormones exerting stimulatory effects on the accumulation of astaxanthin and/or lipids in C. zofingiensis. Considering the unique physiological characteristics of C. zofingiensis in contrast to other microalgal species, the effects of phytohormones on physiological responses of C. zofingiensis are assumed to be species-specific and substantially differentiated from current reports in other microalgal species. Thus, systematic investigations of the ability of different phytohormones to enhance the production of astaxanthin and lipids are required. In particular, deeper insight into the close relations between various phytohormones and the biosynthesis of carotenoids and fatty acids will be critical to guide the economical application of C. zofingiensis for the integrated production of these valuable coproducts 4

through the use of exogenous stimulators in this species. Carotenoids (mainly astaxanthin and keto-lutein) exist in the form of esters with fatty acids and are stored in lipid bodies of C. zofingiensis under stress conditions. They tend to be induced by similar stimuli, including but not restricted to nutrient deprivation, high light irradiance, and ROS(Reactive oxygen species)-generating chemicals (Liu et al., 2016a; Zhekisheva et al., 2002). Thus, coordinated relationships between pigments and fatty acids exist; however, a lack of sufficient evidence to clearly describe these relationships in C. zofingiensis exposed to exogenous phytohormones have continued until now. Cluster analysis and OPLS-DA (orthogonal partial least squares discriminant analysis) are widely used as powerful approaches to reveal all relationships between profile changes of cellular components and environmental conditions in organisms (Lu et al., 2016). Based on these statistical techniques, it is possible to extract useful information from data concerning the composition of cellular metabolites in algal cells and, ultimately, identify potential relations between certain components and exogenous disturbances. The present study aimed to evaluate the feasibility of using phytohormones as stimulators in triggering enhanced astaxanthin and lipid coproduction in C. zofingiensis. Candidate phytohormones were first screened based on their effects on the autofluorescence of C. zofingiensis cells by using flow cytometry, and then several positive phytohormones were selected by quantifying astaxanthin and lipid levels using traditional and accurate analytical methods. Additionally, cluster analysis and multivariable statistical analysis of the data were applied to reveal the relations between phytohormones and the biosynthesis of pigments and fatty acids, greatly contributing to the elucidation of potential mechanisms of phytohormones in enhancing astaxanthin and lipid biosynthesis. This work represents a pioneering effort to facilitate the establishment of a novel strategy for improving coproduction of astaxanthin and lipids from C. zofingiensis by using 5

selected phytohormones as positive stimulators.

2. Materials and methods 2.1. Microalgae and culture conditions The green microalga C. zofingiensis (ATCC 30412) was purchased from the American Type Culture Collection (ATCC, Rockville, USA) and maintained at 4 °C on agar slants of modified Bristol medium (Ip et al., 2004). Cells from the agar slants were transferred into 100 mL of sterilized medium in 250-mL flasks containing 10 g/L glucose and cultivated for four days at 26 °C with orbital shaking at 150 rpm and continuous illumination of ca. 10 μmol m-2 s-1. Algal cells in the exponential phase were used as the seed culture for the inoculum at 10 % (v/v) for subsequent experiments. 2.2. Chemicals and reagents Biochemical-grade phytohormones were mainly purchased from Guangzhou Huaqisheng Biotechnology Co., Ltd. (Guangzhou, Guangdong, China), such as 1-aminocyclopropane-1carboxylic (ACC), 2-chlorobenzoic acid (CA), indole-3-acetic acid (IAA), indole-3-butyric acid (IBA), indole-3-propionic acid (IPA), 2,4-dichlorophenoxy acetic acid (2,4-D), kinetin (KT), 1naphthalene acetic acid (NAA), and diethyl aminoethyl hexanoate (DA). The other phytohormones were purchased from Macklin Biochemical Reagent Co., Ltd. (Shanghai, China). HPLC-grade methanol and MTBE (methyl tert-butyl ether) were purchased from Thermo Fisher Scientific Inc. (Waltham, MA, USA). Pigment standards (astaxanthin, lutein, zeaxanthin, canthaxanthin, chlorophyll a and b) were purchased from Sigma Aldrich Co. (St. Louis, MO, USA). Other analytical-grade chemicals and reagents were purchased from local suppliers. 2.3. Microplate-based cultivation of C. zofingiensis using different phytohormones 6

A two-phase screening strategy was established to identify positive phytohormones for enhancing astaxanthin and lipid accumulation in C. zofingiensis. Seed cells in exponential phase were inoculated into microplate wells and cultivated under white light at 300 ± 30 μmol m-2 s-1 as described previously (Chen et al., 2017b). All these experiments were performed in an incubator to maintain a constant temperature and stable light intensity. In the first phase, the seed culture was inoculated into 24-well microplate to maintain the initial cell density of algal suspensions at 2-3 g DW/L. Stock solutions of phytohormones were prepared in pure water, DMSO (dimethyl sulfoxide) or ethanol accordingly and then directly added into the culture in a microplate at the designated concentrations shown in Table 1 (Bajguz & PiotrowskaNiczyporuk, 2014; Hunt et al., 2010; Jiang et al., 2015; Lee et al., 2016; Liu et al., 2016b; Yu et al., 2015). Three cultures with the addition of the above solvents at a final concentration of 0.1 % (v/v) were set up as the control. After cultivation for 12 days under white fluorescent light at 300 ± 30 μmol m-2 s-1 as described previously (Chen et al., 2017b), the algal samples were harvested to detect the autofluorescence intensity of the cells by flow cytometry. These phytohormones, which efficiently increased the autofluorescence intensity in cells, were regarded as potentially positive for astaxanthin accumulation and subsequently selected for verification in the secondary phase. The secondary phase was conducted to confirm the capability of selected phytohormones from the first stage. Cells were cultivated in microplates under the same conditions described above to obtain sufficient biomass to meet the minimum weight requirements for reliable and accurate measurements. After 12 days of cultivation, the cultivated cells were ultimately collected for accurate quantitation of pigments by HPLC and lipids by gravimetric and GC-MS analysis. 2.4. Analytical methods Algal cells in a certain volume (1.5~2 mL) were collected by centrifugation and washed twice 7

with pure water and dried at 60 °C in an oven until a constant weight was achieved for measuring the dry weight. Total pigments in algal samples were extracted using mixed solvent (methanol/dichloromethane, 3:1, v/v) and analyzed on a reversed-phase HPLC system equipped with a Carotenoid C30 column and a photodiode array detector according to the referenced protocol (Chen et al., 2017b). Cytometric analysis of algal cells was conducted using a BD Accuri ® C6 flow cytometer (Accuri Cytometers, Inc., Ann Arbor, MI, USA) equipped with an air-cooled laser at 488 nm and two interference filters (FL1: 533±15 nm, FL2: 585±20 nm). The detailed procedure and data analysis were carried out according to our previous study (Chen et al., 2017a). All samples were evaluated at least in triplicate. Total lipids in the algal biomass were extracted and analyzed according to our previous work with minor modifications (Chen et al., 2015b). Approximately 20 mg of freeze-dried biomass was mixed with methanol/dichloromethane (3:1, v/v) in a bead-beating tube and repeatedly distributed using a bead beater. After each disruption, the extraction solvent containing lipids was collected by centrifugation. The extraction was repeated several times until the biomass became colorless. The supernatant containing lipids was merged and blow dried by flow gas of nitrogen. The crude oil obtained was weighed to calculate the lipid content. Determination of fatty acids was carried out according to our previous work with minor modifications (Chen et al., 2015a). Briefly, total fatty acids in the biomass were extracted and methyl-esterified to fatty acid methyl esters. They were then analyzed by GC-MS equipped with the DB-23 capillary column (30 m×250 μm, 0.25 μm). The peaks of fatty acids were identified by the NIST mass spectral database and quantified by the internal standard method using nonadecanoic acid C19:0 (1 mg/mL) as the internal standard. 8

