Stage-Specific Light Intensity Optimization for Yield and Energy Efficiency in Plant Factory Potato Pre-Basic Seed Production

Abstract

This study investigated the effects of light intensity regulation on yield and energy efficiency during potato pre-basic seed propagation in plant factories. Using virus-free ‘Favorita’ potato seedlings as experimental material, gradient light intensities (200, 300, and 400 μmol·m²·s⁻¹) were applied at four developmental stages: the seedling stage (SS), tuber formation stage (TFS), tuber growth stage (TGS), and harvest stage (HS), to explore the physiological mechanisms of stage-specific light intensity regulation and energy utilization efficiency. The results revealed that: (1) The per-plant tuber yield of the high yield group reached 72.91 g (T59 treatment), representing a 25% increase compared to the medium yield group and a 168% increase compared to the low yield group. Additionally, the high yield group exhibited superior leaf area, photosynthetic rate, and accumulation of sucrose and starch. (2) The impact of light intensity on tuber development exhibited stage specificity: low light intensity (200 μmol·m⁻²·s⁻¹) during TFS promoted early tuber initiation, while a high light intensity (400 μmol·m⁻²·s⁻¹) enhanced tuber formation efficiency. Increasing the light intensity during TGS facilitated the accumulation of sucrose and starch in tubers. (3) Energy use efficiency (EUE) increased significantly with yield, with the high yield group reaching 3.2 g MJ⁻¹, representing 52% and 88% improvements over the medium yield (2.1 g MJ⁻¹) and low yield (1.7 g MJ⁻¹) groups, respectively. A “stage-specific precision light supplementation” strategy was proposed, involving moderate light reduction (200 μmol·m⁻²·s⁻¹) during TFS and light enhancement (300 μmol·m⁻²·s⁻¹) during TGS to coordinate source-sink relationships and optimize carbohydrate metabolism. This study provides a theoretical basis for efficient potato production in plant factories.

1. Introduction

The potato (Solanum tuberosum L.), owing to its advantages in water use efficiency and per-unit yield, has emerged as a crucial crop for enhancing global food production [1]. Although China ranks first in terms of potato cultivation area, its per-unit yield lags behind countries like the Netherlands due to the low utilization rate of high-quality seed tubers (known as “pre-elite seeds”). Plant factories, leveraging their controllable environmental conditions and year-round production capabilities, offer effective solutions to the challenges in pre-elite seed propagation—such as continuous cropping obstacles, seasonal constraints, and viral infections—thereby improving propagation efficiency and increasing the utilization rate of these seeds [1].

The key to large-scale production of pre-elite seeds in plant factories lies in establishing appropriate light strategies. There has been relatively clear research on photoperiod and light quality regulation strategies for pre-elite seed propagation in plant factories. The photoperiod regulates tuber formation by modulating the expression of SP5G and SP6A. Under short-day conditions, SP5G is suppressed and SP6A is activated to promote tuber formation, while the opposite occurs under long-day conditions [2,3]. At the initiation stage, the expression levels of tuberization-related signals (StBEL5) in leaves under red-blue (RB) light conditions were significantly elevated [4]. However, research on light intensity for potato pre-elite seed propagation in plant factories is currently scarce. Chen et al. [1] explored the impact of light intensity on pre-elite seed propagation under the same daily radiation accumulation. Nevertheless, this study could not accurately reveal the relationship between light intensity and the development and yield of pre-elite seeds because photoperiod was also a variable factor.

Light intensity significantly influences crop growth and development. In plant factories, low light intensity reduces plant leaf area and stem elongation, while high light intensity can easily induce leaf photooxidation [5,6]. Under low light intensity conditions, the content of indole-3-acetic acid (IAA) [7], the activity of phenylalanine ammonia-lyase (PAL) [8], and the photosynthetic rate [9] in plants are all suppressed. Light serves as the fundamental energy source for photosynthesis—the core process governing crop growth and yield—while simultaneously regulating the structural organization and functional efficiency of the photosynthetic apparatus [10]. GO enrichment and KEGG pathway enrichment analyses indicate that genes related to cellular components, biological processes, and molecular functions in rice plants under low light exhibit varying degrees of upregulation or downregulation, with significant downregulation of genes involved in leaf carbon metabolism, plant hormone signal transduction, and photosynthetic carbon fixation [11]. Light intensity markedly regulates leaf stomatal conductance, affecting gas exchange between leaf stomata and the atmosphere [12], which is crucial for the synthesis of leaf photosynthates. Increasing light intensity is beneficial for enhancing crop biomass [13] and secondary metabolite content [14]. Under low light conditions, P_N (net photosynthetic rate) is primarily restricted by the supply of photochemically derived ATP and NADPH via the electron transport rate (ETR) [15]. As the light intensity increases, the dominant limitation shifts from ETR-dependent processes to RuBP carboxylation/regeneration capacity [16]. A large-scale meta-analysis (including 760 plant species) revealed that under high light intensity, the xanthophyll cycle is enhanced in plants, promoting the accumulation of violaxanthin, antheraxanthin, and zeaxanthin. The accumulation of these photoprotective pigments may enhance plant adaptability to high light environments through non-photochemical quenching mechanisms [17]. The synthesis and accumulation of photosynthates such as soluble sugars are highly dependent on light intensity [18]. However, some studies have shown that a high light intensity reduces the soluble sugar and free amino acid contents in lettuce leaves [19]. When light intensity exceeds 500 μmol·m⁻²·s⁻¹, the P_{N max} of lettuce decreases by more than 20% during the late growth stage [20] and the levels of soluble proteins, free amino acids, and DPPH radical scavenging rate in ryegrass plants significantly decrease [21].

