Downregulation of Light-Harvesting Complex II Induces ROS-Mediated Defense Against Turnip Mosaic Virus Infection in Nicotiana benthamiana

The light-harvesting chlorophyll a/b complex protein 3 (LHCB3) of photosystem II plays important roles distributing the excitation energy and modulating the rate of state transition and stomatal response to abscisic acid. However, the functions of LHCB3 in plant immunity have not been well investigated. Here, we show that the expression of LHCB3 in Nicotiana benthamiana (NbLHCB3) was down-regulated by turnip mosaic virus (TuMV) infection. When NbLHCB3 was silenced by tobacco rattle virus-induced gene silencing, systemic infection of TuMV was inhibited. H2O2 was over-accumulated in NbLHCB3-silenced plants. Chemical treatment to inhibit or eliminate reactive oxygen species (ROS) impaired the resistance of the NbLHCB3-silenced plants to TuMV infection. Co-silencing of NbLHCB3 with genes involved in ROS production compromised the resistance of plants to TuMV but co-silencing of NbLHCB3 with genes in the ROS scavenging pathway increased resistance to the virus. Transgenic plants overexpressing NbLHCB3 were more susceptible to TuMV. These results indicate that downregulation of NbLHCB3 is involved in defense against TuMV by inducing ROS production.


INTRODUCTION
Chloroplasts are not only the organelles where photosynthesis takes place, but also the biosynthetic site of defense-related molecules like reactive oxygen species (ROS) and phytohormones like jasmonic acid (JA) (Fonseca et al., 2009) and salicylic acid (SA) (Strawn et al., 2007). ROS are able to kill pathogens directly by oxidation. The accumulation of ROS is known as the pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) response (Park et al., 2012). Increasing evidence shows that ROS produced by plant plasma membrane respiratory burst oxidase homologs (RBOHs) act as secondary messages which are involved in different signal transduction pathways triggering a physiological or programmed pathway for cell death (Baxter et al., 2014;Schieber and Chandel, 2014;Adachi et al., 2015;Waszczak et al., 2018). Recently Wu et al. (2020) found that a leucine-rich-repeat receptor kinase named hydrogen-peroxide-induced Ca 2+ increases (HPCA) is a sensor to perceive extracellular H 2 O 2 and induce activation of Ca 2+ channels to trigger an influx of Ca 2+ ions in Arabidopsis.
To efficiently harvest solar energy, light-harvesting antenna protein complexes bind chlorophylls and carotenoids and then create a number of light-harvesting chlorophyll a/b binding proteins named LHCBs or LHCAs (Jansson, 1999;Klimmek et al., 2005Klimmek et al., , 2006. LHCB proteins are associated with the lightharvesting complexes of photosystem II (PS II), while LHCA proteins are associated with those of photosystem I (PS I). PS I is central to the light-driven conversion of water to molecular oxygen, and consists of a PS II core dimer (C 2 ) bound by four trimeric LHCBs (LHCB1-3) that are surrounded by the monomeric LHCB4 (CP29), LHCB5 (CP26), and LHCB6 (CP24) (Caffarri et al., 2009). By modulating the rate of state transitions, LHCB3 is involved in excitation energy transfer and charge separation and is also required for stomatal response to abscisic acid (Damkjaer et al., 2009;Xu et al., 2012;Adamiec et al., 2015). In Arabidopsis, LHCB3 affects the macrostructure of photosystem II and the rate of state transitions (Damkjaer et al., 2009;Kouřil et al., 2013;Pietrzykowska et al., 2014;Longoni et al., 2015), and in tomato and oil palm, the LHCB3 gene confers continuous light tolerance and enhances yield (Velez-Ramirez et al., 2014;Neoh et al., 2019). LHCB3 is involved in cyclic electron flow and then affects state transition by interacting with PGR5a in cucumber (Wu et al., 2021b). However, it is not known whether LHCB3 functions in plant immunity.
We now report that the expression of Nicotiana benthamiana LHCB3 (NbLHCB3) is downregulated in plants infected with turnip mosaic virus (TuMV). When NbLHCB3 was silenced, systemic infection of TuMV was inhibited. Further analysis showed that the accumulated ROS in the NbLHCB3-silenced plants form an essential part of the plant defense against TuMV infection. The results indicate a mechanism in which the downregulation of NbLHCB3 is involved in defense against TuMV by inducing ROS.

