Proteomic Analysis of Pathogen-Responsive Proteins from Maize Stem Apoplast Triggered by Fusarium Verticillioides

Background: Plant apoplast is the frontline battleeld between the host and the invading pathogen. In response to the pathogen attack, a variety of defense-associated proteins are secreted by the host plant in the apoplast to impede the perceived attack. To better understand the plant response after Fusarium verticillioides attack, proteomic analyses of the maize apoplastic uid was done using LC-MS/MS coupled with label-free quantication. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was done to validate the protein quantication data. In total, 5 secreted proteins including pathogenesis-related 6 (PR6), bowman-birk type bran trypsin inhibitor (100284348), 12-oxo-phytodienoic acid reductase (OPR4), Subtilisin-like protease SBT1.9 (103646447), and chitinase 1 (Zm00001d032947) were randomly selected to conrm the protein quantication results through gene expression analysis of the selected proteins. The expression levels of genes were consistent with the protein quantication results (Fig.

performed to nd out the protein alterations in response to various abiotic and biotic conditions [8]. The proteomic studies during the plant-microbe interactions have been done in Arabidopsis [14], rice [15], and poplar [16]. In maize, cell suspension culture was used to study the extracellular proteomic response of the plant to the elicitor from F. verticillioides [17]. However, inplanta apoplastic proteomics response of maize, in response to the F. verticillioides attack, has not been reported previously. Here in this study, we have employed a common and powerful LC-MS/MS technique coupled with label-free quanti cation and identi ed the apoplastic proteins from maize in response to F. verticillioides infection. Furthermore, the identi ed proteins have also been bifurcated based on their secretory nature. The results obtained during this study will broaden our knowledge about the host plant responses towards the pathogenic fungi.

Interaction of F. verticillioides with the maize roots
The effect of F. verticillioides on the maize root morphology was studied by using a hydroponic system (Fig. 1a). The maize seeds were rst allowed to germinate on the germination papers already soaked in Hoagland's solution. After germination, the seeds were inoculated with the pathogen. The presence of the pathogen had a signi cant effect on the morphology of the roots of the seedlings. The inoculated roots were reduced in length as compared to the un-inoculated roots. Brown discoloration of the inoculated roots was also observed after the pathogen infection (Fig. 1b). Moreover, there was a noticeable difference in the total length of the inoculated and un-inoculated seedlings ( Fig. 1b and c).
The root colonization by F. verticillioides was studied using a mixture of two staining solutions, i.e., WGA-Alexa Fluor ™ 488 mixed with Propidium Iodide (PI) or FM ™ 4-64 Dye. The fungal cell wall was stained with WGA-Alexa Fluor ™ 488 (green colour), whereas the plant cell wall was stained with Propidium Iodide (PI), and the plant plasma membrane was stained with FM ™ 4-64 Dye (Red colour). The pathogen was able to colonize all parts of the roots, especially close to the seed ( Fig. 2a and   b). The tips of the secondary roots and intercellular spaces (apoplast) were also colonized by F. verticillioides ( Fig. 2c and d).