Although research has been conducted on the impact of light intensity on crop growth and development in plant factories, covering crops such as lettuce, tomatoes, and cotton, different species and cultivated varieties respond differently to the same light intensity [22]. Given the shift in growth priorities across potato developmental stages, we hypothesize that: (1) critical tuber development phases (TFS, TGS) exhibit differential responses to light intensity; (2) leveraging these phase-specific light intensity responses will enable the development of a lighting strategy that optimizes the tuber yield–energy consumption trade-off. In this study, we investigated the effects of different light intensities under the same photoperiod on the growth, tuber development, yield formation, and energy consumption of potato plants in a plant factory. Our aim is to establish a light intensity regulation method that can promote the large-scale production of pre-elite seeds in plant factories. Therefore, it is essential to clarify the effects of light intensity on the growth and development, physiological metabolism, and yield formation of potato pre-elite seeds. This will help elucidate the mechanisms by which light intensity regulates the formation of pre-elite seeds, facilitate the establishment of precise light intensity setting protocols, optimize plant lighting strategies, and reduce electricity consumption.

2. Materials and Methods

2.1. Plant Materials and Growth Conditions

This study was conducted from March to July 2024 at the Plant Factory of the College of Agriculture, Nanjing Agricultural University, Nanjing, China. The initial phase (March–April 2024, totaling 30 days) involved cultivating potato tissue-cultured seedlings for experimental use. Subsequently, light intensity experiments were performed to evaluate the morphological, physiological, and photosynthetic parameters of potato plants (totaling 90 days).

This experiment used the potato variety ‘Favorita’ as the test material. Tissue-cultured seedlings were prepared in advance. From inoculation to transplantation, they were cultured for a total of 30 days. During the culture period, the photoperiod was set at 16 h and the light intensity was 100 μmol·m⁻²·s⁻¹ of white light. The tissue-cultured seedlings were transplanted into cultivation frames (with dimensions of 60 × 40 × 18 cm) in a plant factory. A full-factorial experiment was conducted, and different light intensities of 200 μmol·m⁻²·s⁻¹, 300 μmol·m⁻²·s⁻¹, and 400 μmol·m⁻²·s⁻¹ (representing low light intensities, medium light intensities, and high light intensities, respectively) were set for the seedling stage, tuber formation stage, tuber growth stage, and harvest stage, respectively. The specific light treatment schemes are shown in

Table 1. A full-factorial experiment on light intensity treatments.

	Light Intensity (μmol m−2 s−1)				Cluster		Light Intensity (μmol m−2 s−1)				Cluster		Light Intensity (μmol m−2 s−1)				Cluster
Treatment	SS	TFS	TGS	HS		Treatment	SS	TFS	TGS	HS		Treatment	SS	TFS	TGS	HS
T1	200	200	200	200	1	T28	300	200	200	200	2	T55	400	200	200	200	1
T2	200	200	200	300	1	T29	300	200	200	300	2	T56	400	200	200	300	1
T3	200	200	200	400	1	T30	300	200	200	400	2	T57	400	200	200	400	1
T4	200	200	300	200	1	T31	300	200	300	200	2	T58	400	200	300	200	1
T5	200	200	300	300	1	T32	300	200	300	300	2	T59	400	200	300	300	3
T6	200	200	300	400	1	T33	300	200	300	400	3	T60	400	200	300	400	2
T7	200	200	400	200	1	T34	300	200	400	200	2	T61	400	200	400	200	2
T8	200	200	400	300	2	T35	300	200	400	300	2	T62	400	200	400	300	2
T9	200	200	400	400	1	T36	300	200	400	400	2	T63	400	200	400	400	2
T10	200	300	200	200	2	T37	300	300	200	200	1	T64	400	300	200	200	2
T11	200	300	200	300	2	T38	300	300	200	300	1	T65	400	300	200	300	1
T12	200	300	200	400	1	T39	300	300	200	400	1	T66	400	300	200	400	1
T13	200	300	300	200	2	T40	300	300	300	200	1	T67	400	300	300	200	2
T14	200	300	300	300	2	T41	300	300	300	300	3	T68	400	300	300	300	2
T15	200	300	300	400	2	T42	300	300	300	400	2	T69	400	300	300	400	2
T16	200	300	400	200	1	T43	300	300	400	200	2	T70	400	300	400	200	2
T17	200	300	400	300	2	T44	300	300	400	300	1	T71	400	300	400	300	2
T18	200	300	400	400	2	T45	300	300	400	400	2	T72	400	300	400	400	2
T19	200	400	200	200	2	T46	300	400	200	200	2	T73	400	400	200	200	2
T20	200	400	200	300	2	T47	300	400	200	300	1	T74	400	400	200	300	1
T21	200	400	200	400	2	T48	300	400	200	400	3	T75	400	400	200	400	2
T22	200	400	300	200	2	T49	300	400	300	200	2	T76	400	400	300	200	2
T23	200	400	300	300	1	T50	300	400	300	300	3	T77	400	400	300	300	1
T24	200	400	300	400	2	T51	300	400	300	400	3	T78	400	400	300	400	2
T25	200	400	400	200	1	T52	300	400	400	200	3	T79	400	400	400	200	2
T26	200	400	400	300	1	T53	300	400	400	300	2	T80	400	400	400	300	2
T27	200	400	400	400	3	T54	300	400	400	400	2	T81	400	400	400	400	2

Note: Clusters 1, 2, and 3 represent the low yield group, medium yield group, and high yield group, respectively. Treatment T59 was the typical treatment group of the high yield group; treatment T81 was the typical treatment group of the medium yield group; treatment T1 was the typical treatment group of the low yield group.

