Proteomic Analysis of Lotus-Derived NnAP2 Regulation of Soluble Sugar and Starch Content in Potato (Solanum tuberosum)

Abstract

The starch content of lotus (Nelumbo nucifera) rhizomes is a key determinant of their taste and overall quality. In our previous work, a candidate transcription factor, NnAP2, was identified and its coding-region single-nucleotide polymorphisms (SNPs) were significantly associated with rhizome enlargement and carbohydrate-related traits. Owing to limitations in stable genetic transformation systems in lotus, potato (Solanum tuberosum) was employed as a heterologous model to investigate the regulatory role of NnAP2 in starch and soluble sugar metabolism. Overexpression of two allelic variants of the NnAP2 transcription factor (CC and TT) in potato resulted in pronounced differences between CC- and TT-overexpressing lines (NnAP2^CC-OE and NnAP2^TT-OE) in microtuber carbohydrate composition and proteome dynamics, accompanied by divergence in transgene copy number and substantial variation in transgene expression levels among independent lines. Six months after planting transgenic lines NnAP2^CC-OE and NnAP2^TT-OE, the NnAP2^CC-OE micro-tubers exhibited significantly higher starch content and lower soluble sugar levels compared with NnAP2^TT-OE. To uncover the underlying molecular basis, profiling of proteoforms was conducted on leaves, stems and tubers of both genotypes through a label-free proteomic strategy. A total of 51,299 peptides matched to 7292 proteins. Principal component analysis demonstrated clear separation of treatment groups, indicating robust differential accumulation of proteoforms. In total, 1715 differentially expressed proteins (DEPs) were identified across tissues (fold change ≥ 1.5 or ≤0.67, p  <  0.05), of which 1516 (88.4%) were tissue-specific. GO and KEGG enrichment analyses revealed that in leaves, DEPs were enriched for amino sugar metabolism, protein transporter activity and cell-wall macromolecule modification; in stems, enrichment included response to biotic stimulus, defense response and transporter activity; in tubers, DEPs were strongly enriched for carbohydrate metabolic processes, starch and sucrose metabolism, the TCA cycle and nucleotide sugar biosynthesis. Key starch-biosynthetic enzymes (e.g., ADP-glucose pyrophosphorylase, UDP-glucose-4-epimerase) were up-regulated in NnAP2^CC-OE tubers relative to NnAP2^TT-OE, while soluble sugar synthesis pathways (e.g., trehalose-6-phosphate synthase) were down-regulated. Together, these data suggest that elevated NnAP2^CC expression in transgenic potato is associated with allele-dependent shifts in central carbon allocation between starch and soluble sugar pathways, as revealed by comparative analyses between NnAP2^CC-OE and NnAP2^TT-OE. This study provides a comprehensive proteoform framework for allelic variation in an AP2 transcription factor involved in source–sink carbon partitioning and tuber starch accumulation in potato.

1. Introduction

Lotus (Nelumbo nucifera Gaertn.), a perennial aquatic herb of the Nelumbonaceae family, is one of China’s most representative aquatic vegetables, with a cultivation history traceable to the Hemudu culture approximately 7000 years ago [1]. Beyond its dietary, medicinal, and ornamental value, lotus rhizome (the underground stem) constitutes the primary edible organ and holds substantial economic importance in Asian agriculture [2]. According to culinary properties and texture, lotus rhizomes are broadly classified into “crispy” and “floury” types, which differ markedly in soluble sugar and starch composition [3,4]. With the growing differentiation of consumer demand between fresh and processed markets [5], elucidating the molecular mechanisms underlying starch content regulation in lotus rhizomes has become increasingly urgent. Starch accumulation in underground storage organs is a highly coordinated process involving carbon assimilation, transport, and metabolic partitioning. Potato (Solanum tuberosum), a model plant for underground stem organs, has been extensively studied for starch biosynthesis mechanisms.

In potato tubers, starch synthesis involves the coordinated action of multiple enzymes, including ADP-glucose pyrophosphorylase (AGPase), granule-bound starch synthase (GBSS), and soluble starch synthase (SSS) [6]. Soluble sugars such as sucrose are dynamically regulated by sucrose phosphate synthase (SPS) and invertases (INV) [7,8]. Soluble sugars not only serve as carbon sources for starch biosynthesis but also act as signaling molecules that coordinate nutrient partitioning in storage organs. A growing body of evidence indicates that sucrose can positively regulate the expression of key starch biosynthetic genes, such as those encoding AGPase subunits, through sugar signaling pathways [9]. The central step in starch biosynthesis is catalyzed by AGPase, which converts glucose-1-phosphate and ATP into ADP-glucose and pyrophosphate [10]. Subsequently, ADP-glucose acts as a glycosyl donor for starch synthases [11], leading to the formation of starch polymers. This complex process is modulated by multiple transcription factors (TFs); for example, overexpression of the bZIP-type TFs StbZIP63 and StbZIP78 activates several starch synthesis genes and enhances starch accumulation [12]. In lotus, the transcription factor NnNAC100 has been shown to promote amylopectin synthesis by enhancing the transcription of NnSBEII [13]. However, the molecular pathway of starch biosynthesis in lotus rhizomes remains largely unexplored.

The AP2/ERF transcription factor family constitutes one of the largest plant-specific transcription factor families and plays pivotal roles in regulating plant development, stress responses, and primary carbon metabolism. Accumulating evidence across diverse crop species demonstrates that AP2/ERF members function as key transcriptional regulators of soluble sugar homeostasis, starch biosynthesis, and source–sink carbon partitioning in storage organs. These transcription factors typically act by directly binding to GCC-box or DRE/CRT cis-elements in the promoters of metabolic genes, thereby coordinating carbohydrate flux in response to developmental cues and environmental signals [14,15].

In cereal crops, AP2/ERF transcription factors have been repeatedly implicated in starch biosynthesis and storage compound accumulation. In rice, Rice Starch Regulator 1 (RSR1), an AP2/ERF family member, negatively regulates the expression of multiple starch biosynthetic genes, including AGPase and GBSSI, thereby fine-tuning endosperm starch accumulation [16]. Similarly, in maize, ZmABI19, which encodes an AP2-domain protein, modulates the expression of key regulatory genes during seed development and grain filling, thereby affecting carotenoid, soluble sugar, and starch accumulation [17]. In barley, the AP2 transcription factor HvAP2-18 directly binds to the promoters of critical starch synthesis genes HvAGP-S1 and HvSBE1 to activate their transcription, while HvAP2-12 fine-tunes starch biosynthesis by repressing HvAP2-18 expression [18]. Beyond cereals, AP2/ERF transcription factors have also been shown to regulate carbohydrate metabolism in vegetative storage organs. In sweet potato, ACRE (AP2-domain-containing carbohydrate-responsive element-binding factor) transcription factors served as a pivotal link between sugar signaling and the synthesis of storage-related proteins and enzymes [19]. These studies collectively suggest that AP2/ERF transcription factors serve as central nodes integrating developmental signals, sugar signaling, and metabolic gene expression to control starch accumulation and storage organ development. Despite these advances, the role of AP2 transcription factors in coordinating soluble sugar and starch metabolism in lotus rhizomes remains largely unexplored.

