Tissue specific mechanisms of tuber dormancy after 1,4-dimethylnaphthalene treatment in potato

Abstract

1,4-Dimethylnaphthalene (DMN) is gaining popularity as a postharvest sprout suppressant for potato storage. We hypothesized that DMN treatment impact critical phytohormone regulations in a tissue specific manner to suppress sprout growth of potato tubers. This study investigated transcriptional and hormonal responses in cv. ‘Russet Burbank’ tubers subjected to single or multiple DMN treatments during long-term storage. Tuber sprout growth was monitored during storage, and primary and secondary meristems, as well as tuber flesh tissues were collected from DMN treated and untreated tubers at three-week intervals for molecular and phytohormone analyses. Multiple DMN treatments were most effective in suppressing sprout growth up to 18 weeks in storage. Molecular responses to DMN treatments were tissue specific. Treatment of tubers with DMN altered the levels of hormones and their precursors as lower content of active forms of gibberellic acid, cytokinin, and jasmonic acid were observed in meristem tissues particularly with multiple DMN treatments. Aligning with phytohormone analyses results, multiple DMN treatments also altered the abundance of transcripts associated with abscisic acid, gibberellin, cytokinin, auxin, jasmonic acid, salicylic acid, and ethylene biosynthesis and signaling. DMN treatment impacted transcription factor families including APETALA2/ETHYLENE RESPONSIVE FACTOR, MYELOBLASTOSIS TRANSCRIPTION FACTOR, and BASIC HELIX-LOOP-HELIX associated with dormancy maintenance. Results of this research elucidate multiple DMN applications are needed to suppress sprout growth during long-term storage, and the underlying molecular mechanisms of DMN involve tuber tissue specific alteration of phytohormone regulation related to dormancy maintenance.

Graphical abstract

Highlights

• •Multiple 1,4-dimethylnaphthalene treatments were effective for sprout suppression.
• •Tuber tissue specific transcriptional and hormonal changes in response to treatment.
• •Gibberellin, cytokinin, and jasmonic acid were reduced in meristems with treatment.
• •Decreased transcript abundance of genes involved with growth related phytohormones.
• •Treatment altered expression of transcription factors related to meristem dormancy.

1. Introduction

Potato (Solanum tuberosum L.) is one of the leading food crops globally, playing a crucial role in food security and the agricultural economy. Potato tubers are rich sources of carbohydrates, vitamins, essential amino acids, minerals, and antioxidants, making them a dietary staple for a significant portion of the world's population (FAO, 2025). Postharvest losses of potato tubers remain a significant challenge, resulting in substantial crop and economic losses to growers and the potato industry (Olsen et al., 2006; Suttle et al., 2016). Overall, range of physiological alterations during storage that can impact their visual appearance, flavor, texture, nutritional composition, reduce commercial viability and cause economic losses (Jaiswal et al., 2023; Khorramifar et al., 2023; Pinhero et al., 2016; Suttle, 2004; Suttle et al., 2016). Premature sprouting of tubers during long-term storage is one of the physiological challenges that affects overall integrity of tubers and causes deterioration in their nutritional, processing, and other end-use qualities (Dogramaci, Dobry, et al., 2024; Suttle, 2004).

In general, the duration of dormancy and initiation of tuber sprouting is influenced by various factors, with the genetic make-up of potato varieties being a significant determinant (Dogramaci, Dobry, et al., 2024; Visse-Mansiaux et al., 2022). As a common industry practice, potato tubers are kept under cold temperatures to delay sprouting and maintain optimum tuber qualities. While low temperatures (<5 °C) can delay sprouting and minimize postharvest losses, it can also lead to higher levels of reducing sugars and cold-induced sweetening, which affects processing and food quality (Datir et al., 2012; Paul et al., 2016; Toufiq et al., 2024). Hence, it is imperative to explore supplementary strategies, such as application of synthetic or natural sprout suppressants to manage premature sprouting during postharvest storage.

Generally, synthetic sprout suppressants modulate metabolic and cellular processes to delay sprouting while, ideally, maintaining optimum tuber quality (Dogramaci, Dobry, et al., 2024). 1,4-Dimethylnaphthalene (DMN), a naturally occurring compound first isolated from dormant potato tubers (Meigh et al., 1973), and has been proposed as a sprout suppressant for potato tubers (Beveridge et al., 1981). DMN has been increasingly used as a sprout suppressor in potato storage as an alternative to commonly used sprout suppressor Chlorpropham (CIPC), which has been banned in European Union and several other countries (Campbell et al., 2020; Dogramaci, Dobry, et al., 2024; Visse-Mansiaux et al., 2021). Previous studies have demonstrated that DMN regulates starch remobilization and the formation of reducing sugars (Krause et al., 2023; Mendonça et al., 2022) and mitigates sprouting of potato tubers by regulating genes associated with the cell cycle (Campbell, Gleichsner, Hilldorfer, Horvath and Suttle, 2012, Campbell, Gwin, Tai and Adams, 2020; Dobry & Campbell, 2025; Dogramaci, Sarkar, et al., 2024). Decreased abundance of transcripts associated with DNA replication, cell division, and DNA methylation was observed in potato tubers with DMN application (Campbell et al., 2020). Increased abundance of transcripts related to cell cycle inhibitors like KIP-RELATED PROTEIN 1 (KRP1) and KRP2 was observed in meristem tissue of potato tubers when DMN was applied at the ecodormant stage (Campbell et al., 2012). Additionally, DMN is involved in the induction of transcripts associated with plant stress responses (Campbell et al., 2020; Campbell & D'Annibale, 2016; Dobry & Campbell, 2025) and is effective in reducing the incidence and severity of pathogens in potatoes (Campbell et al., 2019; Santos et al., 2023).

Although the use of DMN as a sprout suppressor has increased, the impact of single vs. multiple DMN applications on the metabolic and molecular mechanisms regulating dormancy progression during postharvest storage of potato tubers requires further investigation (Campbell et al., 2012; Dogramaci, Dobry, et al., 2024). Hormonal signaling plays a fundamental role in the establishment, maintenance, and breaking of dormancy in potatoes, and several genes associated with hormone metabolism have been identified during different dormancy stages of potato tubers (Boutsika et al., 2022; Dobry & Campbell, 2025; Suttle, 2004; Suttle et al., 2016). DMN can inhibit sprouting by modifying the abundance of transcripts associated with abscisic acid (ABA) (Campbell et al., 2010) and cytokinin (CK) signaling, auxin biosynthesis, signaling, and response (Dobry & Campbell, 2025), and jasmonate synthesis (Campbell & D'Annibale, 2016). Dobry and Campbell (2025) indicated that response of potato tuber to DMN is dependent on the stage of tuber dormancy and largely cultivar specific. Previous research has also shown that a single DMN application resulted in decreased abundance of transcripts associated with hormones involved in dormancy termination (CKs, gibberellins or GAs) at the early storage stage (Dogramaci, Sarkar, et al., 2024). However, a single application of DMN is not always effective in suppressing sprout growth when tubers are stored for longer periods (Cai et al., 2024; Dogramaci, Sarkar, et al., 2024). Yet, there is a knowledge gap in understanding how multiple applications of DMN treatments at different dormancy stages modulate hormone regulations of potato tubers at the molecular level.

We hypothesized that DMN applications alter phytohormone regulation in a tuber tissue specific manner and subsequently impact sprout suppression in potato tubers. In this research, integrated transcriptome and targeted metabolite analyses were used to determine how single, double, and triple DMN treatments during storage impact hormone profiles, related molecular pathways, and dormancy progression in potato tubers. Additionally, the effects of DMN on tissue specific molecular mechanisms of primary meristems, secondary meristems, and tuber flesh were also investigated by comparing transcriptome and hormone profiles. Furthermore, DMN-induced changes in sugar metabolism in tuber flesh tissues were also determined. The overarching goal was to build comprehensive insights into the underlying molecular mechanisms of DMN related sprout suppression of potato tubers.

2. Materials and methods

2.1. Plant material and experimental design

Russet Burbank potato (S. tuberosum L.) cultivar was chosen for this research because of its agronomic relevance, as it is the most commonly grown potato cultivar in the United States, as well as being among the most studied cultivars. Certified seed tubers of Russet Burbank were obtained from R.D. Offutt Farms (Fargo, ND, USA) and stored in an environmental chamber (Conviron GEN1000-GE, Winnipeg, MB, Canada) at 21 °C and 95% relative humidity (RH) for postharvest curing. After 14 days of curing, the temperature gradually reduced at a rate of ∼1.5 °C per day until reaching the holding storage temperature of 8 °C.