Label-free Quantitative Proteomics Analysis Of Identi ed Maize Proteins
A label-free quantitative proteomics approach was employed to analyze the apoplastic proteins secreted by the maize plant after F. verticillioides infection. After quality validation, 9071 peptides were identi ed. The detailed information of the identi ed peptides, including peptide sequence, peptide length, and others, have been mentioned as Additional le S1. A total of 742 proteins were identi ed from the maize plant that was present in all three biological replicates. The detailed information of the identi ed proteins has been mentioned as Additional le S2. To further understand the putative function, all the identi ed proteins were analyzed for their gene ontology (GO) terms, including biological process (BP), molecular function (MF), and cellular component (CC). The proteins showing fold change ≥ 2 at 0.05 p-value were considered as differentially expressed proteins. The upregulated, down-regulated and uniquely identi ed proteins between the inoculated and un-inoculated plants were collectively termed as differentially accumulated proteins (DAPs). Among the differentially accumulated proteins, 35 proteins were upregulated, whereas 18 proteins were down-regulated (Fig. 3). Besides, 65 and 40 proteins were uniquely identi ed in inoculated and un-inoculated plants, respectively (Additional le S3).
Gene Ontology (GO) terms representing all identi ed proteins and differentially accumulated proteins have been shown in Fig. 4a. The GO enrichment analysis for DAPs showed that the two most enriched BP were L-serine biosynthetic process and protein folding. Moreover, the most enriched MF was the cysteine-type endopeptidase inhibitor activity. For the CC category, protein complex, cytoplasm, extracellular region, and thylakoid lumen were signi cantly enriched (Fig. 4b). In addition to this, the enrichment of the KEGG enriched pathways resulted in the identi cation of biosynthesis of amino acids, protein processing in the endoplasmic reticulum, biosynthesis of secondary metabolites and metabolic pathways related to carbohydrates (Fig. 4c).

Identi cation And Annotation Of Secreted Proteins
We extracted the apoplastic uid from the stem of the maize plants infected by F. verticillioides. The proteins secreted by the maize plants were classi ed based on the conventional and unconventional secretion systems. During the conventional secretion system, the proteins are secreted through ER-Golgi mediated secretory pathways. However, in the unconventional secretion system, the ER-Golgi routes are bypassed. The differentially accumulated proteins (DAPs) were scanned for the presence of the signal peptide. Based on the scanning results, the identi ed secreted proteins were divided into two groups (1) Classical secreted proteins (predicted to follow the conventional secretion system), and (2) Non-classical secreted proteins (predicted to follow the unconventional secretion system), also known as leaderless secreted proteins (LSPs).
Among all the identi ed secreted proteins, more than 50% of proteins were predicted to have a signal peptide, hence followed the conventional secretion system. However, in un-inoculated plants, more proteins were secreted through the unconventional secretion system (Fig. 5). These ndings con rmed that in response to the infection of F. verticillioides, maize plants activated both conventional and unconventional systems for the secretion of the apoplastic proteins. Based on the results, it can be speculated that the maize plant may have activated the ER-Golgi secretory pathway of protein secretion to activate the defense response.
The functional annotation of the secreted proteins was conducted by organizing the proteins based on their GO categories. For the biological process (BP) category, most of the secreted proteins were related to the cellular metabolic process, organic substance metabolic process, catabolic process and, response to stress. The pathogen infection has increased the secretion of the proteins associated with the catabolic process, organic substance metabolic process, nitrogen compound metabolic process, and response to stress. In contrast, the proteins related to cellular metabolic processes and biological regulation were reduced after infection of F. verticillioides. However, the secretion of proteins related to carbohydrate metabolic process, protein oligomerization, response to other organisms, and organic substance biosynthesis were induced as a result of pathogen presence (Fig. 6a). Similarly, for the molecular function (MF) category, a higher number of secreted proteins were related to ion binding, oxidoreductase activity, and hydrolase activity (Fig. 6b). However, their concentration was decreased after the pathogen attack. Besides, the number of proteins related to the hydrolase activity and antioxidant activity was increased as a result of pathogen infection. Moreover, the proteins related to the transferase activity, structural constituent of ribosomes, and enzyme regulator activity were only observed after the inoculation by F. verticillioides. These results indicated that the maize plant changed the secretion of various proteins to deter the invasion of F. verticillioides, especially the proteins related to carbohydrate metabolism.
The identi ed proteins were also grouped based on their putative functions. The major protein groups were (a) proteins involved in redox homeostasis, (b) pathogenesis-related protein (PR), defense signaling, proteinase inhibitors, proteins related to glycosyl hydrolase (GH) activity (responsible for carbohydrate degradation), (c) proteases/peptidases that are involved in breakage of the peptide bonds, (d) proteins involved in binding, proper protein folding, and protein stabilization. Besides, (e) protein involved in energy metabolism pathway, (f) transmembrane transport, and (g) some unknown function proteins were also identi ed (Table 1 and Additional le S1). Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was done to validate the protein quanti cation data.
The expression levels of the genes were consistent with the protein quanti cation results (Fig. 7).

