Two soybean (Glycine max L.) cultivars with contrasting light response characteristics were used. ND12 is a light-adaptive genotype developed by the Nanchong Academy of Agricultural Sciences (Sichuan, China), characterized by strong photoprotective capacity and rapid response to dynamic light. GX7, developed by the Guangxi Academy of Agricultural Sciences (Guangxi, China), is a light-sensitive genotype that tends to exhibit delayed photosynthetic adjustment under fluctuating light (Gao et al., 2024). Seeds were germinated in a 1:3 (v/v) sterilized mixture of vermiculite and nutrient soil (Danish Pinstrup Substrate) pre-treated with carbendazim. Plants were grown under controlled conditions at Sichuan Agricultural University (Chengdu, China), with a 12 h light/12 h dark photoperiod, daytime temperature of 25 °C, nighttime 22 °C, and relative humidity of 60–70%. To establish environmentally relevant treatments, we first conducted field measurements of photosynthetically active radiation (PAR) using an LI-191R line quantum sensor (LI-COR, USA). Measurements were made under clear-sky conditions at 1-meter height within the soybean canopy from 09:00 to 18:00, at one-minute intervals across different maize–soybean relay intercropping row configurations. The baseline low-light intensity of 150 μmol m⁻² s⁻¹ used outside the midday period was chosen to reflect field conditions observed during morning and late afternoon hours in the canopy. Based on our PAR measurements ( Figure 1A ), these periods consistently showed values ranging between 120 and 180 μmol m⁻² s⁻¹, making 150 μmol m⁻² s⁻¹ an appropriate representative intensity for simulating ambient light levels under relay intercropping. These assessments revealed that midday high-light duration (PPFD >1200 µmol m⁻² s⁻¹) increased significantly in widened row configurations: 400.00% in 2:3 and 566.67% in 2:4 compared to the 2:2 configuration, while maintaining optimal light quality and quantity ( Figure 1A ).

These field-derived light dynamics informed a simulation of three distinct midday high-light durations in a growth chamber using a programmable LED system: T30 (30 min at 1200 µmol m⁻² s⁻¹, simulating 2:2), T150 (150 min, simulating 2:3), and T200 (200 min, simulating 2:4). Outside of the midday period, a baseline low-light condition of 150 µmol m⁻² s⁻¹ was maintained. Treatments were applied beginning at 12:00 h, and plants were allowed to recover under low light afterward. Each treatment was performed in two independent experimental batches. In each batch, the treatment was replicated three times, with ten soybean plants per replicate (30 plants total per treatment per batch).

The dynamic response of photosynthesis was assessed using a LI-6800 portable photosynthesis system (LI-COR, Inc., Lincoln, NE, USA) on intact soybean compound leaves. The photosynthetically active radiation (PAR) in the leaf chamber was set to 150 μmol·m⁻² s⁻¹·sm⁻² s⁻¹, and the CO₂ concentration was maintained at 400 μmol·mol m⁻¹. Soybean leaves were clamped into the chamber and acclimated for 20–30 minutes. Once photosynthetic parameters stabilized, the light intensity was increased to 1200 μmol·m⁻²·s⁻¹ for 30 minutes before being reduced back to 150 μmol·m⁻²·s⁻¹. The measurement results were fitted to calculate the response time of the net photosynthetic rate (Pn) to the light intensity during the transitions from low light to high light and from high light to low light. Prior to measurement, five morphologically uniform and healthy plants were selected from each treatment group within each batch. The selection was based on similar plant height, leaf area, and overall appearance, ensuring biological consistency. These five plants were used for all physiological and morphological measurements. The final results were obtained by averaging two independent batches of seedlings.

Calculating the time to 50% or 90% of the rate maximum for Pn, The induction kinetics for a given photosynthetic parameter can be expressed using an exponential model as follows (Mott and Woodrow, 2000):

where Fmin and Fmax represent the minimum and maximum of Pn during photosynthetic induction, and τ represents the time constants for that parameter.

After a 30-minute dark acclimation period, the kinetics of chlorophyll fluorescence induction were measured using a Plant Efficiency Analyzer (HandyPEA, Norfolk, UK). Fluorescence transients were recorded over a 1-second light pulse, and the OJIP curve was analyzed using the JIP-test, as described in previous studies. Measurements were taken from the middle portion of a healthy compound leaf, with 10 plants sampled per treatment. The following fluorescence parameters were recorded: maximum fluorescence intensity (Fm), minimum fluorescence intensity at 20 μs (Fo), and fluorescence intensities at 2 ms (J-step) and 30 ms (I-step). The variable fluorescence yield (Fv), defined as Fm − Fo, was used to calculate the maximum quantum yield of photosystem II (PSII) as Fv/Fm = 1 − Fo/Fm. Additionally, when maximum fluorescence (Fm) was reached, the following parameters were calculated:

ABS/CSm – energy absorbed per unit leaf cross-sectional area (approximated by Fm).

TR₀/CSm – energy captured per unit leaf cross-sectional area.

ET₀/CSm – energy transferred through electron transport per unit leaf cross-sectional area.

DI₀/CSm – energy dissipated as heat per unit leaf cross-sectional area

Photosystem II (PSII) parameters were measured using a Dual-Channel Modulated Chlorophyll Fluorometer (DUAL-PAM-100, WALZ, Germany). A saturating pulse of 20,000 μmol·m⁻²·s⁻¹ was applied for 300 milliseconds. After at least 20 minutes of dark acclimation, the redox state of the photosystem I (PSI) reaction center was analyzed by measuring the absorbance of P700 following a predefined protocol.