Based on ecological adaptation, lotus can be divided into two ecotypes: temperate lotus and tropical lotus [20]. These ecotypes differ in flowering duration, growth rate, and rhizome morphology. Temperate lotus develops enlarged rhizomes suitable for consumption, whereas tropical lotus possesses slender rhizomes and exhibits a flowering period 2–3 months longer than that of temperate lotus [21]. The contrasting rhizome development patterns—particularly the efficient starch accumulation in temperate lotus—make it an ideal natural variant for investigating the molecular mechanisms of lotus rhizome quality formation. Whole-genome resequencing [22] combined with transcriptomic analyses [23] revealed that genes involved in starch biosynthetic pathways of temperate lotus exhibit lower evolutionary rates at the sequence level but more pronounced transcriptional divergence. For instance, an endo-1,4-β-mannanase 7 (MAN7) encoded gene shows low sequence variation yet displays a fourfold higher expression level in temperate lotus rhizomes compared with tropical lotus during rhizome development. Moreover, three GBSS candidate genes are significantly upregulated in early rhizome development of temperate lotus. Conversely, genes such as NADH dehydrogenase-like genes that are downregulated in the late rhizome development stage of temperate lotus harbor higher densities of genetic variations, potentially facilitating early dormancy and sustained starch accumulation [24]. These findings provide crucial clues for identifying genetic determinants underlying rhizome enlargement and efficient starch deposition in lotus.

Quantitative trait locus (QTL) mapping has yielded numerous DNA markers useful for marker-assisted selection (MAS) in crop breeding [25]. A segregating F₂ population derived from a cross between tropical and temperate lotus has been constructed, and a high-density linkage map was generated through genotyping-by-sequencing (GBS), which identified abundant single nucleotide polymorphism (SNP) markers [26,27,28]. Subsequent QTL analysis identified a SNP located within the NnAP2 coding region that is significantly associated with lotus rhizome texture (crispy vs. powdery) [27]. This SNP involves a cytosine-to-thymine (CC/TT) substitution, resulting in an amino acid change from proline (Pro) to leucine (Leu) [27]. The two homozygous allelic variants are hereafter referred to as NnAP2^CC and NnAP2^TT, corresponding to the CC and TT SNP genotypes, respectively. Genotyping of natural lotus populations further revealed that the homozygous CC genotype exhibited significantly higher rhizome circumference, expansion index, and soluble sugar content than TT and heterozygous CT genotypes. Based on these findings, we hypothesize that NnAP2 plays a regulatory role in lotus rhizome enlargement and storage compound accumulation.

Due to the technical limitations associated with stable genetic transformation in lotus, functional validation of NnAP2 was conducted using potato as a heterologous model system for underground storage organ development. To explore the regulatory role of NnAP2 and its allelic variations, we generated transgenic potato (Solanum tuberosum) lines overexpressing the two lotus alleles, and the corresponding overexpression lines, as NnAP2^CC-OE and NnAP2^TT-OE, respectively.

Label-free quantitative proteomic analyses were performed on leaves, stems, and tubers of these transgenic lines to investigate how distinct NnAP2 alleles influence carbon allocation and sugar–starch metabolism in potato. This study situates NnAP2 within a broader cross-species framework of AP2/ERF transcription factors involved in carbohydrate metabolism. Proteoform-level analyses in potato provide functional evidence supporting the regulatory role of a lotus-derived AP2 transcription factor in storage organ development.

2. Results

2.1. The Starch Content in Microtubers of NnAP2^CC-OE Transgenic Plants Was Significantly Higher than That of NnAP2^TT-OE

After six months of cultivation, microtubers were harvested from transgenic potato plants overexpressing the two allelic variants of NnAP2. Microtubers from NnAP2^TT-OE plants exhibited significantly higher soluble sugar contents compared with those from NnAP2^CC-OE plants, whereas starch contents showed the opposite trend between the two allelic overexpression lines (Figure 1). These contrasting carbohydrate profiles were consistently observed across multiple independent transgenic lines.

Genomic DNA–based quantitative PCR analysis showed that NnAP2^TT-OE lines predominantly contained single or low-copy transgene insertions, whereas NnAP2^CC-OE lines harbored one to two copies of the transgene, indicating independent integration events among lines (Table S3). In addition, qRT-PCR analysis demonstrated that NnAP2 transcript abundance was consistently higher in NnAP2^CC-OE lines than in NnAP2^TT-OE lines (Figure S1).

2.2. Overview of Proteoforms in Different Transgenic Plants

To compare proteomic differences between NnAP2^CC-OE and NnAP2^TT-OE transgenic lines with respect to carbohydrate metabolism, leaves, stems, and tubers from transgenic potato plants were subjected to proteomic analysis. Mass spectrometry identified a total of 51,299 peptides corresponding to 7292 proteins. Quality control analysis showed that the majority of identified peptides were 7–25 amino acids in length (Figure 2A), and the precursor ion mass error distribution was highly concentrated (Figure S2), indicating that the mass spectrometry data were of high quality. The molecular weight distribution of the identified proteins showed that most proteins ranged from 10 to 60 kDa (Figure S3).

Label-free quantification was performed to analyze protein abundance. Principal component analysis (PCA) revealed that biological replicates within each treatment clustered closely together, while distinct separation was observed between groups, demonstrating good experimental reproducibility and significant intergroup differences (Figure 2B).

To validate the reliability of the proteomic results, ten genes encoding selected DEPs were randomly chosen for qRT-PCR analysis (Figure S4). The expression trends of mRNA levels were consistent with the proteomic data, confirming the reliability of the proteomic quantification.

2.3. Tissue-Specific Expression of DEPs in Transgenic Plants

To further elucidate key proteins influencing starch and carbohydrate metabolism in transgenic plants, comparative analyses of DEPs across tissues were conducted. DEPs identified between NnAP2^CC-OE and NnAP2^TT-OE were divided into three groups: leaf (Leaf_TT vs. Leaf_CC), stem (Stem_TT vs. Stem_CC), and tuber (Tuber_TT vs. Tuber_CC). Using Fold Change ≥ 1.5 or ≤0.67 and p ≤ 0.05 as thresholds, a total of 1715 DEPs were identified. Among these, 1516 (88.4%) were tissue-specific, comprising 720 up-regulated and 1220 down-regulated proteins.

Overall, 231 proteins were upregulated and 102 downregulated in leaves; 303 upregulated and 322 downregulated in stems; and 186 upregulated and 769 downregulated in tubers. The number of downregulated proteins in tubers was markedly higher than that of the upregulated ones.

2.4. GO and KEGG Enrichment Analyses of Tissue-Specific DEPs

To further explore the potential functions of DEPs in starch and carbohydrate metabolism, GO and KEGG enrichment analyses were performed separately for DEPs in the three tissues (Figure 3 and Figure 4).

In leaves, DEPs were significantly enriched in carbohydrate metabolism–related molecular functions, including the aminoglycan metabolic (GO:0006022) and catabolic processes (GO:0006026), which are linked to sugar precursor supply for starch synthesis (Figure 3A). The enrichment of “protein transporter activity” (GO:0008565) suggested the presence of active transmembrane transport processes involved in the allocation of photosynthetic products. Several processes related to cell wall construction and modification (e.g., GO:0006040, GO:0044036, GO:0006032) were also significantly upregulated. KEGG analysis revealed enrichment in pathways such as “valine, leucine, and isoleucine biosynthesis” (ko00290) and “α-linolenic acid metabolism” (ko00591), suggesting a potential association with precursor and energy metabolism relevant to starch synthesis (Figure 3B). Notably, DEPs were also enriched and upregulated in “pyrimidine metabolism” (ko00240), “DNA replication” (ko03030), “mismatch repair” (ko03430), “nucleotide metabolism” (ko01232), and “nucleotide excision repair” (ko03420) (Figure 4A).

