Fenton Reaction Facilitates The Fungal Infection On The Unicellular Alga Haematococcus Pluvialis and Discovery of a Biosafe and Environment-Friendly Chemical That Blocks The Infection

The green microalga Haematococcus pluvialis is used as cell factories for producing astaxanthin, the high-value natural compound with multiple biological functions. However, H. pluvialis is prone to the infection by a parasitic fungus Paraphysoderma sedebokerense, which is the most devastating threat to the mass culture of H. pluvialis all over the world. Little is known for the mechanisms underlying the infection process, though it is of great essential for developing effective measures to mitigate the pathogen threatening for the natural astaxanthin industry. We observed that there were heat-stable substances with small molecular weight produced during the infection process, which signicantly stimulated the parasitism process and enhanced the susceptibility of H. pluvialis cells to the pathogen. Systematic approaches including multi-omics, biochemical and imaging analysis were deployed to uncover the identity of the metabolites and the underlying mechanisms. Two metabolites, 3-hydroxyanthranilic acid and hordenine were identied and proved to stimulate the infection via driving fenton reaction mediated oxidative stress to H. pluvialis. The reaction generated hydroxyl radicals to disrupt the subcellular components of the algal cells and to make the algal cells more susceptible to the infection. Based on these ndings, a biosafe and environment-friendly antioxidant butylated hydroxyanisole was selected to inhibit the fungal infection, which completely abolished the infection at 12 ppm.


Background
The green microalga Haematococcus pluvialis is used as cell factories for producing astaxanthin, the high-value natural compound with multiple biological functions. However, H. pluvialis is prone to the infection by a parasitic fungus Paraphysoderma sedebokerense, which is the most devastating threat to the mass culture of H. pluvialis all over the world. Little is known for the mechanisms underlying the infection process, though it is of great essential for developing effective measures to mitigate the pathogen threatening for the natural astaxanthin industry.

Results
We observed that there were heat-stable substances with small molecular weight produced during the infection process, which signi cantly stimulated the parasitism process and enhanced the susceptibility of H. pluvialis cells to the pathogen. Systematic approaches including multi-omics, biochemical and imaging analysis were deployed to uncover the identity of the metabolites and the underlying mechanisms. Two metabolites, 3-hydroxyanthranilic acid and hordenine were identi ed and proved to stimulate the infection via driving fenton reaction mediated oxidative stress to H. pluvialis. The reaction generated hydroxyl radicals to disrupt the subcellular components of the algal cells and to make the algal cells more susceptible to the infection. Based on these ndings, a biosafe and environment-friendly antioxidant butylated hydroxyanisole was selected to inhibit the fungal infection, which completely abolished the infection at 12 ppm.

