Functional and regulatory mechanisms of the NtPPO5 gene in the browning reaction of tobacco

Abstract

Tobacco (Nicotiana tabacum L.) is a vital economic crop, yet it frequently suffers excessive enzymatic browning during the curing process, which negatively impacts its leaf color and commercial value. In this study, we constructed NtPPO5 (Nitab4.5_0003171g0010.1) gene overexpression and knock-out vectors and generated transgenic tobacco plants to investigate the role of NtPPO5 in curing-induced browning. Investigation of agronomic traits demonstrated that NtPPO5 plays a positive regulatory role in plant height and leaf area. The curing experiment revealed that overexpression of NtPPO5 exacerbated enzymatic browning, whereas knockout of NtPPO5 effectively controlled the browning reaction. To elucidate the underlying mechanisms, we conducted enzyme activity assays, physiological and biochemical measurements, and non-targeted metabolomics analyses. The results demonstrated that knockout of NtPPO5 significantly reduced polyphenol oxidase (PPO) activity while inducing compensatory upregulation of superoxide dismutase (SOD) activity. Additionally, peroxidase (POD) activity exhibited distinct peak stages compared to wild-type cultivar Cuibi-1 (CB-1). Metabolomics analysis revealed significant changes in the levels of key phenolic substrates involved in enzymatic browning during the curing process, particularly shikimic acid and phenylpropanoic acid compounds, which were predominantly enriched in the Flavonoid Biosynthesis Pathway (map00941). The knockout of NtPPO5 optimized Flavonoid Biosynthesis Pathway by regulating precursor availability and metabolic fluxes, enhancing antioxidant defenses, and reducing curing-induced browning. This study provides theoretical insights and practical evidence for mitigating browning in flue-cured tobacco and offers valuable guidance for optimizing the curing process.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08390-4.

Introduction

Browning is a process in which the color of plant or food products gradually changes to brown or dark brown over time, and this process can have either a positive or negative impact on the quality of these products [1]. The browning reaction is generally categorized into enzymatic and non-enzymatic browning, depending on the underlying mechanism [2]. Non-enzymatic browning involves the formation of brown substances through enzyme-free chemical reactions among food constituents, primarily including the Maillard reaction, caramelization, and ascorbic acid oxidation. These reactions often occur simultaneously in complex food systems [3]. In contrast, enzymatic browning is primarily driven by the action of polyphenol oxidase (PPO), an enzyme naturally present in food [4]. PPO catalyzes the oxidation of phenolic compounds to reactive quinones through a two-step mechanism: first, it can oxidize monophenols to o-diphenols (cresolase activity), and then further converts these o-diphenols to o-quinones (catecholase activity) [5]. Discoloration caused by enzymatic browning is a major issue in many horticultural crops and their minimally processed products, as it negatively affects quality, sensory, and nutritional value, ultimately impacting consumer satisfaction [6]. It is estimated that approximately 50% of fruits and vegetables are wasted each year due to enzymatic browning [7]. This phenomenon frequently occurs in fresh foods and fruits [8], such as apples [9], taro [10], and litchis [11].

Tobacco is an economically important crop. The browning reaction is one of the key factors influencing its economic value. While controlled browning can improve flavor during tobacco processing, excessive browning reduces nutritional content and negatively alters flavor. For instance, the prominent Chinese cultivar Cuibi-1 (CB-1) is highly susceptible to enzymatic browning, frequently resulting in excessive browning reactions. This leads to extensive leaf discoloration, which severely compromises usability and production efficiency [12]. Previous studies on preventing and mitigating enzymatic browning have primarily focused on physical [13] and chemical [14] approaches. These include temperature control [13], ultrasound treatment [15], and the addition of compounds such as 3-mercapto-2-butanol [16] and nitroprusside [17]. However, these methods are often cumbersome to implement and challenging to achieve optimal results. Recently, given the critical role of PPO in enzymatic browning, several studies have explored silencing or altering the PPO gene to reduce PPO enzyme activity, thereby decreasing the degree of browning [5, 18, 19]. Nevertheless, the molecular mechanisms underlying PPO enzyme-catalyzed browning in tobacco leaves remain poorly understood [20], particularly during the curing process.

