The seeds of wild-type tobacco (N. tabacum L. cv. NC89) were surface-sterilized with 10% NaClO for 15 min and sown on 1/2 Murashige and Skoog (MS) medium containing different levels of effectors. Seeds were germinated and grown in a controlled environment room where day and night temperatures were 25 and 23°C under 16-h-light/8-h-dark photoperiod and 70% relative humanity.

For SHAM (Tokyo Chemical Industry, Japan) treatments, the seeds were germinated in 1/2 MS containing 1, 2, 3, 4, and 5 mM SHAM, respectively, under the same conditions as mentioned earlier. In addition, methyl jasmonate (MeJA) and auxin [indoleacetic acid (IAA)] were used to study the restoration of tobacco root development while AOX was inhibited. The different concentrations of MeJA and IAA were added to the media as described earlier with some modifications (Zhu et al., 2006; Ivanova et al., 2014). Dimethylthiourea (DMTU), an H₂O₂ scavenger, was used at different concentrations according to the earlier methods with some modification (Xia et al., 2009; Chandrakar and Keshavkant, 2018).

After the germination of tobacco seeds, at least 30 plants under different treatment conditions were taken every day to measure the length of main roots. Then, the development of root hairs was further recorded using a stereomicroscope (SteREODiscovery.V20, Carl Zeiss AG, Germany).

The reactive oxygen species (ROS) assay was performed according to the method of Xu et al. (2012b). For superoxide ion measurement, the seedlings were transferred into 0.5 mM Nitrotetrazolium blue chloride (NBT; Sigma, St Louis, MO, United States) solution and stained for 2 h under vacuum filtration in the dark. Seedlings were then decolorized in boiling ethanol (95%) for approximately 20 min. The total intensities of NBT staining in the root part were measured in 15 roots with three biological replicates using the Fiji software package of ImageJ.

The H₂O₂ content of seedlings was measured as described by Velikova et al. (2000). Approximately 0.5 g of sample was cut into small pieces and homogenized in an ice bath with 5 ml 0.1% (w/v) trichloroacetic acid. The homogenate was centrifuged at 12,000 × g for 20 min at 4°C, and then 0.5 ml of the supernatant was added to 0.5 ml of 10 mM potassium phosphate buffer (pH 7.0) and 1 ml of 1 M KI. The absorbance of the supernatant was read at 390 nm.

The dead cells on SHAM treatments were visually detected using a trypan blue staining method as described by Han et al. (2017) with some modifications. Tobacco roots were stained with lactophenol–trypan blue solution (i.e., 10 ml of lactic acid, 10 ml of glycerol, 10 g of phenol, and 10 mg of trypan blue, dissolved in 10 ml of distilled water) at boiled water for 1 min and then left staining at room temperature for another 10 min. Then, samples were placed in chloral hydrate solution (i.e., 2.5 g of chloral hydrate dissolved in 1 ml of distilled water) to reduce background staining and then were equilibrated with 70% glycerol for scanning. To observe the root cell death, samples were stained with 10 μg/ml of propidium iodide (PI) for 5 min and then rinsed with ddH₂O. Fluoresce images were obtained using a Leica microscope (EVOS FL Imaging System, Life Technologies Corp., Washington, United States).

The respiration measurement was performed according to the method of Yu et al. (2020). Approximately 50 mg of root tissues were cut into small pieces, and then transferred into air-tight cuvettes containing 20 mM 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) and 0.2 mM CaCl₂ with pH 7.2, and oxygen uptake was measured as a decrease of O₂ concentration in the dark using a Clark-type oxygen electrode (Chlorolab2, Hansatech, United Kingdom). Total respiration rate (V_t) was measured without any inhibitors. Then, 1 mM KCN was added to inhibit cytochrome c pathway respiration (V_cyt), which represents the sum of the AOX alternative pathway respiration rate (V_alt) and the residual respiration rate (V_res). Then, 1 mM n-propyl gallate (nPG) was added to inhibit the V_alt and got the V_res value. Therefore, the V_alt was defined as O₂ uptake rate without V_cyt and V_res.