. LED lamps (Hangzhou Zhuangcheng Lighting Technology Co., Ltd., Hangzhou, China) that emit a composite light consisting of red light (wavelength 620 nm), blue light (wavelength 460 nm), and white light (wavelength 400–700 nm) were used as the artificial light sources. The light intensities ratio of red to blue-white lights was 2:1. To ensure consistent lighting conditions, the distance between the plant canopy and the lamps was maintained at 30 cm for all treatments. Fifteen potato plants were cultivated under each light intensity treatment, with three replicates. During the experiment, the temperature under the light conditions was controlled at 22 ± 1 °C and the temperature under dark conditions was 16 ± 1 °C. The relative humidity was maintained at 70 ± 5%. The CO₂ level inside the plant factory was the same as that of the external atmosphere. Hoagland nutrient solution was used for irrigation to provide the necessary nutrients for plant growth and development, and the pH of the nutrient solution was controlled within the range of 5.5–5.8.

2.2. Growth Parameters and Energy Efficiency

After 90 days of light treatment, three plants were randomly selected from each treatment group. Standard methods were employed to measure plant height (a ruler was used to measure the distance from the base of the plant stem to the stem tip), stem diameter (a vernier caliper was employed to measure the diameter at the base of the plant stem), number of leaves (all compound leaves on the same petiole of a potato plant were counted as one leaf, and the total number of leaves on the plant was then tallied), and leaf area. Leaf area was measured by a disc method at the harvest stage. In briefly, thirty 1-cm² leaves were punched from the leaves with a puncher to measure dry weight as DW₁, and the dry weight of the total leaves were measured as DW₂. Leaf area was calculated according to the formulas:

$$\mathrm{Leaf} \mathrm{area}={\mathrm{DW}}_1 \times 30/{\mathrm{DW}}_2$$

(1)

The plants were then divided into four parts: tubers, leaves, stems, and roots. These parts were placed in an oven at 80 °C and dried to a constant weight. After weighing, the biomass of each plant organ was determined. The electricity consumption for the plant lighting under each treatment was recorded using a DDS863 model electricity meter (Shanghai SII Co., Ltd., Shanghai, China). The effective tuber ratio is the ratio of the number of tubers weighing more than 2 g to the total number of tubers. The energy use efficiency (EUE, g·MJ⁻¹) refers to the tuber biomass (dry weight of tubers, DW_tuber) of potato plants produced per megajoule of electrical energy (E, MJ) consumed by the LED plant growth lights [23]. The calculation formula is as follows:

$$\mathrm{EUE}={\mathrm{DW}}_{\mathrm{tuber}}/\mathrm{E}$$

(2)

2.3. Photosynthetic Parameters

During the seedling stage (light treatment 0–25 d), tuber formation stage (light treatment 26–40 d), tuber growth stage (light treatment 41–65 d), and harvest stage (light treatment 66–90 d) of potato plants, functional leaves from three randomly selected plants in each treatment group were immersed in a chlorophyll extraction solution (acetone: absolute ethanol = 1:1, v/v) to extract chlorophyll (Chl). The contents of Chl and carotenoids (Car) were measured using the following protocol: Fresh leaf samples (0.1 g, denoted as W) were weighed into test tubes and extracted with 10 mL (V) of an acetone: ethanol (1:1, v/v) solution. The tubes were stored in darkness overnight until the leaves became fully bleached. The extracts were then homogenized and their absorbance (OD) at wavelengths of 470 nm, 663 nm, and 645 nm was measured using a UV-vis spectrophotometer (JinPeng Inc., Shanghai, China), with an acetone:ethanol (1:1, v/v) mixture serving as the reference blank. Chlorophyll a (Chl a), chlorophyll b (Chl b), total chlorophyll (Chl a + b), and carotenoid (Car) contents were calculated according to the following equations [24]:

$$\mathrm{Chl} \mathrm{a} (\mathrm{mg} {\mathrm{g}}^{-1})=(12.72\times {\mathrm{OD}}_{663}-2.59\times {\mathrm{OD}}_{645})\times \mathrm{V}/1000 \mathrm{W}$$

(3)

$$\mathrm{Chl} \mathrm{b} (\mathrm{mg} {\mathrm{g}}^{-1})=(22.88\times {\mathrm{OD}}_{645}-4.67\times {\mathrm{OD}}_{663})\times \mathrm{V}/1000 \mathrm{W}$$

(4)

$$\mathrm{Car} (\mathrm{mg} {\mathrm{g}}^{-1})=(1000\times {\mathrm{OD}}_{470}-3.27\times \mathrm{Chl} \mathrm{a}-104\mathrm{Chl} \mathrm{b})\times \mathrm{V}/\mathrm{229,000} \mathrm{W}$$

(5)

$$\mathrm{Chl} (\mathrm{a}+\mathrm{b}) (\mathrm{mg} {\mathrm{g}}^{-1})=(\mathrm{Chl} \mathrm{a}+\mathrm{Chl} \mathrm{b})$$

(6)