PPO activity is a critical internal factor in the browning reaction. Our previous study revealed that PPO enzyme activity varies with different curing temperatures, peaking between 43 °C and 46 °C during the curing process of tobacco leaves. This temperature range is the stage where excessive browning reactions are most likely to occur [21]. This trend may be attributed to the strong reaction between PPO in chloroplasts and the polyphenolic substances released during the rupture of cell vacuoles [12], which exacerbates browning reactions. Therefore, using tobacco as the model organism, it is of great significance to regulate the browning reaction by focusing on PPO gene expression, to analyze the metabolic intermediates and networks of polyphenolic substances during tobacco leaf curing, and to investigate the intrinsic relationship between polyphenol content and PPO activity changes. These insights are crucial for understanding the mechanisms of food browning and developing effective strategies for browning control. Based on our previous study, a total of 12 NtPPO genes were identified in tobacco [22]. Among these, NtPPO5 (Nitab4.5_0003171g0010.1) exhibited a high expression level in mature leaves [23]. In this study, we constructed gene overexpression and knock-out vectors for NtPPO5 and generated transgenic plants. To further explore the function and regulatory mechanisms of the NtPPO5 gene in the browning reaction of tobacco, we employed metabolomics and various physiological and biochemical methods to comprehensively analyze the metabolites, polyphenols, enzymes, and other biochemical substances. The findings will provide valuable insights for the molecular improvement of the browning reaction.

Materials and methods

CRISPR/Cas9-mediated generation of NtPPO5 knockout plants

CRISPR/Cas9 gene-editing technology was employed to generate NtPPO5 knockout plants. A guide RNA targeting sequence (GGTAATTGTGTGCCATCATACGG) of NtPPO5 was designed following the CRISPR/Cas9 gene-editing kit (Catalog No. VK005-1, Vazyme Biotech Co., Ltd., Beijing, China). The synthesized target sequence was cloned into the pHK1-Cas9-U3 vector to generate a recombinant plasmid. This plasmid was propagated in E. coli DH5α, verified by sequencing, and then introduced into Agrobacterium GV3101 for transformation of the tobacco cultivar CB-1. T0 transgenic plants were cultivated in a greenhouse, and genomic DNA was extracted from individual seedlings. The target sequence was PCR-amplified (Supplementary Table S1) for genotyping to identify mutation types, which were confirmed by sequencing. Successful knockout of NtPPO5 was validated based on sequencing results. T0 plants were self-pollinated to produce the T1 and T2 generations (designated as KO), which were subsequently used for further experiments.

Construction and validation of transgenic tobacco lines overexpressing the NtPPO5 gene

To construct NtPPO5 gene overexpression (abbreviated as OE) vector, the full-length coding sequence (CDS) of NtPPO5 gene was amplified by PCR (Supplementary Table S1) from the cDNA of the tobacco cultivar CB-1. After restriction digestion, the PCR amplicon was ligated into the pHK-35 vector to generate the 35 S::NtPPO5 overexpression vector. Transgenic tobacco plants were produced using the Agrobacterium-mediated leaf disc method [24]. Transgenic lines were selected through hygromycin resistance screening and validated by qRT-PCR. The 2^−∆∆CT method was used to calculate relative expression levels [25], with actin serving as the internal reference gene. The NtPPO5 primers used for hygromycin resistance screening and qRT-PCR are shown in Supplementary Table S1.

Planting and morphological phenotype investigation of transgenic lines

Three knockout (KO-1, KO-2, KO-3) and overexpression (OE-1, OE-2, OE-3) lines of NtPPO5 were selected for planting and morphological phenotype analysis. Seeds from the T1 and T2 generations of the transgenic lines were sown under controlled laboratory conditions. After a 65-day seedling stage, 30 plants from each transgenic line, along with the wild-type CB-1, were transplanted into the field for continuous cultivation during the growing seasons, using conventional agricultural methods. The field experiments were conducted in 2024 and 2025 at the experimental farm in Yanping District, Nanping City, Fujian Province, China. Morphological traits, including plant height, stem circumference, number of effective leaves, and maximum leaf area, were systematically measured and analyzed.

Sample preparation

To investigate the regulatory mechanism of NtPPO5 in the browning reaction process, the upper leaves (approximately the 12th to 14th leaf positions) of the wild-type CB-1 and the knockout transgenic line KO-2 (abbreviated as KO), and overexpression transgenic line OE-2 (abbreviated as OE) were selected as experimental materials for curing. Tobacco leaf samples were collected at six critical time stages during the curing process [21]: the starting point (T0), the early yellowing stage at 40 °C (T1), the late yellowing stage at 41 °C (T2), the early color-fixing stage at 43 °C (T3), the color-fixing stage at 45 °C (T4), and the leaf-curling stage at 46 °C (T5). For each sample, six leaves were collected as one sample, with three biological replicates, resulting in a total of 18 samples. The front and back 1/4 parts of the leaves were removed, and the main vein was excised. Each leaf was then divided into two parts: one part was stored at -20 °C for chemical substance content and enzyme activity assays, while the other part was wrapped in tin foil, labeled, placed in liquid nitrogen, and stored in a freezer at -80 °C for subsequent metabolomic analysis.