Plant Materials and Growth Conditions
Nicotiana benthamiana plants were grown in pots with premixed soil in a climate chamber at 26 • C, with 60% humidity and under a 16-/8-h photoperiod (light daily from 08:00 to 24:00) with an average light intensity of 130 µmol m −2 s −1 .

Viral Inoculation and Detection
Agroinfiltration was done as described before (Li et al., 2019) using Agrobacterium tumefaciens transformed with viral infectious clones infiltrated into N. benthamiana leaves at the stage when three or five true leaves were fully expanded. For mechanical inoculation, N. benthamiana leaves previously infiltrated with agrobacteria carrying an infectious clone and displaying viral symptoms were homogenized in 0.1 M phosphate buffer (pH 7.2) using a sterile pestle and mortar and then filtered. Standardized volumes of infectious sap (or phosphate buffer for controls) were then applied to the upper surface of fully expanded leaves which had been sprayed with carborundum.
The distribution of virus in infected tissues or plants was first determined by the intensity of visible green fluorescence under a handheld UV lamp. The transcriptional or translational levels of viral capsid protein were then determined by qRT-PCR, northern blot and western blot using the procedures described below.

Agrobacterium-Mediated Transient Expression in N. benthamiana Leaves
The cDNA fragment of NbLHCB3 was cloned into expression vectors with different tags using ligation-independent cloning (LIC) exactly as described previously (Wang et al., 2019). Recombinant clones were transformed into Agrobacterium GV3101 by electroporation, grown at 28 • C overnight, collected by centrifugation, resuspended in infiltration buffer [10 mM MgCl 2 , 10 mM 2−(N−morpholino) ethanesulfonic acid (MES), and 200 µM acetosyringone, pH 5.6] and kept at room temperature for 4 h. Fully expanded true leaves of N. benthamiana were infiltrated with A. tumefaciens resuspension solution (OD 600 = 1.0) and were harvested at 36∼72 hours postinfiltration (hpi) for further research. For co-expression, equal concentrations and equal volumes of individual Agrobacterium cultures were mixed.

Virus-Induced Gene Silencing
The tobacco rattle virus (TRV)-based virus-induced gene silencing (VIGS) system was used to silence genes in N. benthamiana. This system includes TRV vectors, pYL196 (TRV1) and pYL156 (TRV2), both of which were kindly provided by Prof. Yule Liu at Tsinghua University (Beijing, China) (Liu et al., 2002). Specific primers were designed and used to amplify a ∼300 bp cDNA fragment of the target gene from the N. benthamiana cDNA library. For virus-induced double gene silencing, two fragments were amplified and then linked by homologous recombination. The fragment was then cloned into the vector pYL156 by the LIC method (Wang et al., 2019) to yield the VIGS vector. The negative control vector was prepared using a fragment of the gene encoding glucuronidase. Equal volumes of Agrobacteria carrying pYL196 and pYL156 constructs at OD 600 = 1 were mixed and co-infiltrated into N. benthamiana plants. At least 12 plants were used for each silencing treatment and the same number of plants were used as controls. The infiltrated plants were returned to the climate chamber and used for further analysis at about 15 days post-infiltration (dpi).