Conclusions
This study provide for the rst time, a framework to dissect the functions of secondary metabolites in the interaction between the unicellular algal H. pluvialis and its fungi parasite, indicating that oxidative degradation is a strategy used for the fungal infest. Eliminating the oxidative burst through adding antioxidant butylated hydroxyanisole could be an effective measure to reduce parasitic infection in H. pluvialis mass culture. Consequently, it is di cult to produce bulk volume of microalgal biomass at low cost due to the occurrence of various predators and pathogens in mass cultivation.
The green unicellular microalga Haematococcus pluvialis is a freshwater bi agellate and belongs to the class Chlorophyceae, order Volvocales, which is well known for its ability in accumulating up to 5% of the dry weight biomass of natural bio-active compound astaxanthin under stresses conditions such as nitrogen de ciency and high light irradiation (Damiani et al., 2010;Han et al., 2013;Harker et al., 1996;Ren et al., 2021). Therefore, H. pluvialis is considered as the most sustainable feedstock for the commercial production of astaxanthin (Khoo et al., 2019), which has an estimated market value of USD 240 million in 2021 (Ren et al., 2021). However, the development of H. pluvialis mass culture industry has been retarded by contaminations of fungal parasite, which is often accompanied by reduced biomass yield and astaxanthin productivity (Ding et al., 2020;Shah et al., 2016). A parasitic fungus Paraphysoderma sedebokerense (Blastocladiomycota) infect the H. pluvialis cells in a highly speciesspeci c manner, which is the most devastating threat to the mass culture of H. pluvialis all over the world (Ding et al., 2020;Hoffman et al., 2008;Hwang et al., 2019). Once the pathogen appears in the mass culturing system, the infected algal cells are dying during very short period of time, causing severe economic loss for the natural astaxanthin manufacturing industry (Gutman et al., 2009;Strittmatter et al., 2016). Several strategies have been developed to control the pathogenic fungi, such as maintaining the pH of culture system at the acidic condition and application of sodium dodecylbenzene sulfonate (SDBS) (Ding et al., 2020;Hwang et al., 2019). More environment-friendly and biosafe measures are desired and necessary to improve the sustainability of H. pluvialis mass culture.
Understanding about the mechanisms underlying the infection process is essential for developing effective measures to mitigate the pathogen threatening for the natural astaxanthin industry. Interactions between H. pluvialis and P. sedebokerense have been investigated in previous studies. Sugar moiety on the H. pluvialis cell wall was proposed to play a role as binding molecule for recognition by P. sedebokerense (Gutman et al., 2011). Heat-stable recognition sites on the algal cell were necessary for the fungal attachment and encystment, while signal transduction in H. pluvialis cells was obligated for the fungal sporangium development and epidemic (Asatryan et al., 2019). Interaction between the cell walls of H. pluvialis and carbohydrate activated enzymes in P. sedebokerense was believed to be essential for the parasitism process (Lin et al., 2021). These studies are focused on the speci c recognition of H. pluvialis cells by P. sedebokerense and the warfare between them during the early-stage of infection, but little is known for the mechanisms underlying the post-infection process, especially for the fast material degradation in algal cells.
In this study, we observed that the accumulated secondary metabolites in the infection system signi cantly stimulated the parasitism process and enhanced the susceptibility of the algal cells to the pathogen. With the aim to uncover the identity of them and to understand the underlying mechanisms, systematic approaches including multi-omics, biochemical and imaging analysis were deployed herein. Two metabolites, i.e. 3-hydroxyanthranilic acid (3-HAA) and hordenine, were identi ed and proven stimulating the infection process via non-enzymatic fenton reaction in a congruent manner. The hydroxyl radicals produced from the reaction promoted the infection through causing strong oxidative stresses to the host cells and impairing the algal cell structures along with degradation of the intracellular components. Intentionally, application of the antioxidant butylated hydroxyanisole (BHA) to the algalfungal system reduced the infection ratio effectively, indicating that oxidative degradation is a strategy used for the fungal to successfully infect algal cells while eliminating the oxidative stress was practicable for mitigating fungal infection in H. pluvialis mass culture. This study provided a framework to dissect the functions of secondary metabolites in the interaction between the unicellular algal and its fungal pathogen, and developed a new crop protection measure to improve the sustainability of algal mass cultivation. When the H. pluvialis cells cultivated in 360 L panel photobioreactors were infected by P. sedebokerense, algal cells died rapidly and seriously ( Fig. 1A-B). We collected the supernatant post infection (SPI) by removing both the host and fungal cells via centrifugation, and investigated the effects of SPI on the newly infection process. When the healthy algal cells were pre-treated with SPI and then challenged with the fungal swarmers, the infection process was signi cantly promoted (Fig. 1C). On the second day post inoculation, the color of the control cell culture was dark-green and a few algal cells were attached with the fungal swarmers. By contrast, a large number of dead algal cells were observed in the culture of algal cells pre-treated with SPI, which settled down to the bottom of ask. On the third day post inoculation, the color of the culture of algal cells pre-treated with SPI turned to brownish, whereas the control remained green. Pre-treating the algal cells with SPI prior to the pathogenic challenge caused signi cantly higher infection ratio after the second day post inoculation than the control (Fig. 1D). Additionally, the SPI remained the infection enhancing activity after pre-treated in 98 o C water bath for 15 min, indicating that the activity of SPI was not eliminated by heating. It was further observed that in the 48h-SPI treated algal cells, the pigments were partially degraded and the cellular starch granules disappeared (Fig. 1E).