Enzyme activity measurement

Leaf samples collected at six curing stages (T0–T5) were analyzed for enzyme activity. The enzymatic activities of phenylalanine ammonia-lyase (PAL) [26, 27], polyphenol oxidase (PPO) [6], peroxidase (POD) [28], and superoxide dismutase (SOD) [29, 30] were assayed according to previously described methods with slight modifications. Briefly, approximately 0.1 g of fresh leaf tissue was homogenized in the corresponding extraction buffer. The crude extracts were centrifuged, and the supernatants were used for subsequent activity measurements. Enzyme activities were quantified spectrophotometrically using commercial reagent kits (BoxBio Science & Technology Co., Ltd., Beijing, China) following the manufacturer’s guidelines. Each sample was analyzed in triplicate to ensure the reliability of the data.

Analysis of chemical composition

At the completion of the curing, leaf samples from the wild-type CB-1, KO and OE were subjected to chemical component analyses to evaluate the main quality-related constituents of tobacco leaves. The concentrations of reducing sugars and starch were determined using the phenol-sulfuric acid colorimetric method, as previously described [31]. Total sugar content was quantified using a commercial total sugar assay kit (G0503W; Suzhou Geruisi Biotechnology Co., Ltd., Suzhou, China) following the manufacturer’s protocol. Total nitrogen content was measured by the Kjeldahl method using an automatic nitrogen analyzer (Kjeltec^™ system, Foss, Denmark). In accordance with the Chinese Tobacco Industry Standard YC/T 162–2011, chloride ion concentration was determined via silver nitrate titration. Potassium content was measured by flame photometry, based on the emission intensity of the digested solution. Nicotine levels were quantified using high-performance liquid chromatography (HPLC) following the method outlined in the Chinese Tobacco Industry Standard YC/T 246–2008. All determinations were performed in triplicate to ensure the precision of the data.

Wide-target metabolomic analysis of tobacco leaves

Tobacco leaf samples from the wild-type CB-1 and KO stored at -80 °C were lyophilized under vacuum and finely ground into a homogeneous powder. Approximately 25 mg of each powdered sample was extracted with 1 mL of methanol-water solution (3:1, v/v). The mixtures were vortexed thoroughly and subjected to three rounds of ultrasonic extraction in an ice bath to ensure efficient metabolite release. After overnight incubation at 4 °C, the extracts were centrifuged at 12,000 rpm for 15 min at low temperature. The resulting supernatants were filtered through 0.22 μm membranes and transferred into autosampler vials for UPLC-MS/MS analysis.

Chromatographic separation was carried out on an EXION LC system (SCIEX, USA) equipped with a UPLC Kinetex C18 column (2.1 mm × 100 mm, 2.6 μm). The mobile phase consisted of solvent A (0.01% acetic acid in water) and solvent B (50% acetonitrile/isopropanol). The column oven temperature was maintained at 25 °C, and the autosampler was held at 4 °C. The injection volume was 2 µL, and the flow rate was set to 0.3 mL/min.

Data analysis

Metabolomic data analysis was performed using R software (Version 4.3.0; www.r-project.org) and GraphPad Prism 10. Principal component analysis (PCA) was applied to assess the overall distribution and clustering of samples. Subsequently, orthogonal partial least squares discriminant analysis (OPLS-DA) was employed to identify metabolic differences among sample groups. Based on the OPLS-DA model, significantly differential metabolites were screened according to the criteria of variable importance in projection VIP ≥ 1, |log₂(Fold Change)| ≥ 1, and p < 0.05. Metabolite annotation was conducted using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.kegg.jp/kegg/pathway.html), and the annotated metabolites were mapped to KEGG pathways to identify key metabolic processes. Metabolite set enrichment analysis (MSEA) was subsequently performed to determine significantly enriched pathways, and their statistical significance was evaluated using the hypergeometric test.