2.2. DMN treatments

Subsets of tubers were exposed to single (2 months after harvest), double (6 weeks after the first treatment), or triple (10 weeks after the first and 4 weeks after the second treatment) DMN treatments and compared with the control set (treated with 5% EtOH) following a partially nested experimental design. Tubers (300 per replicate) were placed in 100 L sealed containers, and a liquid solution containing DMN (at the rate of 20 mg kg⁻¹ of potato) was fogged for 60 s using a thermal fogger and maintained in dark at 8 °C. The dose of the DMN treatment was selected based on the standard industry practice for commercial potato storage. Using the same fogging system, 150 tubers per replicate were treated with 5% EtOH as the control in different containers and maintained in dark at 8 °C. The 5% EtOH was selected as the control treatment to match the solvent concentrations used for DMN applications and use of EtOH as control is the standard experimental method with DMN treatment (Dogramaci, Sarkar, et al., 2024). DMN treated and control containers were kept separately, and the container lids were removed 24 h post-treatment, and tubers were placed in environmental chambers (Conviron GEN1000-GE, Winnipeg, MB, Canada) set at 8 °C and 95% RH in the dark. A subset of tubers from the single DMN treatment were used for a second application of DMN (6-weeks after the first application), while a subset of tubers from double DMN treatments were used for a third DMN application (4-weeks after the second application). Each treatment included three biological replicates, and a biological replicate is defined as a batch of 10 tubers from each treatment group. For sampling 10 tubers from each treatment and replication combination were used to collect meristems and tuber flesh tissues at 3-week intervals post- first DMN treatment and until 12-weeks.

2.3. Sprouting measurement

Subsets of DMN-treated and control tubers were maintained in an environmental chamber (Conviron GEN1000-GE, Winnipeg, MB, Canada) at 8 °C to monitor sprouting performance. Each treatment group included three biological replicates, with six tubers per replicate (18 tubers per treatment). Changes in color of tuber meristems, “eye movement” (when at least one potato eye is awakening), “peeping” (when eye pointed out) and sprouting were monitored at 3-week intervals post-treatment. Sprout length (mm) was measured using a graduated ruler and converted to a standardized sprout score using ordinal scale with 5 categories: dormant (no detectable meristem growth, score 0), eye movement (> 0 mm ≤ 0.5 mm; score 1), peeping (> 0.5 mm ≤ 3 mm; score 2), sprout initiation (> 3 mm ≤ 5 mm; score 3), and fully sprouted (> 5 mm; score 4).

2.4. Sample preparation

For transcriptome and hormone analyses, tuber tissue samples were collected at various time points following DMN applications. More specifically, samples were collected at 3-, 6-, 9-, and 12-weeks after single DMN treatment; 3- and 6-weeks after double DMN treatment; and 2-weeks after triple DMN treatment using 10 new tubers (defined as a biological replicate) per sampling time point. Samples of primary and secondary meristems were collected separately using a scalpel to collect the meristematic region with minimal remnants of tuber pith or surrounding flesh tissues. Discs were also obtained from the central area of the tuber flesh by excising tissue in cylindrical shape using a stainless-steel cork borer. Transcriptome and hormone analyses included all three tissue types (primary meristems, secondary meristems, tuber flesh). For sugar analysis, only tuber flesh tissues were used, as there were limited amounts of representative meristem tissues available. The collected tissues were promptly frozen in liquid nitrogen, vacuum freeze-dried, macerated, and transferred to a − 80 °C freezer for subsequent transcriptomic, metabolomic, and sugar analyses. The time points (3-, 6-, 9-, and 12-weeks) referred throughout the manuscript including figures reflect the number of weeks after the first DMN application.

2.5. Determination of reducing sugars, sucrose, and starch

Reducing sugars (glucose and fructose), sucrose, and starch content of tuber flesh tissues were determined using a protocol originally described by Baxter et al. (2003). For reducing sugars and sucrose, frozen tuber flesh tissues (∼20 mg) were extracted in 5 mL of 80% methanol, and extracts were then centrifuged at 3000 ×g for 10 min. Preliminary experiment was conducted to optimize the weight of the tuber tissue for sugar analysis (10, 20, 30, 40, & 50 mg tuber tissues), and 20 mg of tissue was selected for further analyses. For sugar analysis three biological replicates and three technical replicates within each biological replicate were used. Supernatant (10 μL) was mixed with 160 μL of reagent mixture in a 96 well microplate and baseline absorbance was recorded at 340 nm using a SpectraMax microplate reader (Molecular Device, San Jose, CA, USA). After baseline reading, 5 μL hexokinase (1.5 × 10⁶ U L⁻¹), 5 μL phosphoglucose isomerase (0.35 × 10⁶ U kg⁻¹ protein), and 5 μL invertase (0.3 × 10⁶ U kg⁻¹ solid) were added sequentially and absorbance were recorded. For starch determination, the insoluble pellet fraction from the methanol extracts was used. The pellet was resuspended in 400 μL of sodium hydroxide (0.1 M NaOH) and homogenized at 95 °C for 1 h (500 rpm) to carry hydrolysis. The hydrolysis was then stopped by adding 70 μL of 1 M glacial acetic acid and 40 μL of the extracts were mixed with 80 μL of reagent mixture containing 0.5 M sodium acetate (pH 4.9), 1.2 μL of amyloglucosidase (1.4× 10⁷ U kg⁻¹), and 12 μL of α-amylase (4000 U kg⁻¹). The microplate was then incubated at 55 °C for 1 h in a shaker (300 rpm) and absorbance was recorded at 340 nm using a SpectraMax microplate reader (Molecular Device, San Jose, CA, USA). Glucose, fructose, sucrose, and starch content were expressed as mmol kg⁻¹ dry weight (DW).

2.6. Transcriptomics analysis

Total RNA was extracted from primary and secondary meristem samples (∼100 mg) according to Rubio-Piña and Zapata-Pérez (2011), and from tuber flesh samples (∼200 mg) according to Vennapusa et al. (2020). RNA quality and purity were initially assessed using agarose gel electrophoresis (1%) and a NanoDrop One (OD260/280 and OD260/230) (Thermo Scientific, Waltham, MA, USA). RNA integrity was further confirmed using an Agilent 5400 Bioanalyzer at Novogene Inc. (Sacramento, CA, USA). Library construction and RNA sequencing were performed at Novogene using the Illumina NovaSeq 6000 platform, generating 150 bp paired-end reads.

Raw sequencing data were subjected to quality control using FastQC, and low-quality reads and adapter sequences were removed with Trimmomatic v0.39 (Bolger et al., 2014). High-quality reads were aligned to the S. tuberosum reference genome (DM_1-3_516_R44_potato v6.1; Pham et al., 2020) from SpudDB using Bowtie2 v2.5.1. Gene-level read counts were obtained with FeatureCounts and used for differential expression analysis in DESeq2 (Love et al., 2014) within the R software version 4.4.3 (R Core Team, 2025). Normalization of the data and identification of differentially expressed genes (DEGs) were conducted within DESeq2. To ensure the quality of the results following data processing, a stringent selection criterion was applied to identify significant DEGs. Transcripts were considered increased if the log₂ value of their fold change (FC) was greater (+) than 0.6 and decreased if the log₂ value of their FC was lower (−) than 0.6, with significance determined by a false discovery rate (FDR) adjusted p-value (Benjamini & Hochberg, 1995) of <0.05. The rationale for selection of log₂ FC threshold of ±0.6 was to capture biologically meaningful transcriptional responses commonly observed in hormone-mediated sprouting regulation after the sprout inhibitor treatment. Principal component analysis (PCA) and volcano plots were generated using R software. Furthermore, predicted transcription factors (TFs) were identified using iTAK software (http://itak.feilab.net/) and DEGs were classified based on STRING database functional annotation (https://string-db.org/). Downstream analyses for regulatory function were conducted using Metascape (Zhou et al., 2019; https://metascape.org). Raw and processed RNA-seq data files have been deposited in the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) under the accession number GSE297139.

2.7. Targeted metabolome analysis

Targeted metabolite profiles were analyzed at Multi-User Analytical Lab (MUAL; Clemson University, SC, USA) based on the methodology described by Šimura et al. (2018). Hormones were extracted from 20 mg of freeze-dried tissue using two sequential extractions with 1 mL of 1:1 acetonitrile/water (v/v) and stable isotope-labelled internal standards (IS). Preliminary experiment was conducted to determine the optimum weight (20 mg) of the potato tuber tissues for phytohormone profile analysis. Samples from three biological replicates were used for hormone analysis. Samples were homogenized at 6000 rpm for five cycles of 30 s at 4 °C and centrifuged (12,000 rpm, 5 min, room temperature). Supernatants were pooled and purified using Oasis HLB SPE cartridges (60 mg, three cc) (Waters Corporation, Milford, MA, USA), preconditioned with methanol and acetonitrile/water. Purified extracts were dried under nitrogen gas and reconstituted in methanol/water (1:1, v/v). A standard mixture of hormones, including trans-zeatin, gibberellin A3, indole-3-acetic acid, ABA, jasmonic acid (JA), and salicylic acid (SA), was prepared, and calibration curves (R² > 0.99) were generated. Analysis was performed using a Waters Acquity UHPLC (Waters Corporation, Milford, MA, USA) coupled to a Xevo TQ Absolute tandem mass spectrometer (Waters Corporation, Milford, MA, USA) with an HSS T3 C18 column (2.1 × 150 mm, 1.8 μm) at a flow rate of 0.3 mL min⁻¹ with a column temperature of 40 °C. A multiple reaction monitoring method was used with mobile phases of 0.1% formic acid in water and 0.1% formic acid in acetonitrile. Mixture of hormone standards (31.25 ng mL⁻¹) were injected and used for quality control checks and internal standard recoveries were calculated using 0.241 μg mL⁻¹ of standard mixture with sample injection. The mass spectrometer was used in multiple reaction monitoring (MRM) mode. Data processing was carried out using MassLynx v4.1 software (Waters Corporation, Milford, MA, USA) and hormone quantification was performed by comparing results to their corresponding calibration curves of internal standards with minimum five calibration points.

