The genome sequence and gene annotations of P. litchii were obtained from NCBI (BioProject ID: PRJNA290406). Translated protein sequences of all P. litchii genes were analyzed using InterProScan5 (version 5.46) to identify proteins with the cytochrome b₅-like heme/steroid binding domain. The amino acid sequence alignment was generated and adjusted in BioEdit (version 7.0.9.1).

P. litchii wild type (WT) strain SHS3 (Ye et al., 2016), the CK strain, and Δplcb5l1 mutants were cultured on carrot juice agar (CJA) medium (juice from 300 g carrot for 1 L medium, 15 g agar/L for solid media) at 25°C in darkness. The control (CK) strain is a transformant that failed to knockout PlCB5L1. Litchi leaves were harvested from healthy litchi trees in an orchard in South China Agricultural University, Guangzhou, Guangdong province, China. For sporangia production, five 9 mm diameter mycelial plugs were flushed with 2 mL sterilized water, filtering the subsequent suspension with a 100 μm mesh filter. The suspension was incubated at 16°C for 1 h, for zoospores release. After shaking the suspension for 30 s on a vortex oscillator, zoospores were encysted. Cysts were incubated at 25°C 60 rpm for 0.5 h, for cysts germinating. The number of sporangia, release rate of zoospores, and germination rate of cysts were counted under a microscope. The number of oospores was calculated from three 9 mm diameter zones, at 10th days after inoculating on CJA medium at 25°C in the dark (Jiang et al., 2017).

Fungal genomic DNA was extracted from mycelia grown in CJA medium or infected litchi leaves using a Fungal DNA Kit (Omega, America). PCR amplification was performed using Phanta Max Super-Fidelity DNA Polymerase (Vazyme, China). PCR product purification was performed using Cycle Pure Kit (Omega, America) or Gel Extraction Kit (Omega, America). Total RNAs from the life cycle stages of P. litchii, including mycelia, sporangia, zoospores, cysts, germinated cysts, oospores, and stages of infection, were extracted using All-In-One DNA/RNA Mini-preps Kit (Bio Basic, China). FastKing RT Kit (TIANGEN, China) was used for the first-strand cDNA synthesis. The cDNA was stored at –20°C. The expression profile of PlCB5L1 was analyzed with qRT-PCR using SYBR^® Premix Ex Taq™ II (TaKaRa, Japan) and primers PlCB5L1-qRTF/R. P. litchii actin gene (PlActin) (Jiang et al., 2017) was used as a loading control and the relative fold change was calculated using the 2^–ΔΔCT method (Livak and Schmittgen, 2001). Primers used for these analyses were listed in Supplementary Table 1.

A SgRNA was selected and inserted into the sgRNA vector pYF2.3G-RibosgRNA as previously described (Fang and Tyler, 2016; Situ et al., 2020b). To generate gene replacement constructs, 1 kb long upstream/downstream arms of the PlCB5L1 coding region were inserted into pBluescript II KS vector using the ClonExpress MultiS One Step Cloning Kit (Vazyme, China) (Figure 1A). The pYF2.3G-RibosgRNA (PlCB5L1) vector, the hSPCas9 vector pYF2-PsNLS-hSpCas9, and the pBluescript II KS (PlCB5L1) vector were co-transformed into protoplasts of strain SHS3 using PEG-mediated protoplast transformation (Fang and Tyler, 2016). Preliminary transformants were screened by CJA medium containing 50 μg/mL G418. These transformants were further verified by genomic PCR and sequencing. These primers were listed in Supplementary Table 1.

Pathogenicity assays were performed by inoculating 10 μL (20 zoospores per μL) of zoospore suspensions of P. litchii WT, CK, and Δplcb5l1 mutants (M47, M202 and M230) on the abaxial side of litchi leaves at 25°C in the dark. Lesion diameters were measured and photographed 48 h after inoculation. The relative biomass was determined by the ratio of P. litchii DNA to litchi DNA in the inoculated tissues by qRT-PCR using the specific primers for P. litchii and litchi Actin genes (Supplementary Table 1; Zhong et al., 2011). The significant differences were analyzed with Student’s t-test and three independent replicates were set up, with at least 6 leaves in each replicate.

To investigate the sensitivity of PlCB5L1 mutants under different stress conditions, the mycelial plugs (diameter = 9 mm) of Δplcb5l1 mutants were inoculated in the center of the Plich medium (van West et al., 1999) and cultured at 25°C in the dark for 7 days. The Plich media were supplemented with different concentrations of sodium dodecyl sulfate (SDS), Congo Red (CR), Calcofluor White (CFW), H₂O₂, sorbitol, NaCl or CaCl₂. WT and CK strains were used as control. The growth inhibition rate was calculated as: growth inhibition rate (%) = (growth diameter on stress-free plates—growth diameter on stress plates)/growth diameter on stress-free plates × 100%.