Statistical analyses of enzyme activity and physiological-biochemical parameters were performed using Microsoft Excel 2021 and SPSS Statistics 27. All quantitative data were expressed as mean ± standard deviation (± SD), and differences between groups were evaluated using two-way analysis of variance (ANOVA), followed by Duncan’s multiple range test for post hoc comparisons (p < 0.05). Figures were created and refined using GraphPad Prism 10, Adobe Illustrator 2020, and Adobe Photoshop CC 2019 to ensure high-quality graphical presentation.

Results

Validation of genetically modified seedlings

CRISPR/Cas9-mediated gene knockout targets were designed based on the gene structure of NtPPO5. The selected target sequence (GGTAATTGTGTGCCATCATACGG) was located within the first exon of the gene (Fig. 1A). Ten independent transgenic plants (KO-1 to KO-10) were obtained and confirmed as positive transgenic plants (Fig. 1B). Sequence analysis demonstrated that the majority of these plants exhibited deletion-type mutations at the target locus, confirming the successful gene knockout of NtPPO5.

Fig. 1.

Validation of genetically modified tobacco. A Gene structure of NtPPO5 and positions of CRISPR/Cas9 knockout target sites. B Mutation types identified in NtPPO5 knockout lines. C Gel electrophoresis results showing hygromycin resistance in transgenic plants. D Relative expression levels of NtPPO5 in tobacco. Significant differences between groups are indicated by p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****)

The NtPPO5 overexpression transgenic seedlings were selected based on hygromycin resistance and verified through PCR (Fig. 1C), and a total of five plants were confirmed as positive plants (OE-1 to OE-5). The expression level of NtPPO5 in these positive plants and wild-type CB-1 (WT) was quantified by qRT-PCR. The results revealed that all five positive seedlings exhibited significantly higher NtPPO5 expression levels compared to the WT (Fig. 1D), confirming the successful construction and functional expression of the NtPPO5 overexpression material.

Agronomic trait analysis of transgenic tobacco plants

To evaluate the potential effects of NtPPO5 on plant growth and development, field trials were conducted over the 2024 and 2025 growing seasons. Under uniform cultivation conditions, we systematically compared key agronomic traits of wild-type (CB-1) plants with those of three independent NtPPO5 knockout lines (KO-1, KO-2, KO-3) and three overexpression lines (OE-1, OE-2, OE-3). Significant differences in agronomic traits were observed between the wild-type CB-1 and the transgenic lines. Compared with the wild-type CB-1, the KO lines exhibited a notable reduction in both plant height and maximum leaf area, whereas the OE lines showed a significant increase in these traits. In contrast, no significant differences were observed in stem circumference or the number of effective leaves (Fig. 2). These consistent findings across two consecutive growing seasons indicate that NtPPO5 positively regulates plant height and leaf area. Specifically, knockout of NtPPO5 led to growth inhibition, while its overexpression enhanced plant growth and development.

Fig. 2.

Agronomic traits of transgenic tobacco lines in 2024 and 2025. A Plant height. B Maximum leaf area, calculated as the product of the widest leaf length and leaf width. C Stem circumference. D Number of effective leaves suitable for roasting. Significant differences between groups are indicated by p < 0.05 (*), p < 0.01 (**), p < 0.001 (***)

Changes in the appearance of tobacco leaves during the curing process

To investigate the regulatory mechanism of NtPPO5 during the browning reaction, the upper leaves of wild-type CB-1, a knockout line (KO-2, hereafter KO) harboring a frameshift mutation in NtPPO5, and an overexpression line (OE-2, hereafter OE) with the highest transgene expression level were selected for curing experiments. Phenotypic changes in the leaves were systematically observed at various stages (T0-T5) to assess the impact of NtPPO5 modulation (Fig. 3). As the curing process progressed, the yellowing rate of the flue-cured tobacco leaves increased, with the leaves transitioning to a yellowish-brown hue as the temperature rose. By the T5 stage, the upper half of the CB-1 leaves exhibited a darker brown coloration, indicating excessive browning. In contrast, the OE leaves began to darken as early as the T3 stage, displaying severe excessive browning by T5, with browning intensity reaching approximately 90% and significant leaf shrinkage. Conversely, the KO lines demonstrated more uniform coloration, reduced shrinkage, and a milder browning effect. To further investigate the extent of browning, the leaf area was categorized into three levels: browning area > 70%, 70 − 45%, and < 45%. Compared to CB-1, the leaves in the KO line exhibited a significant reduction in browning, while the OE line showed a substantial increase in browning (Supplementary Table S2). Specifically, in CB-1, 17.80% of the leaves had a browning area > 70%, whereas this proportion was only 6.90% in the KO line. In contrast, the OE line reached 41.61%. On the other hand, in CB-1, 49.32% of the leaves had a browning area < 45%. In the KO group, this proportion significantly increased to 68.72% of the total leaf number, while in the OE line, it decreased sharply to 20.13%. These findings suggest that knockout of NtPPO5 inhibits the enzymatic processes driving browning, effectively preventing excessive the browning phenomenon during tobacco leaf curing.

