Sodium alginate was degraded by Aly2 for 12 h, and then AOS with a degree of polymerization (DP) of 2–5 was obtained (Peng et al., 2018) by using a Superdex 30 Increase 10/300 GL column, the mobile phase was 0.20 M NH₄HCO₃ at a flow rate of 0.4 mL/min, and the eluted fractions were monitored at 232 nm using a UV detector. In this study, wild-type Nicotiana benthamiana and wild-type A. thaliana were preserved and propagated. The plants Solanum tuberosum and N. benthamiana were grown at 22°C and 25°C, respectively, under 70% humidity with 16 h of light and 8 h of dark. Arabidopsis thaliana receptor mutants were cultured in an incubator (22°C, 16 h light/8 h dark). RNA extraction kits were purchased from Kangwei Reagent (Taizhou, China), reverse transcription kits were purchased from Hunan Aceri Bioengineering Co. (Changsha, China) and Novozymes (Beijing, China), and HRP-conjugated His₆ was purchased from Sigma Aldrich (St. Louis, MO, USA). Biotin hydrazide, 2-(N-morpholino)-ethanesulfonic acid (MES), and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Escherichia coli strains DH5α and BL21 (DE3) and Pichia pastoris were preserved in our laboratory. Phytophthora infestans strain was grown in the dark at 18°C using rye A agar. The T-DNA insertion mutants SALK_007193C for At3g21630 (AtCERK1) and SALK_206271C for At2g17120 (AtCEBiP-LIKE1) were obtained from EDITGENE Corporation (Guangzhou, China). All the other chemicals and reagents were of the highest quality. Primers were synthesized by Shanghai Shenggong Biotechnology Co. (Shanghai, China; Table 1 ).

Plates with P. infestans were flooded with 5 mL of ddH₂O and scraped to release sporangia. The suspension was poured into a clean Petri dish, placed on ice, and stored at 4°C for 3 h to release zoospores. Then, the sporangia were counted and adjusted to 30,000 sporangia per milliliter. The potato leaves were inoculated with P. infestans (originally isolated from the province of Heilongjiang, north of China) at a concentration of 4 × 10⁵ sporangia mL⁻¹. Droplets of 20 µL of P. infestans zoospores and sporangia suspension were added to the leaves of wild-type N. benthamiana, wild Arabidopsis, and Arabidopsis receptor mutants after AOS (100 μg/mL) treatment for 24 h. The leaves were placed in a plastic dish with ddH₂O sprayed over them regularly. Phytophthora infestans infection was observed at 3 days post-inoculation (dpi), and P. infestans colonization was measured by quantitative real-time PCR. Briefly, total DNA was extracted from diseased leaves, including AOS-treated and untreated leaves, and then quantitative PCR was performed after DNA extraction by using the internal reference gene primers in Table 1 (Nb-actin and P. infestans-O8 primer pairs). The expression levels of the internal reference gene of tobacco and P. infestans were measured to evaluate P. infestans accumulation. The primer sequences for the experiment are presented in Table 1 .

The optimal working concentration of AOS was determined by spraying 0, 25, 50, 100, or 200 μg/mL of gradient AOS aqueous solution on potato Désirée/Eshu 3 leaves for 24 h, followed by inoculation with P. infestans; the other conditions were the same as those described above. There were three biological replicates for each treated sample.

Histochemical staining was performed using 3,3'-diaminobenzidinebutane (DAB) and nitroblue tetrazolium (NBT) to detect hydrogen peroxide (H₂O₂) and superoxide ion (O²⁻) accumulation, respectively, in the leaves of plants subjected to AOS and H₂O treatment. Plant tissues were placed in 1 mg/mL of DAB solution, vacuum-infiltrated for 30 min, washed three times with deionized water, and reacted with H₂O₂ for 12–24 h under light at 28°C. The excess dye solution was washed away by using a boiled solution (ethanol:lactic acid:glycerin = 3:1:1) at 100°C for 10 min, and the leaves were imaged. For NBT staining, N. benthamiana leaves were immersed in 1% (M/V) sodium azide solution, which increases the permeability of cells, vacuum-immersed for 30 min, and then transferred to 0.5 mg/mL of NBT solution, followed by vacuum infiltration for 30 min. O²⁻ reacted with NBT to form a deep blue insoluble complex. The boiled solution (ethanol:lactic acid:glycerin = 3:1:1) was also used to wash off the excess dye solution, after which the leaves were imaged. The strengths of H₂O₂ and O²⁻ were quantitatively analyzed by ImageJ software.