Total RNA Extraction and RNA Analysis
Total RNA was extracted from leaf tissues of N. benthamiana using Trizol reagent (Thermo Fisher, Waltham, MA, United States) as instructed. DNA contamination was removed and cDNA synthesized using ReverTra Ace R qPCR RT Master Mix with gDNA Remover (TOYOBO, Japan). Real-time PCR was performed in a 10 µL volume containing 2 µL of 25-fold diluted cDNA, 5 µM each primer, and 1 × SYBR Green PCR Mix (Vazyme, China) in a LightCycler R 480 II Real-Time PCR System (Roche, Switzerland) and the results were analyzed by the CT method (Livak and Schmittgen, 2001). The N. benthamiana Ubiquitin C gene (UBC, AB026056.1) (Rotenberg et al., 2006;Shi et al., 2016) was used as an internal reference for analysis. Four individual plants per treatment were sampled and each treatment was repeated three times independently.
Northern blot of TuMV RNAs was performed as described before (Wang et al., 2019). The genomic RNA of TuMV was detected by a 260-bp fragment of TuMV CP gene labeled with digoxin (DIG) according to the DIG Oligonucleotide 3 -End Labeling Kit protocol (Roche, Switzerland). Ten micrograms of total RNA per sample was loaded on 1.5% formaldehyde-agarose gels and was separated by electrophoresis at 50 V for about 2 h in 1 × MOPS running buffer, and blotted to Hybond-NX nylon membranes (Amersham Biosciences, United Kingdom) for more than 16 h by capillary transfer. Subsequent procedures including pre-hybridization and hybridization were done following the protocol of the DIG High Primer DNA Labeling and Detection Starter Kit II (Roche).

ROS Detection and Treatment With Chemical Inhibitors
To detect H 2 O 2 or O 2 − by histochemical staining, six leaves of each plant were collected at 22:00, immersed and vacuum-infiltrated in freshly prepared solutions of either 3,3diaminobenzidine (DAB) [1 mg/ml DAB in Tris-HCl buffer (pH 3.8)] or NBT [0.2% (W/V) NBT in 50 mM PBS (pH 7.5)]. After incubation overnight in darkness at room temperature they were destained in 95% (W/V) ethanol by heating in a boiling waterbath to remove the chlorophyll. Images were captured with a Canon digital camera.
Diphenylene iodonium (DPI) (Mei et al., 2017;Zhang et al., 2020) is a chemical inhibitor of RBOHs. A concentrated stock solution of DPI (10 mM) was prepared by adding 10 mg DPI to 3.179 ml anhydrous dimethylsulphoxide (DMSO). This stock solution was then diluted with Agrobacterium infiltration buffer to provide a final working concentration of 50 µM for infiltration into N. benthamiana leaves. Dimethylthiourea (DMTU) (Jiang and Zhang, 2002;Li et al., 2016) is a chemical inhibitor of H 2 O 2 and responsible for scavenging H 2 O 2 . A concentrated stock solution of DMTU (150 mM) was prepared by adding 0.1563 g DMTU to 10 ml ddH 2 O. This stock solution was then diluted with Agrobacterium infiltration buffer to provide a final working concentration of 5 mM for infiltration into N. benthamiana leaves. cMyc-fused beta-glucuronidase protein (GUS-cMyc) was used as control. Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; ***P < 0.001).

The Expression of NbLHCB3 Was Downregulated in TuMV-Infected N. benthamiana
In our previous work, we described the transcriptome of N. benthamiana plants infected with TuMV (Wu et al., 2021a). One gene with downregulated expression had the highest identity to AtLHCB3. For further analysis, we therefore first identified this gene from the genomic sequence of N. benthamiana. Using AtLHCB3 (At5G54270) as query, NbLHCB3 (Sequence ID Niben101Scf07510g00016.1) was identified in the Solanaceae Genomics Network. 1 It was predicted to encode a protein with 265 amino acids. NbLHCB3 had 88.8% protein identity and 79.9% nucleotide identity to AtLHCB3 (Supplementary Figure 1). To confirm its expression response to TuMV infection, N. benthamiana leaves were mechanically inoculated with GFPlabeled TuMV so that viral infection could easily be monitored by detecting GFP fluorescence (Figure 1A). At 4 dpi, the inoculated leaves (IL) were collected and the expression of NbLHCB3 was determined (Figure 1). There was a significant decrease 1 https://solgenomics.net/ in accumulation of both NbLHCB3 transcripts and NbLHCB3 protein (Figures 1B,C) compared to mock-inoculated controls, suggesting that NbLHCB3 was downregulated in the TuMVinfected plants.
To determine which viral protein is responsible for downregulated expression of NbLHCB3, we transiently expressed cMyc-fused TuMV proteins individually by agroinfiltration and detected the expression level of NbLHCB3 in leaves at 2 dpi. In infiltrated leaf areas or regions expressing NIa-cMyc, expression of NbLHCB3 was downregulated significantly, but not in those areas or regions expressing the other TuMV proteins, indicating that TuMV NIa is responsible for downregulation of NbLHCB3 in the infected plants ( Figure 1D).