For the RNA-Seq analysis, tobacco root tissues on the 8th day after treatment with or without 1 mM SHAM were used in this study. The subsequent experiments of tobacco root tissues include the extraction, purification, analysis, and sequencing of total RNA by Novogene Bioinformatics Technology Co. Ltd. (Beijing, China). Sequencing libraries were generated using NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (NEB, Ipswich, MA, United States) according to the instructions of the manufacturer, and index codes were added to attribute sequences to each sample.

HISAT2 was used to count the number of reads mapped to each gene. In addition, the fragments per kilobase of exon model per million mapped reads (FPKM) of each gene were calculated based on the length of the gene and the number of reads mapped to the gene.

The differential expression analysis was performed using the DESeq2 R package (version 1.16.1). DESeq2 provide statistical routines for determining the differential expression in the digital gene expression data using a model based on the negative binomial distribution. The resulting value of p were adjusted using the Benjamini–Hochberg approach for controlling the false discovery rate. Genes with an adjusted p < 0.05 found by DESeq2 were assigned as differentially expressed.

The Gene Ontology (GO) enrichment analysis of differentially expressed genes (DEGs) was implemented by the cluster Profiler R package, in which gene length bias was corrected. GO terms with corrected value of p < 0.05 were considered significantly enriched by DEGs.

The KEGG is a database resource for understanding the high-level functions and utilities of the biological system, such as the cell, the organism, and the ecosystem, from molecular-level information, especially large-scale molecular data sets generated by genome sequencing and other high-throughput experimental technologies.¹ We used the cluster Profiler R package to test the statistical enrichment of DEGs in KEGG pathways.

In order to validate the results from the transcriptome sequencing analysis, a part of genes was confirmed by the real-time quantitative PCR (qRT-PCR), and Actin (Accession number: AB158612) gene was used as an internal control. All the primers are listed in Supplementary Table S1. The qRT-PCRs were prepared with the SYBR Green Master Mix Reagent (Applied Biosystems, Waltham, MA, United States), following the instructions of the manufacturer. The reactions were carried out in Applied Real-Time System (ABI 7500). All samples were performed in triplicate, and the relative expression levels were calculated using the 2^−ΔΔCT method of the system.

The statistical analysis of the results from three independent experiments with nine measurements used a one-way ANOVA, followed by Tukey’s HSD post hoc test or Dunnett’s HSD test. Asterisks or different letters in graphs indicate the level of significance (p < 0.05).

To investigate the effects of AOX inhibition on the tobacco root development, different concentrations of SHAM were applied and the root growth was compared (Figure 1; Supplementary Figure S1). As shown in Figure 1, the root development including main roots and root hairs of tobacco was greatly inhibited by SHAM treatments compared with control. Under the conditions of 1 and 2 mM SHAM, the root hair development was markedly blocked and the length of the main root was obviously shorter than the control (Figure 1A). When the higher concentrations of SHAM were applied, the development of tobacco root systems was further inhibited, especially under the condition of 5 mM SHAM (Figures 1A,B). The microscopic observations revealed that the SHAM-treated tobacco showed lacking of root hair around the root system, and the main roots of tobacco showed wilting and browning, especially at the tip part (Figure 1C). Moreover, it should be noted that when the concentration of SHAM treatment reached 4 mM, the root development was completely blocked (Figure 1A).

Considering that AOX mediates branch respiration in the mitochondrial electron transport chain, the total respiration (V_t) and alternative pathway respiration (V_alt) of root were determined under different concentrations of SHAM. As shown in Figure 2A, SHAM treatments suppressed the V_t and V_alt of tobacco roots and this inhibition became more obvious with the increase of SHAM concentration. Under the conditions of 1 mM SHAM, the V_t of tobacco roots was four times lower than that under the normal conditions (Figure 2A). When 2 mM SHAM was applied, the V_t decreased to 10 times lower than that of the control. Certainly, more serious respiration inhibition happened when seedlings suffered from 3 to 4 mM SHAM treatment (Figure 2A). Similarly, the V_alt of tobacco roots was significantly suppressed by SHAM treatments, especially under the conditions of 3 and 4 mM SHAM, where the V_alt of tobacco roots was almost undetectable (Figure 2B).