In stems, DEPs were enriched in biological processes and pathways related to starch biosynthesis and nutrient transport (Figure 3B and Figure S5A). GO terms such as “response to biotic stimulus” (GO:0009607) and “defense response” (GO:0006952) suggested that DEPs participated in adaptive responses influencing sugar transport (Figure 3C). KEGG analysis showed significant enrichment in the “motor protein” pathway (ko04814), involved in intracellular material transport, as well as in “starch and sucrose metabolism” (map00500) and “global metabolic pathways” (map01100), underscoring the central role of stems in carbon allocation and metabolism (Figure 4B).

Tuber DEPs were enriched in starch biosynthesis-related pathways (Figure 3D, Figure 4C and Figure S5B). GO analysis highlighted “single-organism metabolic process” (GO:0044710) and “carbohydrate metabolic process” (GO:0005975) as significantly enriched, suggesting that tubers play a key role in carbohydrate metabolism and starch accumulation. Enrichment of “amylase activity” (GO:0016160) indicated that DEPs are directly involved in starch metabolism. KEGG enrichment showed that DEPs were highly enriched in “biosynthesis of secondary metabolites” (ko01110) (p < 1.0 × 10⁻⁶, FDR < 0.001), suggesting coordinated changes in secondary metabolite pathways that are accompanied by altered carbon allocation patterns in tubers. The “global metabolic pathway” (ko01100) exhibited the highest enrichment, reflecting the overall reprogramming of central carbon metabolism by NnAP2. Furthermore, “citrate cycle (TCA cycle)” (ko00020) enrichment suggested enhanced ATP and NADPH supply for starch biosynthesis. Enrichment in “nucleotide sugar biosynthesis” (ko01250) indicated elevated intracellular levels of UDP-glucose and ADP-glucose, providing substrates for starch chain elongation. Additional enrichment in “arginine and proline metabolism” (ko00330), “phenylalanine metabolism” (ko00360), “tyrosine metabolism” (ko00350), and “zeatin biosynthesis” (ko00908) implied coordinated regulation of nitrogen-carbon partitioning and cytokinin signaling, jointly promoting tuber growth and starch accumulation.

Notably, proteins annotated to the “carbon metabolism” (ko01200) pathway showed overall lower abundance in NnAP2^TT-OE tubers relative to NnAP2^CC-OE tubers, suggesting differential proteomic allocation of central carbon metabolic components between the two genotypes.

2.5. Key KEGG Pathways Associated with Starch Accumulation

We further screened and analyzed the key KEGG pathways enriched for tissue-specific proteins and the expression profiles of DEPs within these pathways (

Table 1. Key pathways enriched for tissue-specific proteins.

Origin	Class	ID	Description
Leaf	KEGG	KO00480	Glutathione metabolism
	KEGG	KO00195	Photosynthesis
	KEGG	KO00051	Fructose and mannose metabolism
	KEGG	KO00710	Carbon fixation by Calvin cycle
	GO	GO:0044036	cell wall macromolecule metabolic process
	GO	GO:0008565	protein transporter activity
	GO	GO:0006032	chitin catabolic process
	GO	GO:0009605	response to external stimulus
Stem	KEGG	ko04145	Phagosome
	KEGG	ko04146	Peroxisome
	KEGG	ko04148	Efferocytosis
	KEGG	ko04136	Autophagy-other
	KEGG	ko04144	Endocytosis
	KEGG	ko00520	Amino sugar and nucleotide sugar metabolism
	GO	GO:0098771	inorganic ion homeostasis
	GO	GO:1901071	glucosamine-containing compound metabolic process
	GO	GO:1901072	glucosamine-containing compound catabolic process
	GO	GO:1901135	carbohydrate derivative metabolic process
	GO	GO:1901136	carbohydrate derivative catabolic process
	GO	GO:0050801	ion homeostasis
	GO	GO:0055065	metal ion homeostasis
	GO	GO:0055072	iron ion homeostasis
	GO	GO:0055076	transition metal ion homeostasis
	GO	GO:0055080	cation homeostasis
	GO	GO:0022804	active transmembrane transporter activity
	GO	GO:0030001	metal ion transport
Tuber	KEGG	KO01200	Carbon metabolism
	KEGG	KO00020	Citrate cycle (TCA cycle)
	KEGG	KO00908	Zeatin biosynthesis
	KEGG	KO00400	Phenylalanine, tyrosine and tryptophan biosynthesis
	GO	GO:0004553	hydrolase activity, hydrolyzing O-glycosyl compounds
	GO	GO:0016798	hydrolase activity, acting on glycosyl bonds
	GO	GO:0016831	carboxy-lyase activity
	GO	GO:0003855	3-dehydroquinate dehydratase activity
	GO	GO:0016160	amylase activity
	GO	GO:0005996	monosaccharide metabolic process
	GO	GO:0005975	carbohydrate metabolic process

and Tables S4–S6, Figure 5).

In leaves, 24 DEPs were enriched in photosynthesis-related pathways (Table S4), among which XP_006351518.1 (α-1,4-glucan-protein synthase) is one of the key enzymes in polysaccharide biosynthesis, forming a material basis for starch accumulation. In stems, 57 DEPs were mainly enriched in 12 KEGG pathways related to signal transduction and nutrient transport (Table S5). The enrichment analysis indicated that phagosome, peroxisome, efferocytosis, autophagy, and endocytosis pathways (ko04145, ko04146, ko04148, ko04136, ko04144) together constitute a vesicular transport system across membranes; amino sugar and nucleotide sugar metabolism (ko00520) provides activated sugars for cell wall synthesis and remodeling; inorganic and metal ion homeostasis (GO:0098771, GO:0050801, GO:0055065, etc.) maintains osmotic and charge balance; and active transmembrane transporter activity (GO:0022804) and metal ion transport (GO:0030001) directly mediate long-distance transport of sucrose and minerals. These functions may contribute to the nutritional transport role of stems in starch accumulation.

In tubers, 107 DEPs were enriched in pathways directly related to starch–sugar metabolism (

Table 2. Top 20 GO enrichment terms for DEPs across tissues—retained as per original structure.