During the same four growth stages, three potato plants with similar growth vigor were randomly selected from each treatment group. Between 9:00 a.m. and 11:30 a.m., a photosynthesis system LI-6400XT (LI-COR, Lincoln, NE, USA) equipped with a fluorescence leaf chamber was used to measure the photosynthetic parameters (net photosynthetic rate, P_N; intercellular CO₂ concentration, Ci), fluorescence parameters (maximal photochemistry efficiency, Fv/Fm; practical photochemical efficiency, PhiPSII; electron transport rate, ETR), and light response curves of the functional leaves of the plants. The measured data were fitted for light response curves using Ye’s methods [25], through which the dark respiration rate (R_d) was determined. When measuring the photosynthetic parameters, the light intensity setting of the instrument was consistent with that of the plant growth environment. The leaf temperature was set at 22 °C, the CO₂ concentration was set at 390 ± 10 μmol mol⁻¹, and the relative humidity was set at 60–70%. For measuring the light response curves, the light intensity gradients were set as 1200 μmol m⁻² s⁻¹, 1000 μmol m⁻² s⁻¹, 800 μmol m⁻² s⁻¹, 600 μmol m⁻² s⁻¹, 400 μmol m⁻² s⁻¹, 200 μmol m⁻² s⁻¹, 150 μmol m⁻² s⁻¹, 100 μmol m⁻² s⁻¹, 75 μmol m⁻² s⁻¹, 50 μmol m⁻² s⁻¹, 25 μmol m⁻² s⁻¹, and 0 μmol m⁻² s⁻¹. The light use efficiency (LUE), carboxylation efficiency (CE), the photorespiratory electron transport flux (J_o) and the non-cyclic electron flow for photosynthetic carbon fixation (J_c) were calculated using the following formulas:

$$\mathrm{LUE}={\mathrm{P}}_{\mathrm{N}}/\mathrm{Photosynthetic} \mathrm{Photon} \mathrm{Flux} \mathrm{Density}$$

(7)

$$\mathrm{CE}={\mathrm{P}}_{\mathrm{N}}/\mathrm{Ci}$$

(8)

$$\begin{array}{c}{\mathrm{J}}_{\mathrm{c}}=\left[\mathrm{ETR}+8({\mathrm{P}}_{\mathrm{N}}+{\mathrm{R}}_{\mathrm{d}})\right]/3\end{array}$$

(9)

$$\begin{array}{c}{\mathrm{J}}_{\mathrm{o}}=2\left[\mathrm{ETR}-4\left({\mathrm{P}}_{\mathrm{N}}+{\mathrm{R}}_{\mathrm{d}}\right)\right]/3\end{array}$$

(10)

2.4. Measurement of Photosynthates

During the seedling stage, tuber formation stage, tuber growth stage, and harvest stage of potato plants, three potato plants with similar growth vigor were randomly selected from each treatment group. The plants were dug out, washed, and then placed in an oven at 80 °C until a constant weight was achieved. The contents of photosynthates (sucrose and starch, mg g⁻¹) in each organ of the dried plants were determined using the method proposed by Chen et al. [1].

On the 30th day of treatment (during the tuber formation stage), at 10:30 a.m., three potato plants with similar growth vigor were selected from each treatment group. The potato plants were labeled with ¹³CO₂ according to the method of He et al. [26]. Samples were taken from the plants before the lights were turned on the next day. The samples were then dried in an oven at 80 °C until a constant weight was reached and ground into powder. Four milligrams of the sample to be tested were taken, and the isotope abundance of the sample was measured using a stable isotope ratio mass spectrometer (Isoprime100, Elementar, Oberursel, Germany). The tubers at the time of sampling were arranged in descending order of volume and divided into two equal parts, labeled as large tubers and small tubers, respectively, to investigate the distribution of ¹³C-photosynthates between tubers of different sizes. Proportions of large tubers and proportion of small tubers were calculated using the following formulas:

$$\mathrm{Proportion} \mathrm{of} \mathrm{large} \mathrm{tubers}=\mathrm{The} \mathrm{number} \mathrm{of} \mathrm{large} \mathrm{tubers}/\mathrm{the} \mathrm{total} \mathrm{number} \mathrm{of} \mathrm{tubers}$$

(11)

$$\mathrm{Proportion} \mathrm{of} \mathrm{small} \mathrm{tubers}=\mathrm{The} \mathrm{number} \mathrm{of} \mathrm{small} \mathrm{tubers}/\mathrm{the} \mathrm{total} \mathrm{number} \mathrm{of} \mathrm{tubers}$$

(12)

2.5. Carbon Metabolism Enzymes and Relative Expression Level of StSP6A

During the tuber formation stage, leaves and tubers were collected from potato plants under the T1, T81, and T59 treatments. The activities of SPS (sucrose phosphate synthase) and AGPase (ADP-glucose pyrophosphorylase) in the leaves and tubers, as well as the relative expression level of StSP6A in the leaves were determined following the method of He et al. [26]. The Rubisco content in the plant leaves was measured according to the method of Cai et al. [27]. Appendix A presents the determination methods for SPS, AGPase, StSP6A, and Rubisco.

2.6. Statistical Analyses

The impacts of light intensity were evaluated through analysis of variance, with the statistical analysis conducted using Statistical Product and Service Solutions (SPSS) software, version 20.0 (IBM, New York, NY, USA), followed by Tukey’s test at p < 0.05 level. The correlations between different parameters were determined using Pearson’s correlation analysis based on the measured data from the light treatments. Differences were considered to be statistically significant at p-values below 0.05. The graphs were generated using Microsoft Excel 2016 software (Microsoft, Redmond, WA, USA).

Based on the tuber yield data of potato plants under different treatments, cluster analysis was conducted using SPSS 20.0 software to classify the experimental treatments into three categories: high yield group, medium yield group, and low yield group. Additionally, typical treatment groups at each yield level were screened out by the software.