It was observed that, moreover, there was a large amount of O2^.− accumulation in the root system during tobacco root development under normal growth conditions, especially at the root tip (Figure 2C). In contrary, tobacco seedlings grown under different concentrations of SHAM showed less O2^.− content than the control, especially under the higher concentrations of SHAM conditions such as 4 mM SHAM. When the higher concentration of SHAM was applied, the lower level of O2^.− was detected in the root (Figures 2C,D). However, the H₂O₂ assay showed that there were more H₂O₂ contents accumulated in SHAM-treated seedlings than that in the control seedlings (Figure 2E).

In addition to ROS burst, it was worth noting that the inhibition of AOX with different concentrations of SHAM caused an obvious cell death in the root system based on the trypan blue assay, and this effect became more serious as higher concentrations of SHAM were applied (Figure 3). Results showed that tobacco roots were stained light blue by trypan blue under normal culture condition, whereas the color became more obvious under SHAM treatments (Figure 3A). In addition, the microscopic observation of the roots after PI treatment showed that the root cell damage was the same as the results of trypan blue staining. When compared with control, the cell death of tobacco roots under the SHAM treatment conditions was more obvious, especially the part embedded in the medium (Figure 3B). These results together with the ROS analysis implied that AOX inhibition led to an excessive increase in ROS, which then caused cell death in the root system.

Given that both the main roots and root hairs were obviously inhibited by 1 mM SHAM, this concentration was used in the RNA-Seq analysis to reveal the regulation of gene expression affected by AOX inhibition (Figure 4A). As shown in Figure 4, a total of 5,855 DEGs were identified, of which 3,517 were upregulated genes and 2,338 were downregulated genes (Figure 4B). If fold change ≥2 was considered as a basic threshold, 2,279 DEGs were detected, of which 1,580 were upregulated and 699 were downregulated by SHAM treatment (Figure 4C). To confirm the reliability of the transcriptome sequencing, parts of DEGs were investigated by qRT-PCR. As shown in Supplementary Figure S2, the results of qRT-PCR were generally consistent with the transcriptome data, suggesting a strong positive correlation between the qRT-PCR and transcriptome data.

To further identify the functions of DEGs regulated by SHAM treatment, the GO and KEGG analyses were then performed. As shown in Figure 4D, the DEGs of SHAM vs. control (CK) were enriched in GO terms of the cell wall, cell periphery, and external encapsulating structure, etc. The KEGG pathway analysis showed that most of DEGs were markedly enriched in “Carbon metabolism”, “Biosynthesis of amino acids”, “Glyoxylate and dicarboxylate metabolism”, and “Cysteine and methionine metabolism” (Figure 4E). It should be noted that several DEGs (|log₂FC ≥ 1|) were commonly enriched in these pathways, i.e., LOC107808059 and LOC107804363, which encode serine acetyltransferase, were enriched in the KEGG pathways of “Carbon metabolism”, “Biosynthesis of amino acids”, and “Cysteine and methionine metabolism”, and showed upregulation on SHAM treatment (Figures 5A–C). In addition, the gene expression of LOC107826261, which encodes aminotransferase and belongs to KEGG pathways of “Biosynthesis of amino acids” and “Cysteine and methionine metabolism”, was downregulated by SHAM treatment (Figures 5B,C). Similarly, the gene expression of LOC107787646, which encodes acetate/butyrate-CoA ligase, was also significantly downregulated by SHAM treatment (Figures 5A,D). Notably, LOC107774406, which encodes 1-aminocyclopropane-1-carboxylate oxidase and is a key enzyme of ethylene biosynthesis pathway, showed the upregulated gene expression on SHAM treatment. These findings indicate that AOX inhibition affects the anabolic and catabolic processes directly or indirectly.