Origin	ID	Class	Description
Leaf	GO:0008565	Molecular Function	protein transporter activity
	GO:0004568	Molecular Function	chitinase activity
	GO:0006022	Biological Process	aminoglycan metabolic process
	GO:0006026	Biological Process	aminoglycan catabolic process
	GO:0006030	Biological Process	chitin metabolic process
	GO:0006032	Biological Process	chitin catabolic process
	GO:0006040	Biological Process	amino sugar metabolic process
	GO:0016998	Biological Process	cell wall macromolecule catabolic process
	GO:0044036	Biological Process	cell wall macromolecule metabolic process
	GO:0046348	Biological Process	amino sugar catabolic process
	GO:1901071	Biological Process	glucosamine-containing compound metabolic process
	GO:1901072	Biological Process	glucosamine-containing compound catabolic process
	GO:0030904	Cellular Component	retromer complex
	GO:1901136	Biological Process	carbohydrate derivative catabolic process
	GO:0045735	Molecular Function	nutrient reservoir activity
	GO:0006606	Biological Process	protein import into nucleus
	GO:0009605	Biological Process	response to external stimulus
	GO:0034504	Biological Process	protein localization to nucleus
	GO:0044744	Biological Process	protein targeting to nucleus
	GO:0051170	Biological Process	nuclear import
Stem	GO:0009607	Biological Process	response to biotic stimulus
	GO:0006952	Biological Process	defense response
	GO:0003924	Molecular Function	GTPase activity
	GO:0045735	Molecular Function	nutrient reservoir activity
	GO:0004568	Molecular Function	chitinase activity
	GO:0042578	Molecular Function	phosphoric ester hydrolase activity
	GO:0016020	Cellular Component	membrane
	GO:0006022	Biological Process	aminoglycan metabolic process
	GO:0006026	Biological Process	aminoglycan catabolic process
	GO:0006030	Biological Process	chitin metabolic process
	GO:0006032	Biological Process	chitin catabolic process
	GO:0006040	Biological Process	amino sugar metabolic process
	GO:0016998	Biological Process	cell wall macromolecule catabolic process
	GO:0044036	Biological Process	cell wall macromolecule metabolic process
	GO:0046348	Biological Process	amino sugar catabolic process
	GO:1901071	Biological Process	glucosamine-containing compound metabolic process
	GO:1901072	Biological Process	glucosamine-containing compound catabolic process
	GO:0009073	Biological Process	aromatic amino acid family biosynthetic process
	GO:0016462	Molecular Function	pyrophosphatase activity
	GO:0016817	Molecular Function	hydrolase activity, acting on acid anhydrides
Tuber	GO:0008152	Biological Process	metabolic process
	GO:0044710	Biological Process	single-organism metabolic process
	GO:0035091	Molecular Function	phosphatidylinositol binding
	GO:0055114	Biological Process	oxidation-reduction process
	GO:0005975	Biological Process	carbohydrate metabolic process
	GO:0016160	Molecular Function	amylase activity
	GO:0004097	Molecular Function	catechol oxidase activity
	GO:0004867	Molecular Function	serine-type endopeptidase inhibitor activity
	GO:0048037	Molecular Function	cofactor binding
	GO:0030170	Molecular Function	pyridoxal phosphate binding
	GO:0004553	Molecular Function	hydrolase activity, hydrolyzing O-glycosyl compounds
	GO:0016798	Molecular Function	hydrolase activity, acting on glycosyl bonds
	GO:0016831	Molecular Function	carboxy-lyase activity
	GO:0009112	Biological Process	nucleobase metabolic process
	GO:0046112	Biological Process	nucleobase biosynthetic process
	GO:0003855	Molecular Function	3-dehydroquinate dehydratase activity
	GO:0004764	Molecular Function	shikimate 3-dehydrogenase (NADP+) activity
	GO:0019238	Molecular Function	cyclohydrolase activity
	GO:1901362	Biological Process	organic cyclic compound biosynthetic process
	GO:0016682	Molecular Function	oxidoreductase activity, acting on diphenols and related substances as donors, oxygen as acceptor

and Table S6). The relative upregulation of proteins involved in starch biosynthetic branches and the downregulation of proteins associated with starch degradation are consistent with proteomic profiles favoring starch-oriented carbon allocation in NnAP2^CC-OE tubers. Overall, DEPs in leaves were primarily enriched in photosynthesis-related KEGG pathways, those in stems in signal transduction and transport pathways, and those in tubers in starch and carbohydrate metabolism pathways.

2.6. Potential Protein Regulatory Network Associated with Enhanced Starch Accumulation in NnAP2^CC-OE Tubers

Based on the proteomic data, the starch and sucrose metabolism pathway (map00500) was analyzed in detail. Differentially expressed proteins between NnAP2^TT-OE and NnAP2^CC-OE were enriched in this pathway, indicating that proteomic alterations in NnAP2^CC-OE tubers are associated with coordinated changes across multiple metabolic nodes relevant to starch metabolism. Key carbohydrate metabolic pathways in tubers were extracted and analyzed using heatmaps of protein abundance (Figure 6, Table S7).

In the trehalose biosynthesis pathway, the abundance of trehalose-6-phosphate synthase decreased, reducing the flux toward trehalose, an important soluble sugar. Similarly, the expression of key enzymes involved in the synthesis of raffinose family oligosaccharides was downregulated. Conversely, enzymes responsible for the synthesis of ADP-glucose (ADPG) and UDP-glucose, the essential precursors for starch biosynthesis, were significantly upregulated. At the level of nucleotide sugar metabolism, the abundance of UDP-glucose-4-epimerase (NP_001275004.1) increased, enhancing the conversion of galactose to UDP-glucose.

In addition, a notable global trend was observed: the majority of enzymes in glycolysis and the tricarboxylic acid (TCA) cycle exhibited increased abundance. Moreover, in the zeatin biosynthesis pathway, several cytokinin glycosyltransferases (e.g., XP_015167987.1, XP_006362320.1, XP_015164939.1, XP_015167613.1) were markedly upregulated, with XP_015167987.1 and XP_015170294.1 showing extremely high expression differences (log₂FC > 10). In summary, NnAP2^CC-OE tubers exhibited coordinated proteomic differences in sucrose- and starch-centered metabolic pathways relative to NnAP2^TT-OE, accompanied by altered abundance of enzymes related to energy production and precursor supply.

3. Discussion

3.1. Relationship Between Allelic Variation and Expression Divergence of NnAP2 in Transgenic Potato

Variable transgene expression levels across independent transformants have been widely documented in plant systems. Such expression divergence is commonly observed in plant transformation experiments and is frequently attributed to position effects, copy number variation, and local chromatin environments at transgene insertion sites, rather than to intrinsic differences in promoter sequences [29,30].

In our previous association analysis in lotus, a coding-region SNP of NnAP2 was significantly correlated with rhizome enlargement and carbohydrate-related traits, suggesting a potential functional divergence between allelic variants at the genetic level [27]. In the present study, heterologous overexpression of the two NnAP2 alleles (CC and TT) in potato resulted in pronounced differences in microtuber starch and soluble sugar contents. Although crop plants with single-copy insertions can exhibit relatively uniform expression, many early observations have shown order-of-magnitude variation among events with different insertion contexts [29].

Published studies on potato tuber carbohydrate composition consistently report wide yet overlapping ranges of starch and soluble sugar contents across cultivars and growing conditions. Wu et al. (2020) systematically evaluated 67 potato cultivars grown in China and reported starch contents ranging from 56.0 to 75.5 g/100 g dry weight, highlighting substantial cultivar-dependent variation in tuber carbohydrate composition under standard cultivation conditions [31]. In particular, Wu et al. (2020) reported that the widely cultivated potato cultivar Longshu No. 7, grown in Shaanxi Province, China, exhibited a dry matter content of approximately 20 g per 100 g fresh weight, with a corresponding starch content of about 63 g per 100 g dry weight [31], equivalent to approximately 126 mg g⁻¹ fresh weight. Notably, the starch contents measured in microtubers of both NnAP2^CC-OE and NnAP2^TT-OE lines in the present study were lower than those reported for mature Longshu No. 7 tubers in the field-grown system. This difference is likely attributable to disparities in growth environment (field cultivation versus in vitro microtuber induction), developmental stage (mature tubers versus microtubers), and cultivation duration, rather than to anomalous starch accumulation in the transgenic lines.