3. Results

3.1. Potato Among Different Yield Groups

3.1.1. Yield Traits

Significant differences were observed in potato yield components across yield groups. Tuber yield per plant in the high yield group 75 ± 5 g exceeded the medium yield group 60 ± 5 g by 25.0% and surpassed the low yield group 38 ± 5 g by 97.4%, while medium yield plants produced 57.9% more than low yield plants (Figure 1A). Although tuber number interquartile ranges (IQR) overlapped (low: 3.00–4.00; medium: 2.50–4.23; high: 3.17–4.75), the high yield group’s median tuber count was 13.5% higher than the low yield group (Figure 1B). Crucially, average tuber weight in the low yield group was >50% lower than both the medium and high yield groups (shared IQR: 12.98–23.78 g), while the medium and high groups showed a <2% difference (Figure 1C). The effective tuber rate reached 100% in the high yield group (Figure 1D). The coefficient of variation for tuber weight across yield levels (~42%) was 3.2 times as large as tuber number (~13%), confirming weight as the primary yield determinant.

3.1.2. Pearson Correlation Analysis Between Yield Traits and Light Intensity

Pearson correlation analysis (

Table 2. Pearson correlation analysis between the light intensity and tuber yield related traits of potato plants. * indicates significant correlations at p < 0.05 level.

		Tuber Weight of per Plant			Number of Tubers per Plant			Average Tuber Weight
		Low Yield Group	Medium Yield Group	High Yield Group	Low Yield Group	Medium Yield Group	High Yield Group	Low Yield Group	Medium Yield Group	High Yield Group
Number of tubers per plant		0.16	0.30	0.27
Average tuber weight		0.38 *	0.27	−0.12	−0.80 *	−0.88 *	−0.95 *
Light intensity	SS	0.33	0.12	0.29	−0.23	0.20	0.49	0.42	−0.11	−0.28
	TFS	0.53 *	−0.04	−0.23	0.27	−0.03	−0.79 *	0.04	0.14	0.68
	TGS	0.34	0.39 *	−0.54	0.49 *	0.20	0.20	−0.27	−0.19	−0.42
	HS	−0.20	0.10	0.02	0.25	0.15	−0.27	−0.30	−0.13	0.38

) revealed a significant negative correlation between tuber number per plant and average tuber weight. Yield-limiting factors differed across yield levels: In medium yield groups, both average tuber weight and tuber number per plant positively correlated with yield per plant. In high yield groups, tuber number per plant positively correlated with yield, while average tuber weight negatively correlated with yield per plant.

Light intensity impacts on potato yield traits exhibit dual specificity (developmental stage and yield level). In the low yield groups, tuber weight correlated with TFS light intensity, while tuber number correlated with TGS PPFD. The medium yield groups showed a tuber weight–TGS light intensity correlation. The high yield groups displayed a tuber number–TFS light intensity correlation. These correlations validate a dynamic light intensity strategy (200 μmol·m⁻²·s⁻¹ at TFS for tuber initiation and 300–400 μmol·m⁻²·s⁻¹ at TGS for sink allocation) to synergistically maximize yield.

3.1.3. Growth Parameters

Potato morphology showed yield group specificity: Low yield plants had greater height across stages (Figure 2A), while high yield plants were significantly shorter at the seedling stage. Stem diameter and leaf number were reduced at the seedling stage without intergroup differences (Figure 2B,C). Leaf area peaked at the tuber growth stage, with the leaf areas of high yield groups 50% larger than those of the low yield groups (Figure 2D). The medium and high yield groups maintained greater leaf areas from TFS to harvest. Minimal leaf number variation indicated that yield divergence was driven by organ size.

3.1.4. Photosynthetic Characteristics

Chl (a + b) content showed no significant differences among the low, medium, and high yield potato groups during the seedling, tuber growth, and harvest stages (Figure 3A). At the tuber formation stage, high yield plants exhibited lower Chl (a + b) than the low and medium yield groups. During the tuber growth stage, Chl (a + b) in high yield plants displayed a pronounced positive skewness (indicating higher-value concentration), while low yield plants showed a distinct negative skewness. The Chl a/b ratio was lower in high yield plants versus other groups at the seedling stage but higher during the tuber growth and harvest stages (Figure 3B). At the tuber formation stage, Chl (a + b) in the low yield plants demonstrated a marked negative skewness, with no significant intergroup differences.

Leaf P_N showed no significant differences between the low and medium yield groups from the seedling to harvest stages (Figure 3C). At the tuber formation stage, the high yield plants exhibited slightly lower P_N than the low and medium yield groups, but exceeded both during the tuber growth and harvest stages.

Car content peaked during the tuber growth stage and declined sharply at harvest across all yield groups (a significant decrease of more than 50%, Figure 3D). During the seedling and tuber formation stages, the low yield groups showed lower Car content than the medium and high yield groups. At the tuber growth stage, the high yield plants exhibited a pronounced positive skewness in Car content. During harvest, both the mean and median Car contents were higher in the high yield versus the low and medium yield groups.

From the seedling to harvest stages, leaf sucrose content in the low, medium, and high yield potato groups exhibited an initial rise, subsequent decline, and ultimate rebound (Figure 3E). During each growth period, leaf sucrose content increased with the population yield level. From tuber formation to harvest, the tuber sucrose content in high yield groups progressively decreased (Figure 3F). Across all periods, tuber sucrose content was higher in the low yield than in the medium and high yield groups, peaking during tuber formation in the low and medium yield groups. Leaf starch content mirrored the sucrose fluctuation pattern (rise- decline-rebound) in all groups, while tuber starch content showed progressive accumulation (Figure 3G). Leaf starch content increased with yield level during each period. Unlike the low and high yield groups, the medium yield plants exhibited lower tuber starch content during tuber growth versus the formation stage (Figure 3H).