Results And Discussion
Quantitative analysis revealed that when compared to the control, the contents of the total cellular carbohydrates and pigments (i.e. carotenoids and chlorophyll) were reduced in the algal cells treated with SPI by 50% and 20%, respectively (Fig. 1F). Additionally, after ltered with 3000 Da cut-off membrane, the activity of the ltrate did not signi cantly altered in degrading carbohydrates and pigments in the algal cells, suggesting that the activity in the SPI was most likely attributable to small molecules. These results together suggested that some heat-stable substances with small molecular weight produced during the infection process and were capable of enhancing the susceptibility of H. pluvialis cells to the pathogen by affecting the algal cell integrity.

SPI induced oxidative stresses within algal cells
Alterations in the subcellular structures of the algal cells challenged with both P. sedebokerense and SPI were observed with transmission electron microscopy (TEM) (Fig. 2A). Degradation of the subcellular membrane systems were observed in the algal cells treated with SPI without the involvement of fungus.
Additionally, the algal cell walls were loosened after being treated with SPI. These results indicated that the substances in SPI degraded the algal cellular components and destructed membranes.
To uncover the identity of the secondary metabolites in SPI, transcriptomic analysis was conducted to facilitate understanding the effects of SPI on the algal cells. A total of 998 and 490 genes were up-and down-regulated in H. pluvialis, respectively, after the SPI treatment for 24 h. Expression of many genes involved in biotic stresses responses were signi cantly altered in the algal cells treated with SPI (Fig. 2B). Several genes coding for the anti-oxidative enzymes were found to be signi cantly up-regulated while genes coding for synthesis and transportation were down-regulated. Up-regulation of the genes involved in oxidative stress responses indicated that SPI may contain substances that can cause the generation of ROS (Hasanuzzaman et al., 2020;Torres et al., 2006). To test this hypothesis, the oxidative activities of SPI were measured by using the thiobarbituric acid (TBA) assay with fenton reagent as the positive control because it is a known reaction that generates oxidative stress through small molecules (Arantes et al., 2012; Eastwood et al., 2011). The results showed that the SPI possessed strong oxidative activity in vitro (Fig. 2C). In addition, the SPI showed lipid peroxidation activity when acting on the algal cellular membranes, leading to formation of malondialdehyde (Fig. 2D). To further identify the ROS produced by SPI, the dimethyl sulfoxide trapping method was used and the results suggested that SPI could produce hydroxyl radical in vitro (Fig. 2E).
Based on the transcriptomic results and a suite of observations and biochemical assays, it can be concluded that SPI contained substances that exerted oxidative stresses via generation of ROS in the algal cells. Oxidative degradation of the algal subcellular structures might be the cause of decreased resistance to fungal infection.

Secondary metabolites mediated fenton reaction facilitates the fungal infection
Metabolomic analysis was performed for identi cation of the small molecules causing the oxidative stresses. The SPIs were collected at different infection stages, i.e., 1, 3 and 5 day post inoculation of the fungus into the algal cell cultures. It was found that the degradation activity of SPI collected on Day (D) 5 was signi cantly higher than that on D1, suggesting that the concentration of the metabolites of target increased over 5 days. Based on this, 62 metabolites which showed over 2-fold increases on D5 than that on D1 were selected, most of which were organic acid, dipeptide, amino acid and derivatives. Ten metabolites, including tyramine, trimethoprim, indole-3-carboxylic acid, hordenine, deoxycytidine, 4pyridoxic acid, lumichrome, 3-hydroxyanthranilic acid (3-HAA), baclofen and cyclohexylamine, with the phenol/quinone/aromatic structure, were retrieved manually (Fig. 3), since such types of compounds are known to be able to mediate the fenton reaction producing hydroxyl peroxide (Arantes et al. When these substances were added into the algal cell culture, it was rstly observed that hordenine and 3-HAA signi cantly reduced the contents of carbohydrates while tyramine, hordenine and cyclohexylamine caused degradation of the pigments in the treated algal cells (Fig. 4A). Secondly, the potential infectionprompting effect of the candidate metabolites was checked. On the 3 rd dpi (day post infection), the algal cells pre-treated with either hordenine or 3-HAA showed signi cantly enhanced infection ratio than that of the control and other compounds (Fig. 4B).
The fenton reaction is initiated from the reduction of Fe 3+ to Fe 2+ , which is the key factor for driving fenton reaction (Kameshwar & Qin, 2018). Thus the reducing activities of the 10 candidate metabolites were tested. Among them, 3-HAA showed the strongest activity in reducing Fe 3+ to Fe 2+ at 4 h (Fig. 4C).
Additionally, 3-HAA and hordenine generated hydroxyl radical in the assay with DMSO as substrate (