Discussion
The presence of the fungus induced two signi cant changes, including the retardation of growth and brown discoloration of the roots (Fig. 1). Besides, F. verticillioides was able to colonize all parts of the roots, including intercellular spaces (apoplast), the surface of the roots, and the tips of the secondary roots ( Fig. 2). Retardation of growth is a common phenomenon after the pathogen attack [18]. The browning of the root tissues indicated the possible accumulation of phenolic compounds by the host, in response to the pathogen infection. The accumulation of the phenolic compounds is a primary stress response for the remodeling of the cell wall [19]. It can also be speculated that the fungus has secreted secondary metabolites such as melanin onto the root surface or may have blocked the oxygen for the plant, which causes the discoloration of the root cells. Therefore, we were interested to know the molecular strategy of the maize plant in the apoplastic region against the pathogen attack.
The vacuum in ltration-centrifugation (VIC) technique has been regarded as an e cient method of apoplastic uid extraction and previously been successfully used to study the interaction of Arabidopsis [14], rice [20], and maize [21] with various pathogens. However, previous studies have reported the presence of proteins from other cellular compartments during the cell wall or apoplast preparations, using various extraction methods in different plant species [22][23][24][25]. In addition to the apoplastic proteins, many proteins related to the other compartments of the cell have also been identi ed in this study. The detection of a higher number of intracellular proteins indicated the disruption of the plasma membrane during the in ltration process [26]. It is well established that the plants use both conventional (ER-Golgi dependent) and unconventional (ER-Golgi independent) secretory pathways for the secretion of the apoplastic proteins [12,27]. Generally, LSPs can account for more than 50% of all the proteins present in the plant secretome [8]. Surprisingly, after F. verticillioides attack, the maize plant increased the secretion of proteins through the conventional secretory pathway. After the pathogen attack, the number of conventionally secreted proteins was increased from 42-61%, whereas the number of LSPs was decreased from 59-34%. In a previous study, the number of LSPs was increased to 54% after Trichoderma virens attack the maize [21]. As high as 60% of proteins identi ed from the apoplast during the interaction of rice plants with Magnaporthe oryzae followed non-classical secretory pathways [15]. It was interesting to note that during this study, the number of LSPs was decreased in response to the pathogen attack. These results suggested that, in response to F. verticillioides, the maize plant may have switched to the classical secretory pathway for the secretion of the apoplastic proteins.
Here, the expression of three peroxidases such as B4FU88, B4FH68, B6THG0 was upregulated after the pathogen attack. Besides, two peroxidase proteins (A0A1D6E530 and A0A1D6IMZ0) were unique to the apoplastic uid of the inoculated plants. However, there was no effect of the pathogen infection on the concentration of one peroxidase protein (A0A1D6KK78) (Additional le S3). These ndings indicated that the former two peroxidase proteins might be particularly crucial during the response of maize to F. verticillioides infection. Additionally, the other proteins related to Reactive oxygen species (ROS) activity, including glutathione s-transferase GSTU6 (B6TLM5), 2-cys peroxiredoxin BAS1 (C4J9M7) and peroxiredoxin (B4FN24) were also identi ed in response to the pathogen infection. Uclacyanin-3 (A0A1D6I054), which is implicated to play a role in the redox reaction during plant-pathogen interaction [28], was also induced by the pathogen attack. Defense-related proteins, including peroxidases, were induced in maize embryos, with varying resistance levels, after the infection of F. graminearum [29]. Reactive oxygen species (ROS) is a ubiquitous and complex early response to the pathogen attack by the plants, required for cell wall strengthening and also has a major role in regulating the programmed cell death [30]. The plant secretes peroxidases as a primary component of defense against the invading pathogens that play a key role in cell wall reinforcements and in producing reactive oxygen species (ROS) [31], and also help to the host plant to tolerate the oxidative stress [32]. It is well known that peroxidases are essential for the activation of ROS during PAMP-triggered immunity (PTI) [33].
The extent of ROS accumulation is governed by the antioxidant system of the plant, which maintains ROS balance for proper cellular functions [34] and the function of ROS as signaling molecules [35]. H 2 O 2 is a major player in signal transduction for downstream defense responses [33]. For the signal speci city of the ROS, redox homeostasis is governed by the presence of low molecular weight antioxidants that absorb and buffer reductants and oxidants [33,36]. The antioxidant status has pivotal importance during the response to the various environmental in uencers, including the pathogen attack. Antioxidants status modulates the ROS balance and can set the benchmarks for general plant defense responses. In this study, the change in the number of antioxidant enzymes, after perceiving the pathogen, points out that the maize plant may have activated the other downstream defense pathways such as cell wall reconstruction and defense-related gene expression.