In addition, Saar-Reismaa et al. (2020) showed that total sugar contents in potato tubers ranged from 10.3 to 47.1 mg g⁻¹ dry weight, with considerable variation among genotypes, even when grown under comparable conditions [32]. Bruno et al. (2016) summarized that typical potato tuber composition across cultivars includes starch contents of approximately 10–18% and total sugars of 1–7% on a fresh weight basis, serving as a widely accepted baseline for wild-type potato tubers [33]. Sampaio et al. (2021) analyzed 50 potato genotypes and found that total sugar contents varied widely among flesh-color groups, reaching up to 700–800 mg per 100 g fresh weight, underscoring strong genotype-driven variability in tuber sugar accumulation [34]. Sim et al. (2023) demonstrated that starch (12.13–18.15% fresh weight) and sucrose contents vary dynamically with both cultivar and cultivation period, indicating that tuber carbohydrate composition is influenced by developmental stage as well as genetic background [35]. Overall, starch contents typically range from approximately 10–18% on a fresh weight basis and 56–76% on a dry weight basis, while total sugar contents vary from ~1–7% fresh weight or 10–47 mg g⁻¹ dry weight. Notably, these values are strongly influenced by genetic background, developmental stage, and cultivation period. The starch and soluble sugar levels observed in both NnAP2^CC-OE and NnAP2^TT-OE lines fall within these reported ranges, indicating that neither transgenic phenotype represents an extreme or aberrant state. Importantly, despite residing within the typical physiological range of potato tubers, the consistent and opposite shifts in starch and soluble sugar accumulation between the two allelic overexpression lines support an allele-dependent regulatory effect of NnAP2 rather than simple background variation.

Genomic DNA–based qPCR analysis revealed that NnAP2^TT-OE lines predominantly harbored single or low-copy insertions, whereas NnAP2^CC-OE lines contained one to two copies of the transgene. Consistently, quantitative RT-PCR demonstrated that NnAP2 transcript abundance in NnAP2^CC-OE lines was approximately two orders of magnitude higher than that in NnAP2^TT-OE lines, and this expression pattern was reproducible across independent transgenic events. Transgene copy number has been reported as a determinant of expression level, with multiple copies sometimes correlating with higher expression but also increased risk of silencing [36].

Collectively, these results indicate that, within the heterologous potato system, phenotypic divergence between NnAP2^TT-OE and NnAP2^CC-OE plants is closely associated with differences in transgene expression levels. Multiple factors, including insertion site, local genomic environment, and copy number, must therefore be considered when interpreting expression and phenotypic variation in transgenic studies [37].

Nevertheless, the SNP-associated variation identified in lotus cannot be excluded as biologically relevant. It remains plausible that allelic differences in the NnAP2 coding sequence may influence transcriptional efficiency, mRNA stability, protein conformation, or transcriptional activation capacity, thereby interacting with expression dosage effects to modulate downstream carbohydrate metabolism. Under the current experimental framework, however, it is not possible to unequivocally attribute the observed phenotypic differences solely to the coding-region SNP between the two NnAP2 alleles. Disentangling expression-level effects from intrinsic allelic functional divergence will require future analyses using comparable expression backgrounds. Therefore, interpretations regarding allele-specific functional divergence should be made with caution within heterologous overexpression systems.

3.2. NnAP2-Mediated Reprogramming of Carbon Allocation and Starch Metabolism

Although wild-type plants were not included as a control in the current experimental setup, both overexpression lines were generated under identical transformation conditions, allowing the analysis to focus primarily on differences between NnAP2^CC-OE and NnAP2^CC-OE lines.

In this study, starch and soluble sugar contents were selected as primary phenotypic indicators to evaluate the effects of NnAP2, as these traits represent direct metabolic readouts of carbon allocation and storage. Although tuber development also involves changes in agronomic traits and ultrastructural features, the inclusion of these parameters was beyond the primary scope of the present study and would require dedicated experimental designs. This limitation is acknowledged and will be addressed in future work.

Our findings show that NnAP2^CC-OE lines exhibit markedly higher starch accumulation in potato tubers compared with NnAP2^TT-OE lines. These results are consistent with previous evidence showing that AP2/ERF transcription factors play pivotal roles in regulating starch biosynthetic genes such as AGPase, SS, GBSS, and SBE. For instance, Rice Starch Regulator 1 (RSR1), an AP2/EREBP transcription factor, directly represses starch biosynthetic genes in rice endosperm, thereby modulating starch content [16]. Similarly, NnAP2 may alter carbohydrate allocation by regulating key starch metabolic enzymes, thereby shaping tuber phenotypes. Earlier studies also support this interpretation. Tiessen et al. (2002) [38] showed that AGPase activity is subject to redox regulation, coupling sucrose availability to starch synthesis. Similarly, Ballicora et al. (2004) [39] identified AGPase as the primary regulatory point controlling starch accumulation in higher plants.

Based on these insights, we propose a conceptual model in which NnAP2 overexpression may influence starch metabolism at multiple regulatory levels, potentially including transcriptional regulation and post-translational modulation. Comparative proteomic analyses further indicate that NnAP2 reprograms central carbon allocation and reshapes starch biosynthesis in potato tubers. In NnAP2^CC-OE plants, the enrichment of “carbon metabolism” (ko01200) and the “tricarboxylic acid cycle” (ko00020) suggests enhanced energetic and biosynthetic capacity, supporting ADP-glucose formation and starch accumulation [40]. Additionally, the significant enrichment of “carbohydrate metabolic process” (GO:0005996; GO:0005975) highlights the differential regulation of sugar interconversion and nucleotide-sugar biosynthesis, both directly linked to starch production [41]. Collectively, these results suggest that NnAP2^CC-OE promotes starch accumulation by enhancing energy metabolism, sink strength, and substrate availability, while reducing starch degradation. In contrast, NnAP2^TT-OE appears to favor metabolic diversion and turnover, resulting in reduced starch deposition. Taken together, these findings indicate that NnAP2 coordinates tissue-specific source–sink transport efficiency, contributing to enhanced starch accumulation in NnAP2^CC-OE tubers.

3.3. Tissue-Specific Proteomic Changes and Carbon Partitioning

A striking observation was that more than 88% of differentially expressed proteins (DEPs) were tissue-specific, highlighting the distinct functional roles of leaves, stems, and tubers in carbon partitioning. Plants coordinate carbon–nitrogen balance, sugar signaling, redox status, and hormonal networks to regulate source–sink relationships between leaves and tubers [42,43]. Future strategies for high-yield potato breeding will likely center on optimizing source–sink cooperation by enhancing both the photosynthetic “source” and the storage “sink” capacity of tubers [44,45,46]. This conceptual framework aligns closely with the tissue-specific proteomic reprogramming observed in our study.