2.5. Data analysis All experimental results shown in the figures are expressed as the mean value ± SD (standard deviation). All experiments were performed in triplicate. The statistical significance of the results was tested by one-way analysis of variance (ANOVA) or the Student’s t-test at significant levels of p<0.05 and p<0.01. Hierarchical cluster analysis was performed using SPSS software (SPSS Inc., Chicago, IL, USA). The orthogonal partial least squares discriminant analysis (OPLS-DA) was employed to identify multivariate relationships between phytohormones and the abundance of metabolites (pigments and fatty acids) in algal cells. This multivariate analysis was performed using SIMCA-P 14.1 software (Umetrics AB, Umea, Sweden) for pattern recognition.

3. Results and discussion 3.1. Screening of positive phytohormones in the first phase Natural pigments, especially carotenoids, are important fluorescent compounds in algal cells due to their symmetrical tetraterpene skeleton, which can emit both green and yellow fluorescence when excited by a laser at 488 nm (Lagorio et al., 2015). Based on the accurate correlation of the autofluorescence intensity and astaxanthin content in C. zofingiensis established in a previous study (Chen et al., 2017a), we developed a fast screening method by FCM (flow cytometry) for rapid estimation of astaxanthin formation, and then applied this method in a preliminary screening of potential astaxanthin-stimulating phytohormones. Twelve phytohormones were chosen to assess their potential stimulatory effects on astaxanthin accumulation (Table 1), and two dosages (between 2.3 mmol/L and 2.5 mol/L) of each phytohormone based on previous literature were investigated to estimate their contribution to the autofluorescence intensity of astaxanthin. As shown in Fig. 1, all phytohormones exerted a significant influence on the mean 9

fluorescence intensity (MFI) of algal cells (p<0.01). Particularly, auxins (i.e., IAA and IBA) were the most efficient triggers, resulting in a 25 % increase of MFI in FL2 at their optimal dosages, followed by GA, ABA and 2,4 D with increases of 15.3 %, 19.9 % and 9.5 %, respectively. In contrast, ACC, CA and IPA seemed to have no obvious stimulatory effects on MFI in FL2, even though they have been reported to have a superior ability to efficiently induce carotenoid and/or lipid biosynthesis in other microorganisms (Li et al., 2015; Liu et al., 2016b). Additionally, no significant effects in terms of MFI in FL1 and FL2 were observed following the treatments with KT, ETA, DA and NAA compared with the control. These results were clearly contradictory to previous studies, which suggested their positive effects on the production of valuable metabolites in microalgal cells (Anahas & Muralitharan, 2019; Cui et al., 2018; Jiang et al., 2015; Li et al., 2015; Liu et al., 2016b). This discrepancy may be explained by the differences in metabolic responses and regulatory mechanisms in C. zofingiensis compared with other astaxanthin producers. Chromochloris zofingiensis has unique capabilities of multiple tropical growth by utilizing organic and inorganic nutrients (Liu et al., 2014); thus, it may exhibit species-specific metabolism to regulate carotenogenesis processes in response to exogenous phytohormones. Based on these results, the desired dosages per phytohormones were estimated according to their effects on MFI of FL2, as presented in Table 1. Phytohormones (e.g., IAA, IBA, ABA, GA and 2,4D) that increased MFI of Fl2 were selected as potential positive triggers for further evaluation in the second stage. Other phytohormones (ACC, CA and IPA) with potential abilities to favor carotenoid or lipid accumulation in other microorganisms were also selected for further confirmation to avoid possible false negatives caused by interferences during flow cytometric analysis. Notably, the stability and robustness of the flow cytometry-based screening approach are highly dependent on the pigment composition and their respective contents in C. zofingiensis (Chen 10

et al., 2017a). Thus, diverse changes in pigment composition caused by different phytohormone treatments would certainly affect the accuracy of this rapid method. Altogether, a series of phytohormones (Table 1) with stimulatory effects on MFI of FL2 were selected in this phase. 3.2. Identification of selected phytohormones in the secondary phase In the following experiments, C. zofingiensis was cultivated in a microplate by adding selected phytohormones at optimal dosages for 12 days. The collected biomass and extracted astaxanthin were used to evaluate positive effects on cell growth and astaxanthin and lipid accumulation in C. zofingiensis. As shown in Fig. 2, different phytohormones exhibited significantly different effects on biomass (p<0.01), either stimulatory or inhibitory. The addition of auxin IAA significantly promoted biomass production to the maximum concentration of 8.3 g/L, demonstrating a 26.7 % increase compared with the control (p<0.01). This result was consistent with previous research that reported similar stimulatory effects of various auxins on microalgal cell growth and cellular composition (Liu et al., 2016b). Conversely, CA, IPA and ABA all exerted significantly detrimental effects on biomass with a decrease exceeding 47 % compared with the control (p<0.01). It was hypothesized that different phytohormones were able to implement both positive or negative effects on the biomass of C. zofingiensis, probably by regulating the nutrient assimilation and activating redox systems (Han et al., 2018). The changes in pigment and fatty acid compositions affected by phytohormones are presented in Fig. 3. Almost all of the selected phytohormones were capable of inducing astaxanthin accumulation, which was consistent with the above screening results. Fig. 3a shows that IPA, the most effective phytohormone, greatly increased the astaxanthin content in C. zofingiensis to as high as 13.1 mg/g, which was 48.9 % higher than the control (p<0.01). Similarly, ABA increased astaxanthin content to 12.6 mg/g with extremely significant difference with the control (p<0.01); 11

however, these increases were achieved at a low biomass concentration. This can be explained through the mechanisms taking place in higher plants, where ABA primarily functions as a protective signaling molecular through shifting physiological responses to resist abiotic stresses but at the expense of cell growth, which could possibly t true for microalgae also (Kozlova et al., 2017; Lu & Xu, 2015). In comparison to the results that both IPA and ABA considerably depressed biomass production and then significantly reduced the final yield and productivity of astaxanthin (p<0.01), several phytohormones had the ability to trigger astaxanthin accumulation and simultaneously promote biomass production. In IAA-added cultures, the maximum biomass concentration of 8.3 g/L could be obtained in parallel to high astaxanthin content at 10.8 mg/g. Accordingly, the highest yield (89.9 mg/L) and productivity (7.5 mg/L/d) of astaxanthin were obtained in these cultures, which were significantly higher than other treatments, although the astaxanthin content was significantly reduced compared with the IPA- or ABA-added cultures (p<0.01). It is worth noting that although tremendous efforts had been made worldwide to increase the efficiency of astaxanthin production by C. zofingiensis in the past decades, the relatively low astaxanthin accumulation still restrained the development of this microalga for commercial applications (Chen et al., 2017b; Liu et al., 2014); while our present study made considerable contribution to this research field by markedly increasing astaxanthin yield and productivity far beyond the current levels reported to date (Liu et al., 2013; Zhang et al., 2017). Additionally, Fig. 3b shows that there were no significant differences in pigment profiles between IAA- and IPA- or ABA-added cultures. In these cultures, astaxanthin, keto-lutein and canthaxanthin accounted for 86.5 % of the total pigments, in which 68.3 % to 74.9 % of the total pigments consisted of astaxanthin in all cultures with no significant variation among the above three phytohormones. However, as an exception, the chemical synthetic phytohormone 2,4 D showed a significantly 12