During the plant-pathogen interactions, induction of pathogenesis-related (PR) proteins has been regarded as a primary defense response that can also help to activate the downstream defense response. Generally, the PR proteins are induced in response to the signaling compounds such as jasmonic acid (JA), salicylic acid (SA), or ethylene [37]. During the interaction of maize with F. graminearum, the most abundant defense proteins were related to PR-10 [38]. In this study, several proteins from PR and PR-like families have been identi ed. These proteins include win1 (B6SH12), chitinase 1 (A0A1D6KV12), basic endochitinase B (C0P3M6), pathogenesis-related protein 1 (A0A1D6HRU2), pathogenesis-related protein 6 (Q2XXB3). Two cysteine proteinase inhibitors (K7VLF2, Q30KW0) and two proteins such as A0A1D6L886 and K7UAX3 related to the germinlike activity were also identi ed. Moreover, proteins belonging to the GH family, including K7V329, C0HEI0, and B4G1J7, were induced after the F. verticillioides infection. The GH family proteins play a role in the maintenance of the plant cell wall. The induction of B6U1 × 3 (PMR5N domain-containing protein) [39] and B4G233 (Dirigent protein) [40] further strengthened the idea of cell wall remodeling during the attack of the pathogen. Other defense-related proteins include putative beta-D-xylosidase 5 (B8A1R0), glucan endo-1,3 beta-glucosidase 7 (B6TU78), and polygalacturonase QRT3 (B4FAQ3) were also induced after the attack of the pathogen. The handling of the samples has caused the secretion of one bowman-birk type bran trypsin inhibitor protein (C0PNX6), that has been secreted by the host plant in response to the mechanical damage [37]. However, the expression of this protein was down-regulated in the presence of F. verticillioides.
Moreover, caffeoyl-CoA o-methyltransferase 1 (A0A1D6ES04), which is involved in the production of phenolic compounds [41,42], was also identi ed after the pathogen infection, which supports the notion that the change in root colour was due to the possible accumulation of phenolic compounds by the host plant. Another protein, ATP-dependent (S)-NAD(P)H-hydrate dehydratase (B4FJJ9), related to the repair of the metabolic pathways for stress adaption, was also identi ed after the infection of the pathogen [43]. In rice, the resistance to the Magnaporthe grisea was enhanced through the activation of pathogenicity related genes involved in the defense hormone Jasmonic Acid (JA) pathway [44]. Here, two proteins, including 12-oxo-phytodienoic acid reductase (Q49HE1) and peroxisomal-CoA synthetase (E7DDV3), are putatively related to the Jasmonic acid (JA) pathways, were also activated. The expression level of the former was increased, whereas the latter was detected only in response to the F. verticillioides attack. Both proteins have a major role in the conversion of JA precursor to JA after three rounds of beta-oxidation [45][46][47]. These results indicated that the maize plants might have activated the JA pathways to defend the invading pathogen.
Proteases/peptidases are the proteins secreted by the plants to cope with the invading pathogen by maintaining the cell wall.
About 15% of the apoplastic proteins, secreted by the tobacco plant, were associated with the peptidase/protease family [48]. During the interaction of rice with M. oryzae, the rice plant induced 5 proteins related to proteases family [15]. Aspartic proteases have been reported to play a critical role in defense signaling in plants [49]. In this study, we have identi ed eight proteins belonging to the protease/peptidase family. These proteins include two subtilisin-like proteases, three aspartyl proteases, two cysteine protease family proteins, and one carboxypeptidase (Table 1). In addition to this, the proteins involved in binding (RNA-, heme-, and metal ion-binding), proper protein folding, and synthesis of other proteins were also observed during this study. These proteins include histone H4 (implicated to bind with other proteins during plant-pathogen interactions) [50], cytochrome B5 isoform D (involved in electron transfer activity during the lignin synthesis) [51], leucine-rich repeat family protein (involved in pathogen recognition) [52], two small heat shock proteins (sHSP) (required for proper folding of the proteins) [53].
Additionally, various ribosomal proteins (involved in protein synthesis, protein stabilization, and binding with other proteins) such as 60S ribosomal protein L2, ribonuclease 2, putative elongation factor 1-gamma 2, phosphoserine aminotransferase, and two nascent polypeptide-associated (NAC) complex alpha subunit-like protein were also identi ed in this study. The ribosomal proteins play an important role in the defense of the host plants against various stimuli [8]. Moreover, the proteins involved in energy production pathways were also present in the apoplastic uid of the maize. The proteins which are involved in the energy metabolism pathway include B6TF15, B6T7D4, Q8S4W8, K7UAI4, B6TF15 and E9NQE4, which is implicated to play a role in photosynthesis [54]. The proteins responsible for the transport of solutes and water, namely early nodulin-like protein 3 and aquaporin PIP1-2 and unknown functions including C0P566 and B4G0J0, were also in uenced in the presence of F. verticillioides.