To analyze the expression of PlCB5L1 under oxidative stress, the WT strain was cultured in liquid Plich medium for 3 days at 25°C in the dark. The mycelia were immersed in the liquid medium supplemented with 5 mM H₂O₂ for 0, 5, 15 or 55 min. All samples were harvested and the expressional levels of PlCB5L1 were evaluated by qRT-PCR.

DAB staining was performed to visualize the accumulation of reactive oxygen species (ROS) in the infected leaves. The infected leaves were stained with 1 mg/mL DAB solution at room temperature in the dark for 8 h, and then decolorized in 96% ethanol for 48 h (Molina and Kahmann, 2007). ImageJ was used to record the grayscale values of all pixels within the brown areas in the infected leaves.

We screened all proteins encoded in the genome of P. litchii (Ye et al., 2016) using InterProScan and identified 11 Cyt-b₅ superfamily members. In addition to the cytochrome b₅-like heme/steroid binding domain (Cyt-b₅), some proteins also contain additional domains, including the flavin adenine dinucleotide domain (FAD), fatty acid desaturases domain (FA_desaturase), molybdopterin cofactor oxidoreductase dimerization domain (Mo-co_dimer), cGMP-specific phosphodiesterases, adenylyl cyclases, and FhlA (GAF) domain, and oxidoreductase molybdopterin binding domain (Oxidored_molyb) domain (Supplementary Figure 1). Among them, Pl109805 (named PlCB5L1) showed the highest transcriptional levels during infection among the 11 Cyt-b₅ superfamily members, and is dramatically up-regulated in the stages of infection (based on unpublished transcriptome data). Furthermore, the expression profile of the PlCB5L1 were determined by quantitative reverse-transcription polymerase chain reaction (qRT-PCR). Compared with mycelial stage, PlCB5L1 had much higher expression levels in zoospores, cysts, cyst germination, and the early stages of plant infection (1.5 and 3 h post-inoculation, hpi) (Figure 2), suggesting that PlCB5L1 may have important roles in the colonization and early infection of P. litchii. Therefore, we characterized the function of PlCB5L1 in this study.

PlCB5L1 encodes a protein of 160 amino acids (aa) and its Cyt-b₅ domain is located in the C-terminal of the protein (64–160 aa). Sequence analyses showed that a single ortholog of PlCB5L1 is present in each of the 19 sequenced oomycetes and the levels of protein sequence similarity between PlCBL5L1 and its orthologs range between 51.70 and 96.30%, indicating that PlCB5L1 is well-conserved in oomycetes (Supplementary Table 2 and Figure 3). We also compared PlCB5L1 with its most similar homologs in other eukaryotes including fungi, animals, and plants, they possess high levels of sequence divergence (protein sequence similarity < 45%) (Supplementary Table 2).

We generated three PlCB5L1 knockout mutants (M47, M202, and M230) using the CRISPR/Cas9 system, as previously described (Fang and Tyler, 2016; Situ et al., 2020b; Figure 1A). Genomic PCR assays and sequencing results proved that PlCB5L1 was replaced with the NPTII gene in the three mutants (Figures 1B,C). Subsequently, qRT-PCR analysis confirmed that PlCB5L1 was not expressed in these mutants (Figure 1D). A transformant that failed to knockout PlCB5L1 was selected as the control (CK) strain.

To investigate the biological functions of PlCB5L1, the sexual and asexual phenotypes of the Δplcb5l1 mutants, CK and wild-type strains (WT) were examined. We cultured the Δplcb5l1 mutants, WT and CK strains on CJA medium for 5 days at 25°C in the dark, and measured the colony diameter of each strain. As shown in Figures 4A,B, the average growth rates (mm/day) of the mutants (10.45–11.20) were significantly lower than that of the WT (12.19) and CK (12.05), suggesting that PlCB5L1 contributes to the vegetative growth of P. litchii.

Previous studies have shown that Phytophthora spp. cannot synthesize sterols on their own; instead, they use sterol carrier protein elicitins to absorb and metabolize many kinds of sterols from host plants (Boissy et al., 1999; Dahlin et al., 2017). β-sitosterol is one of the most abundant sterols within plant tissues and is useful for mycelial growth of P. infestans in V8 agar medium (Medina and Platt, 1999; Klingberg et al., 2008; Stong et al., 2013). We examined the function of PlCB5L1 on β-sitosterol utilization by culturing Δplcb5l1 mutants, WT, and CK on Plich medium supplemented with 10 mg/L or 20 mg/L β-sitosterol. Their colony diameters were measured 5 days after inoculation at 25°C in the dark. Results showed that the growth-promoting rates of WT and CK were significantly (p < 0.05) higher than that of mutants on Plich medium with 10 or 20 mg/L β-sitosterol (Figure 4C), suggesting that the Δplcb5l1 mutants have impaired capability of β-sitosterol utilization compared with WT and CK.