Silencing of NbLHCB3 Reduced the Infection of TuMV in N. benthamiana
To investigate the function of NbLHCB3 in plants infected by TuMV, we used the TRV-induced gene silencing system (VIGS), creating a VIGS vector (TRV:LHCB3) to silence NbLHCB3 in at least 12 plants and then mechanically inoculated these plants with TuMV-GFP. At 12 dpi of TRV:LHCB3, the expression of NbLHCB3 was decreased to 30% of the normal (TRV: 00 control) level (Figures 2A,B). We also determined the The chlorophyll contents [measured as described before (Yang et al., 2020)] were reduced in the NbLHCB3-silenced plants.
(D) Detection of TuMV RNAs and CP protein accumulation in plants inoculated with TRV:LHCB3 or TRV:00 at 4 dpi. Polyclonal antibody to TuMV CP and a CP probe tagged with DIG were used for analysis. Bars represent the standard errors of the means from three biological repeats. The relative intensity of the blot signal quantified by IMAGE J software is shown on the lanes. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; ***P < 0.001).
Frontiers in Microbiology | www.frontiersin.org expression of other members of the NbLHCB family in the TRV:LHCB3-infected plants but found no obvious difference in their expression compared to that in the control TRV:00-infected plants, indicating that NbLHCB3 had been specifically silenced in the TRV:LHCB3-infected plants (Supplementary Figure 2). The silenced plants showed mild yellowing with reduced chlorophyll content (Figures 2A,C). The plants inoculated with TRV:LHCB3 and TRV:00 were then mechanically inoculated with TuMV-GFP. At 4 dpi of TuMV-GFP, there were fewer fluorescent infection foci on the IL of the NbLHCB3-silenced plants than on non-silenced plants (Figure 2A). TuMV RNAs and CP protein in both IL and systemically infected leaves accumulated less in the NbLHCB3-silenced plants than in non-silenced plants ( Figure 2D), showing that silencing of NbLHCB3 inhibited TuMV-GFP infection.

ROS Accumulated in the NbLHCB3-Silenced Plants
Light-harvesting chlorophyll a/b binding proteins play a key role in the electron transport chain in chloroplasts, which are the major sites of ROS production (Galvez-Valdivieso and Mullineaux, 2010). In previous reports, ROS levels increased in all the Arabidopsis thaliana lhcb T-DNA insertion mutants (Xu et al., 2012). We next used DAB and NBT staining to determine the H 2 O 2 and O 2 − levels in the NbLHCB3-silenced plants before and after inoculation with TuMV. H 2 O 2 and O 2 − accumulated (E) Identification of transcriptional levels of genes in ROS production and scavenging pathways. The relative DAB and NBT stain intensity was quantified by IMAGE J software. Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; **P < 0.01).
Frontiers in Microbiology | www.frontiersin.org The relative intensity of the blot signal was quantified by IMAGE J software. Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; **P < 0.01; ***P < 0.001). The relative intensity of the blot signal quantified by IMAGE J software is shown on the lanes. Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; **P < 0.01; ***P < 0.001).