For aniline blue staining, 20 mL of lactic acid, 20 mL of phenol, 40 mL of 20% glycerol, and 20 mL of deionized water were mixed evenly; the volume was adjusted to 100 mL; and then anhydrous ethanol was added at a volume ratio of ~2:1. Nicotiana benthamiana leaves were put into the above solution under vacuum for 30 min and then treated at 60°C for 30 min. The 0.01% aniline blue solution was added after washing with deionized water, and the leaves were kept at room temperature overnight without light. Finally, the leaves were preserved in 50% glycerol and were observed and photographed under a fluorescence microscope.

Nicotiana benthamiana leaf epidermis strips were soaked in MES, pH 6.0 buffer under light for 3 h to open the stomata, and then a final concentration of 20 µmol/L of Fluo-3AM was added at 4°C for 2.5 h. The excess fluorescent dye was washed by using the MES buffer and then kept at room temperature for 1 h. The epidermis strips were treated with ddH₂O and 100 μg/mL of AOS for 24 h, and then fluorescence was observed using a confocal microscope. Fluo-3AM was used to analyze Ca²⁺ accumulation in the guard cells. Each treatment included an investigation of at least three epidermis strips, and the experiment was repeated three times. Images of the stomatal aperture were captured with an Olympus BX43 microscope (Olympus, Tokyo, Japan) using the cellSens Standard software, and the diameters of 50 randomly selected stomata were measured. Each assay was repeated three times.

Total RNA was extracted from tobacco leaves with TRIzol reagent (TaKaRa, Shiga, Japan) according to the manufacturer’s instructions. The cDNA was synthesized from 1 μg of total RNA using a FastKing gDNA Dispelling RT SuperMix kit (Tianjin, Beijing, China). Quantitative real-time PCR (qRT-PCR) was performed by using a Talent SYBR Green Kit (Tianjin, Beijing, China). Each reaction was conducted in triplicate and repeated three times. Bio-Rad CFX Manager software (Bio-Rad, California, USA) was used to analyze the data. The relative expression levels of the ROS-scavenging enzymes catalase (CAT), superoxide dismutase (SOD), ascorbate peroxidase (APX), and ROS-generating-related genes, including the respiratory burst oxidase homolog genes (RbohA and RbohB), were measured by qRT-PCR after AOS treatment for 0, 2, 4, 8, 12, and 24 h, respectively. Primer sequences for the experiment are presented in Table 1 .

The transcriptome was sequenced by Shanghai OE Biotech Co., Ltd. (Shanghai, China). RNA samples were taken from five- or six-leaf-stage leaves treated with H₂O (0 h) or 100 μg/mL of AOS for 24 h. Each sample was analyzed three times. The samples were selected depending on quality (RIN score ≥ 7). All differential gene expression data were based on the following criteria: an absolute log₂ ratio ≥1 and an FDR ≤0.001.

The two PCR products of the AtCERK1 gene and the AtCEBiP-LIKE1 gene were individually cloned into the pET-30a (+) vector and SacI vector. The proteins AtCERK1 (extracellular domain, At3g21630) and AtCEBiP-LIKE1 (At2g17120) were expressed by using BL21(DE3) and yeast with a His₆ tag at the C-terminus, respectively. For AtCERK1 expression, E. coli cells harboring the recombinant plasmid were initially cultured in LB broth. When the cell density reached an OD₆₀₀ of 0.8–1.0, the broth was supplemented with the inducer isopropyl 1-thio-β-D-galactopyranoside at a final concentration of 0.05 mM to initiate the expression of AtCERK1. AtCEBiP-LIKE1 was expressed in a similar manner, and yeast cells harboring AtCEBiP-LIKE1 were cultured in BMGY broth and then in BMMY broth. The broth was supplemented with 1% methanol to induce AtCEBiP-LIKE1 expression. Then, the proteins were purified by Ni²⁺ chelation chromatography according to the method provided by Peng et al. (2018).

Molecular interactions were analyzed by using biotin-labeled AOS and CERK1/CEBiP-LIKE1 based on the ELISA method. AOS (8 mg/mL) was biotinylated in 0.1 M of MES (Sigma-Aldrich) (pH 5.5) biotin LC-hydrazide solution (Deepa et al., 2002; Peng et al., 2021). EDC (1 mg) was added to the reaction mixtures and reacted overnight at room temperature, and each reaction mixture was desalted three times with PBS and centrifuged (4°C, 2,000 rpm, 2 min) to obtain the biotin-labeled AOS. Then, 50 μL of 1 mg/mL streptavidin was added to a 96-well plate and sealed at 4°C overnight, 1% BSA was added for 1 h, and biotin-labeled AOS was added at room temperature for 2 h. The binding reaction of chitin-binding proteins (AtCERK1 and AtCEBiP-LIKE1) was carried out at 4°C overnight, and the above process was avoided from light and washed with PBS. Then, the antibody was added for 30 min, the sample was washed with PBST, TMB was used to develop the color, and the absorbance was measured at 450 nm.