In leaves, several proteins—including ribulose-1,5-bisphosphate carboxylase/oxygenase activase (XP_006339014.1), dihydrolipoyl dehydrogenase (XP_006344750.1), dihydroxy-acid dehydratase (XP_006349441.1), tryptophan synthase (XP_006366042.1), aquaporin PIP1-1 (XP_006362987.1), and glucose-6-phosphate/phosphate translocator (GPT) (XP_006366041.1)—were linked to photosynthesis, amino acid biosynthesis, carbohydrate metabolism, and mitochondrial import. These enzymes are involved in sustaining metabolic flux in chloroplasts and mitochondria.

Previous studies support similar functional roles. In Ipomoea batatas, heterologous expression of a GPT1 gene in Arabidopsis increased starch content in leaves, seeds, and root tips while reducing soluble sugars in leaves, suggesting that enhanced GPT1 expression promotes starch accumulation via altered carbon partitioning [47]. Likewise, overexpression of a pea GPT (PsGPT) together with an adenylate transporter (AtNTT1) in potato tubers markedly increased tuber starch content and yield [48]. Although this manipulation targets sink tissues, it demonstrates that boosting Glc-6-P transporter capacity enhances carbon flux into starch biosynthesis, implying that GPT and related transporters co-limit carbon supply to starch-producing plastids. In non-photosynthetic tissues such as pea roots, maize endosperm, and potato tubers, GPT imports Glc-6-P into plastids, feeding the carbon skeletons required for starch biosynthesis or the oxidative pentose phosphate pathway [49]. This confirms the substrate specificity and biochemical potential of GPT proteins to channel carbon toward starch formation. Overexpression of StPIP1 aquaporins in potato improves leaf water status, maintains stomatal conductance and photosynthesis under stress, and increases non-structural carbohydrate levels. Enhanced hydraulic conductivity and photosynthetic stability maintain assimilate export to sinks, thereby supporting tuber biomass and carbohydrate content even under adverse conditions [50]. Together, these effects provide a plausible mechanistic link between elevated PIP1 abundance in NnAP2^CC-OE leaves, enhanced source performance, and increased carbon allocation to tuber starch accumulation.

In stems, enrichment of DEPs in “starch and sucrose metabolism” and “motor protein” pathways points to intensified transport activity, consistent with the stem’s pivotal role in carbon allocation between source and sink. Sucrose transporters such as StSUT1 are essential for phloem loading and for tuber development in several plant species, including potato [43,51,52]. Many stem-specific DEPs identified in this study may facilitate the efficient translocation of assimilates from aerial tissues to underground storage organs. For example, the tonoplast dicarboxylate transporter (tDT, XP_006352352.1) and ZIP family transporter (XP_006353722.1) are integral membrane proteins that can influence vacuolar or symplastic partitioning of metabolites and metal cofactors within tuber tissues. Functional characterization of SlTDT in tomato has shown its involvement in vacuolar storage and remobilization of organic acids, processes that may indirectly affect starch biosynthesis in sink organs [53]. Similarly, ZIP-family transporters such as ZIP3 and ZIP5 are known to participate in metal ion transport and distribution in vascular tissues [54]. The presence of zinc transporter 5-like (XP_006353722.1) in stems may thus potentially influence metal cofactor availability for enzymes or mark vascular transport capacity. In transgenic potatoes overexpressing sucrose synthase (SUSY), elevated Susy activity increased UDP-glucose, ADP-glucose, and total starch content by 55–85%, establishing a causal relationship between sucrose metabolism and starch deposition [55]. Comparable evidence from transcriptomic analyses of Castanea mollissima seed development also revealed co-expression of key starch biosynthesis and carbohydrate metabolism genes (e.g., SUSY, GPT, AGPase) with starch content [56]. Similarly, in our study, the upregulation of sucrose synthase (SUSY, XP_006345244.1) in NnAP2^CC-OE relative to NnAP2^TT-OE tubers likely strengthens sink capacity and flux into starch biosynthetic pathways.

3.4. Comparative Regulatory Mechanisms of NnAP2 and AP2/ERF Transcription Factors in Starch Metabolism

The AP2/ERF transcription factor family plays multiple roles in plant growth and development. Its members exhibit a high degree of functional diversification. These functions include organ development, stress responses, and the regulation of carbon metabolism [57,58,59]. Studies in cereals and model crops have shown that AP2/ERF proteins participate in starch biosynthetic regulatory networks by binding to cis-elements such as DRE and GCC motifs [60]. Representative examples include the rice starch regulator RSR1 and barley HvAP2-18. These studies demonstrate that AP2/ERF transcription factors can regulate the expression of starch biosynthetic genes and reshape metabolic fluxes to promote starch accumulation or degradation. RSR1 functions as a negative regulator of starch biosynthetic gene expression in rice [16], whereas barley HvAP2-18 directly activates the promoters of starch biosynthesis genes in developing endosperm [18]. Our proteomic results suggest that NnAP2 may regulate carbohydrate allocation by altering central carbon metabolic pathways. In contrast, direct regulation of starch biosynthetic genes or enzymes by NnAP2 was not examined in the present study. Nevertheless, previous studies provide valuable mechanistic frameworks for further investigation.

In addition, specific ERF-type transcription factors have been reported to directly bind the promoters of key starch metabolism genes, as shown for barley HvERF72 [61] and maize ZmEREB94 [62]. In potato, ERF group VII transcription factor StRAP2.3 has been shown to regulate soluble sugar levels (reducing sugars and sucrose) during cold storage, without significantly affecting the expression of starch-related enzymes [63].

In this study, the two allelic variants of NnAP2 exhibited clear allele-dependent functional differentiation, which was accompanied by differences in NnAP2 expression levels, suggesting a potential gene dosage–dependent effect on carbon partitioning. NnAP2^CC-OE lines favored starch accumulation, whereas NnAP2^TT-OE lines were associated with relatively higher soluble sugar levels. Notably, our proteomic analysis revealed coordinated changes in central carbon metabolism rather than alterations in individual starch biosynthetic enzymes. These changes involved glycolysis, the tricarboxylic acid cycle, and nucleotide sugar biosynthesis. Together, these results suggest that NnAP2 may indirectly regulate starch accumulation by reprogramming carbon flux and metabolic capacity. Whether NnAP2 functions as a direct transcriptional activator or repressor of starch biosynthetic genes will be a major focus of our future work.

4. Materials and Methods

4.1. Materials

4.1.1. Plant Materials

Potato (Solanum tuberosum cv. Longshu No. 7).

4.1.2. Reagents

Bradford Protein Quantification Kit, dithiothreitol (DTT), iodoacetamide (IAM), sodium dodecyl sulfate (SDS), urea, sequencing-grade trypsin, ammonium bicarbonate, tris(2-carboxyethyl) phosphine (TCEP), chloroacetamide (CAA), n-octane, LC-MS-grade ultrapure water, triethylammonium bicarbonate buffer (TEAB), LC-MS-grade acetonitrile, LC-MS-grade formic acid, acetone, ammonia, ProteoMiner low-abundance protein enrichment kit, and trifluoroacetic acid (TFA).