NbLHCB3-Silenced Plants
To determine whether the accumulation of ROS plays a role against TuMV infection in the NbLHCB3-silenced plants, we treated the silenced plants with 50 µM DPI and 5 mM dimethylthiourea (DMTU) to suppress the production of ROS and eliminate existing ROS, and then examined its effect on TuMV infection (Figure 4). The amount of ROS in the NbLHCB3-silenced leaves was indeed decreased after being treated with DPI and DMTU ( Figure 4A). When TuMV-GFP was inoculated, more fluorescent infection foci appeared and there was stronger GFP fluorescence on the IL of the NbLHCB3-silenced plants treated with DPI and DMTU, compared to those of the non-treated NbLHCB3silenced plants (Figures 4B-D). Furthermore, TuMV RNAs and CP protein accumulated at higher levels in the NbLHCB3silenced plants treated with DPI and DMTU than in the nontreated NbLHCB3-silenced plants (Figure 4E), indicating that the accumulated ROS in the NbLHCB3-silenced plants plays a role against TuMV infection.
To further confirm this, we co-silenced NbLHCB3 and each of the individual genes involved in ROS production: AO, RBOHA, RBOHB, RBOHC, and RBOHE. The downregulated expression of these genes in the co-silenced plants was confirmed by qRT-PCR (Supplementary Figures 5A-E). It was expected that silencing of these genes would impair the accumulation of ROS in the NbLHCB3-silenced plants. Indeed, DAB and NBT staining showed a significantly lower accumulation of ROS in these co-silenced plants than in those where only NbLHCB3 was silenced (Figures 4F-H). TuMV-GFP was then mechanically inoculated onto these plants. At 5 dpi of TuMV-GFP infection, GFP fluorescence was stronger on the systemic leaves (SL) of TRV:LHCB3 + AO, TRV:LHCB3 + RBOHA, TRV:LHCB3 + RBOHB, TRV:LHCB3 + RBOHC, and TRV:LHCB3 + RBOHE compared with the TRV:LHCB3 plants ( Figure 4F). Consistently, TuMV RNAs and CP protein accumulated at higher levels in both the IL and SL of these cosilenced plants (Figure 4I and Supplementary Figure 3). These results further demonstrate that the accumulated ROS in the NbLHCB3-silenced plants plays a role against TuMV infection.

Co-silencing of Genes in the ROS Scavenging Pathway With NbLHCB3
Increased the Resistance of

NbLHCB3-Silenced Plants to TuMV Infection
To further confirm the relationship between ROS and defense against TuMV in the NbLHCB3-silenced plants, we cosilenced genes in the ROS degradative pathway, CAT (CAT functions in H 2 O 2 scavenging), SOD(CuZn), SOD(Fe), SOD(Mn) (SODs play pivotal roles in metabolizing O 2 − and producing H 2 O 2 ), tAPX (encodes ascorbate peroxidase which has a similar function to CAT and exists in thylakoids), GR1&GR2 and GR3&GR4, with NbLHCB3, and examined TuMV infection on the silenced plants. The expression of these genes was shown to be significantly downregulated in the co-silenced plants by qRT-PCR ( Supplementary  Figures 5F-L). We expected that ROS would accumulate more in these co-silenced plants than in those where only NbLHCB3 was silenced, and this was indeed confirmed by DAB and NBT staining (Figures 5A-C). TuMV-GFP was then inoculated onto these plants. At 8 dpi of inoculation, there were fewer fluorescent spots on the IL of TRV:LHCB3 + CAT, TRV:LHCB3 + SOD(CuZn), TRV:LHCB3 + SOD(Fe), TRV:LHCB3 + SOD(Mn), TRV:LHCB3 + tAPX, TRV:LHCB3 + GR1&GR2, and TRV:LHCB3 + GR3&GR4 plants than on the control TRV:LHCB3 plants. Consistently, fluorescent areas on the systemically infected leaves of these plants were small, while the fluorescence intensity was not affected significantly. TuMV RNAs and CP also accumulated at lower levels in these co-silenced plants ( Figure 5D and Supplementary Figure 4). These results give further support to the conclusion that the accumulated ROS in the NbLHCB3-silenced plants plays roles against TuMV infection.

Overexpression of NbLHCB3 in N. benthamiana Increased the Accumulation of TuMV RNAs and CP
To further explore the biological function of NbLHCB3 in TuMV-GFP infection, we obtained transgenic N. benthamiana overexpressing NbLHCB3, and detected the accumulation of TuMV in these plants. These transgenic plants contained the full-length ORF of NbLHCB3 driven by the 35S promoter of cauliflower mosaic virus (CaMV) and were obtained by agrobacterium-mediated transformation. Three independent transgenic lines (OE:NbLHCB3 −4, −8 and −9) were identified with about three times increased expression of NbLHCB3 compared to wild type N. benthamiana (Figure 6B) with no obvious developmental phenotype compared to WT (Figure 6A). In these plants, ROS levels were decreased (Figures 6A,C,D). After inoculation with TuMV-GFP, there were more fluorescent spots on the IL at 5 dpi than on the controls ( Figure 6A) and fluorescence was also more extensive on their SL (Figure 6A). Blotting analysis showed a significantly increased accumulation of TuMV RNAs and CP on the three transgenic lines, further The relative NBT stain intensity in these three transgenic lines compared with WT calculated by IMAGE J software. (E,F) Accumulation of TuMV RNAs and CP in IL (E) and SL (F). The relative intensity of the blot signal quantified by IMAGE J software is shown on the lanes. Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences (*P < 0.05; **P < 0.01; ***P < 0.001).
demonstrating that decreased ROS together with increased expression levels of NbLHCB3 resulted in greater susceptibility to TuMV (Figures 6E,F).