Fig. 3.

Appearance changes of tobacco leaves during curing

Physiological and biochemical changes during the curing process

To examine the relationship between tobacco leaf browning and the antioxidant enzyme system, the activities of four key enzymes—polyphenol oxidase (PPO), peroxidase (POD), superoxide dismutase (SOD), and phenylalanine ammonia-lyase (PAL) —were measured throughout the curing process. Compared to the wild-type CB-1, PPO activity was significantly reduced in the KO line but markedly increased in the OE line across all six stages (T0-T5) of curing (Fig. 4A). PAL activity in both KO and OE steadily increased with rising temperatures, whereas in CB-1, it peaked at T4 and then decreased (Fig. 4B). Interestingly, POD activity exhibited an initial increase followed by a decrease, but the peak stages were different: CB-1 and KO peaked at T4, while OE peaked at T2 (Fig. 4C). For SOD activity, CB-1 and KO exhibited an initial increase followed by a decrease, with the KO showing significantly higher activity than CB-1. In contrast, OE showed a continuous increase in SOD activity with rising temperatures (Fig. 4D).

Fig. 4.

Dynamic changes in enzyme activity of tobacco leaves during the six stages of the curing process. A PPO. B PAL. C POD. D SOD. Error bars represent the mean ± SD (n = 3). Significant differences between groups are indicated by p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****)

To assess the impact of NtPPO5 on tobacco quality, the biochemical substance content of cured leaves was analyzed (Fig. 5). The carbohydrate content, including sugar, reducing sugar, and starch, remained relatively similar among the CB-1, OE, and KO, with no significant differences observed. In contrast, the nicotine, nitrogen, and potassium contents showed notable variations: KO exhibited higher nicotine and nitrogen levels, while OE showed significantly elevated potassium and nicotine contents.

Fig. 5.

Biochemical substance content at the end of baking. A Carbohydrate content. B Trace elements; error bars represent mean ± SD (n = 3). Significant differences between groups are indicated by p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****)

Statistical analysis of tobacco leaf metabolites

To investigate the influence of NtPPO5 on metabolites during the curing process, leaf samples from CB-1 and KO at five curing stages (T1-T5) were selected for metabolomic profiling analysis. Principal component analysis (PCA) of the 30 sample groups revealed that the first principal component (PC1) accounted for 33.07% of the total variance, while the second principal component (PC2) contributed 12.09% (Fig. 6A). Although metabolites from adjacent curing stages exhibited relative similarity, with partial overlap among samples, the three replicates of each sample group were closely clustered. This clustering indicated that the samples in each group had good repeatability and consistent metabolite profiles. A total of 694 metabolites were identified and classified into eight categories based on their chemical structures and biosynthetic pathways (Fig. 6B). The largest proportion of metabolites consisted of shikimic and phenylpropanoic acids, which accounted for 32.13%, followed by terpenoids at 21.04%, and alkaloids, which comprised 14.99%.

Fig. 6.

Metabolite profiling in curing tobacco leaves. A PCA analysis of samples. B Percentage of different types of metabolites

Dynamic analysis of metabolites during tobacco leaf curing

To investigate the dynamic changes in metabolites during the curing process in CB-1 and KO, metabolite profiles were analyzed across five curing stages (T1-T5). Four comparison groups were established: T1 vs. T2, T1 vs. T3, T1 vs. T4, and T1 vs. T5. In CB-1, 22, 102, 130, and 176 significantly differential metabolites were identified in these comparisons, respectively. In KO, 146, 248, 277, and 300 differential metabolites were detected across the same comparisons (Fig. 7A). Among these, shikimic acid, phenylpropanoids, and terpenoids exhibited the most significant fluctuations (Fig. 7B). K-Means clustering analysis of the differential metabolites across all comparisons revealed distinct dynamic patterns. In CB-1 (Fig. 7C), 169 metabolites (Cluster 2) were continuously upregulated, while 64 metabolites (Cluster 3) exhibited continuous downregulation. Clusters 1, 4, and 6 exhibited an initial increase followed by a decrease as temperatures rose. In contrast, KO exhibited a similar but more complex pattern (Fig. 7D), with 156 metabolites (Cluster 3) continuously upregulated, 63 metabolites (Cluster 4) continuously downregulated, and Clusters 2, 6, and 8 (147 metabolites) showed an initial rise followed by a decline. Compared to CB-1, KO cured leaves displayed richer dynamic metabolite changes during the curing process, with a substantially higher number of continuously upregulated metabolites. Notably, phenylpropanoids and terpenoids in KO showed greater diversity, higher content levels, and more pronounced fluctuations. These findings suggest that the knockout of the NtPPO5 gene may activate a phenylpropanoid pathway-mediated response to curing-induced stress. However, the underlying regulatory network and its association with aroma and other quality traits require further investigation.