elevated chlorophyll content (6.6 % DW) than the other phytohormone-treated cultures (below 4.0 % DW), which is consistent with a report showing that a low dosage of 2,4-D had stimulatory effects on photosynthesis and chlorophyll synthesis in the microalga S. quadricauda (Wong, 2000). Notably, the steady composition of cellular carotenoids showed an accurate correlation with the stability and robustness of the FCM method (Chen et al., 2017a). In the present study, changes in pigment profiles were caused by the addition of phytohormones, and these variations would affect the intensities of algal autofluorescence with the accuracy of the FCM method, which may explain to some extent the false-positive errors in the first phase of the screening assay. These false-positive errors could be minimized to a large extent by identification in the second phase, thus the FCM approach could still be utilized as an efficient and powerful method for rapid screening of stimulatory phytohormones with acceptable accuracy. The effects of different phytohormones on the accumulation of lipids and fatty acid in C. zofingiensis determined in the subsequent study are presented in Fig. 3c. Chromochloris zofingiensis cells were not only able to synthesize astaxanthin but also other valuable lipids (mainly fatty acids), as well as to resist adverse conditions (e.g., high light and nutrient stress). These results indicated that the selected autofluorescence-enhancing phytohormones had no significant effects on lipid accumulation. In particular, upon IAA addition, algal cells could accumulate lipids up to 64.5 % DW, demonstrating no substantial difference from other cultures, while the maximum lipid yield (8.3 g/L) and productivity (445.7 mg/L/d) were markedly improved by IAA addition due to its stimulatory effect on biomass production. These results were partly concordant with a previous report showing that auxins (e.g., IAA, IBA, IPA) had strong stimulatory effects on algal cell growth and lipid production in Chlorella pyrenoidosa and Scenedesmus quadricauda (Liu et al., 2016b). In contrast, IPA, favoring lipid accumulation in these two fresh-water microalgae, exhibited no 13

significant or even negative effects on lipid accumulation in C. zofingiensis. The observed differences might be a consequence of not only the different culture systems and stress conditions but also, and most importantly, the specific growth and metabolic properties of these algal species. Additionally, the permeability of algal membrane and its binding sites for hormone uptake are also responsible for the specific effects of phytohormones on algal cells, since auxins with different dosage concentrations have variable affinities towards their corresponding cellular receptors, and subsequently affect gene expression and cellular physiological processes (Kozlova et al., 2017; Sun et al., 2019). Fig. 3d shows the changes in fatty acid abundance in C. zofingiensis cells in response to the addition of different phytohormones for 12 days. It has been emphasized that the nutritional value of microalgal oils is highly dependent on the fatty acid composition in biomass feedstocks (Ghazali et al., 2015); therefore, it is necessary to investigate variations in fatty acid profiles in algal cells in response to phytohormone addition. Although the effects of different phytohormones on the accumulation of fatty acids might be highly variable, C16:0, C18:1, C18:2 and C18:3 were found to be the predominant fatty acids in C. zofingiensis cells in all of these cultures, accounting for approximately 90 % of the total fatty acids. Additionally, similar to previous studies demonstrating the positive effects of phytohormones on fatty acid when expressed as the percentage of total fatty acids or lipids (Kozlova et al., 2017; Sun et al., 2019), our study also had shown similar observation that these phytohormones had different degrees of effects on the fatty acid profiles. For instance, IPA, the most efficient phytohormone for astaxanthin accumulation, resulted in a percentage increase of C16 and C20 reaching as high as 27.8 % and 5.6 %, respectively; in contrast, the percentage of total C18 was reduced to 66.3 %, which was significantly lower than that observed with IAA and IBA (p<0.01). These results demonstrated that, unlike other auxins, IPA partially 14

inhibited the desaturation and elongation of C16 to C18 in lipid biosynthetic pathways. The discrepancies of fatty acid profiles under auxins might be due to structural differences of these auxins and their participated signal transduction pathways (Kozlova et al., 2017; Liu et al., 2016b; Sun et al., 2019). These findings provide important informative evidence for in-depth studies of the physiological responses of C. zofingiensis to exogenous phytohormone disturbances. At present, numerous reports have investigated commercial mass production of natural astaxanthin using microalgae; in contrast, economic astaxanthin production with a highly competitive cost still faces enormous challenges. By combining the stress of nutrient starvation and high light irradiation as well as phytohormone induction, the microplate-based two-step cultivation strategy described in the present study substantially improves the potential of C. zofingiensis in astaxanthin and lipid production by enhancing its productivity. To highlight the superiority of phytohormone induction in this study, the current production levels of astaxanthin and lipids by C. zofingiensis in comparison to H. pluvialis are compared and discussed (E-supplementary data of this work can be found in online version of the paper). The highest astaxanthin yield and productivity of 89.8 mg/L and 7.5 mg/L/d were achieved in IAA-induced cultures, respectively, and high levels of lipid content and productivity were also obtained at 64.5 % DW and 445.7 mg/L/d by employing our integrated strategy. These levels were markedly higher than those previously reported in C. zofingiensis. For instance, the present work showed a significantly increased astaxanthin yield by approximately 22.6 % compared with the highest astaxanthin yield of 73.3 mg/L using the two-step cultivation strategy (Zhang et al., 2017), which, to the best of our knowledge, has been the highest yield of astaxanthin from C. zofingiensis reported to date. More importantly, astaxanthin content was substantially promoted from below 1 mg/g to higher than 10.8 mg/g using our strategy, and along with astaxanthin accumulation, high productivity (445.7 mg/L/d) and content (64.5 % DW) of 15

lipids were achieved, with levels slightly lower than the 473.0 mg/L/d obtained using the continuous culture strategy (Liu et al., 2016a). Although there have been numerous studies of using microalgae as biodiesel feedstocks in recent years, the economic production of algal biodiesel still remains challenging owing to the high production cost. Thus, the integrated production of lipids with astaxanthin will contribute to the economic production of biodiesel by C. zofingiensis (Liu et al., 2016a; Mao et al., 2018). The above results clearly demonstrated that the strategy of phytohormone induction, integrated with typical nitrogen deprivation and high light irradiation, significantly improved current astaxanthin productivity in C. zofingiensis along with relatively high lipid productivity. The highly efficient coproduction of astaxanthin and lipids facilitates the economic feasibility of using C. zofingiensis for commercial application. The astaxanthin content obtained in the present study was still lower compared with H. pluvialis, but the volumetric biomass (8.3 g/L) and astaxanthin productivity (7.5 mg/L/d) were higher or comparable (E-supplementary data of this work can be found in online version of the paper). Although the maximum content of astaxanthin in H. pluvialis reported so far is up to 49 mg/g (Sun et al., 2015), the relatively low biomass productivity ranged from 0.05 to 0.41 mg/L/d leads to the limitation of its mass cultivation for the economic and efficient production of natural astaxanthin (Cheng et al., 2016; Park et al., 2014; Sun et al., 2015; Wen et al., 2015; Zhao et al., 2018). In addition, when considering the nutritional value of lipids and fatty acids as potential coproducts, C. zofingiensis may be an ideal producer for economic coproduction of astaxanthin and algal oil. This pioneering work provided a feasible approach for integrated coproduction of carotenoids and lipids by microalgae using the phytohormone addition strategy, which have a considerable contribution to highlight the promising potential of C. zofingiensis for future commercial applications. 3.3. Multivariable data analysis of pigment and fatty acid abundance 16