2.8. Statistical analysis

The experimental design was partially nested as tubers receiving double and triple DMN treatments were subsets of the single DMN treatment groups. Statistical comparisons were conducted separately within each treatment stage to avoid non-independence across treatment exposure histories. For disruptive tuber tissues sampling, batch of 10 new tubers from each biological replicate and treatment combinations were used, and statistical analysis for phytohormone and transcript abundance results compared treatment effect separately for different tissue types and the statistical significance presented was time point specific. Tuber sprout scores were analyzed for each time point and tissue type combination, with one-way ANOVA compared treatments against controls, with biological replicates (n = 3) as independent observations. Normality (Shapiro-Wilk test) and homogeneity of variance (Levene's test) were verified prior to parametric testing (p > 0.05 for assumption validation). Significant ANOVA results (p < 0.05) underwent Dunnett's post-hoc contrasts, comparing treatments exclusively to control while controlling family-wise error. Missing data were excluded via pairwise deletion. Analyses were conducted using R v4.4.3 (R Core Team, 2025).

For transcriptome data, functional annotation and enrichment analyses of DEGs were performed to identify significantly enriched Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. These analyses were conducted in Metascape (Zhou et al., 2019; https://metascape.org). Principal component analysis (PCA), heatmaps with hierarchical clustering, and enrichment plots were generated to visualize the transcriptomic and metabolomic results, as well as the statistical analyses were conducted using R v4.4.3.

3. Results and discussion

3.1. Suppression of sprout growth and changes in reducing sugars, sucrose, and starch content with DMN treatment

Sprout growth results revealed distinct responses to DMN from the primary and secondary bud meristems during the 18-weeks of storage observations (Fig. 1). In control tubers, primary meristems initiated growth (“eye movement”) 6-weeks after the experiment start date. Whereas, DMN significantly suppressed this growth activation, with increasing efficacy from single to triple applications (Fig. 1). Significant reductions in sprout scores of primary meristems were observed from week-6 onwards (p < 0.01), and significant inhibition was achieved with double and triple treatments by week-9 and week-12, respectively, with suppression maintained through week-18 (p < 0.0001). Although sprout scores were ordinal, the scale reflects dormancy progression and sprouting developmental stages, which met the standard parametric assumptions.

Fig. 1.

Sprout score of potato tubers at 3-, 6-, 9-, 12-, 15-, and 18-weeks after first DMN application. Sprouting dynamics in primary meristem and secondary meristem were determined by converting actual sprout length (mm) to a standardized scale. Different letters denote significant differences between DMN-treated tubers [single (DMN 1×), double (DMN 2×) and triple (DMN 3×) DMN applications] vs. control by Dunnett's test (* p < 0.05, ** p < 0.01, *** p < 0.001). Data points represent mean sprout scores ± SE (n = 6).

In contrast, secondary meristems responded differently than primary meristems (Fig. 1). While growth from primary meristems were suppressed, a single DMN treatment stimulated sprouting in secondary meristems between 6- and 15-weeks (p < 0.01). Whereas double and triple applications resulted in delayed and significant suppression, evident at week-18 (p = 0.0039 and p = 0.0014). These results suggest that while the first DMN treatment suppressed growth from primary meristems, it may have impacted the apical dominance of the primary meristem and consequently stimulated growth from the secondary meristems. Hormone analysis results supported this finding as active form of GA (GA₄), different forms of CKs (trans-zeatin-riboside-O-glucoside, dihydrozeatin-riboside-O-glucoside, and 2-methylthio-cis-zeatin) and auxin biosynthesis precursor (tryptophan) reduced in primary meristems with single DMN treatment but increased in secondary meristem under same treatment (Fig. 3, Fig. 5, Fig. 6). However, double and triple DMN treatment, which suppressed sprout growth, also reduced concentration of these sprouting related hormones in secondary meristems. Higher concentrations of these hormones (GA, CK, auxin) in meristem tissues are associated with reactivation of meristem activity and sprout growth in potato tubers (Hartmann et al., 2011; Sonnewald & Sonnewald, 2014; Suttle, 2004, Suttle, 2007). Auxin level is generally higher in primary bud tissues during apical dominance, while once apical dominance is lost, the higher auxin level is found in secondary meristems (Eshel & Teper-Bamnolker, 2012). Similarly, higher CK is required for bud activation and our results suggest that multiple DMN treatments are needed to reduce level of different forms of CKs in secondary meristem and to suppress sprout growth. Transcriptomic analysis also highlighted this trend as transcript abundance of CYTOKININ OXIDASE 5 (CKX5) and ARABIDOPSIS RESPONSE REGULATOR 1 (ARR1) increased in secondary meristems with single DMN application, while multiple DMN applications resulted in decreased transcript abundance. Overall, these findings help explain the need for repeated DMN applications for long-term inhibition of sprout growth in potato tubers.

Fig. 3.

Overview of gibberellin biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in gibberellin biosynthesis (Diterpenoid biosynthesis pathway, ko00904) and signaling (Plant hormone signal transduction pathway, ko04075) in primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and sampling time points. (B) Heatmap represents the relative accumulation of gibberellin precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*). Metabolites with no detection are represented in black.

Fig. 5.

Overview of cytokinin biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in cytokinin biosynthesis (Zeatin biosynthesis pathway, ko00908) and signaling (Plant hormone signal transduction pathway, ko04075) in primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and storage periods. (B) Heatmap representing the relative accumulation of cytokinin precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*). Metabolites with no detection are represented in black.

Fig. 6.

Overview of auxin biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in auxin biosynthesis (Tryptophan metabolism pathway, ko00380) and signaling (Plant hormone signal transduction pathway, ko04075) in the primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and storage periods. (B) Heatmap represents the relative accumulation of auxin precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*).

Collectively, our sprout growth results align with previous reports showing that DMN suppresses growth by interfering with cell cycle progression in primary meristems (Campbell, Gleichsner, Hilldorfer, Horvath and Suttle, 2012, Campbell, Gwin, Tai and Adams, 2020; Dobry & Campbell, 2025; Dogramaci, Sarkar, et al., 2024) and that single applications are often insufficient for long-term sprout suppression, specifically to restrict growth of secondary meristems (Dogramaci, Sarkar, et al., 2024). The multiple dose-dependent nature of these effects also corroborates with previous findings by de Weerd et al. (2010), who reported enhanced sprout suppression with elevated DMN concentrations across potato cultivars.

Sprout suppression by synthetic or natural inhibitors involves alteration of sugar metabolism in meristem and tuber flesh tissues (Dogramaci, Sarkar, et al., 2024, Dogramaci et al., 2025; Paul et al., 2016; Suttle, 2004; Suttle et al., 2016). Previous research highlighted the role of DMN in reduction of starch remobilization and lowering reducing sugar content in tuber meristem tissues (Krause et al., 2023; Mendonça et al., 2022). In this study, sugar analysis results indicated significant reduction in glucose and fructose content in tuber flesh tissues with single and double DMN treatments when compared to control tubers at 3-, 6-, and 12-weeks (Supplementary Fig. 1 A & 1B). The single DMN treatment also reduced sucrose content in tuber flesh tissues at 3- and 6-weeks (Supplementary Fig. 1C). Interestingly, reduced starch content was observed in flesh tissues of tubers treated with triple DMN treatments at 12-weeks (Supplementary Fig. 1D).

These changes in reducing sugars and starch content in tuber flesh with DMN treatment have relevance in sprout suppression. Breakdown of starch to smaller sugar units such as glucose and mobilization of sugars from source (parenchyma tissue or tuber flesh) to sink (meristem tissues) is important for biochemical regulation during sprouting (Dogramaci et al., 2025; Haider et al., 2023; Suttle et al., 2016). The alteration of reducing sugar content and starch in tuber flesh with DMN applications might act as an indicator of DMN-induced changes in sugar metabolism and its subsequent impact on suppression of meristematic growth. Beyond just altering sugar metabolism, sprout suppressors also modulate hormones, stress responses, and cell cycle regulation in tuber tissues (Considine & Considine, 2016). Therefore, we examined how single and multiple DMN applications modulate gene expression and hormone profiles across meristems (primary and secondary meristems) and tuber flesh tissues.