DISCUSSION
Viral infections usually cause plant chlorosis, stunting, necrosis, or other symptoms (Shi et al., 2016). Chlorophyll pigmentation can be reduced following virus infection (Grisham et al., 2010), chloroplast structures and functions may be changed (Otulak et al., 2015;Rong et al., 2018), and the expression of nuclear-encoded chloroplast photosynthesisrelated genes repressed (Mochizuki et al., 2014;Das et al., 2019;Zhao et al., 2019). All these suggest that viruses and chloroplasts are inextricably linked. Many viral proteins interact directly with chloroplast proteins thus linking chloroplasts to viruses (Ma et al., 2008;Sun et al., 2013;Balasubramaniam et al., 2014;Kong et al., 2014;Gnanasekaran et al., 2019). TuMV also associates with chloroplasts and its successful infection depends on the fusion of virus-induced vesicles with chloroplasts (Wei et al., 2013). In this study we found that chloroplast antenna protein NbLHCB3 was downregulated by TuMV infection and played roles in the interaction between TuMV and plants, providing further evidence of the association of chloroplasts with TuMV infection (Figures 1B,C).
Many studies have shown that viruses from different genera significantly downregulate chloroplast-related genes (ChRGs) including light harvesting genes (Bhattacharyya et al., 2015;Shi et al., 2016;Das et al., 2019;Zhao et al., 2019;Leonetti et al., 2021). The mechanism by which viruses regulate expression of these ChRGs is not well understood. One reported mechanism is RNA silencing, in which small RNAs derived from the viral genome or satellite RNAs target ChRGs for silencing (Shimura et al., 2011;Smith et al., 2011). We here found that the NIa protein of TuMV was responsible for NbLHCB3 reduction, indicating a potential role for NIa in regulating chloroplast function ( Figure 1D).
Light-harvesting chlorophyll a/b complex protein 3 interacts with PsbS and affects the macrostructure of photosystem II (Damkjaer et al., 2009;Kouřil et al., 2013;Pietrzykowska et al., 2014;Gerotto et al., 2015;Longoni et al., 2015). PsbS (NPQ4) plays a major role in activating the photoprotection mechanism known as "non-photochemical quenching (NPQ)" which dissipates chlorophyll excited states exceeding the capacity for photosynthetic electron transport (Fan et al., 2015;Gerotto et al., 2015;Correa-Galvis et al., 2016;Tibiletti et al., 2016;Sacharz et al., 2017). The latest research finds that PsbS can also activate photoprotection by some mechanism independent of NPQ (Redekop et al., 2020). OsPsbS-deficient rice generated more superoxide and hydrogen peroxide (Zulfugarov et al., 2014). Consistent with this, silencing of NbLHCB3 here caused ROS accumulation and it seems possible that this was a consequence of the impaired NPQ in the NbLHCB3-silenced plants. Meanwhile, the NbLHCB3-silenced plants also had upregulated expression of genes in ROS production and downregulated expression of genes in the ROS scavenging pathway ( Figure 3E). Although the mechanism by which NbLHCB3 regulates the expression of these genes is not yet clear, there is clearly some indirect mechanism to cause ROS accumulation in the NbLHCB3-silenced plants.
Reactive oxygen species produced by RBOHs, and especially RBOHD, appear to be involved in plant immune responses (Sagi and Fluhr, 2001;Yoshioka et al., 2003;Torres et al., 2005;Asai et al., 2008;Lee et al., 2020;Ngou et al., 2021;Yuan et al., 2021). RBOHD is highly responsive to MAMP or pathogen treatments (Morales et al., 2016) and is negatively regulated through phosphorylation and degradation by ubiquitination (Lee et al., 2020). NbRBOHC is homologous to AtRBOHD. Here, the expression of NbRBOHC was upregulated in the NbLHCB3-silenced plants ( Figure 3E). Co-silencing of NbRBOHC significantly impaired the inhibition of TuMV in the NbLHCB3-silenced plants, compared to other genes in the ROS production pathway, which indicates that NbRBOHC plays an essential role in the accumulation of ROS in the NbLHCB3silenced plants (Figure 4).
It has been reported that light deficiency and photosystem impairment increased the susceptibility of N. benthamiana to TuMV infection, suggesting that both light and optimal chloroplast function influence virus infection by limiting the cellular resources needed by TuMV to establish replication complexes (Manfre et al., 2011). However, the effect of the photosystem on viral infection is likely to be complicated since many essential molecules, including ROS and various sugars that all play roles in plant defense, are produced from the photosystem (Bhattacharyya and Chakraborty, 2018;Littlejohn et al., 2021). Deficiencies of different components in the photosystem would have different effects on these molecules, which might result in plants showing different responses to pathogens. In our previous report, silencing of genes in the Calvin cycle limited rice stripe virus (RSV) infection (Bi et al., 2019). We supposed that the energy produced in the Calvin cycle was necessary for normal RSV infection. Here, silencing of NbLHCB3 caused the accumulation of ROS, which was demonstrated to regulate TuMV infection negatively (Figures 3, 4). Both results suggest the different roles of photosystem products in response to viral infection and reflect the complicated functions of chloroplasts in the interaction between viruses and plants.
Reactive oxygen species themselves have also been reported to be harnessed by viral infection and transmission. Red clover necrotic mosaic virus and brome mosaic virus both harness host NbRBOHB-dependent O 2 − for robust genome replication (Hyodo et al., 2017a,b). For TuMV, ROS is reported to be beneficial to aphid-borne spread (Berthelot et al., 2019;Guo et al., 2019). Taken together with the results of the present study, this suggests an interesting hypothesis that ROS accumulates in TuMV-infected N. benthamiana by down-regulation of NbLHCB3 to inhibit TuMV infection, while the accumulated ROS enhances its transmission by aphids. We will examine this hypothesis next.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
SQ, JC, and FY designed the experiments. SQ, XC, YZ, WC, XA, and SR performed the experiments and interpreted the data. SQ and FY drafted the manuscript. FY and JC revised the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
This work was financially supported by the Chinese Agriculture Research System (CARS-24-C-04) and the K. C. Wong Education Foundation.