In this study, the effects of phytohormones on changes in pigment and fatty acid abundance in C. zofingiensis were analyzed by cluster analysis and OPLS-DA. These powerful approaches provide an intuitive illustration by clustering the selected phytohormones into groups based on distinguished similar effects on algal cells, this supplying informative clues for phytohormoneinduced responses and underlying regulatory mechanisms in this microalga. Fig. 4 shows the dendrogram based on the hierarchical cluster analysis of astaxanthin and lipid abundance by phytohormone addition. Clearly, when the rescaled distance between groups was 20, the selected phytohormones were clustered into three main groups. GA, IAA and IBA were in the same group, in which they enhanced astaxanthin and lipid accumulation; in contrast, IPA and ABA belonged to the same group in which astaxanthin biosynthesis was significantly stimulated but lipid biosynthesis was substantially inhibited in cells. Both ACC and 2,4-D belonged to the same group as the control, which indicated that the metabolic sites in which they acted might not help trigger the biosynthesis of astaxanthin. Therefore, it was inferred that phytohormones in the same group might share the same regulatory mechanism and/or affect identical interaction sites, thus providing insights for the coproduction of astaxanthin and lipids in C. zofingiensis. The results also demonstrated that under certain treatments, C. zofingiensis might have the unique property of accumulating astaxanthin efficiently without the synergistically coordinated accumulation of lipids, even though many previous studies have shown that astaxanthin and lipids accumulate synergistically to resist adverse growth conditions (Liu et al., 2016a; Liu et al., 2016b). The findings further support a previous study in which a special interaction mechanism likely participated in regulating the fraction division of fatty acids utilized for the formation of TAG and carotenoid esters in nitrogen-deprived cultures of C. zofingiensis (Mulders et al., 2014). Our findings provide new clues concerning the presence of unique regulatory mechanisms modulating the separate 17

biosynthesis of astaxanthin and lipids in microalgae, processes that lack in-depth investigations and require greater efforts to fully understand the potential underlying relationships. In this study, variations of pigment and fatty acid compositions in response to phytohormone induction were assessed. As shown in Fig. 5, the pigment and fatty acid abundance in total lipids in all the cultures were applied in a multivariable data analysis coupled to OPLS-DA to recognize underlying patterns. The score scatter plot shown in Fig. 5a demonstrates that the selected phytohormones could be separated according to the scores of their effects on the accumulation of pigments and fatty acids in C. zofingiensis. Phytohormone-added algal cells almost clustered together in the positive region of t1, which separated with the control clusters. Particularly, phytohormones (e.g., IAA, IBA, GA3) located closer to each other in the same cluster had similar stimulatory effects and resulted in nearly identical compositions of pigments and fatty acids, suggesting that these selected phytohormones had significant impacts on the physiological status and metabolite profiles compared with the control. Thus, OPLS-DA and HCA analyses were found to provide a powerful approach to visualize the data for grouping phytohormones and discriminating the origins of the proportions of pigments and fatty acids. To further examine the correlations between exogenous phytohormones and endogenous metabolites (particularly pigments and fatty acids), the OPLS-DA score-loading biplot was constructed as shown in Fig. 5b. This figure charts the scores and loadings for identifying the influential metabolites responsible for the classification of the selected phytohormones. In principle, the metabolites situated far from the plot origin usually had strong impacts on the classification of these phytohormones, while those phytohormones that clustered together and closed to the plot origin had similar properties. In this case, C14:0, C16:0 and chlorophyll a had a reduced influence on the clustering since they were located very close to the origin; in contrast, astaxanthin, 18

ketolutein, and fatty acids including C16:1, C18:1, C18:2 and C18:3 were the dominant variables determining the classification of the selected phytohormones. Fatty acids (e.g., C16:1, C18:1, C18:2, C18:3) clustered very close together and were located in the upper-left quadrant, but secondary carotenoids such as astaxanthin and ketolutein were located at opposite position, indicating that the controlled conditions (high light and nitrogen deprivation) were more favorable for fatty acid biosynthesis than astaxanthin and ketolutein. IPA and ABA were situated opposite to these fatty acids in the lower-right quadrant of the plot, located near astaxanthin and carotene-like carotenoids. The results demonstrated that IPA and ABA correlated negatively with these fatty acids but had positive relationships with the biosynthesis of astaxanthin and carotene-like carotenoids. Unexpectedly, IAA, IBA and GA were clustered together and situated in positive values of both axes (upper-right quadrant) of the plot, although they showed relatively long distances from either astaxanthin or primary fatty acids. These results indicated that three types of phytohormones were, to a certain extent, able to induce C. zofingiensis cells to simultaneously accumulate astaxanthin and fatty acids, in accordance with the above results shown in Fig. 3. Our results confirmed that OPLS-DA was a powerful and reliable tool for discriminating the selected phytohormones and providing new insights for cellular metabolic regulation in response to exogenous disturbances. 3.4. Hypothetic mechanisms of the selected phytohormones on C. zofingiensis To date, numerous research studies have reported positive effects of phytohormones on the biosynthesis of carotenoids and/or lipids in microalgae. However, systematic investigations of the effects of these phytohormones on the coproduction of astaxanthin and lipids in microalgae have been rarely reported, especially on the involved hypothetical regulatory mechanisms. Based on our aforementioned results and previous reports, we propose a series of hypothetic mechanisms of these 19

positive phytohormones in the regulation of the biosynthesis of astaxanthin and lipids in C. zofingiensis cells, as presented in Fig. 6. ROS (reactive oxygen species) and phytohormones are two known types of signaling molecules that regulate algal responses to exogenous stresses. They can either activate their particular signal transduction cascades or interact with each other by activating second messengers or phosphorylation cascades to regulate downstream metabolic events (Lu & Xu, 2015; Zhao et al., 2019). Conditions such as the control, high light irradiation and nitrogen deprivation primarily lead to increases in cellular ROS in chloroplasts, mitochondria and the cytosol, subsequently resulting in the regulation of downstream targets for astaxanthin and lipid accumulation. Conversely, phytohormones are another important kind of signaling molecule that can activate cellular antioxidant systems and increase antioxidase and antioxidant levels to resist stress conditions. They are also involved in specific signal transduction cascades via different types of regulatory mechanisms and interact with other signaling pathways (e.g., NO, calcium signaling, ROS signaling, MAPK signaling and other plant hormones) to form a complex signaling network acting on downstream metabolic events (Zhao et al., 2019). Thus, the coupling of phytohormones and ROS-mediated abiotic stress facilitates the exploitation of C. zofingiensis for the enhanced coproduction of valuable carotenoids and lipids beyond the sole use of either phytohormones or abiotic stresses. Noticeably, several synthetic analogs or cell growth regulators, despite belonging to phytohormones, have effects on cellular physiological activities by virtue of ROS signaling pathways. For example, 2,4-D, an efficient cell growth regulator, has the unique ability to stimulate S-nitrosylation of endogenous antioxidant enzymes, inducing ROS formation and lipid peroxidation (Ortega-Galisteo et al., 2012); thus, to some extent, it has dosage-dependent stimulatory effects on the accumulation of secondary carotenoids and lipids. The combined induction of 2,4-D and conditions in the control (i.e., high light irradiation and nitrogen deprivation) further increases the 20