3.2. Overview of the RNA-seq analysis

Differential expression analyses were conducted for each tissue type with a significance threshold of p < 0.01, revealing clear transcriptional divergence between DMN-treated and control tubers across all sampling time points. In the primary meristems, transcripts for an average of 23,900 genes were detected, with ∼9.5% exhibiting increased and ∼ 7.9% decreased transcript abundance (Fig. 2A). The strongest transcriptional shifts were observed at 12-weeks, particularly in tubers exposed to double (29.8% DEGs) and triple (20.4% DEGs) DMN applications. PCA confirmed these dynamic responses, separating samples into three main groups (Fig. 2B). The first group included control samples and those from 3- and 6-weeks, while the second group comprised tissues of DMN-treated tubers collected at 9- and 12-weeks. Notably, tubers treated with triple DMN treatment formed a distinct third cluster at 12-weeks, highlighting the strong effect of repeated applications.

Fig. 2.

Transcriptomic analysis of potato tuber tissues treated with DMN. Bar charts showing the total number of increased (red), decreased (blue), and unchanged (green) transcript abundance in the primary meristem (A), secondary meristem (C), and tuber flesh (E), with p-value <0.01. Principal Component Analysis (PCA) plots of the RNA-seq data for the primary meristem (B), secondary meristem (D), and tuber flesh (F). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In the secondary meristems, transcripts for an average of 24,289 genes were identified (Fig. 2C) and the proportion with decreased transcript abundance (10.7%) exceeded those with increased abundance (9.4%). The largest transcriptional response occurred at 6-weeks, when 33.4% of genes were differentially expressed. PCA analysis separated the samples into six distinct clusters, reflecting both number of treatment applications and storage duration (Fig. 2D). These results indicate a unique regulatory trajectory for the secondary meristems, with pronounced responses at earlier storage stages.

In the tuber flesh, transcripts for 21,734 genes were identified (Fig. 2E). Only a small fraction exhibited altered transcript abundance (3.2% increased, 3.5% decreased), indicating lower responsiveness compared to meristematic tissues. However, marked transcriptional changes were detected at 12-weeks, especially in tubers treated once (19.8% DEGs) or thrice (10.1% DEGs) with DMN. PCA revealed four major clusters, with the two principal components explaining 83.6% of the variation (Fig. 2F). The most distinct group corresponded to triple DMN-treated tubers sampled at 12-weeks, underscoring the cumulative impact of repeated treatments.

These findings emphasize the tissue-specific nature of DMN responses. The primary meristems exhibit the strongest and most consistent transcriptional reprogramming, consistent with its anatomical role as the primary initiation site for sprouting (Hu et al., 2023). The secondary meristems exhibit a distinct regulatory profile, dominated by reduced transcript abundance at mid-storage, while the tuber flesh is comparatively less responsive to the treatments. Collectively, our results indicate that multiple DMN applications are required to suppress meristematic activity across tissues and preserve tuber integrity during long-term storage.

3.3. GO and KEGG enrichment highlight hormone-related regulation by DMN

Functional enrichment analysis using GO and KEGG pathways revealed the biological roles of DEGs across tissues, DMN treatments, and sampling times. GO analysis exhibited significant overrepresentation of biological processes related to shoot system development (GO:0048367), regulation of hormone levels (GO:0010817), and hormone-mediated signaling pathways (GO:0009755) in all three tissue types examined (Supplementary Fig. 2 A–C). KEGG analysis highlighted enrichment in plant hormone signal transduction (ath04075) (Supplementary Fig. 3 A–C). In the tuber flesh, the strongest enrichments (both increased and decreased transcript abundance) were observed at 12-weeks in tissues of DMN-treated tubers (Supplementary Fig. 3C). These results reinforce the hypothesis that DMN modulates dormancy and postharvest traits through hormone-related signaling pathways.

Clearly, hormones are central regulators of plant development and tuber dormancy in potato (Sorce et al., 2009). The enrichment of hormone biosynthesis and signaling pathways across GO and KEGG categories highlights the role of DMN in modulating dormancy regulation, consistent with previous studies (Campbell, Gleichsner, Alsbury, Horvath and Suttle, 2010, Campbell, Gwin, Tai and Adams, 2020; Campbell & D'Annibale, 2016; Dobry & Campbell, 2025; Dogramaci, Sarkar, et al., 2024). Overlap analysis revealed that the largest set of unique DEGs occurred in the primary meristems of tubers sampled at 12-weeks with double and triple DMN treatments (Supplementary Fig. 2 A, 3 A), and in the tuber flesh with a single application (Supplementary Fig. 2C, 3C). In contrast, the secondary meristem had the highest unique gene count at 6-weeks following a single DMN application (Supplementary Fig. 2B, 3B). More substantial overlaps in DEGs and enriched pathways were observed in the meristems than in the tuber flesh, with most functional categories converging on hormone-related processes.

Overall, these findings suggest that DMN modulates a network of hormone-associated pathways, and that crosstalk among different tissues contributes to the coordinated suppression of sprouting during storage. Consistent with this, functional classification of DEGs further highlighted the strong impact of DMN on genes related to hormone biosynthesis and signaling, with effects that varied across sampling time and tissue types (Supplementary Fig. 4; Supplementary Table 1). The greatest DEG counts were observed in the primary meristems of tubers subjected to double DMN treatment and in the flesh tissues from tubers with single or triple treatments at 12-weeks, whereas in the secondary meristem, DEGs peaked after a single DMN applications at both 6- and 12-weeks. Notably, most DEGs were linked to ABA, GA, CK, auxin, ethylene, JA, and SA pathways. In the following subsections, we detail how DMN modulated each of these hormonal networks, integrating DEGs and hormone profile data.

3.4. GA and ABA pathways under DMN treatment

Following the termination of dormancy in potato tuber, GAs plays a pivotal role in initiating sprout development (Sonnewald & Sonnewald, 2014; Suttle, 2004). The application of DMN resulted in differential transcript abundance of three genes [ENT-COPALYL DIPHOSPHATE SYNTHASE 1 (GA1), GIBBERELLIN 20-OXIDASE 1 (GA20OX1), GIBBERELLIN 3-OXIDASE 1 (GA3OX1)] involved in diterpenoid biosynthesis (ko00904) and three genes [GIBBERELLIN RECEPTOR GID1B (GID1B), GIBBERELLIC ACID INSENSITIVE (GAI), PHYTOCHROME-INTERACTING FACTOR 4 (PIF4)] in the hormonal signal transduction pathway (ko04075) (Fig. 3A). Transcript abundance of GA1, which is involved in first committed step of GA biosynthesis, increased in all tuber tissues, except in the secondary meristems of tubers exposed to single or triple DMN applications. In contrast, transcript abundance of GA3OX1, which facilitates the conversion of inactive precursors into bioactive forms (Di et al., 2024), decreased in primary and secondary meristems with multiple DMN applications at 9- and 12-weeks of sampling.

Genes involved in GA signal transduction also exhibited tissue-specific responses to DMN. In meristem tissues, transcript abundance of GID1B and PIF4 decreased, whereas transcript abundance of GAI, which encodes a DELLA repressor that antagonizes GA signaling, increased with DMN applications, particularly in primary meristems. However, in tuber flesh tissues, transcript abundance of GID1B and PIF4 increased under DMN treatment, while transcript abundance of GAI decreased. These contrasting responses indicate that DMN alters GA signaling inversely between meristem and flesh tissues, potentially reducing GA responsiveness in meristems (where sprout growth is initiated) but not in tuber flesh.

Metabolite analyses further supported these transcriptomic trends. A decrease in GA₄, an active GA form, was observed in secondary meristems of DMN-treated tubers at 12-weeks (Fig. 3B), coinciding with stronger suppression of secondary sprout growth (Fig. 1). Additionally, reduced contents of GA₄₄, an intermediate of GA biosynthesis, were detected in primary meristems with DMN treatment at 12-weeks. Additionally, contents of GA₄₄ and GA₅₁, inactive forms of GA were significantly higher in flesh tissues of DMN-treated tubers at 12-weeks. These findings indicate that DMN effects on GA pathways vary depending on the tissue type. Furthermore, our results align with previous reports that DMN adversely influences the transcript abundance of genes involved in GA biosynthesis and signaling in potato tubers (Dobry & Campbell, 2025; Dogramaci, Sarkar, et al., 2024). Overall, reduced transcript abundance of GA-related genes together with lower contents of active GAs in meristems suggest that DMN treatment suppresses GA biosynthesis and signaling, thereby influencing tuber sprouting through potential suppression of cell division and cell elongation, which is associated with elevated GA response.