Next, we evaluated the sporangia number, sporangia size, the rate of zoospores release and cysts germination, as well as the number of produced oospores. Our results showed that the knockout of PlCB5L1 did not have a significant impact on these phenotypes (Supplementary Figure 2).

To determine the role of PlCB5L1 in the pathogenicity of P. litchii, the abaxial surface of tender litchi leaves were inoculated with zoospores suspensions (20 per μL) of WT, CK, and Δplcb5l1 mutants (M47, M202, and M230), and kept at 25°C. At 2 days post-inoculation (dpi), we measured the lesion diameter (Figure 5A) and found that the lesions caused by Δplcb5l1 mutants were significantly (p < 0.05) smaller than WT and CK strains (Figure 5B). Biomass quantification confirmed that the amounts of Δplcb5l1 mutants DNA in litchi leaves were at least 40% lower than that of WT and CK (Figure 5C). These results suggest that PlCB5L1 is required for the full virulence of P. litchii.

To investigate whether PlCB5L1 is related to various abiotic stress responses of P. litchii, the Δplcb5l1 mutants, WT and CK strains were cultured on Plich medium supplemented with different concentrations of sodium dodecyl sulfate (SDS), Congo red (CR), calcofluor white (CFW), H₂O₂, sorbitol, NaCl or CaCl₂. Colony diameter was measured after 7 days of growth at 25°C in the dark (Figure 6A).

The growth inhibition rates of Δplcb5l1 mutants were significantly (p < 0.01) lower than that of WT and CK under the cell wall stress caused by 25 μg/mL SDS and the osmotic stresses caused by 0.2 M sorbitol (Figures 6A,B). On the other hand, Δplcb5l1 mutants were more sensitive to the cell wall stress caused by 350 μg/mL CR and 100 μg/mL CFW, the oxidative stress caused by 2.0 mM H₂O₂, and the salt stress caused by 0.05 M NaCl and 0.1 M CaCl₂ (Figures 6A,B). These results suggest that the function of PlCB5L1 is related to osmoregulation, cell wall integrity, and tolerance to salt and H₂O₂.

Our abovementioned results showed that Δplcb5l1 mutants were more sensitive to H₂O₂ than WT and CK (Figures 6A,B). To investigate the expression levels of PlCB5L1 under oxidative stress, the hyphae of WT were exposed to 5 mM H₂O₂-added Plich medium for 0, 5, 15, and 55 min, in order to simulate oxidative stress imposed by the host upon infection. The qRT-PCR analysis showed that PlCB5L1 expression is significantly up-regulated (p < 0.05) at 15 and 55 min after H₂O₂ treatment (Figure 7A).

Since ROS are known to play a key role in many plant-pathogen interactions (Lamb and Dixon, 1997), ROS accumulation was detected by DAB staining. There is a higher level of H₂O₂ accumulation in litchi leaves inoculated with Δplcb5l1 mutants compared with WT and CK strains at 18 h post-inoculation (Figures 7B,C), suggesting that the Δplcb5l1 mutants showed lower capability of scavenging ROS.

To further investigate the role of PlCB5L1 in scavenging host-derived ROS, we examined the expression levels of five peroxidases and six cytochrome P450 (CYP) genes in WT strain and the Δplcb5l1 mutants. These genes possess highly expression levels in WT strain, based on RNA-seq data (Unpublished). After soaking mycelia in 5 mM H₂O₂-added Plich medium for 5 min, the expression levels of three peroxidase genes (Pl101341, Pl110273, Pl100432) (Figure 7D) and five CYP genes (Pl113076, Pl110055, Pl103820, Pl103856, Pl112304) were significantly (32∼83 and 36∼86%, respectively) lower in Δplcb5l1 mutants comparing with WT (Figure 7E). These results suggest that PlCB5L1 can affect the expression of these peroxidase genes and CYP genes under oxidative stress.

Extracellular laccase activity has been shown to be an important component of plant pathogens defense against oxidative stress (Mayer and Staples, 2002; Yang et al., 2012), therefore we analyzed the laccase activity in Δplcb5l1 mutants, WT and CK strains following an established protocol (Sheng et al., 2015). In the 0.2 mM 2,2-azobis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)-added lima bean agar (LBA) medium, the three mutants accumulated significantly (p < 0.05) lower amounts of oxidized ABTS than WT and CK strains, at 7 days after inoculation, suggesting that Δplcb5l1 mutants had lower laccase activity (Figures 8A,B).

We then examined whether this reduction in laccase activity was due to the down-regulation of laccase genes. We focused on eight laccase genes (Pl103272, Pl104952, Pl106181, Pl106183, Pl106923, Pl106924, Pl111416, and Pl111417), which were selected because they are highly expressed in WT during oxidative stress (5 mM H₂O₂) and their proteins possess signal peptides (Huang et al., 2021). In Δplcb5l1 mutants, the expression levels of four (Pl103272, Pl106181, Pl111416, and Pl111417) laccase-encoding genes were significantly decreased (Figure 8C).