Fig. 7.

Analysis of differential metabolites during the curing process. A Volcano plots showing the distribution of significantly differential metabolites. Red circles represent metabolites with p < 0.05, while blue circles indicate p ≥ 0.05. Numerical labels denote the total number of significant metabolites. The Y-axis represents the log₂ (Fold Change), with positive values indicating upregulated metabolites and negative values indicating downregulated ones. B Classification of primary metabolites. C K-means clustering line charts of CB-1. D K-means clustering line charts of KO

Integrative analysis of differential metabolites between CB-1 and KO during the curing process

To further elucidate the mechanism regulated by NtPPO5 in the browning reaction, the dynamic differences in metabolites between CB-1 and KO tobacco leaves were analyzed during the curing process. Five comparisons were conducted: C1 (CB-1T1 vs. KOT1), C2 (CB-1T2 vs. KOT2), C3 (CB-1T3 vs. KOT3), C4 (CB-1T4 vs. KOT4), and C5 (CB-1T5 vs. KOT5). The analysis identified 74, 157, 136, 56, and 136 significantly differential metabolites in these respective comparisons (Fig. 8A). All differential metabolites were systematically classified using the KEGG pathway database. The analysis revealed that the majority of these metabolites were significantly enriched in the global metabolic pathway category, underscoring its critical role throughout the curing process. Furthermore, the overview maps (Fig. 8B) highlighted key pathways, including membrane transport, cofactor and vitamin metabolism, amino acid metabolism, and carbohydrate metabolism, among others. To investigate the temporal dynamics of metabolic changes, the relative abundances of all differentially abundant metabolites were normalized and analyzed using K-means clustering. A total of 362 significant different metabolites across the five comparative groups (C1-C5) were identified, which were further categorized into six distinct expression profiles. These profiles were visualized in a clustering heatmap (Fig. 8C). The results indicated that while the trends of these differential metabolites during the curing process were similar between CB-1 and KO, their content and magnitude of change differed significantly. Specifically, a total of 182 metabolites included in Cluster 3, Cluster 4, and Cluster 5 exhibited higher levels in CB-1 compared to KO, whereas 180 metabolites included in Cluster 1, Cluster 2, and Cluster 6 showed lower levels in CB-1 than in KO. Focusing specifically on phenolic compounds, 62 phenolics were identified among the 362 differential metabolites (Fig. 8D). About half of these compounds were upregulated in KO, exhibiting higher abundance compared with CB-1. KEGG pathway enrichment analysis of the differential metabolites revealed significant enrichment in the Flavonoid biosynthesis (map00941) and Glutathione Metabolism (map00480) pathways (Fig. 8E). Additionally, in the early stages of curing (C2, C3), differential metabolites were enriched in pathways such as Pantothenate and CoA Biosynthesis (map00770). In contrast, in the later stages of curing (C4, C5), metabolites were primarily enriched in pathways related to Tyrosine Metabolism (map00350), Nicotinate and Nicotinamide Metabolism (map00760), and Arginine and Proline Metabolism (map00330), among others. These findings suggest that metabolic shifts occur throughout the curing process, with distinct pathways being activated in the early and later stages, potentially contributing to the flavor, color, and overall quality of cured tobacco leaves.

Fig. 8.