Among major phytohormones, ABA is well recognized for its critical role in inducing and maintaining dormancy in potato tubers during postharvest storage (Destefano-Beltrán et al., 2006; Dogramaci et al., 2026; Dogramaci, Dobry, et al., 2024; Foukaraki et al., 2016; Suttle, 2004). ABA levels decline progressively through inactivation and catabolism during storage, leading to dormancy release (Sonnewald & Sonnewald, 2014; Suttle, 1995, Suttle, 2004). In our study, DMN impacted transcript abundance of four genes [LYCOPENE Β-CYCLASE 1 (LCY1), 9-CIS-EPOXYCAROTENOID DIOXYGENASE 4 (CCD4), ABSCISIC-ALDEHYDE OXIDASE 1 (AAO1), ABSCISIC ACID DEFICIENT 2 (ABA2)] involved in carotenoid biosynthesis pathway (ko00906) and three genes [PYRABACTIN RESISTANCE 1-LIKE 4 (PYL4), HIGHLY ABA-INDUCED PP2C PROTEIN 1 (HAB1), SNF1-RELATED PROTEIN KINASE 2 (SnRK2)] of ABA signal transduction pathway (ko04075) (Fig. 4A). Transcript abundance of AAO1, which is involved in the final step of ABA biosynthesis, increased in meristem tissues at 9-weeks under single and double DMN applications, while transcript abundance of CCD4, primarily involved in carotenoid breakdowns, increased in all three tissues at 9-weeks under double DMN treatment. However, by 12-weeks, CCD4 transcript abundance decreased in meristems, suggesting a temporal shift in ABA biosynthesis under multiple DMN treatments.

Fig. 4.

Overview of abscisic acid (ABA) biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in ABA biosynthesis (Carotenoids biosynthesis pathway, ko00906) and signaling (Plant hormone signal transduction pathway, ko04075) in primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and storage periods. (B) Heatmap representing the relative accumulation of ABA precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*).

Hormone analyses corroborated these results as phaseic acid, a major catabolic product of ABA, was lower in primary meristems with multiple DMN treatments at 12-weeks, while it was greater in tuber flesh under single and triple DMN treatments (Fig. 4B). Dihydrophaseic acid, another catabolic product, was also detected at greater concentrations in tuber flesh, while it was lower in secondary meristem tissues under triple DMN treatment (Fig. 4B). These findings suggest that DMN suppresses ABA catabolism in meristems, supporting dormancy maintenance, while ABA breakdown is more evident in tuber flesh tissues.

Transcript abundance of LCY1, which codes for a carotenoid pathway enzyme supplying precursors for ABA biosynthesis, was predominantly higher in tuber flesh with DMN treatment. In contrast, transcript abundance of PYL, HAB1, and SnRK2 decreased in meristem tissues, particularly at 12-weeks sampling time point. Dogramaci, Sarkar, et al. (2024) previously reported higher transcript abundance of PYL4 in meristems under methyl jasmonate (MeJA) and MeJA + DMN treatments, suggesting that DMN alone and in combination with other regulators can differentially influence ABA signaling. Our findings, together with prior studies (Sorce et al., 2000), emphasize that DMN modulates ABA biosynthesis, catabolism, and signaling in a tissue-specific manner, sustaining ABA's dormancy maintenance role in meristems while altering ABA turnover in flesh tissues.

The DMN-induced adjustments in GA and ABA pathways exemplify how sprout inhibition is accompanied by broad hormonal rebalancing. With DMN treatment, not only are GA and ABA levels impacted, but other growth regulators are also modulated. Notably, auxins and CKs exhibit significant DMN-responsive changes in transcript abundance and metabolite levels, reinforcing the knowledge of integrated hormonal crosstalk in tuber dormancy control (Dogramaci et al., 2026).

3.5. Tissue-specific regulation of CK and auxin pathways under DMN treatment

CKs are intricately associated with reserve accumulation and tuberization processes in potatoes (Cheng et al., 2020; Raspor et al., 2021). They act as pivotal signals for the reactivation of meristematic activity, ultimately triggering dormancy release (Hartmann et al., 2011; Suttle et al., 2014). CK concentrations are tightly controlled through genes linked to their biosynthesis and signaling (Campbell et al., 2014; Suttle et al., 2014). In our results, DMN treatment influenced transcript abundance of four genes [ISOPENTENYLTRANSFERASE 1 (IPT1), CYTOCHROME P450 735 A2 (CYP735 A2), UDP-GLYCOSYLTRANSFERASE 73C3 (UGT73C3), CKX5] involved in the zeatin biosynthesis pathway (ko00908) and four genes [ARABIDOPSIS HISTIDINE KINASE 3 (AHK3), ARR1, ARABIDOPSIS RESPONSE REGULATOR 17 (ARR17), ARABIDOPSIS HISTIDINE-CONTAINING PHOSPHOTRANSFER PROTEIN 1 (AHP1)] in the hormone signal transduction pathway (ko04075) (Fig. 5A). Transcript abundance of IPT1, a gene involved in CK biosynthesis (Miyawaki et al., 2006), decreased in meristem tissues following DMN applications. Overexpression of IPT has previously been correlated with increased sprouting rates in vitro (Hartmann et al., 2011). Consistent with these findings, IPT1 transcript abundance decreased across different tissues and time points in DMN-treated tubers in our result, supporting earlier observations (Dogramaci, Sarkar, et al., 2024; Hartmann et al., 2011).

Meristem tissues under DMN treatment also had decreased transcript abundance of UGT73C3, a UDP-glycosyltransferase critical for CK biosynthesis (Alameldin & Montgomery, 2023; Hou et al., 2004). Another gene, CYP735A2, involved in converting CK precursors into active forms, displayed a dynamic regulation pattern: decreased transcript abundance during early storage, followed by increased abundance at 12-weeks, regardless of number of DMN applications or tissue type. Results of hormone profiles further supported these findings. Concentrations of trans-zeatin-riboside-O-glucoside, dihydrozeatin-riboside-O-glucoside, and 2-methylthio-cis-zeatin significantly declined in primary meristems after single and multiple DMN applications, particularly at 12-weeks (Fig. 5B). In contrast, single DMN treatment led to increased concentrations of these CK forms in secondary meristems at 12-weeks, whereas double and triple DMN treatments reduced their levels in the same tissues. These findings suggest that multiple DMN applications are required to suppress secondary meristem growth, and lower concentrations of CKs are related to sprout growth cessation.

Transcript abundance of CK signaling genes also reflected tissue-specific responses to DMN. At 12-weeks sampling time point, transcript abundance of AHK3, ARR1, ARR17, and AHP1 decreased in primary and secondary meristems, whereas transcript abundance of these same genes increased in tuber flesh (Fig. 5A). Dogramaci, Dobry, et al. (2024) reported similar reductions in ARR1 transcript abundance in meristems of cv. Russet Burbank under single DMN treatment. Likewise, Dobry and Campbell (2025) reported decreased transcript abundance of CK signaling repressors KISS ME DEADLY 1 (KMD1) and KMD2 in meristems of DMN-treated tubers. Taken together, these findings indicate that DMN exerts tissue-specific modulation of CK biosynthesis and signaling, with distinct regulatory patterns dependent on tissue type and storage duration.

Auxin also plays a central role in tuber dormancy regulation. Indeed, auxin directly and indirectly impacts all stages of plant development (Kolachevskaya et al., 2019). Auxin concentrations are highest in freshly harvested potato tubers especially in the meristems, and decrease progressively until dormancy ends (Sorce et al., 2000; Sorce et al., 2009). Exogenous auxin application also has dose-dependent effects: high concentrations inhibit sprouting, whereas low concentrations promote it (Kolachevskaya et al., 2019; Suttle, 2003, Suttle, 2007). Suttle (2007) also suggested that auxin is a prerequisite for sprout growth rather than a direct regulator of dormancy maintenance.

In our results, tissue-specific and storage-time-dependent changes in auxin-related transcript abundance were observed, including genes involved in tryptophan metabolism and hormone signal transduction pathways (Fig. 6A). Five genes [TRYPTOPHAN AMINOTRANSFERASE RELATED 2 (TAR2), FLAVIN-CONTAINING MONOOXYGENASE 6 (YUC6), ALDEHYDE DEHYDROGENASE 2B7 (ALDH2B7), AMIDASE (AMIE), DIOXYGENASE FOR AUXIN OXIDATION (DAO)] within the tryptophan metabolism (ko00380) and nine genes [AUXIN INFLUX CARRIER 1 (AUX1), TRANSPORT INHIBITOR RESPONSE 1/AUXIN SIGNALING F-BOX PROTEINS (TIR1/AFB), AUXIN/INDOLE-3-ACETIC ACID (AUX/IAA), AUXIN RESPONSE FACTOR (ARF), INDOLE-3-ACETIC ACID-AMIDO SYNTHETASE GH3.1 (GH3.1), SMALL AUXIN-UP RNA (SAUR), TRANSMEMBRANE KINASE 1 (TMK1), TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS 1 (TAA1), ABA INSENSITIVE2 (ABI2)] in plant hormone signal transduction pathways (ko04075) exhibited differential transcript abundance across sampling time points, with notable tissue-specific variation (Fig. 6A). In primary meristems, transcript abundance of TAR2, YUC, ALDH2B7, and AMIE increased with single and multiple DMN applications at 12-weeks; all these genes participate in the conversion of tryptophan to indole-3-acetate. In contrast, in secondary meristems, transcript abundance of YUC, which is involved in final rate limiting step of auxin biosynthesis, initially increased at 9-weeks, but significantly decreased at 12-weeks under DMN treatment, aligning with suppressed sprout growth at that time point (Fig. 6A & Fig. 1). Dobry and Campbell (2025) also reported decreased YUC transcript abundance in meristems following DMN treatment.