Application of antioxidant to inhibit the fungal infection
As the oxidative stresses caused by the SPI impaired the algal cell structures and promoted the fungal infection process, an exogenous antioxidant was introduced to relieving such oxidative stress in the culture to inhibit the infection. BHA is one of the most commonly used synthetic antioxidants in food and biodiesel fuels to prevent oxidation for its low cost, high stability and effectiveness (Rodil et , 1999). BHA was added into the infection system at different nal concentrations (i.e. 2 ppm, 7 ppm and 12 ppm) (Fig.  5). The infection ratio of the newly infected algal cells was calculated to re ect the infection inhibitory effect. Compared to the untreated H. pluvialis culture, addition of BHA at 2 ppm delayed the complete fungal infection for 1 day. Elevating the concentration of BHA to 7 ppm decreased the infection greatly, and the infection was only about 30% on day 5, while the untreated group was 100% infected. Application of 12 ppm BHA to the system completely suppressed the fungal infection.
According to all the results described above, a model was proposed herein to illustrate the major ndings of this study. Production of secondary metabolites such as 3-HAA and hordenine in the infection system mediated the generation of hydroxyl radical via the fenton reaction, the most reactive free radicals among various ROS, which disrupt the subcellular components of the H. pluvialis cells and make the algal cells more susceptible to the infection. However, by adding 12 ppm of the antioxidant BHA to the culture, the fungal infection was completely abolished, indicating the oxidative burst is essential for the pathogens to infest.

Conclusion
In this study, two secondary metabolites (i.e. 3-HAA and hordenine) were identi ed from the fungal infection system, which can drive fenton reaction thereby causing severe oxidative stresses on the host cells. Based on these ndings, the antioxidant BHA was used to successfully block the fungal infection. This proof-of-concept study indicated utilization of antioxidants represents a novel strategy to reduce the fungal infection in microalgal mass culture.

Strains and culturing conditions
H. pluvialis cells were maintained in our laboratory (Lin et al., 2021) and cultured in the BG11 growth medium at 21-23°C under continuous illumination (20 μmol·m -2 ·s -1 ). The fungal parasite P. sedebokerense used in this study was isolated in the previous study (Lin et al., 2021). P. sedebokerense cells were grown in the fungal growth medium supplemented with yeast extract and peptone (Hoffman et al., 2008) on an orbital shaker at a speed of 150 rpm maintained at 30°C.

Preparation of the supernatant post infection (SPI) and SPI incubation assay
When the OD 600 of the fungal cell cultures reached 0.03 within 5 days, 1% (v/v) of the fungal cells was inoculated in the algal cell cultures (about 3.0 × 10 5 algal cells mL -1 ). The mixture was then incubated on an orbital shaker at a speed of 150 rpm at 30°C. Samplings were performed on a daily basis and the infection ratios (Gutman et al., 2011) were determined by cell counting under microscope (Olympus, BX53 with a DP70 CCD camera). To prepare for the SPI, when the algal cells were completely infected by the parasitic fungus, the culture mixture was centrifuged at 2,100 g for 5 min to collect the supernatants and was ltrated by using double-layered 1.2 μm glass micro ber lters (Whatman, UK), heated at 95°C in water bath for 15 min and were stored at 4°C, such liquid was termed as supernatant post infection (SPI).
For detecting the infection enhancing activity of the SPI, 100 mL of algal cells with density of 5.0 × 10 5 cells mL -1 from exponential growth phase was harvested at 700 g for 3 min, the pellet was pre-treated with 100 mL SPI or BG11 medium for 48 h at 30 °C, under light intensity of 20 μmol·m -2 ·s -1 at 150 rpm. The liquid was removed via centrifugation at 700 g for 3 min, and the cell pellet was re-suspended in 100 mL of BG11 medium. Fungal cells were inoculated in the algal cell cultures, samples of each experiment were checked with microscope to count the prevalence of infection every day as described above. Three independent biological replicates were conducted for each sample and the quantitative data were presented as mean ± S.D. (n=3). **, p < 0.01 (Student's t-test).