Analysis of differential metabolites during the curing process. A Volcano plots showing the distribution of significantly differential metabolites. Red circles represent metabolites with p < 0.05, while blue circles indicate p ≥ 0.05. Numerical labels denote the total number of significant metabolites. The Y-axis represents the log₂ (Fold Change), with positive values indicating upregulated metabolites and negative values indicating downregulated ones. B KEGG pathway classification of the differential metabolites. C K-means clustering analysis of all differential metabolites, visualized as line plots and corresponding heatmap profiles. D Heatmap showing the relative abundance of phenolic compounds among the differential metabolites. E KEGG pathway enrichment bubble plot, where deeper blue indicates smaller p-values and larger circle diameters represent higher enrichment levels

Notably, the differential metabolites from four comparisons (C2-C5) were all enriched in the pathway of flavonoid biosynthesis (map00941). Phenolic compounds serve as key substrates for browning reactions. Among these, flavonoids—a prominent class of phenolic compounds—exhibit strong antioxidant and antitoxin properties. These properties enable flavonoids to effectively scavenge reactive oxygen species (ROS) and enhance stress resistance during fermentation [32]. Naringin chalcone is a critical intermediate and precursor in the biosynthesis of flavonoids. Under the catalysis of chalcone synthase (CHS), p-coumaroyl-CoA and malonyl-CoA are condensed to form naringin chalcone, a reaction that serves as the rate-limiting step in flavonoid production [33]. In this study, the content of Naringin chalcone in the Flavonoid Biosynthesis Pathway increased with rising curing temperatures. However, its level was significantly lower in KO compared to CB-1 (Fig. 9). Following the knockout of NtPPO5, the accumulation of Naringenin chalcone significantly decreased, accompanied by an accelerated rate of consumption. This phenomenon may indicate an enhanced production of downstream polyphenolic metabolites, thereby reinforcing plant stress resistance and antioxidant capacity. Additionally, Caffeoylshikimic acid levels increased with elevated curing temperatures and were notably upregulated in KO relative to CB-1. As a polyphenolic compound, Caffeoylshikimic acid exerts antioxidant properties by scavenging free radicals, mitigating oxidative stress-induced damage in plants, and potentially modulating phytohormone signaling pathways through the generation of reactive intermediates to enhance stress tolerance [34]. The elevated Caffeoylshikimic acid content suggests that NtPPO5 knockout triggers the accumulation of polyphenolic substances, strengthening antioxidant defenses to counteract curing-induced abiotic stress and attenuate tissue browning.

Fig. 9.

Differential analysis of secondary metabolism. Mapping of flavonoid biosynthetic pathway genes in KEGG database

From the perspective of metabolite dynamics and metabolic pathway regulation, NtPPO5 knockout appears to optimize the Flavonoid Biosynthesis Pathway by modulating precursor availability and upstream metabolic fluxes, thereby increasing flavonoid content. This metabolic reprogramming establishes a more efficient antioxidant defense system, enabling tobacco leaves to better withstand curing-related stress, reduce curing-induced damage and browning, and ultimately improve the usability of cured tobacco leaves.

Discussion

The curing process of tobacco leaves involves a series of complex biochemical reactions, driven by enzyme catalysis and the transformation of inherent substances, ultimately yielding flue-cured tobacco leaves with uniform color and a distinctive aroma [21, 31]. Previous studies have demonstrated that key enzymes associated with browning—polyphenol oxidase (PPO) [35] and peroxidase (POD) [36]—exhibit distinct response patterns during curing. While PPO directly catalyzes the oxidation of phenolic compounds to quinones, leading to browning, POD contributes to this process by oxidizing phenolic substrates in a H₂O₂-dependent manner [36]. In contrast, superoxide dismutase (SOD) acts as a key inhibitor of browning by catalyzing the dismutation of superoxide anions (·O₂⁻), thereby reducing the generation of reactive oxygen species (ROS) and preventing the oxidation of phenolic compounds [37]. Consistent with previous study [5], this study observed a significant downregulation of PPO activity along with a marked upregulation of both SOD and POD activities in the KO line. This enzymatic reconfiguration—reduced PPO and enhanced antioxidant capacity—likely contributed to the observed phenotypic improvements. Compared to the control CB-1, the KO line exhibited superior color uniformity, significantly reduced browning, and an overall more desirable appearance after curing. We propose that the coordinated modulation of PPO and ROS-scavenging enzymes (SOD and POD) helps maintain redox homeostasis during curing, thereby suppressing excessive browning reactions in the KO line. Notably, while POD can participate in browning under certain conditions, its elevated activity in this context—coupled with enhanced SOD action—may collectively mitigate oxidative stress and limit the accumulation of browning precursors. This insight not only underscores the importance of the antioxidant enzyme system in the curing process but also suggests a potential regulatory strategy for optimizing tobacco curing and improving the visual and biochemical quality of cured leaves.