cellular ROS level. Similarly, ACC, as an ethylene precursor, also directly or indirectly induces free radical-involved lipid oxidation and membrane breakdown and then regulates plant metabolism due to its close relationship with ROS formation, perception and signaling (Kim et al., 2016). It is worth noting that if stress-induced free radicals exceed the neutralization capabilities of the algal cells, algal growth will be negatively impacted. This phenomenon suggests that, by virtue of the ROS signaling pathways, the addition of these phytohormones could not further promote the accumulation of astaxanthin and lipids but, conversely, inhibit their biosynthesis. In comparison to ROS-mediated stresses, the integrated strategy of ROS- and phytohormone-mediated stresses would be more favorable for coordinating the efficient production of astaxanthin and lipids in C. zofingiensis. In addition to the separate signal transduction and molecular mechanisms of ROS and phytohormones, the different types of phytohormones are not isolated in signal transduction pathways but interact with each other in complex networks and work synergistically or antagonistically to respond to exogenous environmental stresses (Han et al., 2018; Lu & Xu, 2015). For example, phytohormones such as GA and ABA evaluated in the present study, have been reported to be important signaling molecules that are involved in algal defense responses and in the regulation of astaxanthin and lipid accumulation by mediation of their separate signal transduction pathways, although their bioactivities vary somewhat in the literature, in which different microalgal species and associated culture conditions were employed (Gao et al., 2013; Lu & Xu, 2015; Shan et al., 2012; Yu et al., 2015). They also interact each other in complex networks to modulate physiological processes by stimulating signaling pathways or certain coordinators with close relationships (Lu & Xu, 2015; Shan et al., 2012). In a previous study, synergistic and antagonistic interactions between GA and ABA have been reported, which act to regulate algal responses to 21

stress conditions; their crosstalk has been shown to be mediated by a special SnRK2-APC/CTE module that regulates the proteasome-mediated degradation of ABA receptors (Lin et al., 2015). It was additionally found that NPF (nitrate/peptide family transporter) as a GA influx transporter is not only responsible for regulating GA distribution and its biological activities, but also partly involves in the GA-ABA crosstalk at the transport level in vitro (Tal et al., 2016). Consistent results were obtained in the present study, although solid information is lacking to explain the discrepancy in the stimulatory effects triggered by auxin family phytohormones with similar structures and signal transduction routes, especially for the typical auxin family members IAA and IBA, which showed significantly different induction trends from IPA. This phenomenon might be caused by structural differences of auxins and their affinities towards corresponding cellular receptors, which may directly affect the expression of targeted metabolic genes and ultimately exert influences on the biosynthesis of astaxanthin and lipids (Kozlova et al., 2017; Sun et al., 2019). As shown in Fig. 6, we inferred that IAA and IBA might identically involve in SCFTIR1/AFB-mediated auxin signaling pathways, but IPA probably interacted with other receptors (e.g., indole butyric acid response 5(IBR5) phosphatase) involving in different signaling pathways such as the mitogen-activated protein kinase (MAPK). Additionally, the similar enhanced effects of the different phytohormones suggested same coordinators or direct crosstalk of specific signal transduction pathways involved in regulating astaxanthin and lipid production. Based on the above cluster analysis, IAA and IBA might also trigger signaling pathways or certain coordinators with close relationships with GA since they were clustered together due to their physiological effects. This result is consistent with a previous study demonstrating that auxin could act upstream of the GA pathway and crosstalk could occur through interactions between the GA-signaling repressor DELLA and auxin signaling components (Hu et al., 2018). The possible mechanism underlying the crosstalk between auxins and 22

gibberellin signaling might be that the GA-signaling repressor SlDELLA and the SlIAA9 and SlARF7 complex antagonistically regulate the transcriptional expression of GA- and auxin-response genes (Hu et al., 2018), and plant-specific transcriptional factors of the GRAS gene family were also reported to involve in the crosstalk of auxin and GA signaling pathways (Habib et al., 2019). These findings provided strong molecular evidences to our results demonstrating that auxins (i.e., IAA and IBA) and GA3 exerted similar stimulatory effects on the compositions of pigments and fatty acids in C. zofingiensis. Thus, certain phytohormones might share similar signal transduction pathways and regulate the same target genes. However, the above auxin mechanism failed to explain all auxin responses in our study demonstrating that the typical auxin family members IAA and IBA showed significantly different induction trends from another auxin IPA. Currently, the canonical TIR1/AFB pathway is the wellcharacterized auxin signaling pathway; while, this auxin signaling mechanism could not clarify the discrepancy in physiological responses of all auxins. In addition, currently there are additional and emerging auxin signaling pathways for the clarification of the differences of transcriptional responses to auxins. For example, there are several different types of auxin receptors (e.g., TIR1/AFB, ABP1, IBR5, SKP2A, etc) in algal cells, which coexist in microalgal cells to perceive auxins with different concentrations but involve in specific signaling cascades and subsequent regulation of gene expression (Alsenani et al., 2019). Particularly, IBR5 (indole butyric acid response 5), a dual-specificity MAP kinase phosphatase, was reported to promote auxin responsiveness in a unique manner that bypasses the requirement for the well-established auxinresponse component TIR1/AFB, which does not stimulate the degradation of Aux/IAA repressors and downstream gene transcription (Alsenani et al., 2019; Monroe-Augustus et al., 2003; Strader et al., 2010). Thus, it was inferred that unlike IAA and IBA, IPA might be more likely correlated to 23

ABA-mediated signaling pathways and downstream regulatory targets. This result was also consistent with a previous study showing that certain auxins trigger ABA biosynthesis and further regulate ABA signaling components and ABA-specific downstream targets (Alsenani et al., 2019). As shown in Fig. 6, both of IPA and ABA activated members of the MAPK family and then interacted with each other via MAPK signaling pathways to participate in subsequent metabolic regulation and stress adaptation. According to previous studies, IBR5 appears to implicate in ABA signaling pathways by mediation of mitogen activated protein kinase12 (MPK12) which participates in MAPK cascades and subsequent signal transduction (Jagodzik et al., 2018; MonroeAugustus et al., 2003). In other words, in comparison with the negative role of MAPKs (particularly MPK12) in the regulation of auxin signal transduction, MAPK cascades might play stimulatory roles in ABA signaling through a series of complex signal transduction pathways involving ABA receptors (i.e., PYR, PYL & RCAR) and protein kinase/phosphatase cascades (i.e., PP2Cs and SnRK2s) (Jagodzik et al., 2018). Admittedly, current knowledge of MAPK cascades and their interactions with auxin and ABA signaling is still comparatively limited, and their interactions and associated MAPK-dependent molecular mechanisms remain to be deciphered. The signaling roles of IBR5 and MAPK cascades contribute convincing molecular evidences for explaining the discrepancy in the stimulatory effects of different auxin family phytohormones, and provide new insights into the investigation of the interaction between auxin (particularly IPA) and ABA signal transduction pathways. Besides, the superior effects of IPA over ABA might be explained by its special ability to improve cellular enzymatic and nonenzymatic antioxidant activities by evaluating lipid peroxidative products and active levels of antioxidative enzymes (Piotrowska-Niczyporuk & Bajguz, 2014). Noticeably, ROS signaling also interacts with auxin and ABA signaling pathways through intracellular secondary messengers and mutual actions, and it cooperates to induce the 24