3.2. Typical Treatment

3.2.1. Biomass and Yield Traits

Using SPSS 20.0 software for cluster analysis, the system identified the T59 treatment as the representative of the high yield group, the T81 treatment as the representative of the medium yield group, and the T1 treatment as the representative of the low yield group. At the seedling stage, plant biomass under T59 was significantly lower than under the T1 and T81 treatments (Figure 4A). From tuber formation to harvest, T1’s biomass remained lower than those of the other treatments. Crucially, the T59 treatment exhibited significantly reduced biomass allocation to non-tuber parts but elevated allocation to tubers during the tuber formation–harvest stages (Figure 4B). The T59 treatment achieved higher yields per plant, tuber number, and small tuber proportion. T1 showed the lowest average tuber weight. T81 had lower a effective tuber ratio but higher large tuber proportion (

Table 3. Response of yield-related traits in potato plants from typical treatment groups to dynamic light intensity. Different lowercase letters indicate significant differences among groups (p < 0.05, n = 3).

	Number of Tubers per Plant	Average Tuber Weight (g)	Tuber Weight per Plant (g)	Effective Tuber Ratio (%)	Proportion of Large Tubers (%)	Proportion of Small Tubers (%)
T1	2.83 b	9.74 b	27.31 c	95.68 a	35.34 ab	64.66 ab
T81	4.00 b	14.70 a	58.15 b	81.71 b	50.00 a	50.00 b
T59	6.50 a	11.31 ab	72.91 a	98.335 a	23.08 b	76.92 a

).

3.2.2. Typical Treatment Photosynthetic Characteristics

From the seedling stage to the harvest stage, the P_N of potato plant leaves under the T1 treatment was significantly lower than those found under other treatments (Figure 5A). During the tuber formation stage, the P_N of potato plant leaves under the T81 treatment was notably higher than those found under other treatments. During the tuber growth and harvest stages, the P_N of plants under the T59 treatment was significantly greater than those found under other treatments. From the seedling stage to the harvest stage, the CE of potato plant leaves under the T59 treatment remained at a high level, while the CE of potato plant leaves under the T1 treatment was lower than those found under other treatments (Figure 5B).

During the seedling, tuber growth, and harvest stages, the sucrose content in the leaves of potato plants increased significantly with the rise in yield levels among typical treatment groups, with the T81 treatment showing notably higher leaf sucrose content than the T59 treatment (Figure 5C). The sucrose content in the stems of potato plants under the T81 treatment was significantly lower than those found under the T1 and T59 treatments. From the tuber formation stage to the harvest stage, the sucrose content in the tubers of potato plants under the T1 treatment was significantly higher than those found under the T81 and T59 treatments.

Starch content exhibited organ- and stage-specific patterns: In leaves, T1 was significantly lower than T81 and T59 at the seedling and tuber formation stages but higher during the tuber growth stage. T59 was lower than T81 at the harvest stage (Figure 5D). In stems, T59 exceeded T1 and T81 at the seedling stage, T81 was lower than the others during the tuber growth stage, and T81 was significantly higher at the formation and harvest stages. In tubers, T81 surpassed the others at the formation and harvest stages, with no intergroup differences during the tuber growth stage.

3.2.3. ¹³C Metabolism of Potato Plants at the Tuber Formation Stage

Under the same experimental conditions, the δ¹³C value in potato plants under the T81 treatment was significantly lower than those found under other treatments (Figure 6A), indicating that the photosynthetic carbon fixation capacity of potato plants under the T81 treatment was inferior to those found under other treatments. The distribution proportion of δ¹³C in the leaves of potato plants under the T1 treatment was higher than those found under other treatments (Figure 6B–D), while the distribution of δ¹³C in the tubers was lower than those found under other treatments. Within the T1 treatment, the distribution proportion of δ¹³C in small tubers was higher than that found in large tubers, whereas the opposite phenomenon was observed under the T81 and T59 treatments (Figure 6B–D).

3.2.4. Carbon Metabolism Enzymes and Related Genes at the Tuber Formation Stage

Under the T81 treatment, the Rubisco content in the leaves of the potato plants (Figure 7A) was significantly higher than that under other treatments, while the SPS activity in the leaves (Figure 7B) and the relative expression level of StSP6A (Figure 7D) were lower than those found under other treatments. The AGPase activity in the leaves under the T59 treatment (Figure 7C) was significantly higher than that found under the T1 treatment. Under the T1 treatment, the SPS activity in the tubers of potato plants (Figure 7E) was significantly higher than that found under other treatments, but the AGPase activity (Figure 7F) was significantly lower than that found under other treatments.

3.2.5. Fluorescence Parameters at the Tuber Formation Stage and Tuber Growth Stage

During the tuber formation stage, the Fv/Fm value under the T81 treatment was significantly lower than that found under other treatments, while no significant difference was observed between the T1 and T59 treatments (

Table 4. The fluorescence parameters in typical treatment groups during tuber formation stage and tuber growth stage. Different lowercase letters indicate significant differences among groups (p < 0.05, n = 3).