Conclusions
This study clari es that maize plants responded to F. verticillioides attack by secreting complex arsenals of proteins involved in a variety of pathways. The increase in peroxidase proteins after the pathogen attack indicated the activation of strong innate immune responses by the maize plant. The activation of PR proteins is an obvious and earliest response by the host to the pathogen attack. However, the involvement of carbohydrate metabolism, protein binding, protein stabilization, and other defense-related proteins indicated that the host plants make a rigorous protein reprogramming in the apoplastic region. The results presented here will provide a better understanding of the apoplastic immune response from the host plant in response to the pathogens.

Plant and Fungal Materials
The seeds of maize (B73), obtained from the Chinese Crop Germplasm Information System (CGRIS), Chinese Academy of Agricultural Sciences (CAAS), Beijing, were surface sterilized by dipping in 2% Sodium Hypochlorite (NaOCl) solution followed by treatment with 70% Ethanol for 5 minutes each and washed at least three times with sterilized water. Sterile germination papers (30 × 45 cm; Anchor Paper Company, MN, USA) were used for the germination of the seeds. The germination papers were soaked in sterile Hoagland's solution [55]. The seeds were allowed to germinate in a contamination-free and humidity-

Isolation of apoplastic proteins
For the extraction of apoplastic uid, the stem of the maize plants was cut into small sections (5 cm) using a sterile razor blade. Three replicates were used for each condition: inoculated and un-inoculated plants at 7 days post-inoculation. The apoplastic uid was collected using the vacuum in ltration-centrifugation (VIC) methodology with slight modi cations [58]. Brie y, the samples were washed using sterile water to remove any contamination from the surface and dried by gently blotting with tissue paper. The weight of the sample was measured, and the samples were then placed in a 20 ml syringe and lled with distilled water. The air in the syringe was removed by pushing the plunger. After that, the tip of the syringe was covered with a piece of para lm, and negative pressure was generated by pulling the plunger. Then the plunger was carefully released to avoid any cell lysis and cytoplasmic contamination. Now the syringe was unplugged to eject any air inside and replugged. Then a modest positive pressure was created inside the syringe by pressing the plunger carefully. The process was repeated until the whole plant tissue was in ltrated. The sample was removed from the syringe and gently blotted with the piece of absorbent paper to remove the liquid outside the surface of the samples. The sample was weighed again, and approximate volume was calculated by the difference in weight before and after the in ltration. The samples were placed in a 20 ml syringe without plunger and inserted into a 50 ml centrifuge tube, and centrifuged at 2000×g for 15 minutes at 4 o C (Eppendorf, Hamburg, Germany). The harvested apoplastic uid was further ltered through a cellulose acetate membrane (0.22 µm pore size) to ensure the removal of any cells or particulate matter in the uid. The apoplastic uids from uninoculated and inoculated plants were concentrated by freeze-drying (Thermo Fisher Scienti c, USA), and rehydration of the samples was achieved by adding a 20 µl buffer solution (25 mM Tris-HCL and 100 mM NaCl, pH=7.4). The samples were stored at -80 o C until the next use.