Since SHAM treatment affected the accumulation of ROS in roots, then the ROS-related genes were analyzed further. As shown in Figure 6, the genes related to antioxidant systems such as superoxide dismutase (SOD), peroxidase (POD), ascorbate peroxidase (APX), glutathione peroxidase (GPX), and glutathione S-transferase (GST) were affected by AOX inhibition. Among these, the expression of a small amount of DEGs related to SOD and GPX were upregulated, but a large amount of DEGs related to POD and APX were downregulated by SHAM (Figures 6A–E). It should be noted that the DEGs related to GST, which plays a key role in glutathione binding reaction, and their expressions were significantly upregulated by SHAM treatment (Figure 6F). Notably, further analysis showed that the DEGs related to polyphenol oxidase (PPO) and polyamine oxidase (PAO) exhibited downregulation after SHAM treatment (Figures 6G,H).

Many studies have shown that auxin is the main signaling molecule involved in regulating root development and root hair growth. In this study, auxin pathway-related genes were downregulated or upregulated by SHAM treatment, such as tryptophan aminotransferase (TAA), flavin-containing monooxygenases (YUCs), and transport inhibitor response 1 (TIR1; Figure 7). As shown in Figure 7A, one TAA (LOC107802923) and three YUCs (LOC107825840, LOC107810383, and LOC107825958) genes were involved in auxin synthesis, and their expressions were significantly downregulated by SHAM treatment. In addition, the expression of TIR1 (LOC107764726), a gene of auxin receptor, was also downregulated by SHAM treatment (Figure 7B). Moreover, it should be noted that six auxin binding proteins (ABPs), which are also believed to function as an IAA receptor, showed downregulation on SHAM treatment (Figure 7B).

Further analysis showed that the expression of 11 AUX/IAA genes was affected by SHAM treatment, and most of these genes were upregulated (Figure 7C). Likewise, the DEGs related to auxin transport (efflux carrier) such as ABC transporter B family member (ABCB) and multidrug resistance protein (MFS) showed upregulation on SAHM (Figure 7D). Moreover, many genes involved in auxin response were also regulated by SHAM treatment (Figure 7E). Of these, five auxin response factor (ARF) genes showed significant downregulation after SHAM treatment. Besides, seven genes of Gretchen Hagen 3 (GH3) belonging to indole-3-acetic acid-amido synthetase were identified, and the expression of five GH3 genes was significantly upregulated by SHAM treatment (Figure 7E). In comparison, the gene expressions of small auxin-up RNA (SAUR) were up- and downregulated on SHAM treatment (Figure 7E). These results indicate that AOX inhibition affected the tobacco root formation by suppressing auxin synthesis and signaling transduction.

Considering that ethylene is a positive regulator of root hair development (Tanimoto et al., 1995; Pitts et al., 1998; Lee and Cho, 2013) and some of the key genes related to ethylene biosynthesis were assigned to KEGG pathway (sly00270, p < 0.05), we further analyze the effects of SHAM treatment on the gene expression of ethylene biosynthesis and signaling transduction pathway. As shown in Figure 8, a total of 14 DEGs involved in key steps of ethylene biosynthesis such as 1-aminocyclopropane-1-carboxylate oxidase (ACO) were identified and 12 of them showed significant upregulation on SHAM treatment (Figure 8). Moreover, the expression of two genes of ethylene receptor (ETR), five genes of ethylene insensitive 3 (EIN3), and more than 20 genes of ethylene-responsive transcription factor (ERF) were affected differentially by SHAM treatment. Of these, the expressions of EIN3 genes and EFR1 genes were upregulated by SHAM treatment (Figure 8). These results suggest that AOX inhibition-mediated defects in root development did not occur through impaired ethylene biosynthesis and signal transduction.

During root development, especially the root hair initiation, the local cell wall loosening is required (Tominaga-Wada et al., 2011). In addition, it has been demonstrated that root meristem size is controlled by root meristem growth factor 1 (RGF1), which affects the root architecture (Yamada et al., 2020). Interestingly, in this study, we found a total of 25 DEGs related to the cell wall, including the genes involved in xyloglucan endotransglucosylase/hydrolase protein and pectinesterase, showed significant downregulation on SHAM treatment (Table 1). Conversely, the gene expressions of pectinesterase inhibitor were upregulated by SHAM treatment (Table 1). It is of interest to note that, moreover, the expression of two RGF1 genes (LOC107788696, log₂FC = −7.15; LOC107823248, log₂FC = −2.84) were significantly downregulated by SHAM treatment (Table 1). These results indicate that AOX-mediated root development defects were involved in the cell wall loosening and root tip meristem formation.