biosynthesis of astaxanthin and lipids to resist adverse conditions via particular mechanisms of crosstalk between ROS- and phytohormone-mediated signaling pathways (Zhao et al., 2019). Admittedly, these is still lack of information about these signaling pathways of phytohormones and their crosstalk in regulating astaxanthin and lipid production nowadays; further investigation of the signaling pathways of these molecules and their resulting downstream genes will provide new clues with regards to the coordinated accumulation of astaxanthin and lipids. To our knowledge, this is the first study to systematically evaluate differences in auxin-responsive behavior involving the biosynthesis of astaxanthin and lipids in C. zofingiensis cells. The present work provides new insights and inspiration to establish an integrated production process of astaxanthin and lipids in microalgae. It is worth mentioning that the mechanisms underlying the roles of phytohormones in C. zofingiensis were hypothesized based on the limited results of this work. Further studies providing direct evidence are needed to gain clearer insights into phytohormone-induced physiological responses and regulatory mechanisms in C. zofingiensis.

Conclusions The present study demonstrates that the established two-stage screening strategy can be used to select and identify positive phytohormones for enhancement of astaxanthin and lipid coproduction in C. zofingiensis. Statistical analyses could benefit the clustering of phytohormones into three groups and provide new insights into their regulatory biofunctions and hypothetical mechanisms. Therefore, the present results confirm that C. zofingiensis has promising abilities for accumulating high levels of astaxanthin and lipids via an ideal induction process combining phytohormones, high light irradiation and nitrogen deprivation, facilitating novel technology development for the economical coproduction of astaxanthin and lipids from this microalga. 25

Acknowledgments The work was supported by the program of Science and Technology of Guangzhou (Grant no. 201704030084) and the SinoPec Technology Development Program (218017-1, 36100002-19FW2099-0035). This work was partly supported by the 111 Project (B17018).

Conflicts of Interest The authors declare no conflicts of interest.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version.

References [1] Alsenani, F., Wass, T.J., Ma, R.J., Eltanahy, E., Netzel, M.E., Schenk, P.M., 2019. Transcriptome-wide analysis of Chlorella reveals auxin-induced carotenogenesis pathway in green microalgae. Algal Res. 37, 320-335. [2] Ambati, R.R., Phang, S.M., Ravi, S., Aswathanarayana, R.G., 2014. Astaxanthin: Sources, extraction, stability, biological activities and its commercial applications-a review. Mar. Drugs 12, 128-152. [3] Anahas, A.M.P., Muralitharan, G., 2019. Central composite design (CCD) optimization of phytohormones supplementation for enhanced cyanobacterial biodiesel production. Renew.Energ. 130, 749-761. [4] Bajguz, A., Piotrowska-Niczyporuk, A., 2014. Interactive effect of brassinosteroids and cytokinins on growth, chlorophyll, monosaccharide and protein content in the green alga Chlorella vulgaris (Trebouxiophyceae). Plant Physiol. Biochem. 80, 176-183. [5] Chen, J., Liu, X., Wei, D., Chen, G., 2015a. High yields of fatty acid and neutral lipid production from cassava bagasse hydrolysate (CBH) by heterotrophic Chlorella protothecoides. Bioresour. Technol. 191, 281-290. [6] Chen, J., Wei, D., Pohnert, G., 2017a. Rapid estimation of astaxanthin and the carotenoid-to-chlorophyll ratio in the green microalga Chromochloris zofingiensis using flow cytometry. Mar. Drugs 15, 231-253. [7] Chen, J.H., Liu, L., Wei, D., 2017b. Enhanced production of astaxanthin by Chromochloris zofingiensis in a microplate-based culture system under high light irradiation. Bioresour. Technol. 245, 518-529. [8] Chen, T., Liu, J., Guo, B., Ma, X., Sun, P., Liu, B., Chen, F., 2015b. Light attenuates lipid accumulation while enhancing cell proliferation and starch synthesis in the glucose-fed oleaginous microalga Chlorella zofingiensis. Sci. Rep. 5, 14936. [9] Cheng, J., Li, K., Yang, Z., Zhou, J., Cen, K., 2016. Enhancing the growth rate and astaxanthin yield of 26

Haematococcus pluvialis by nuclear irradiation and high concentration of carbon dioxide stress. Bioresour. Technol. 204, 49-54. [10] Cui, J.Y., Diao, J.J., Sun, T., Shi, M.L., Liu, L.S., Wang, F.Z., Chen, L., Zhang, W.W., 2018. C-13 Metabolic Flux Analysis of Enhanced Lipid Accumulation Modulated by Ethanolamine in Crypthecodinium cohnii. Front. Microbiol. 9, 1-10. [11] Gao, Z., Meng, C., Gao, H., Li, Y., Zhang, X., Xu, D., Zhou, S., Liu, B., Su, Y., Ye, N., 2013. Carotenoid genes transcriptional regulation for astaxanthin accumulation in fresh water unicellular alga Haematococcus pluvialis by gibberellin A3 (GA(3)). Indian J. Biochem. Biophys. 50, 548-553. [12] Ghazali, W.N.M.W., Mamat, R., Masjuki, H.H., Najafi, G., 2015. Effects of biodiesel from different feedstocks on engine performance and emissions: A review. Renew. Sust. Energ. Rev. 51, 585-602. [13] Habib, S., Waseem, M., Li, N., Yang, L., Li, Z.G., 2019. Overexpression of SlGRAS7 Affects Multiple Behaviors Leading to Confer Abiotic Stresses Tolerance and Impacts Gibberellin and Auxin Signaling in Tomato. International Journal of Genomics 2019, 1-16. [14] Han, X., Zeng, H., Bartocci, P., Id, F., Yan, Y., 2018. Phytohormones and Effects on Growth and Metabolites of Microalgae: A Review. Fermentation 4, 1-15. [15] Hu, J.H., Israeli, A., Ori, N., Sun, T.P., 2018. The Interaction between DELLA and ARF/IAA Mediates Crosstalk between Gibberellin and Auxin Signaling to Control Fruit Initiation in Tomato. Plant Cell 30, 1710-1728. [16] Hunt, R.W., Chinnasamy, S., Bhatnagar, A., Das, K.C., 2010. Effect of Biochemical Stimulants on Biomass Productivity and Metabolite Content of the Microalga, Chlorella sorokiniana. Appl. Biochem. Biotechnol. 162, 2400-2414. [17] Ip, P.F., Wong, K.H., Chen, F., 2004. Enhanced production of astaxanthin by the green microalga Chlorella zofingiensis in mixotrophic culture. Process Biochem. 39, 1761-1766. [18] Jagodzik, P., Tajdel-Zielinska, M., Ciesla, A., Marczak, M., Ludwikow, A., 2018. Mitogen-Activated Protein Kinase Cascades in Plant Hormone Signaling. Front. Plant Sci. 9, 56-64. [19] Jiang, L.Q., Pei, H.Y., Hu, W.R., Han, F., Zhang, L.J., Hou, Q.J., 2015. Effect of diethyl aminoethyl hexanoate on the accumulation of high-value biocompounds produced by two novel isolated microalgae. Bioresour. Technol. 197, 178-184. [20] Kim, S.H., Lim, S.R., Hong, S.J., Cho, B.K., Lee, H., Lee, C.G., Choi, H.K., 2016. Effect of Ethephon as an Ethylene-Releasing Compound on the Metabolic Profile of Chlorella vulgaris. J. Agric. Food Chem. 64, 48074816. [21] Kozlova, T.A., Hardy, B.P., Krishna, P., Levin, D.B., 2017. Effect of phytohormones on growth and accumulation of pigments and fatty acids in the microalgae Scenedesmus quadricauda. Algal Res. 27, 325-334. [22] Lagorio, M.G., Cordon, G.B., Iriel, A., 2015. Reviewing the relevance of fluorescence in biological systems. Photoch. Photobio. Sci. 14, 1538-1559. [23] Lee, C., Choi, Y.E., Yun, Y.S., 2016. A strategy for promoting astaxanthin accumulation in Haematococcus pluvialis by 1-aminocyclopropane-1-carboxylic acid application. J. Biotechnol. 236, 120-127. [24] Li, J.H., Niu, X.F., Pei, G.S., Sui, X., Zhang, X.Q., Chen, L., Zhang, W.W., 2015. Identification and metabolomic analysis of chemical modulators for lipid accumulation in Crypthecodinium cohnii. Bioresour. Technol. 191, 362368. [25] Lin, Q., Wu, F., Sheng, P., Zhang, Z., Zhang, X., Guo, X., Wang, J., Cheng, Z., Wang, J., Wang, H., Wan, J., 2015. The SnRK2-APC/C(TE) regulatory module mediates the antagonistic action of gibberellic acid and abscisic acid pathways. Nat Commun 6, 7981. [26] Liu, J., Mao, X., Zhou, W., Guarnieri, M.T., 2016a. Simultaneous production of triacylglycerol and high-value carotenoids by the astaxanthin-producing oleaginous green microalga Chlorella zofingiensis. Bioresour. Technol. 214, 319-327. 27