Flavonoids, as widely distributed natural phenolic compounds in plants, play a pivotal role in regulating plant growth and development, stress responses, and secondary metabolism through their unique antioxidant mechanisms [38]. As a major branch of secondary metabolism, flavonoid biosynthesis significantly contributes to the plant’s antioxidant defense by scavenging clearing excess ROS, thereby playing a key protective role in the plant’s response to environmental stresses [39]. KEGG pathway enrichment analysis across all five curing stages consistently identified the flavonoid biosynthesis pathway (map00941) as significantly enriched, indicating that this pathway may be actively involved in the metabolic response associated with NtPPO5 knockout. This finding is consistent with previous studies demonstrating that knocking down the NtPPO genes in tobacco leads to the enrichment of metabolites primarily in the flavone and flavonol biosynthesis pathways, triggering the specific accumulation of purines and flavonoid compounds in tobacco pistils [40]. Taken together, these observations support the notion that PPO-mediated metabolic regulation may influence key branch points in the flavonoid biosynthesis pathway, thereby regulating the dynamic balance of secondary metabolites in tobacco tissues. Additionally, in the flavonoid biosynthesis pathway, we observed distinct metabolic patterns in KO compared to CB-1. Specifically, the content of naringenin chalcone increased with rising curing temperatures but remained significantly lower in KO than in CB-1. Conversely, the content of caffeoyl shikimic acid increased in KO. Naringenin chalcone, a key intermediate in flavonoid biosynthesis, is not only essential for the pathway but also possesses significant antioxidant activity [41, 42]. Similarly, caffeoyl shikimic acid demonstrates strong antioxidant properties, effectively scavenging free radicals and inhibiting lipid peroxidation, thereby enhancing the plant’s antioxidant capacity [34, 43]. These differential patterns suggest that NtPPO5 knockout may alter the allocation and turnover of phenolic intermediates, thereby promoting downstream antioxidant capacity and contributing to reduced browning during curing. This interpretation is also consistent with the elevated antioxidant enzyme activity observed in KO leaves during curing, supporting a coordinated enzymatic–metabolic adjustment that favors browning suppression.

PPO activity is a key internal determinant of enzymatic browning during tobacco curing. During the temperature-sensitive phase, vacuolar disruption releases phenolic substrates that come into contact with chloroplast-associated PPO, thereby accelerating quinone formation and pigment darkening [12].Our study reveals that NtPPO5 plays a pivotal role in the browning process of tobacco leaves, where its overexpression intensifies enzymatic browning, while its knockout effectively mitigates the browning reaction. It can be speculated that after knocking out the NtPPO5 gene, tobacco leaves during the curing process may establish a redox buffering system. This system likely operates through a dual defense mechanism: (1) directly interrupting the oxidative chain reaction via a cascade-like free radical scavenging effect, and (2) blocking the enzymatic browning pathway by inhibiting polyphenol oxidase activity. Together, these mechanisms form a robust defense against oxidative browning (Fig. 10).

Fig. 10.

Schematic overview of the browning mechanism, gene editing application, and mechanisms for alleviating browning

Conclusion

This study provides a comprehensive analysis of the dynamic changes in differential metabolites, enzymes, and biochemical compounds in different tobacco cultivars during the flue-curing process. We found that, after knocking out the NtPPO5 gene, PPO enzyme activity was significantly reduced, effectively slowing down the excessive browning of the cured tobacco leaves. By integrating metabolomics, enzyme activity assays, and biochemical analysis, we inferred that the knockdown of NtPPO5 alleviated the degree of browning in the cured tobacco leaves through a coordinated or antagonistic effect of enzymes in the antioxidant enzyme system. Furthermore, an effective defense mechanism was established by regulating the biosynthesis pathways of compounds such as flavonoids, which further reduced the browning during curing. This study provides theoretical insights and practical evidence for mitigating the browning of cured tobacco leaves and offers valuable guidance for optimizing the curing process. Further research is needed to explore the broader effects of NtPPO5 knockout on related genes and metabolic pathways.

Authors’ contributions

TW and BZ analyzed the data and contributed to writing the manuscript; AC, HC, YS, GG and NZ conducted the experiments; XX and WL conceived and designed the research, wrote and revised the manuscript. All authors have read and approved the manuscript.

Funding

This research was financially supported by Fujian Tobacco Company (2024350000240023; 2025350000240080) and Guiding Project of Fujian Provincial Department of Science and Technology (2025N0023), and the authors declare that they have no competing interests with the funding company.

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.