These transcriptional patterns were supported by phytohormone analyses. Tryptophan content in primary meristems was lower at 3-weeks under a single DMN application but increased at 12-weeks under double and triple treatments. Conversely, in secondary meristems, tryptophan levels were significantly lower at 12-weeks under triple DMN treatment (Fig. 6B), coinciding with reduced sprout growth scores (Fig. 1). This contrasting behavior highlights that primary and secondary meristems respond differently to single vs. multiple DMN treatments; as single DMN treatment reduced tryptophan level and suppressed growth from primary meristems at 3-weeks, while multiple DMN treatments, which was effective in suppressing growth from secondary meristem, reduced auxin precursor content and decreased YUC transcript abundance at 12-weeks. These findings are consistent with previous studies demonstrating the importance of auxin gradients as signals for potato tuber sprouting (Dogramaci, Sarkar, et al., 2024; Sorce et al., 2000; Sorce et al., 2009). Transcriptome analyses in potato have also shown that amino acid metabolism is subject to strong tissue-specific regulation under changing physiological conditions (Tiwari et al., 2020), supporting the idea that tryptophan dynamics under DMN may reflect broader tissue-dependent metabolic regulation.

Studies in seed dormancy and other plant models suggest auxin interacts with ethylene and ABA in regulating dormancy maintenance (Chao et al., 2017; Dogramaci et al., 2017; Jhanji et al., 2024; Li et al., 2017), and this crosstalk may also apply to potato tubers (Dogramaci, Sarkar, et al., 2024). Previously, Suttle (2003) reported that exogenous auxin treatment impacted the ethylene-mediated inhibition of sprout growth in cv. Russet Burbank. Furthermore, tryptophan may function beyond its role as an auxin precursor. Recent evidence suggested that exogenous tryptophan can act as a signaling molecule in potato, inducing broad-spectrum defense pathways including SA- and JA/ET-related responses (Zhao et al., 2022). Such multifunctional roles of tryptophan may explain why its accumulation under certain DMN treatments does not directly parallel auxin-related gene expression, reinforcing the idea that tryptophan contributes to the broader hormonal network controlling dormancy and sprouting.

In this research, transcripts involved in auxin signaling (including AUX1, TIR1/AFB, ARF, and GH3.1) exhibited increased abundance in primary meristems under DMN between 9- and 12-weeks (Fig. 6A). In contrast, transcript abundance of AUX/IAA, a negative regulator of auxin signaling, decreased in all three tuber tissues under DMN. In tuber flesh, decreased transcript abundance of GH3.1 and AUX1 was observed under single and triple DMN treatments at 12-weeks. Together, these results suggest that DMN induces tissue-specific adjustments in auxin signaling, reinforcing auxin's critical role in dormancy regulation and sprouting.

Altogether, the modulation of CK and auxin pathways by DMN highlights their coordinated roles in dormancy regulation and sprout initiation in a tissue-dependent manner.

3.6. Regulation of JA and brassinosteroid (BR) pathways under DMN treatment

JA is another important hormone implicated in regulating potato morphogenesis, with its levels varying depending on the developmental stage and the specific organ analyzed (Abdala et al., 2002). In the context of potato tuber meristems, endogenous jasmonate levels are comparatively low during the deepest dormancy phase but exhibit a substantial increase at the onset of sprouting, subsequently decreasing once sprout development progresses (Suttle et al., 2011). In our results, nine DEGs [LIPOXYGENASE 2 (LOX2), ALLENE OXIDE CYCLASE 2 (AOC2), AMP-DEPENDENT SYNTHETASE AND LIGASE (OPCL1), ACYL-COA OXIDASE (ACX), ENOYL-COA HYDRATASE/3-HYDROXYACYL-COA DEHYDROGENASE (AIM1), ALLENE OXIDE SYNTHASE (CYP74A), JASMONATE RESISTANT 1 (JAR1), JASMONATE-ZIM-DOMAIN PROTEIN 5 (TIFY10A), BASIC HELIX-LOOP-HELIX (bHLH) DNA-BINDING FAMILY PROTEIN (MYC2) were identified in the jasmonate biosynthesis and signal transduction pathways (Fig. 7A). DMN treatment led to increased transcript abundance of LOX2 and AOC2, which are involved in initiation and rate limiting steps of JA biosynthesis, across all three tuber tissues at 12-weeks, indicating DMN induced stimulation of stress responses (Dobry & Campbell, 2025). However, in the meristems, OPCL1, ACX, and AIM1, which are involved in downstream of the JA biosynthesis pathway, exhibited decreased transcript abundance. Additionally, transcripts of JAR1, TIFY10A, and MYC2 exhibited decreased abundance across all tissues and most sampling time points. Collectively, these results indicate that DMN treatment was associated with decreased abundance of transcripts involved in jasmonate biosynthesis (OPCL1, ACX, AIM1) and signaling (JAR1, TIFY10A, MYC2), corroborating previous findings (Dogramaci, Sarkar, et al., 2024). The content of JA-isoleucine, the main active form of JA, was also lower in primary meristem tissues at 12-weeks under single and multiple DMN applications, while single DMN treatment enhanced JA-isoleucine content in secondary meristems (Fig. 7B). Suttle et al. (2011) posited that JA-isoleucine content is significantly lower in dormant tubers than in sprouted tubers. Decreased transcript abundance of genes involved in JA biosynthesis and signaling, together with lower JA-isoleucine content in primary meristem tissues, indicates that DMN suppresses JA biosynthesis and signaling and subsequently influences tuber dormancy and sprouting.

Fig. 7.

Overview of jasmonate biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in jasmonate biosynthesis (alpha-linolenic acid metabolism pathway, ko00592) and signaling (Plant hormone signal transduction pathway, ko04075) in the primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and storage periods. (B) Heatmap representing the relative accumulation of jasmonate precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*). Metabolites with no detection are represented in black.

BRs are associated with plant stress responses and can function as hormones regulating various morphogenic processes, including seed germination (Chakraborty et al., 2025). In potatoes, BRs have been implicated in tuberization (Huang et al., 2021), wound healing (Han et al., 2022), and the induction of sprouting (Li et al., 2020; Liu et al., 2023). The application of BRs has been shown to enhance soluble sugar accumulation and protein phosphorylation while simultaneously decreasing the transcript abundance of ABA-related genes, thus reducing dormancy duration and promoting sprouting (Li et al., 2020). Conversely, overexpression of StSN2, a key gene associated with dormancy regulation in potatoes (Li et al., 2017; Liu et al., 2023), promotes increased transcript abundance of BIN2, a known BR signaling inhibitor, resulting in prolonged dormancy. In this research, two DEGs [CYTOCHROME P450 75B1 (CYP75B1), CYTOCHROME P450 90C1 (CYP90C1)] were identified in the BR biosynthesis pathway and six [BRASSINOSTEROID-INSENSITIVE 1 (BRI1), BR-SIGNALING KINASE 5 (BSK5), GSK3/SHAGGY-LIKE PROTEIN KINASE 7 (ASK7), BRASSINAZOLE-RESISTANT 1/2 (BZR1/2), XYLOGLUCAN ENDOTRANSGLUCOSYLASE/HYDROLASE 23 (XTH23), CYCLIN D3 (CYCD3)] in the BR signaling pathway (Fig. 8A). DMN treatment induced a decrease in transcript abundance of CYP75B1 and CYP90C1, which are involved in BR biosynthesis (Ohnishi et al., 2009), in meristem tissues, while transcript abundance of CYP90C1 increased in flesh tissues. Regarding the BR signaling pathway, transcript abundance of BSK5, ASK7, BZR1/2, and CYCD3 increased in the meristems, particularly in the primary meristem under DMN treatment (Fig. 8A). In contrast, transcript abundance of BRI1 and XTH23 increased in the tuber flesh and decreased in the meristems during most sampling times with DMN treatment (Fig. 8A). Notably, DMN-treated tubers exhibited a significant reduction in the concentrations of two BR-related compounds (dolichosterone and castasterone) in the primary meristem tissues (Fig. 8B). Prior studies have indicated that reduced BRI1 transcript abundance in transgenic potato varieties impairs BR signaling, adversely affecting tuber formation (Huang et al., 2021). Our findings also illustrate variations in BRI1 transcript abundance, which plays critical role in BR signaling cascade, depending on tissue type and storage duration. Interestingly, transcript abundance of BRI1 decreased in the primary meristem at 3- and 12-weeks under DMN treatment. Collectively, our transcriptome and hormone analyses suggest that multiple DMN treatments suppress JA- and BR-related responses in meristem tissues in order to cease the growth of the sprout at 12-weeks.