The potato cultivar Désirée and P. infestans were used as materials to examine the activity of AOS with a DP of 2–5 ( Figure 1A ) against late blight. The potato leaves were inoculated with P. infestans after being sprayed with different concentrations of AOS for 24 h at 0, 25, 50, 100, and 200 μg/mL. Images were photographed at 4 dpi, and AOS could significantly enhance the resistance of potato to late blight. The infected area and disease index was gradually decreasing with increasing AOS concentration, but it was greater when the concentration was increased to 200 μg/mL ( Figures 1B, C ). We hence used 100 μg/mL of AOS in the next experiment.

To investigate whether AOS regulates ROS accumulation, DAB and NBT staining were applied to evaluate the H₂O₂ and O²⁻ levels in tobacco leaves that were detached from the water-spraying group and the 100-μg/mL AOS-spraying group. Figures 2A, B show that DAB staining was first strengthened and then weakened with the extension of time. DAB staining was the deepest after 24 h of spraying AOS, which indicated that the accumulation of H₂O₂ was elevated. Similarly, NBT staining was the deepest after 24 h of spraying AOS, suggesting that AOS could also promote O²⁻ accumulation in plants.

The mRNA levels of several important genes encoding the ROS-scavenging enzymes CAT, SOD, APX, and ROS-generating-related genes, including the respiratory burst oxidase homolog genes (RbohA and RbohB), were determined by qRT-PCR analysis and monitored before and after AOS treatment. After AOS treatment, there was a significant increase in the expression level of the ROS generation-related gene RbohB ( Figure 2C ). Conversely, the expression level of the CAT gene decreased significantly, while there was no significant change in the expression levels of the SOD and APX genes ( Figure 2D ). These results suggest that AOS may enhance hydrogen peroxide accumulation by inhibiting CAT gene expression and promoting RbohB gene expression.

Ca²⁺ is an important secondary messenger that triggers plant defense response. Ca²⁺ usually stays at a low concentration in the plant cell cytoplasm, whereas biotic stresses, pathogen infection, and elicitor promote Ca²⁺ influx from the extracellular to the cytoplasm, thus leading to a rapid transient cytoplasmic Ca²⁺ increase (Zang et al., 2019). To investigate whether AOS leads to Ca²⁺ influx, the fluorescent-labeled Fluo-3AM was used to evaluate the cytoplasmic Ca²⁺ levels. ddH₂O and COS were used as negative and positive controls, respectively (Zhang et al., 2009; Iriti and Varoni, 2015). There was no obvious fluorescence in the ddH₂O-treated plant tissues, while the COS- and AOS-treated guard cells showed obvious fluorescence in the cells, indicating that AOS treatment significantly promoted Ca²⁺ influx in the guard cells. Moreover, AOS promoted stomatal closure, suggesting the prevention of pathogen infection ( Figure 3 ).

To ascertain the ability of AOS to induce callose deposition in plants, wild-type N. benthamiana was subjected to 100 μg/mL of AOS spray treatment, while H₂O was used as a control. Subsequently, aniline blue staining was conducted after 24 h. Microscopic examination ( Supplementary Figure S1A ) of the stained samples revealed conspicuous callose deposition surrounding the veins of N. benthamiana treated with AOS compared to the control group. The fluorescence intensity ( Supplementary Figure S1B ) was quantified and found to be consistent with the observed phenotype. These results indicate that AOS can effectively enhance callose deposition in plants and consequently improve their resistance against pathogen infections.

Plants synthesize SA mainly by the isochorismate synthase (ICS) and phenylalanine ammonia lyase (PAL) pathways. Thus, we detected the gene expression level of the ICS and PAL genes by using qRT-PCR. The results showed that the PAL gene was upregulated by AOS, whereas the ICS gene basically remained unchanged ( Figure 4A ). These results suggested that AOS promotes SA synthesis by improving the PAL transcription level.

Pathogenesis-related (PR) proteins play an important role in plant defense. They can improve plant disease resistance by inhibiting pathogen reproduction and are mainly involved in plant-acquired systemic resistance. PR-1A and NPR are the key factors in the SA pathway, involved in N. benthamiana’s resistance to pathogens or other pathogens, such as Phytophthora (Zang et al., 2019). The expression levels of genes related to these signaling pathways were examined by qRT-PCR in N. benthamiana leaves after AOS treatment for 24 h. The expression levels of the PR1a and NPR1 genes, key genes in the SA pathway, were upregulated significantly ( Figure 4B ). This suggests that AOS can promote the expression of the NPR1 and PR1 proteins via the SA signaling pathway, enabling plants to acquire systemic resistance and enhance their resistance to late blight.