Transmission electron microscopy and starch staining
Samples of healthy H. pluvialis cells that grown at logarithmic phase, or incubated with SPI for 3 days were centrifuged at 700 g for 3 min, respectively. The cell pellet of each sample was washed twice with fresh BG11 medium, and was xed overnight in 2.5% (w/v) glutaraldehyde at 4°C. The owing sample processing procedures included osmium xation, dehydration, in ltration, polymerization, section and staining (Wayama et  (MDA) kit (Jiancheng Bioengineering Institute, Nanjing, China) was introduced to measure the in vivo lipid peroxidation in the SPI treated H. pluvialis cells. Ferrozine assay was conducted to measure the ferric ions reducing activity (Gibbs, 1976).

Analysis of the hydroxyl radical and hydrogen peroxide production
The hydroxyl radicals oxidize dimethyl sulfoxide (DMSO) and generate formaldehyde, which can react with 2, 4-dinitrophenylhydrazine (DNPH) and form hydrazone (HCHO-DNPH). HCHO-DNPH can be analyzed by HPLC, eluted peaks of HCHO-DNPH with maximum absorbance at 355 nm was targeted (Tai et al., 2004). Fenton reagent containing 0.2 mM Fe 2+ and 8 mM H 2 O 2 was used as the positive control to generate hydroxyl radicals. To measure the intracellular hydrogen peroxide (H 2 O 2 ), the uorescence dye 2', 7'-dichlorodihydro uorescein diacetate (DCFH-DA, Sigma, USA) was introduced to stain the treated algal cells (Gwak et al., 2014). The stained cells were analyzed by using the ow cytometer and the uorescence intensity (excitation wavelength at 488 nm) of the cells re ected the intracellular H 2 O 2 level.

Metabolomics
The SPI on day 1, 3 and 5 were collected respectively, and the supernatants were used for the ultra-high performance liquid chromatography coupled with hybrid quadrupole time-of-ight mass spectrometry (UHPLC-QTOF-MS) analysis. Two independent biological repeats were conducted for each sample. Fold change (peak intensity of D5/D1) > 3 was screened and compounds with phenol/quinone/aromatic moiety (Korripally et al., 2013) were particularly targeted. Target reagents were purchased from Sigma-Aldrich, cyclohexylamine was dissolved as 2% (v/v) solution in BG11 medium and the other 9 reagents were dissolved as 0.2% (w/v) solution in BG11 medium. These solutions were put in a 95°C water bath for 15 min and cooled down and applied appropriate ultrasonic treatment to enhance the dissolution, the supernatants of each solutions were collected to further test their activity.

RNA extraction and RNA-seq analysis
For analysis of algal response to SPI treatment, 100 mL of the H. pluvialis cell culture (5.0 × 10 5 cells mL -1 ) were incubated in SPI or BG11 medium (as control), respectively for 24 h. One mL of the algal cell cultures was then centrifuged to collect the pellet, and was frozen with liquid nitrogen immediately. The RNA extraction and RNA-seq analysis were performed according to the methods described by Ma et al. (Ma et al., 2020). Differentially expressed (DE) genes with Log 2 FC > 1.5 was considered, and the upand down-regulated gens was checked and analyzed manually. Three independent biological repeats were conducted for each sample. (Biotechnology Co., Ltd. Shanghai, China) provided the support on metabolomics analysis. We thank Engineer Zhenfei Xin (Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China) for assistance in electron microscopy and Dr. Hu Jin (Center for Microalgal Biotechnology and Biofuels, Institute of Hydrobiology, Chinese Academy of Sciences) for offering kind advice in submitting the manuscript.

Data availability
Transcriptome data is available at NCBI Sequence Read Archive (https://www.ncbi.nlm.nih.gov/sra) with accession number PRJNA720251.

Declaration of Competing Interest
The authors declare that they have no competing nancial interests or personal relationships that could have appeared to in uence the work reported in this paper.    Identi cation of the metabolites with putative functions in causing oxidative stresses in the host algal cells through comparative metabolomics analysis. The formula of each metabolites was referred to Sigma-Aldrich and KEGG data-base (https://www.genome.jp/kegg/), the relative content of each metabolites from D1: D3: D5 was also indicated below.