[27] Liu, J., Sun, Z., Gerken, H., Liu, Z., Jiang, Y., Chen, F., 2014. Chlorella zofingiensis as an alternative microalgal producer of astaxanthin: biology and industrial potential. Mar. Drugs 12, 3487-3515. [28] Liu, J., Sun, Z., Zhong, Y., Gerken, H., Huang, J., Chen, F., 2013. Utilization of cane molasses towards costsaving astaxanthin production by a Chlorella zofingiensis mutant. J. Appl. Phycol. 25, 1447-1456. [29] Liu, J.Y., Qiu, W., Song, Y.M., 2016b. Stimulatory effect of auxins on the growth and lipid productivity of Chlorella pyrenoidosa and Scenedesmus quadricauda. Algal Res. 18, 273-280. [30] Lu, N., Chen, J.H., Wei, D., Chen, F., Chen, G., 2016. Global Metabolic Regulation of the Snow Alga Chlamydomonas nivalis in Response to Nitrate or Phosphate Deprivation by a Metabolome Profile Analysis. Int. J. Mol. Sci. 17, 1-19. [31] Lu, Y.D., Xu, J., 2015. Phytohormones in microalgae: a new opportunity for microalgal biotechnology? Trends Plant Sci. 20, 273-282. [32] Mao, X.M., Wu, T., Sun, D.Z., Zhang, Z., Chen, F., 2018. Differential responses of the green microalga Chlorella zofingiensis to the starvation of various nutrients for oil and astaxanthin production. Bioresour. Technol. 249, 791798. [33] Monroe-Augustus, M., Zolman, B.K., Bartel, B., 2003. IBR5, a dual-specificity phosphatase-like protein modulating auxin and abscisic acid responsiveness in Arabidopsis. Plant Cell 15, 2979-2991. [34] Mulders, K.J.M., Janssen, J.H., Martens, D.E., Wijffels, R.H., Lamers, P.P., 2014. Effect of biomass concentration on secondary carotenoids and triacylglycerol (TAG) accumulation in nitrogen-depleted Chlorella zofingiensis. Algal Res. 6, 8-16. [35] Ortega-Galisteo, A.P., Rodriguez-Serrano, M., Pazmino, D.M., Gupta, D.K., Sandalio, L.M., Romero-Puertas, M.C., 2012. S-Nitrosylated proteins in pea (Pisum sativum L.) leaf peroxisomes: changes under abiotic stress. JExB 63, 2089-2103. [36] Park, J.C., Choi, S.P., Hong, M.E., Sim, S.J., 2014. Enhanced astaxanthin production from microalga, Haematococcus pluvialis by two-stage perfusion culture with stepwise light irradiation. Bioproc. Biosyst. Eng. 37, 2039-2047. [37] Piotrowska-Niczyporuk, A., Bajguz, A., 2014. The effect of natural and synthetic auxins on the growth, metabolite content and antioxidant response of green alga Chlorella vulgaris (Trebouxiophyceae). Plant Growth Regulation 73, 57-66. [38] Shah, M.M.R., Liang, Y., Cheng, J.J., Daroch, M., 2016. Astaxanthin-producing green microalga Haematococcus pluvialis: From single cell to high value commercia products. Front. Plant Sci. 7, 1-28. [39] Shan, X.Y., Yan, J.B., Xie, D.X., 2012. Comparison of phytohormone signaling mechanisms. Curr. Opin. Plant Biol. 15, 84-91. [40] Strader, L.C., Chen, G.L., Bartel, B., 2010. Ethylene directs auxin to control root cell expansion. Plant J. 64, 874884. [41] Sun, H., Kong, Q., Geng, Z., Duan, L., Yang, M., Guan, B., 2015. Enhancement of cell biomass and cell activity of astaxanthin-rich Haematococcus pluvialis. Bioresour. Technol. 186, 67-73. [42] Sun, H.Y., Zheng, M., Song, J.Y., Huang, S.Y., Pan, Y.Z., Gong, R.C., Lin, Z.F., 2019. Multiple-species hormetic phenomena induced by indole: A case study on the toxicity of indole to bacteria, algae and human cells. ScTEn 657, 46-55. [43] Tal, I., Zhang, Y., Jorgensen, M.E., Pisanty, O., Barbosa, I.C., Zourelidou, M., Regnault, T., Crocoll, C., Olsen, C.E., Weinstain, R., Schwechheimer, C., Halkier, B.A., Nour-Eldin, H.H., Estelle, M., Shani, E., 2016. The Arabidopsis NPF3 protein is a GA transporter. Nat Commun 7, 11486. [44] Wen, Z., Liu, Z., Hou, Y., Liu, C., Gao, F., Zheng, Y., Chen, F., 2015. Ethanol induced astaxanthin accumulation and transcriptional expression of carotenogenic genes in Haematococcus pluvialis. Enzyme Microb. Technol. 78, 10-17. 28