4.1.3. Instruments

EASY-nLC™ 1200 nano UHPLC (Thermo Fisher, LC140, Waltham, MA, USA), nanoElute nano UHPLC system (Bruker, Bremen, Germany), Q Exactive™ HF-X mass spectrometer (Thermo Fisher, Waltham, MA, USA), Orbitrap Exploris 480 mass spectrometer (Thermo Fisher, Waltham, MA, USA), timsTOF Pro2 mass spectrometer (Bruker, Bremen, Germany), L-3000 HPLC system for peptide fractionation, refrigerated centrifuge (Scilogex D3024R, Göttingen, Germany), thermomixer (Allsheng MSC-100, Hangzhou, China), freeze dryer (Labogene ScanSpeed 40, Labogene A/S, Roskilde, Denmark), electrophoresis apparatus and tank (Bio-Rad, Hercules, CA, USA), analytical balance (Sartorius BSA124S, Sartorius AG, Göttingen, Germany), vortex mixer (Guanghe HY-6B, Changzhou, China), microplate reader (Thermo Multiskan FC, Waltham, MA, USA), ice maker (Xueke, Ningbo, China), tissue homogenizer (Shanghai Jingxin, 24-well, Shanghai, China), and ultrasonic cell disruptor (Ningbo Xinzhi JY92-11N, Ningbo, China).

4.2. Methods

4.2.1. Potato Transformation and Cultivation

The two NnAP2 alleles with homozygous SNP variants (C/C and T/T), hereafter referred to as NnAP2^CC and NnAP2^TT, were cloned into the PBI121 vector. The overexpression constructs were introduced into Agrobacterium tumefaciens and subsequently transformed into potato tissues via Agrobacterium-mediated transformation, as described below.

• Vector construction

The full-length coding sequences (cDNAs) of two allelic variants of NnAP2 were amplified from lotus (Nelumbo nucifera) and designated as NnAP2^CC and NnAP2^TT, respectively. Both alleles were cloned into the binary vector pBI121 under the control of the Cauliflower mosaic virus (CaMV) 35S promoter. To ensure translational fusion consistency, the native stop codon of NnAP2 was removed during primer design, and transcription termination was provided by the nopaline synthase (NOS) terminator present in the vector backbone.

Sequence alignment confirmed that the two NnAP2 cDNA alleles shared high overall sequence identity but differed at several single-nucleotide polymorphism (SNP) sites within the coding region (Table S1). Among these, a focal SNP located at nucleotide position 1220 resulted in a nonsynonymous amino acid substitution at position 399, corresponding to Proline (Pro) in NnAP2^CC and Leucine (Leu) in NnAP2^TT. In addition to this focal SNP, two other nonsynonymous SNPs were present between the alleles, which were retained as part of the naturally occurring allelic sequences.

All constructs were verified by Sanger sequencing to confirm sequence integrity, correct orientation, and in-frame insertion. The resulting expression vectors, pBI121::35S::NnAP2^CC and pBI121::35S::NnAP2^TT, were introduced into Agrobacterium tumefaciens strain EHA105 for subsequent genetic transformation of potato (Solanum tuberosum cv. Longshu No. 7).

2.

Genetic transformation

A. tumefaciens containing the recombinant vector was cultured in 20 mL of LB medium supplemented with 15 mg/L hygromycin and 50 mg/L rifampicin at 28 °C with shaking at 240 rpm for 24 h. Then, 1 mL of culture was inoculated into 50 mL LB medium containing 15 mg/L hygromycin and incubated at 28 °C, 240 rpm until the OD600 reached 0.5. The culture was centrifuged (5000 rpm, 3 min), and the pellet was resuspended in 20 mL hormone-free MS liquid medium (3% sucrose).

Microtuber slices (1–2 mm thick, diameter ~0.5 cm, pre-cultured for 2 months) were immersed in the Agrobacterium suspension for 10 min, blotted on sterile filter paper, and co-cultured on MS medium containing 3% sucrose (w/v), 0.2 mg/L IAA, 0.2 mg/L GA₃, 0.5 mg/L 6-BA, and 2 mg/L ZT at 25 ± 1 °C in darkness for 48 h. The slices were then transferred to the shoot induction medium (P2) supplemented with 15 mg/L hygromycin and 400 mg/L cefotaxime.

Cultures were maintained under a 16 h light/8 h dark photoperiod (light intensity 2000 lx) at 23 ± 1 °C. After approximately one week, lateral buds were excised to prevent nutrient competition. New shoots emerging from the central region were identified as resistant transformants. After ~4 weeks, 1–2 cm resistant shoots were excised and transferred to MS rooting medium containing 15 mg/L hygromycin and 200 mg/L cefotaxime to obtain complete transgenic plants.

3.

Detection of Selectable Marker (Resistance) Gene

Genomic DNA was extracted from young leaves of putative transgenic and wild-type (WT) plants using the cetyltrimethylammonium bromide (CTAB) method. The quality and concentration of DNA were assessed using a NanoDrop spectrophotometer and agarose gel electrophoresis. The presence of the selectable marker gene was verified by polymerase chain reaction (PCR) analysis. PCR amplification was performed using gene-specific primers designed for the resistance gene (e.g., nptII for G418/kanamycin resistance). Each PCR was carried out in a 20 μL volume containing 50 ng genomic DNA, 10 μL 2× PCR Master Mix, 0.5 μM of each primer, and nuclease-free water. The PCR program consisted of an initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 55–60 °C for 30 s, and extension at 72 °C for 30–45 s, with a final extension at 72 °C for 7 min. PCR products were separated on a 1.0% agarose gel stained with ethidium bromide (or GelRed) and visualized under UV illumination. WT plants were used as negative controls, while plasmid DNA served as a positive control. The presence of the selectable marker gene (nptII, conferring G418 resistance) in regenerated plants was verified by PCR using gene-specific primers, consistent with previous studies employing nptII PCR for molecular confirmation of transgenic plants [64,65].

4.

Total RNA extraction and quantitative RT–PCR analyses

To evaluate the expression levels of NnAP2 in overexpression (OE) lines, in vitro–grown plants of the potato cultivar Longshu No. 7 cultured for three weeks were used. Whole plants were harvested and pooled for RNA extraction. Total RNA was isolated from whole-plant samples of Longshu No. 7 and NnAP2-OE plants following previously reported procedures [66,67]. Gene expression levels were normalized using the potato Actin gene (PGSC0003DMT400010174) as an internal reference [68,69]. Relative transcript abundance was calculated using the 2^−ΔCq method with multiple internal controls, according to the manufacturer’s instructions provided by Bio-Rad. Quantitative RT–PCR analysis was performed using three independent transgenic lines as biological replicates. For each biological replicate, three technical replicates were conducted for each gene.

5.

Determination of AP2 copy number by gDNA-qPCR

Genomic DNA was extracted from leaves of CK (WT, non-transgenic potato cv. Longshu No. 7), NnAP2^CC-OE and NnAP2^TT-OE potato plants using a CTAB-based method. Genomic DNA copy number estimation was performed by quantitative PCR using a validated single-copy reference gene (Elongation factor-1alpha, StEF1α) in potato [66,67]. StEF1α has been widely used as a stable reference gene in potato RT-qPCR analyses under diverse biological conditions (e.g., abiotic stress, biotic stress) due to its consistent expression and reliability as an internal control in Solanum tuberosum studies [70,71].

Standard curves were generated using serial dilutions of CK genomic DNA for StEF1α and NnAP2^TT-OE genomic DNA for NnAP2, respectively. Quantitative PCR was performed using gene-specific primers (Table S2), and Cq values were obtained for both StEF1α and NnAP2 in each sample.