ACKNOWLEDGMENTS
We thank Mike Adams for manuscript correction.

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2021.690988/full#supplementary-material Bars represent the standard errors of the means from three biological repeats. A two-sample unequal variance directional Student's t-test was used to test the significance of the differences ( * P < 0.05).

Supplementary Figure 3 | Northern blotting showing the accumulation of
TuMV-GFP RNAs at 5 dpi in the plants where NbLHCB3 was co-silenced with individual genes in the ROS production pathway, AO, RBOHA, RBOHB, RBOHC, and RBOHE. The relative intensity of the blot signal quantified by IMAGE J software is shown on the lanes.
Supplementary Figure 4 | Northern blotting showing the accumulation of TuMV-GFP RNAs at 8 dpi in the plants where NbLHCB3 was co-silenced with individual genes in the ROS scavenging pathway, CAT, SOD(CuZn), SOD(Fe), SOD(Mn), tAPX, GR1&GR2, and GR3&GR4. The relative intensity of the blot signal quantified by IMAGE J software is shown on the lanes.
Supplementary Figure 5 | The expression of LHCB3 and genes in the ROS production and scavenging pathways were all downregulated in these co-silenced plants. (A-E) The expression of LHCB3 and genes in the ROS production pathway (AO, RBOHA, RBOHB, RBOHC, and RBOHE) were downregulated in these co-silenced plants. (F-L) The expression of LHCB3 and genes in the ROS scavenging pathway [CAT, SOD(CuZn), SOD(Fe), SOD(Mn), tAPX, GR1&GR2, and GR3&GR4] were downregulated in these co-silenced plants.