		T1	T81	T59
TFS	Fv/Fm	0.86 a	0.75 b	0.85 a
	PhiPSII	0.76 a	0.77 a	0.77 a
	ETR	64.07 b	129.73 a	64.52 b
	Jc	31.62 c	54.01 a	34.65 b
	Jo	32.44 b	75.72 a	29.87 c
TGS	Fv/Fm	0.79 a	0.78 a	0.78 a
	PhiPSII	0.77 a	0.78 a	0.78 a
	ETR	64.9 c	130.62 a	97.99 b
	Jc	33.87 b	56.06 a	54.52 a
	Jo	31.03 c	74.56 a	43.47 b

). The ETR, J_o, and J_c values found under the T81 treatment were significantly higher than those under other treatments. The J_c value found under the T1 treatment was significantly lower than that found under the T59 treatment, whereas the J_o value exhibited an opposite trend. No significant differences in the PhiPSII values were observed among the leaves of potato plants under different treatments.

During the tuber growth stage, leaf Fv/Fm and PhiPSII showed no significant differences among treatments. However, ETR under the T81 treatment was significantly higher than others. Concurrently, J_c and J_o values under the T1 treatment were significantly lower than other treatments, with no significant difference in J_c between T59 and T81 (Table 4).

3.2.6. Light Response Curve at Tuber Formation Stage

Under the T1 and T59 treatments, the light response curves of the plants exhibited an inflection point at a light intensity of 800 μmol m⁻² s⁻¹, whereas under the T81 treatment, the P_N of the plant leaves still showed an increasing trend (Figure 8A). Within the light intensity range of 0–125 μmol m⁻² s⁻¹, the LUE of potato plant leaves was similar across all treatments. When the light intensity exceeded 125 μmol m⁻² s⁻¹, the LUE of potato plant leaves gradually decreased under all treatments. Under the T81 treatment, the LUE of potato plant leaves declined more significantly but followed a similar trend to that under the T1 treatment, while the decline under the T59 treatment was relatively gradual (Figure 8B).

3.2.7. Tuber Development at Tuber Formation Stage

Tuber formation primarily occurred during the tuber formation stage (25–40 days) across treatments, stabilizing thereafter. T1 initiated earliest but produced fewer tubers. T81 initiated latest yet exceeded T1 in final tuber count. Crucially, T59 exhibited prolonged continuous tuber formation (from initiation to peak) during this stage, yielding significantly more tubers than the others (Figure 9).

3.3. EUE

As the yield level increases, the median of EUE shows an upward trend, and the EUE value of the high yield group is higher than those of the other treatments (Figure 10A). The median EUE value of the high yield group reached 3.2 ± 0.2 g MJ⁻¹, representing a 52% increase compared to the medium yield group (2.1 ± 0.2 g MJ⁻¹) and an 88% increase over the low yield group (1.7 ± 0.2 g MJ⁻¹). The T59 treatment had the highest EUE, while the T1 treatment had the lowest, with significant differences observed among the treatments (Figure 10B). Notably, the medium yield group consistently showed transitional EUE values across treatments. These findings confirm that optimizing light capture and conversion is pivotal for improving energy efficiency in plant factory potato production.

4. Discussion

4.1. Stage-Specific Light Intensity Modulation Enhances Potato Yield by Coordinating Photosynthetic Efficiency and Source-Sink Dynamics

Crop light requirements vary developmentally [28]. Matching light conditions to stage-specific growth traits maximizes potato productivity and energy efficiency. High yield groups showed significantly greater tuber weight and number (Figure 1), demonstrating profound light modulation effects on yield [29,30]. Yield formation involves agronomic traits, photosynthesis, and light environment [31,32]. Larger leaf areas in the medium and high yield groups (Figure 2) enhanced photosynthetic capacity (Figure 3), boosting photoassimilates. Consistently taller plants in the low yield groups (Figure 1) suggested elevated gibberellin (GA) [33]. While GA induced StSUT4 expression to promote sucrose transport to tubers [34] (aligning with higher tuber sucrose in low yield, Figure 3), it activated vacuolar acid invertase, accelerating sucrose degradation while restricting tuber expansion [35]. This impaired expansion reduced starch accumulation (Figure 3) and yield [36]. Thus, the “high-sugar sink status” favored tuber initiation but disrupted starch metabolism, ultimately reducing average tuber weight (Table 2) and per-plant yield in the low yield groups (Figure 1).

Car and Chl a + b content followed similar trends across all yield groups from the seedling to harvest stages (Figure 3), indicating that photosynthetic pigment dynamics are primarily governed by developmental progression [37]. During the tuber growth and harvest stages, the high yield group exhibited elevated Chl a/b ratios (Figure 3), reflecting a higher proportion of PSII core complexes (Chl a-enriched) in light-harvesting chlorophyll-protein complexes (LHCP). This enhances light energy absorption and conversion [38], consistent with the higher P_N observed in high yield populations (Figure 3).

The high yield group demonstrated superior net photosynthetic rate, sucrose, and starch accumulation [39,40,41], indicating robust photosynthetic and assimilate production capacity. Medium yield groups showed intermediate parameter values, while low yield groups underperformed across metrics, suggesting optimized photosynthesis and carbohydrate metabolism are pivotal for yield enhancement. Pearson analysis revealed a significant negative correlation between tuber number and average tuber weight (Table 2), implying increased tuber numbers in high yield conditions may come at the cost of individual tuber size [42]. Notably, light intensity during tuber initiation negatively correlated with tuber number in high yield groups (Table 2), indicating moderate light (e.g., 200 μmol·m⁻²·s⁻¹) at this stage prevents premature sink competition to promote tuber formation. These findings provide direct evidence for a “stage-specific precision lighting” strategy, addressing previous confounding of light intensity and photoperiod effects [1].