Protein Digestion
The protein sample for each treatment was mixed with a lysis buffer containing 4% SDS, 100 mM Tris-HCL, and 1mM DTT (pH 7.6) (Bio-Rad, USA) and boiled for 15 minutes. The debris was removed by centrifugation at 14000×g for 40 minutes. After centrifugation, the supernatant was retained and quanti ed by the help of the BCA Protein Assay Kit (Bio-Rad, Beijing, China). The protein solution was added with 30 µl SDT buffer containing 4% SDS, 150mM Tris-HCL and 100mM DTT (pH=8.0) (Bio-Rad, USA). The sample volume was reduced by ultra ltration (Microcon Units, 10 kD) after adding UA buffer (8 M Urea, 150mM Tris-HCL, pH=8.0) to the samples. For alkylation, the samples were treated by 100µl iodoacetamide and were placed in complete darkness for 30 minutes at room temperature. The lters were washed with 100 µl UA buffer three times, followed by washing with 100 µl of 25 mM ammonium bicarbonate (NH 4 HCO 3 ). Finally, the protein suspension was diluted by adding 40 µl of 25 mM NH 4 HCO 3 buffer and digested by adding trypsin (Promega, Madison, USA). The sample to trypsin mass ratio was adjusted as 50:1 for the rst round of digestion at 37 o C overnight, and 100:1 for the second round of digestion for 4 hours.
The resulting peptides were collected by ltration. About 200 µg of each protein sample was digested by trypsin. The peptides for each sample were desalted using Empore C18 solid-phase extraction column (Sigma-Aldrich, Beijing, China) and concentrated by vacuum centrifugation. The resulting peptides were reconstituted in 40 µl of 0.1% (v/v) of formic acid before analysis.

HPLC and LC-MS/MS Analysis
The HPLC and LC-MS/MS analysis were done by Shanghai Applied Protein Biotechnology, Ltd. Shanghai, China. Brie y, the peptides mixture was added to the buffer A (0.1% Formic acid) and loaded onto a C18 trap column (Thermo Scienti c Acclaim PepMap100, 100 μmx2 cm, nanoViper C18, 3 μm, 100 Å). The reverse-phase trap column was connected to the C18 analytical column (Thermo scienti c EASY column, 10 cm, ID 75 μm, 3 μm, C18-A2). For the separation of peptides, a linear gradient of buffer B (84% acetonitrile and 0.1% Formic acid) was injected at a ow rate of 300 nl/min using IntelliFlow TM technology. The time for the linear gradient of buffer B was set to 120 minutes (0-55% for 110 min, 55-100% for 5 min, hold in 100% for 5 min).
For LC-MS/MS analysis, the Q Exactive TM mass spectrometer coupled with Easy nLC Liquid Chromatograph (Thermo Fisher Scienti c Co. Ltd., Shanghai, China) was used for 120 minutes. The instrument was operated on positive ion mode and run with peptide recognition enabled. MS data was acquired by series of cyclic scans at a high resolution of 70,000 using a datadependent method dynamically choosing the most abundant precursor ions from the survey scans (300-1800 m/z) followed by scans at a relatively low resolution of 17,500 at 200 m/z. The survey scan width was set as 2 m/z. The automatic gain control (AGC) target was set to 3e6, whereas the maximum time for injection was set to 10 ms. The dynamic exclusion duration was set as 40 sec. The normalized collision energy was 30 eV, and the under ll ratio was de ned as 0.1%. The under ll ratio speci es the minimum percentage of the target value to reach the maximum ll time.