Considering that the root development and root hair formation are related to cell proliferation, we also tried to analyze the effect of SHAM treatment on the expression of cell cycle-related genes. As shown in Table 2, some DEGs related to cyclin-dependent kinase inhibitor protein (CDI), cell division cycle, and cell cycle regulatory genes were identified. Of these, all five genes of CDIs, whose proteins function as constraining the activities of cyclin-dependent kinases (CDKs), showed a significant upregulation under the condition of SHAM treatment (Table 2). In contrast, the expressions of cell division cycle (e.g., LOC107832361, log₂FC = −2.42) and cell cycle regulatory genes including cyclin D3-1 (CYCD3;1, LOC107776168) and cyclin D3-2 (CYCD3;2, LOC107792202) were downregulated by SHAM treatment (Table 2).

Since the cell apoptosis of the root was clearly observed on SHAM treatment, the DEGs involved in apoptosis were further analyzed. As shown in Figure 9A, three genes of apoptosis-inducing factors were identified. Of these, it is of interest to note that the gene expressions of apoptosis-inducing factor were significantly (p < 0.05) upregulated by SHAM treatment, especially LOC107831784 (log₂FC = 1.37) and LOC107827516 (log₂FC = 1.76). In addition, the qRT-PCR analysis also showed that these genes were upregulated on SHAM treatment (Figure 9B), indicating that the impairment of the tobacco root development was associated with cell apoptosis induced by AOX inhibition.

Considering that SHAM was reported as an inhibitor of lipoxygenase, which is a key enzyme of jasmonate (JA) biosynthesis, the DEGs related to lipoxygenase were analyzed. As shown in Supplementary Table S2, five DEGs related to lipoxygenase were regulated by SHAM. Of these, three genes, namely, LOC107800317, LOC107760301, and LOC107763327, were downregulated, while LOC107803018 and LOC107770253 were upregulated by SHAM treatment. In order to study whether MeJA helps to reduce the inhibitory effect of SHAM, the different concentrations of MeJA were introduced into the media containing 1 mM SHAM. The results showed that 100–500 nM MeJA did not restore the root development defects mediated by AOX inhibition, but aggravated the root inhibition (data not shown). Subsequently, we reduced the concentrations of MeJA to 10–50 nM and observed the same results, i.e., there were some cumulative damages to the root development (Figure 10). When the applied concentrations of MeJA were decreased to 1–5 nM, no restoration effects were still occurred (Figure 10). These results indicate that MeJA cannot restore root development defects mediated by AOX inhibition. It should be noted that, furthermore, the gene expression of TIFY 6B [also known as JA ZIM-domain (JAZ3) protein] was significantly downregulated (log₂FC = −4.75) by SHAM treatment (Supplementary Table S2).

Given that auxin biosynthesis and ROS overaccumulated in the SHAM-treated roots, different concentrations of IAA and ROS scavenger (DMTU) were introduced into the media. As shown in Figure 11, when 50–100 nM IAA were added to the media containing 1 mM SHAM, the development of these roots was further inhibited compared with roots treated with 1 mM SHAM. When the concentration of IAA reduced to 10–20 nM, some restoration effects were occurred, e.g., the root length was longer, and a small amount of root hairs appeared (very short) in the zone of differentiation than the 1 mM SHAM-treated roots (Figure 11A). However, there was no significant difference from the roots treated with 1 mM SHAM when the concentration of IAA reduced to 1–5 nM (Figure 11A).

When DMTU was introduced into the media containing 1 mM SHAM, different effects were observed. It should be noted that 5 mM DMTU did not change the root inhibitory effect caused by 1 mM SHAM. When 10 mM DMTU was used, a small amount of root hairs (very short) appeared in the zone of differentiation. When the concentration was further increased to 20 mM DMTU, the restoration effect was not further improved (Figure 11B). These results indicate that the inhibition of auxin and the excessive accumulation of ROS are only part of the reasons why AOX inhibition mediated the hindrance of root development.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.