The MEK2 (MAPK kinase)-SIPK/WIPK cascade, an N. benthamiana mitogen-activated protein kinase (MAPK) cascade, is an essential signaling pathway for plant immunity and is involved in the hypersensitive response (HR) accompanied by cell death. Figure 4B shows that MEK2 genes were also upregulated significantly, suggesting that AOS may activate plant immunity through the MAPK cascade.

Transcriptome sequencing analysis was performed on the leaves treated with AOS to reveal the role of AOS in inducing plant resistance. The leaves were harvested after 24 h of 100 μg/mL AOS treatment. The differentially expressed genes (DEGs) after AOS and H₂O treatment (AOS0H) were analyzed. The results revealed 2,595 DEGs in the AOS24H group vs. the AOS0H group, of which 1,219 genes were upregulated and 1,376 genes were downregulated ( Figure 5A ).

Most of the AOS-regulated genes at 24 hpt were annotated with a wide range of Gene Ontology (GO) terms in the molecular functions ( Figure 5B ). Molecular function results suggested receptor activity and receptor regulator activity, and the transcriptome DEG showed that the CERK1 gene expression level was upregulated, consistent with the results in Supplementary Figure S2B . The enriched GO terms ( Table 2 ) of the biological process category included positive regulation of defense response, chitinase activity, and signal transduction. Within the KEGG classification, plant hormone signal transduction genes showed the greatest changes in expression ( Figure 5C ), and 155 DEGs were upregulated ( Figure 5D ). These results further verified that resistance to P. infestans was induced by the SA signaling pathway. Plant chitinases are described as pathogen-associated proteins because they are induced in response to invasion by plant pathogens. The genes related to the chitin catabolic process and chitinase (Niben101Scf01789g03003, Niben101Scf02041g00002, and Niben101Scf02171g00007) were also significantly upregulated ( Table 2 ). Based on the above results, we speculate that the receptors in the plants of AOS may be related to chitin elicitor receptor proteins.

To determine the AOS receptor protein in plants, several Arabidopsis receptor loss mutants, including the T-DNA insertion mutants for At3g21630 (AtCERK1) and At2g17120 (AtCEBiP-LIKE1), were inoculated with P. infestans. There was no significant difference in mortality in the Arabidopsis mutant group after AOS or ddH₂O treatment, while AOS improved the resistance to P. infestans in the Columbia wild-type Arabidopsis group ( Supplementary Figure S2A ). The levels of the CEBiP-LIKE1 gene and the AtCERK1 gene were significantly increased after AOS treatment for 24 h ( Supplementary Figure S2B ).

To analyze the interaction between AOS and the receptor proteins, the receptors AtCERK1 (extracellular domain, At3g21630) and AtCEBiP-LIKE1 (At2g17120) were expressed and purified. SDS-PAGE showed that the molecular weights of AtCERK1 (with amino acids ranging from 26 to 230) with a His₆ tag at the C-terminus and AtCEBiP-LIKE1 (with a secretory peptide at the N-terminus and a His₆ tag at the C-terminus) were approximately 23 kDa ( Figure 6A ) and 47 kDa ( Figure 6B ), respectively, which are consistent with the theoretical molecular weights.

Then, ELISA was used to analyze the interaction between the biotin-labeled AOS and the different concentrations of AtCEBiP-LIKE1, as well as between the biotin-labeled AOS and the different concentrations of AtCERK1. The OD₄₅₀ absorption value was greatest (0.60) when the AtCERK1 concentration was 1 ng/μL but was only 0.056 when the AtCEBiP-LIKE1 concentration was 1 ng/μL. The data suggested that AOS binds tightly to AtCERK1, while AOS cannot bind tightly to AtCEBiP-LIKE1. Furthermore, COS was used as a control and had a binding affinity to AtCERK1 and AtCEBiP-LIKE1, but the affinity was lower than that of AOS ( Figures 7A, B ). For further confirmation of AOS combined with CERK1, different concentrations of AtCERK1 and AtCEBiP-LIKE1 and a mix of AtCERK1 and AtCEBiP-LIKE1 were used to check the affinity to AOS. The results ( Figure 7C ) showed that with the increase of CERK1 concentration, the absorption value was higher to a certain extent; however, the concentration of AtCEBiP-LIKE1 had no significant effect on the absorption value. Identical molar values of AtCERK1 and AtCEBiP-LIKE1 were used to bind AOS, and the absorption value was lower when only AtCERK1 was used. The above results confirmed that AOS binds tightly to AtCERK1, indicating that AtCERK1 is the main plant receptor of AOS and is involved in the resistance induced by AOS. Moreover, the ELISA process and a schematic diagram of the interaction between AOS and AtCERK1/AtCEBiP-LIKE1 are shown in Figure 7D .