[45] Wong, P.K., 2000. Effects of 2,4-D, glyphosate and paraquat on growth, photosynthesis and chlorophyll-a synthesis of Scenedesmus quadricauda Berb 614. Chemosphere 41, 177-182. [46] Yu, X., Niu, X., Zhang, X., Pei, G., Liu, J., Chen, L., Zhang, W., 2015. Identification and mechanism analysis of chemical modulators enhancing astaxanthin accumulation in Haematococcus pluvialis. Algal Res. 11, 284-293. [47] Zhang, Z., Huang, J.J., Sun, D.Z., Lee, Y., Chen, F., 2017. Two-step cultivation for production of astaxanthin in Chlorella zofingiensis using a patented energy-free rotating floating photobioreactor (RFP). Bioresour. Technol. 224, 515-522. [48] Zhao, Y.T., Wang, H.P., Han, B.Y., Yu, X.Y., 2019. Coupling of abiotic stresses and phytohormones for the production of lipids and high-value by-products by microalgae: A review. Bioresour. Technol. 274, 549-556. [49] Zhao, Y.T., Yue, C.C., Ding, W., Li, T., Xu, J.W., Zhao, P., Ma, H.X., Yu, X.Y., 2018. Butylated hydroxytoluene induces astaxanthin and lipid production in Haematococcus pluvialis under high-light and nitrogen-deficiency conditions. Bioresour. Technol. 266, 315-321. [50] Zhekisheva, M., Boussiba, S., Khozin-Goldberg, I., Zarka, A., Cohen, Z., 2002. Accumulation of triacylglycerols in Haematococcus pluvialis is correlated with that of astaxanthin esters. J. Phycol. 38, 40-41.

29

Figure Captions: Fig. 1 Effects of phytohormones on the mean fluorescence intensities (MFI) of C. zofingiensis cells. For convenient comparison, MFI at 533 nm (FL1) and 585 nm (FL2) in a flow cytometer are presented as a ratio divided by MFI in the control. Abbreviations represent phytohormones with different dosages as shown in Table 1. Fig. 2 Biomass concentration of mixotrophic C. zofingiensis in the control and cultures with the addition of phytohormone dosages. All cultures were performed in microplates with nitrogen-deprived medium under 300 μmol m-2 s-1 white light. Fig. 3 Production of astaxanthin (a) and lipids (c) in response to mixotrophic C. zofingiensis in microplate cultivations with phytohormone addition. The percentages of pigments (b) and fatty acids (d) are presented in the total pigments and total fatty acids, respectively. Fig. 4 Hierarchical cluster analysis dendrogram of astaxanthin and lipid abundance in C. zofingiensis cells by phytohormone addition. Fig. 5 OPLS-DA score plot (a) derived from the pigment and fatty acid abundance in C. zofingiensis by phytohormone addition; biplot from OPLS-DA (b) summarizing the effects of phytohormones on the accumulation of pigments and fatty acids. Fig. 6 Hypothetical mechanisms underlying the effects of positive phytohormones on C. zofingiensis. Different phytohormones either activate their particular signal transduction cascades or interact with each other by activating second messengers or phosphorylation cascades to regulate downstream metabolic events in astaxanthin and lipid biosynthesis. Red solid arrows represent stimulatory effects on downstream targets, and green solid arrows 30

represent inhibition of targeted pathways. Yellow lightning arrows indicate the induction of ROS accumulation by the combined stress of high light irradiation and nitrogen deprivation. ROS: reactive oxygen species; CM: cell member; ER: endoplasmic reticulum; CAT: catalase; SOD: superoxide dismutase; GSH-Px: glutathione peroxidase; SCF: skp1-cullin-F-box protein; TIR/AFB: transport inhibitor response protein1/auxin-related F-box; AUX/IAA: auxin-responsive element; ARF: auxin response factors; GID: gibberellin insensitive dwarf; NPF: nitrate/peptide family transporter; APC/CTE: anaphase promoting complex/cyclosome complex; PP2C: protein phosphatase 2c family protein; SnRK2: sucrose nonfermenting related kinase group 2; MAPK: mitogen-activated protein kinase; IBP5: indole-3-butyric acid response 5; CTRs: calcitonin receptors; ETRs: ethylene receptors; AHPs: His phosphotransfers; AHKs: His kinases; NO: nitrite oxide; TCA: tricarboxylic acid cycle; TAG: triacylglyceride; LBs: lipid droplets.

31

Fig. 1

32

Fig. 2

33

(b)

(a)

(d)

(c)

Fig. 3 34

Fig. 4

35

(a)

(b)

Fig. 5

36

Fig. 6 37

Table 1 Dose-dependent effects of phytohormones on MFI in C. zofingiensis a Types

Phytohormones

Abbreviation

Cosolvents

Ethylene precursor

1-aminocyclopropane-1carboxylic acid

ACC

2-Chlorobenzoic acid

Dosages (mg/L)

Effects on MFI (%) c

Low dosage (L)

High dosage (H)

MFI in FL1

MFI in FL2

Water

1.0

5.0 b

- 10.9

- 5.0

CA

Ethanol

5.0 b

10.0

+ 3.2

+ 2.1

Abscisic acid

ABA

Ethanol

10.0 b

25.0

+ 3.7

+ 19.9

Gibberellin

Gibberellic Acid (GA3)

GA

Ethanol

3.5

6.9 b

+ 14.5

+ 15.3

Cytokinin

Kinetin

KT

Ethanol

0.5 b

5.0

+ 2.0

- 5.7

Precursors

Ethanolamine

ETA

Water

76.8

153.7 b

- 38.7

- 40.6

Indole-3-acetic acid

IAA

Dimethyl sulfoxide

7.8 b

15.7

+ 20.3

+ 26.2

Indole-3-butyric acid

IBA

Dimethyl sulfoxide

9.9 b

49.9

+ 21.2

+ 24.9

Indole-3-propioponic acid

IPA

Ethanol

10.0 b

25.0

- 9.3

- 1.2

1-Naphthylacetic acid

NAA

Ethanol

5.3 b

10.7

- 45.7

- 56.1

2,4-Dichlorophenoxy acetic acid

2,4-D/24D

Dimethyl sulfoxide

5.0 b

10.0

+ 17.2

+ 9.5

Diethyl aminoethyl hexanoate

DA-6/DA

Water

2.1

21.5 b

- 14.5

- 11.7

Signal transducers

Auxins and synthetic analogs

Plant growth regulator

Notes: MFI: mean autofluorescence intensity; FL1: 533 nm±15 nm; FL2: 585 nm±20 nm; a Comparison of the MFI of C. zofingiensis in different phytohormone-induced cultures vs.

control cultures (with/without cosolvents); b Optimal dosages of specific phytohormones were selected based on their effects on the MFI of algal cells in FL1 and FL2; c Calculated by comparing the MFI of algal cells in cultures with optimal dosages of phytohormones with control cultures.

Highlights: 38

A two-stage screening strategy was established to identify positive phytohormones Auxins were the most efficient phytohormones for astaxanthin accumulation Maximum contents of astaxanthin (13.1 mg/g) and lipids (64.5 % DW) were obtained Relations between phytohormones and astaxanthin and lipid biosynthesis were revealed

Declaration of Interest Statement

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled “Enhanced coproduction of astaxanthin and lipids by the green microalga Chromochloris zofingiensis: Selected phytohormones as positive stimulators”. 39

Dong Wei, PhD (on behalf of all authors)

Professor School of Food Science and Engineering South China University of Technology 381 Wushan Rd, Guangzhou, 510640, P.R. China Tel: +86-20-87113849 E-mail: [email protected]

40