Fig. 8.

Overview of brassinosteroid biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in brassinosteroid biosynthesis (ko00905) and signaling (Plant hormone signal transduction pathway, ko04075) in the primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and storage periods. (B) Heatmap representing the relative accumulation of brassinosteroid precursors and related metabolites, quantified in the primary meristem, secondary meristem, and tuber flesh. Compounds with significant differential accumulation (p-value <0.05) are marked with an asterisk (*). Metabolites with no detection are represented in black.

3.7. DMN modulation of ethylene and SA pathways

Ethylene can act either as a promoter of sprouting or as a dormancy-prolonging hormone depending on its endogenous levels in tuber tissues and the concentration of external treatments (Dobry & Campbell, 2025; Dogramaci et al., 2026; Dogramaci, Dobry, et al., 2024; Foukaraki et al., 2016; Suttle, 2004). In our transcriptome analyses, ten DEGs [TYROSINE AMINOTRANSFERASE 7 (TAT7), AMINOCYCLOPROPANE CARBOXYLATE OXIDASE 1–2 (ACO1–2), ACIREDUCTONE DIOXYGENASE 4 (ARD4), ETHYLENE RESPONSE 2 (ETR2), CONSTITUTIVE TRIPLE RESPONSE1(CTR1), EIN3-BINDING F-BOX PROTEIN 1-LIKE (EBF2), ETHYLENE-INSENSITIVE 2 (EIN2), ETHYLENE-INSENSITIVE 3 (EIN3), ETHYLENE-RESPONSIVE TRANSCRIPTION FACTOR 1B (ERF1B), MITOGEN-ACTIVATED PROTEIN KINASE KINASE 5 (MKK5)] associated with ethylene biosynthesis and signaling were identified (Supplementary Fig. 5). DMN treatment resulted in increased transcript abundance of TAT7, ACO1–2, ETR2, and EBF2 across all three tuber tissues, particularly after 12-weeks of storage.

Previously, Dobry and Campbell (2025) also reported increased transcript abundance of several ethylene-related genes in DMN-treated tubers during ecodormancy, suggesting a potential interaction between DMN's mode of action and ethylene in dormancy regulation. Similarly, Dogramaci, Sarkar, et al. (2024) observed increased transcript abundance of ACC OXIDASE (ACCO) in meristem tissues exposed to MeJA and MeJA+DMN treatments, highlighting the potential crosstalk between ethylene and jasmonate pathways under DMN effect.

In our study, DMN also caused decreased transcript abundance of EIN3 in primary meristems. Increased transcript abundance of EIN3 and other members of EIN family have been correlated with dormancy break or germination in other plant systems (Corbineau, 2024; Dogramaci et al., 2010, Dogramaci, Horvath, Christoffers and Anderson, 2011, Dogramaci et al., 2013, Dogramaci et al., 2014), while suppression of sprout growth is associated with decreased EIN3 transcript abundance, which was reported in onion following exogenous ethylene application (Cools et al., 2011). Thus, the DMN-induced reduction in EIN2 and EIN3 transcript abundance in primary meristems may represent a key mechanism for sprout suppression in potato tubers. These results are also supported by Dogramaci et al. (2026), where continuous ethylene exposure in potato tubers increased transcript abundance of ACS6, ACO4, and several ERF family members (ERF1A, ERF5, ERF9), while subsequent removal of ethylene rapidly decreased transcript abundance of EIN3 and ERFs, highlighting the dynamic and reversible nature of ethylene signaling in tuber dormancy.

In addition to modulating ethylene-related pathways, DMN also affected SA signaling. Transcript abundance of NPR1-LIKE PROTEIN 3 (NPR3) and TGA TRANSCRIPTION FACTOR (TGA) decreased in meristem tissues at 9-and 12-weeks, whereas NPR3 transcript abundance increased in tuber flesh tissues under DMN treatment, particularly at 12-weeks (Supplementary Fig. 6). These results indicate that DMN suppresses SA signaling in a tissue-specific manner. Previously, Liu et al. (2015) reported that overexpression of genes involved in JA and SA biosynthesis and signaling and upregulation of stress response are associated with dormancy release in potato tubers.

Overall, suppression of JA, BR, and SA pathways suggests that DMN reduces stress-related hormonal responses in meristem tissues, which may contribute to the suppression of sprout growth during late storage stages. The combination of increased transcript abundance of several ethylene biosynthesis and response genes (TAT7, ACO, ETR2, EBF2) with decreased EIN2-EIN3 transcript abundance further highlights the complexity of DMN-mediated ethylene signaling. Collectively, these results reinforce the idea that DMN alters distinct nodes of the ethylene pathway to sustain dormancy while limiting pathways that promote sprout initiation.

3.8. TFs and signaling pathways under DMN treatment

TFs are pivotal regulators of gene networks that control plant growth, development, and stress responses (Marathe et al., 2024; Yuan, Kagale, & Ferrie, 2024). In our study, DMN treatment significantly affected transcript abundance of TFs across the tuber tissues examined, with the greatest number of TFs detected in the primary and secondary meristems compared to the tuber flesh. The spatial localization and tissue-specific regulation of TFs are known to play critical roles in shaping metabolic, molecular, and morphogenetic processes (Huang et al., 2018; Rohrmann et al., 2012). This tissue-dependent distribution of TFs is consistent with the central role of meristematic tissues in controlling dormancy and developmental reactivation, as also highlighted by Rohde et al. (2007).

Overall, 664 TFs were identified in the primary meristems, 756 in the secondary meristems, and 281 in the tuber flesh (Supplementary Table 2). The number of TFs with altered transcript abundance varied across sampling time points. In the primary meristem and tuber flesh tissues, the number of TFs (with increased or decreased transcript abundance) rose progressively with storage duration; whereas in the secondary meristems, TF regulation declined during the extended storage. These TFs were classified into 67 families, with the most abundant being APETALA2/ETHYLENE-RESPONSIVE FACTOR (AP2/ERF), C2H2 ZINC FINGER PROTEIN (C2H2), MYELOBLASTOSIS TRANSCRIPTION FACTOR (MYB), BASIC HELIX-LOOP-HELIX (bHLH), WRKY DNA-BINDING PROTEIN (WRKY), BASIC LEUCINE ZIPPER (bZIP), GRAS TF (GRAS), and C3H ZINC FINGER PROTEIN (C3H).

Heatmap analyses illustrated the expression patterns of the top 10 TFs with significant changes for each tissue type and treatment (Supplementary Fig. 7–9). Among these, AP2/ERF, bHLH, and MYB were consistently the most abundant families across tissue types examined. AP2/ERF members, associated with hormonal signaling and stress responses (Feng et al., 2020; Gao & Dubos, 2024; Wu et al., 2024; Yuan, Kagale, & Ferrie, 2024), displayed consistent clustering and decreased transcript abundance in primary meristem tissues under DMN treatment. Interestingly, transcript abundance of AP2/ERF increased in secondary meristems with single DMN treatment at 6-weeks of storage, but decreased abundance was observed with multiple DMN treatments, which also suppressed growth of secondary meristems. Results of hormonal and transcriptomic analyses also corroborated with these findings as levels of CKs and transcript abundance of genes related to zeatin biosynthesis and CK signal transduction that are involved in tuber sprouting decreased in secondary meristems with multiple DMN applications. Our transcriptomic analysis result suggests involvement of AP2/ERF members in dormancy maintenance or sprout suppression in potato tubers and its potential crosstalk with phytohormone such as CKs. AP2/ERFs are known to regulate meristem development (Rohde et al., 2007; Yordanov et al., 2014), and decreased transcript abundance of specific AP2/ERFs was shown to delay reactivation of cell division in poplar (Populus tremula × Populus alba) vegetative meristems, prolonging dormancy (Yordanov et al., 2014). In tuber flesh, however, AP2/ERFs exhibited a transient increase in transcript abundance during early sampling time point, suggesting an initial adaptive response to DMN that was later suppressed with prolonged storage duration.