Database search and label-free quanti cation analysis
The MS data were searched against the Maize Uniprot proteome database (https://www.uniprot.org/proteomes/UP000007305), including other contaminants (total number of entries 132460) using Andromeda connected with the MaxQuant Software version 1.3.0.5 (Max Plank Institute of Biochemistry, Martinized, Germany) using the default parameters. To quantify the proteins, Carbamidomethylation was set as a xed modi cation, whereas the oxidation as a variable modi cation. A mass error of 20 ppm was allowed with 2 min retention time for shift tolerance. False Discovery Rate (FDR) threshold for proteins and peptides was less than 1%, the peptides having less than 7 amino acids were not included. Besides, proteins having unique peptides were considered for quanti cation.

Protein annotation
For protein annotation, the Blast2GO bioinformatics platform was used to search the sequences. For pathways analysis, the online Kyoto Encyclopedia of Genes and Genomes (KEGG) (https://www.genome.jp/kegg) database was searched, and the proteins were mapped for KEGG pathways. Subsequently, the corresponding GO terms and KEGG pathways were extracted.

Reverse transcription-quantitative PCR validation (RT-qPCR)
To validate the protein quanti cation results, RNA from each sample was extracted using the commercial RNA extraction Kit (Easy Spin, NBFB, Beijing, China) by following the manufacturer's instructions. The rst-strand cDNA was synthesized from 3 µg of total RNA with the help of a commercial cDNA synthesis kit (Takara, Dalian, China) according to the manufacturer's instructions. RT-qPCR was done using the SYBR Green kit (NovoStart® SYBR qPCR SuperMix Plus) on QuantStudio (TM) 6 Flex System (Thermo Fisher Scienti c, Shanghai, China). The experiment was conducted in triplicate. The maize actin gene was used as an internal standard to calculate relative fold-changes based on comparative cycle threshold (2 −ΔΔCt ) values. The primers were designed using the DNASTAR laser gene version 7.

Consent for publication
Not applicable.

Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional les.
Competing interests WG conceived and designed the experiments. HAH and JZ performed the experiments. JZ, HAH, and WG analyzed the data and wrote the manuscript. WG, YSG, and MXG coordinated the project. The nal manuscript was read and approved by all the authors.   Differentially expressed maize proteins in response to the F. verticillioides infection Heat map showing the differentially expressed (upregulated and down-regulated) maize proteins in response to the F. verticillioides infection after 7 days of inoculation. Bar colours representing the protein expression. Red colour means upregulated expression, and the blue colour means down-regulated expression. Hierarchal clustering was done based on protein expression (p-value ≤ 0.05) between uninoculated (Un1, Un2, Un3) and inoculated (In1, In2, and In3) maize plants. Classical and non-classical secretory proteins of maize Comparison of the proteins secreted by F. verticillioides inoculated and un-inoculated maize plants at 7 days post-inoculation. The presence of the signal peptide was con rmed using the SignalP website. SecretomeP was used to identify the unconventionally secreted or leaderless secreted proteins (LSPs).