4.2. Precision Regulation of Light Intensity on the Mechanism of Potato Tuber Formation

This study provides comprehensive insights into how light intensity affects potato plant growth, yield formation, and photosynthetic performance across developmental stages through analysis of typical yield groups. We elucidate key physiological mechanisms, particularly carbon metabolism, to advance theoretical frameworks for light-regulated potato pre-basic seed production in plant factories.

Pearson analysis confirms that light intensity significantly impacts tuber development and yield during the initiation and growth stages (Table 2) [43]. The T1 treatment showed reduced plant and tuber biomass allocation from growth to harvest (Figure 4), limiting the carbon supply for tuber expansion and reducing the tuber number, average weight, and yield per plant (Table 3) [44,45]. The higher small tuber ratio and lower average weight under T1 and T59 versus T81 support our hypothesis that an increased tuber count may reduce individual tuber size [46,47].

During tuber initiation, identical light intensity the in T1/T59 treatments (lower than T81) resulted in higher P_N for T81 (Figure 5). However, its lower δ¹³C (Figure 6) suggests photorespiration suppression under high light [1]. Elevated Rubisco in T81 (Figure 7) with reduced carboxylation efficiency, coupled with restricted stomatal conductance (GS, T1, T81, T59 were 0.063, 0.059, 0.070 mol m⁻² s⁻¹, respectively) may explain the P_N-δ¹³C paradox. High light increased the P_N, ETR, and electron fluxes (Figure 5, Table 4), but only 41.63% of electrons fixed carbon, while 58.36% fueled photorespiration (Table 4), reducing net assimilation (Figure 4) and aligning with reduced LUE under T81 (Figure 8).

Tuber formation timing, duration, and efficiency collectively determine yield [1]. Higher StSP6A in T1 (Figure 7) enabled earlier initiation (Figure 9), yet limited carbon allocation to sinks (7.47%, Figure 6) restricted sustained growth. Though T81 initiated later with a shorter duration, its higher efficiency (the tuber formation efficiencies were 0.2, 0.5, and 0.35 tubers per day, respectively) suggests that light intensity drives strategic divergence—low light (T1 and T59) favors early initiation, while high light (T81) enhances efficiency. Reduced sucrose in T81 tubers (Figure 5) implies impaired translocation, inhibiting the transport of floral integrators (CONSTANS/FLOWERING LOCUS T) [48] and tuber expansion. T59’s 50% light reduction (non-stress per Fv/Fm) boosted sucrose accumulation and LFY expression [48], accelerating tuber formation. Low sucrose inhibits initiation [49]; T81’s deficient photosynthate and sucrose translocation (Figure 5) delayed sink sucrose accumulation, postponing formation (Figure 9). T59 matched T1’s StSP6A levels (Figure 7) but excelled in carbon fixation and allocation (Figure 6), explaining its superior tuber count and duration [50]. Higher tuber AGPase activity in T59 (Figure 7) indicates enhanced starch metabolism, whereas T1’s high sucrose but low starch metabolism (Figure 7) favored early yet unsustainable formation, consistent with developmental patterns (Figure 9) [51].

In the growth stage, sucrose declined in all treatments, steeply in T1/T81 (Figure 5), indicating reduced photosynthetic capacity [52]. The T81 treatment’s high J_o/J_c ratio was 1.40 (T1 and T59 were 0.94 and 0.78, respectively) shows excessive photorespiration, impairing light use. These findings validate “stage-specific precision light supplementation” strategies.

4.3. Energy Efficiency Analysis

High energy consumption constitutes a critical bottleneck restricting plant factory development [53]. For potato pre-basic seed production, balancing the tuber yield–energy consumption trade-off is essential to enhance energy efficiency and foster sustainable industry growth [54]. We demonstrated that static lighting regimes exhibit inherent limitations: prolonged low intensity compromises yield potential despite reduced energy use, whereas sustained high intensity diminishes LUE (Figure 8) and overall EUE (Figure 10), while accelerating fixture degradation and elevating operational costs [55].

Leveraging phase-specific responses to light intensity, our T59 lighting protocol delivers a breakthrough solution—simultaneously achieving triple objectives: 167% higher tuber yield versus controls, 36–71% energy savings, and peak breeding efficiency (3.2 g MJ⁻¹ EUE). This strategy establishes a high yield, low-carbon paradigm for industrial-scale pre-basic seed production in plant factories.

5. Conclusions

This study underscores the critical role of stage-specific light intensity modulation in optimizing potato yield and EUE within plant factories. By analyzing yield groups and physiological responses across developmental stages, we demonstrated that dynamic light adjustment—particularly during the tuber initiation and growth phases—enhances photosynthetic efficiency, coordinates source-sink dynamics, and improves carbohydrate metabolism. High yield groups exhibited superior leaf area development, chlorophyll a/b ratios, and net photosynthetic rates, which collectively boosted sucrose and starch accumulation. However, excessive light intensity during tuber formation increased photorespiration, reduced carboxylation efficiency, and impaired LUE, highlighting the need for balanced light management. Notably, the T59 (the light intensity was set to 400, 200, 300, and 300 μmol m⁻²·s⁻¹ for the seedling stage, tuber formation stage, tuber growth stage, and harvest stage, respectively) light protocol emerged as optimal, achieving high yield while minimizing energy consumption by avoiding the pitfalls of sustained low or high light regimes.

Our findings validate a “stage-specific precision lighting” strategy, which provides a theoretical and practical framework for advancing sustainable, energy-efficient potato pre-basic seed production in controlled environments.