MYB TFs, another large family with conserved regulatory functions, are central to morphogenesis, secondary metabolism, hormonal signaling, and stress responses (Ma et al., 2023; Wu et al., 2024). The temporal transcriptional changes of MYBs in DMN-treated tuber meristems, characterized by increased abundance until the 9-weeks in primary meristems followed by a decrease, suggesting their role in coordinating dormancy maintenance and the transition to sprouting. In secondary meristems, increased transcript abundance of MYB was observed with multiple DMN treatments at 12-weeks of storage, coinciding with suppression of secondary meristem growth under same treatment. At the same 12-weeks of storage time-point, decreased transcript abundance of GA3OX1 and reduced concentration of GA₄ were also observed in secondary meristems with multiple DMN treatments. Previous research highlighted the role of MYB TF in phytohormone crosstalk, specifically in GA and ABA signaling in plants under abiotic stresses (Yan et al., 2025). These results align with findings in other plant systems where decreased MYB transcript abundance accelerated dormancy release in tree peony (Paeonia suffruticosa) meristems (Yuan, Zeng, et al., 2024) and Arabidopsis seeds (Lee et al., 2015), in both cases associated with altered ABA and GA levels.

In our results, the bHLH TF family also emerged as an important regulator under DMN treatment. Increased transcript abundance of bHLHs in meristems during later sampling times may be linked to delayed dormancy release. In our results, transcript abundance of bHLH increased in primary meristems with DMN treatments at 12-weeks of storage, but with only triple DMN applications in secondary meristems (Supplementary Fig. 7, 8). Prior studies reported that SPATULA (SPT) and PHYTOCHROME-INTERACTING-FACTOR-LIKE 5 (PIL5), both bHLH members, repress GA3OX while activating ABA-related genes, thereby promoting dormancy in Arabidopsis seeds (Gabriele et al., 2010; Oh et al., 2009; Penfield et al., 2005). In our study, transcript abundance of GA3OX1 also decreased in meristem tissues with DMN treatment, indicating an inverse relationship between bHLH TF and GA biosynthesis and its impact on tuber dormancy maintenance. Furthermore, SPT was reported to fine-tune the ABA/GA balance, maintaining dormancy or promoting germination depending on ecotype (Vaistij et al., 2013). These findings reinforce the potential of bHLHs as modulators of dormancy regulation in potato meristems under DMN treatment.

Collectively, these results demonstrate that DMN exerts a considerable impact on TF networks in potato tubers. DMN, especially multiple DMN applications, may orchestrate a transcriptional environment that maintains dormancy and suppresses premature sprouting by modulating AP2/ERF, MYB, and bHLH families, key regulators of hormonal and developmental pathways.

4. Conclusion

This study elucidates the complex and dynamic nature of DMN-induced hormone regulation in potato tubers. The effects of DMN are tissue-specific and time-dependent, shaping hormonal signaling and dormancy progression. Sprout measurements revealed meristem-type specificity; primary meristems responded immediately to single DMN treatment, while secondary meristems showed transient stimulation followed by suppression under repeated applications. By integrating phenotyping, hormone and transcriptome profiling and functional enrichment analyses (GO and KEGG), we observed that DMN treatment suppresses GA, CK, JA, SA, and BR related hormonal responses in meristem tissues while stimulates ethylene related responses. Additionally, crucial pathways associated with dormancy progression were identified as distinct clustering of AP2/ERF, MYB, and bHLH transcription factor families were determined under DMN treatment. Collectively, these results provide valuable insights into the molecular regulatory framework by which DMN modulates transcriptional networks, hormonal crosstalk, and dormancy regulation in a tissue specific manner. Overall, this work advances the fundamental understanding of potato tuber dormancy and sprout growth and provides a framework for developing more effective dormancy management strategies.

The following are the supplementary data related to this article.

Fig. S1.

Effect of DMN treatment on sugar and starch content in potato tuber flesh during storage. Glucose (A), fructose (B), sucrose (C), and starch (D) content are shown for control and DMN-treated tubers at 3, 6, 9, and 12 weeks of sampling time points. Concentrations are expressed as mmol kg⁻¹ DW. Bars represent means ± SE (n = 3). Different uppercase letters indicate significant differences among sampling times, and different lowercase letters indicate significant differences between treatments within each sampling time (p < 0.05; Tukey test).

Fig. S2.

GO enrichment analysis of hormone-related DEGs in potato tuber tissues under DMN treatments. Bubble plots show enriched biological processes associated with genes of increased and decreased transcript abundance (p-value <0.01) in primary meristem (A), secondary meristem (B), and tuber flesh (C) across storage times.

Fig. S3.

Top 10 most representative KEGG pathways for genes with increased and decreased transcript abundance and the overlap plots of DEGs and KEGG pathways KEGG enrichment analysis of hormone-related DEGs in potato tuber tissues under DMN treatments. Bubble plots show enriched KEGG pathways associated with genes of increased and decreased transcript abundance (p-value <0.01) in primary meristem (A), secondary meristem (B), and tuber flesh (C) across sampling time points.

Fig. S4.

Distribution of hormone-related DEGs with increased and decreased transcript abundance in primary meristem (A), secondary meristem (B), and tuber flesh (C) under DMN treatments at different sampling time points.

Fig. S5.

Overview of ethylene biosynthesis and signaling pathways under DMN treatment. Heatmap showing the differential expression of genes involved in ethylene biosynthesis within the cysteine and methionine metabolism pathway (ko00270) and signaling (Plant hormone signal transduction pathway, ko04075) in the primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across the tissues and sampling time points.

Fig. S6.

Overview of salicylic acid biosynthesis and signaling pathways and associated differential expression and metabolite accumulation in potato tuber tissues under DMN treatment. (A) Heatmap showing the differential expression of genes involved in salicylic acid signaling (Plant hormone signal transduction pathway, ko04075) in primary meristem, secondary meristem, and tuber flesh after DMN treatment. Only differentially expressed genes (DEGs) with p-value <0.01 are shown. The heatmap highlights increased and decreased transcript abundance of genes across tissues and sampling time points.

Fig. S7.

Heatmap of transcription factor (TF) gene expression changes in the primary meristem.

Fig. S8.

Heatmap of transcription factor (TF) gene expression changes in the secondary meristem.

Fig. S9.

Heatmap of transcription factor (TF) gene expression changes in the tuber flesh tissue.

Supplementary material 1

List of identified DEG related to hormone biosynthesis and signaling (Plant hormone signal transduction pathway, ko04075) on KEGG pathway analysis. It includes genes involved in the biosynthesis pathways of auxins (Tryptophan metabolism pathway, ko00380), cytokinin (Zeatin biosynthesis pathway, ko00908), gibberellins (Diterpenoid biosynthesis pathway, ko00904), abscisic acid (Carotenoids biosynthesis pathway, ko00906), ethylene biosynthesis within the cysteine and methionine metabolism pathway (ko00270), brassinosteroids (brassinosteroid biosynthesis ko00905), jasmonic acid (alpha-linolenic acid metabolism pathway, ko00592), and salicylic acid (Plant hormone signal transduction pathway, ko04075). Transcripts are classified as increased if their log2 fold change (FC) ≥ 0.6 and decreased if ≤ -0.6, with an adjusted p-value < 0.01.

Supplementary material 2

List of putative transcription factors identified in potato tuber tissues under DMN treatment. Transcripts were considered increased with a log₂ fold change (FC) greater than 0.6 and decreased with an FC less than −0.6, with an adjusted p-value of <0.01.

Disclaimer statement

The findings and conclusions in this publication are those of the authors and should not be construed to represent any official USDA or U.S. Government determination or policy. Mention of trade names or commercial products in this publication is solely for the purpose of providing scientific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. USDA is an equal opportunity provider and employer.

Author contribution

M.D. conceived the research plan and designed the experiment. M.D. and F.F. applied treatments and collected and processed tissue samples for transcriptome and metabolome analyses. T.R.O. and E.F. extracted RNA for transcriptomics. N.T. and E.L. performed targeted metabolome assays. T.R.O. and N.W. performed transcriptomic data processing. T.R.O. and E.F. conducted statistical analyses and carried out all comparative analyses of the transcriptomic data, and prepared the tables and figures, deposited the transcriptome data to NCBI. T.R.O., E.F., D.S., N.W., N.T., E.L., F.F., and M.D. contributed to developing the manuscript draft, and all authors read and approved the final version of the manuscript.

CRediT authorship contribution statement

Munevver Dogramaci: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Tadeu dos Reis de Oliveira: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Investigation, Formal analysis, Data curation. Evandro Alexandre Fortini: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Investigation, Formal analysis, Data curation. Dipayan Sarkar: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Investigation, Formal analysis, Data curation. Nathan Wyatt: Writing – review & editing, Writing – original draft, Validation, Software, Formal analysis. Elizabeth Leonard: Writing – review & editing, Validation, Methodology, Data curation. Nishanth Tharayil: Writing – review & editing, Validation, Investigation, Data curation. Fernando Finger: Writing – review & editing, Validation, Resources, Methodology, Investigation, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.