The genome equivalent of each sample was calculated based on the StEF1α standard curve. The relative copy number of NnAP2 was calculated by dividing the NnAP2 quantity by the corresponding genome equivalent, assuming one copy of StEF1α per haploid genome [72,73]. Each reaction was performed with three technical replicates, and at least three biological replicates were analyzed.

The absolute quantity of each target gene was calculated according to the corresponding standard curve using the equation:

log₁₀ (Quantity) = (Cq − b)/a

where a represents the slope and b the intercept of the standard curve. The initial quantity of template DNA was then calculated as:

Quantity = 10 ^(Cq−b)/a

The copy number of NnAP2 in each transgenic line was estimated by normalizing its absolute quantity to that of StEF1α, assuming one copy of StEF1α per haploid potato genome:

Copy number of NnAP2 = Quantity_NnAP2/Quantity_StEF1α

All reactions were performed with three biological replicates, each containing three technical replicates. Mean Cq values were used for copy number calculation.

4.2.2. Protein Extraction and MS Analysis

Frozen leaf, stem, and tuber tissues (−80 °C) were ground in liquid nitrogen and transferred into pre-chilled centrifuge tubes. Total protein extraction was conducted according to previous studies [74,75,76,77,78]. Briefly, SDT lysis buffer (containing 100 mM NaCl) was added, and samples were sonicated for 5 min in an ice bath. The lysates were heated at 95 °C for 8–15 min, cooled on ice for 2 min, and centrifuged at 12,000× g for 15 min at 4 °C. The supernatant was incubated with IAM at room temperature in the dark for 1 h. Four volumes of pre-chilled acetone (−20 °C) were added to precipitate proteins for ≥2 h, followed by centrifugation (12,000× g, 15 min, 4 °C). The precipitate was washed once with −20 °C acetone, air-dried, and dissolved in DB buffer.

Protein concentration was determined using the Bradford assay. BSA standards (0–0.5 µg/µL) were prepared and measured at 595 nm. The standard curve was used to calculate protein concentrations. Twenty µg of each sample were separated by 12% SDS–PAGE (stacking gel: 80 V, 20 min; resolving gel: 120 V, 90 min), stained with Coomassie Brilliant Blue R-250, and destained until clear bands appeared.

Protein samples were adjusted to 100 µL in DB buffer (8 M urea, 100 mM TEAB, pH 8.5). Trypsin and 100 mM TEAB were added and incubated at 37 °C for 4 h, followed by addition of trypsin and CaCl₂ for overnight digestion. The pH was adjusted below 3 with formic acid, centrifuged (12,000× g, 5 min, room temperature), and the supernatant was desalted using C18 columns. Columns were washed three times (0.1% formic acid, 3% acetonitrile), eluted (0.1% formic acid, 70% acetonitrile), and lyophilized [79].

Mobile phase A: 2% acetonitrile, 98% water, pH 10 (adjusted with ammonia). Mobile phase B: 98% acetonitrile, 2% water, pH 10 (adjusted with ammonia). Lyophilized peptides were reconstituted in phase A and centrifuged (12,000× g, 10 min, room temperature). Fractionation was performed using an L-3000 HPLC system equipped with a Waters BEH C18 column (Milford, MA, USA, 4.6 × 250 mm, 5 µm) at 45 °C. Elution was performed following the gradient in

Table 3. LC elution gradient for peptide fractionation.

Time (min)	Flow Rate (mL/min)	Phase A (%) 1	Phase B (%) 2
0	1	97	3
10	1	95	5
30	1	80	20
48	1	60	40
50	1	50	50
53	1	30	70
54	1	0	100

¹ Mobile phase A: 0.1% formic acid in water. ² Mobile phase B: 0.1% formic acid in 80% acetonitrile.

. Fractions were collected every minute, combined into 10 portions, lyophilized, and redissolved in 0.1% formic acid.

Lyophilized peptides were dissolved in 10 µL phase A, centrifuged (14,000× g, 20 min, 4 °C), and 1 µg was injected for LC–MS/MS analysis using an EASY-nLC™ 1200 nano UHPLC system. The precolumn (4.5 cm × 75 µm, 3 µm) and analytical column (15 cm × 150 µm, 1.9 µm) were maintained at 55 °C. The elution gradient is shown in

Table 4. LC elution gradient for EASY-nLC™ 1200.

Time (min)	Flow (nL/min)	Phase A (%)	Phase B (%)
0	600	94	6
2	600	90	10
45	600	70	30
48	600	65	35
50	600	50	50
51	600	0	100
60.5	600	95	5
61.5	600	95	5
62	600	5	95
67	600	5	95
70	600	95	5

.

4.2.3. Database Search

MS data were acquired on a Q Exactive™ HF-X mass spectrometer equipped with a Nanospray Flex™ (ESI) ion source. Parameters: spray voltage = 2.1 kV, capillary temperature = 320 °C. Data were collected in data-dependent acquisition (DDA) mode with a full MS scan range of m/z 350–1500, resolution = 60,000 (at m/z 200), maximum AGC = 3 × 10⁶, maximum injection time = 20 ms. The top 40 precursor ions were selected for higher-energy collisional dissociation (HCD, NCE = 27%), MS/MS resolution = 15,000, AGC = 1 × 10⁵, injection time = 45 ms, intensity threshold = 2.2 × 10⁴, and dynamic exclusion = 20 s. Raw files (.raw) were generated for analysis.

4.2.4. Proteomic Data Analysis

Spectra were searched against the Solanum tuberosum protein database (1583799-Solanum_tuberosum fasta, 37,960 sequences) using Proteome Discoverer 2.5.

Parameters: precursor mass tolerance = 10 ppm, fragment mass tolerance = 0.02 Da; fixed modification = Cys carbamidomethylation; variable modifications = Met oxidation and N-terminal Met loss; up to two missed cleavages allowed.

PSMs with ≥99% confidence were retained. Only proteins containing at least one unique peptide were considered valid. FDR filtering was applied at ≤1% at both peptide and protein levels.

Protein quantification data were analyzed using Student’s t-test. Proteins with significant expression differences between treatment and control (p < 0.05 and FC > 1.5 or FC < 0.67) were defined as differentially expressed proteins (DEPs).

GO and IPR functional annotations (Pfam, PRINTS, ProDom, SMART, ProSite, and PANTHER databases) were performed using InterProScan. COG and KEGG analyses were conducted to assign functional categories and pathways [80]. DEPs were further visualized via volcano plots, hierarchical clustering heatmaps, and GO/IPR/KEGG enrichment analyses [81]. Protein–protein interaction networks were predicted using STRING (http://string.embl.de/, accessed on 15 January 2025) [82].

5. Conclusions

In summary, the overexpression of the NnAP2^CC allele in potato significantly enhances tuber starch content by remodelling carbon partitioning at the proteome level. Compared with NnAP2^TT, the NnAP2^CC allele exhibited allele-dependent effects that were correlated with differences in NnAP2 expression levels, suggesting a gene dosage–dependent influence on carbon partitioning. Proteomic analyses indicate that NnAP2^CC-OE plants display enhanced metabolic capacity and sink strength, favoring assimilate flux toward starch accumulation while maintaining lower soluble sugar levels. Together, these findings reveal an underexplored allelic and dosage-related regulatory mechanism by which AP2-type transcription factors modulate carbohydrate allocation. This study provides proteomic evidence and candidate targets for improving starch traits in potato and other tuber crops through NnAP2 editing.