To determine the dark-reduction kinetics of P700, measurements were performed immediately after obtaining the dark-reduction curve, without re-acclimating the plants to darkness. The plastoquinone (PQ) pool size was then measured using a sequential light exposure protocol: the measurement protocol began with the activation of far-red light, followed by a 10-second waiting period to allow for initial stabilization. Next, a single-turnover saturating flash (ST) was applied for 10 seconds to induce a rapid electron transfer response. This was followed by the initiation of multiple-turnover saturating flashes (MT) for 30 seconds to further assess the plastoquinone (PQ) pool dynamics. Finally, the far-red light was deactivated to conclude the measurement. Following data collection, PSI and PSII parameters were calculated as Supplementary Table 1 to the Supplementary Information (SI) section.

Using the Dual PAM-100 (Walz) equipped with a P515/535 emitter-detector module, the ECS signal was monitored as the absorbance change at 515 nm. After the plants were fully dark-adapted, the measurement was initiated. When the curve leveled off, the actinic light was turned on. After 600 seconds of illumination with the actinic light, it was turned off. At this time, the P515 value decreased rapidly and then increased. After the curve stabilized, the measurement program was terminated. From the dark-light-dark curve changes of the 550–515 nm signal, two components of the transmembrane proton motive force (pmf) could be obtained: the proton gradient (ΔpH) and the membrane potential (ΔΨ). All ΔpH and ΔΨ levels were normalized according to the amplitude of the ECS and calculated using the DIRK analysis.

After the plants were fully dark-adapted, the rapid responses of the PSI and PSII parameters were measured at 25°C using the Dual-PAM 100 measurement system (Heinz Walz, Effeltrich, Germany). The leaves were illuminated with 150 μmol·m⁻²·s⁻¹ for 40 seconds, followed by illumination with 1200 μmol·m⁻²·s⁻¹ for 80 seconds, and then with 150 μmol·m⁻²·s⁻¹ for 80 seconds. During the alternation of low-intensity, high-intensity, and low-intensity light, the PSI and PSII parameters were measured (the formulas are consistent with those in section 2.5).

The results are presented as mean values obtained from five biological replicates per treatment per batch (n = 5), averaged across two independent experimental batches (total n = 10). Statistical significance was determined using LSD tests (p < 0.05) in SPSS 26.0. GraphPad Prism8.0 was used for graph plotting.

The impact of these light regimes on morphological traits was first evaluated ( Figures 2A, B ). Both cultivars showed significant reductions in plant height under prolonged midday high light, with ND12 decreasing by 47.52% and 81.46% under T200 compared to T150 and T30, respectively, and GX7 showing even greater reductions of 75.62% and 96.90%. Conversely, stem diameter increased in T200: ND12 showed increases of 6.16% and 14.48% over T150 and T30, while GX7 increased by 26.32% and 15.60%.

Through photosynthetic induction analysis spanning low-high-low light transitions, we found no significant differences in rapid photosynthetic response capacity (time to 50% Pnmax) between the two soybean cultivars across treatments. While ND12 showed comparable photosynthetic acclimation capacity (time to 90% Pnmax) among treatments, GX7 exhibited significant enhancements of 206.53% and 152.69% under T150 and T200, respectively, compared to T30 ( Figure 3 ; Table 1 ).

ND12 demonstrated 4.07% and 2.09% increases under T150 and T200 compared to T30, whereas GX7 showed a 10.05% enhancement exclusively under T150. Energy allocation patterns differed markedly between cultivars. ND12 displayed higher energy absorption per reaction center (ABS/Csm) and trapping efficiency (TRo/Csm) in T30 than in T150/T200 ( Figure 4 ), while GX7 maintained comparable values across treatments. Under the T150 and T200 treatments, the electron transfer energy per reaction center (ETo/Csm) of both ND12 and GX7 decreased significantly, while the dissipated heat energy per reaction center (DIo/Csm) increased. The longer the duration of intense midday light, the lower the reduction energy per reaction center (REo/Csm) of ND12 and GX7. There were no significant differences in this regard between the two varieties.

Both soybean cultivars exhibited common photoprotective responses to prolonged midday high-light exposure, including a significant increase in regulated energy dissipation [Y(NPQ)] and a concurrent decline in non-regulatory dissipation [Y(NO)] ( Figure 5 ). These adjustments occurred alongside largely stable quantum yields of PSII and PSI [Y(II), Y(I)] and photochemical quenching (qP) across treatments, indicating a conserved baseline capacity to regulate excess excitation energy under variable light durations(SI.fig1). Despite these similarities, cultivar-specific differences emerged, particularly under short-term high-light conditions (T30). GX7 showed significant reductions in Y(II), Y(I), and qP under T30 compared with T200 (30.25%, 38.33%, and 40.50% decreases, respectively; P < 0.05), whereas ND12 maintained higher overall photosystem efficiency, with combined PSII–PSI quantum yields 22.6% greater than GX7 under T30.

Electron transport dynamics further distinguished the two genotypes ( Supplementary Figure 1 ). ND12 displayed stable ETR(II) and ETR(I) across treatments, while GX7 exhibited pronounced increases in both parameters under T200 and T150, by 41.46% and 31.99%, respectively, relative to T30. CEF activity in GX7 increased substantially under prolonged high-light (by 60.47%), as did the CEF/Y(II) ratio (+40.20%), contrasting with the minimal variation observed in ND12. Analysis of PSI donor-side limitation [Y(ND)] revealed higher stress in GX7 under T30, with a 46.46% increase relative to ND12. Moreover, plastoquinone (PQ) pool capacity in GX7 increased by 21.96% from T30 to T200, whereas ND12 maintained consistent PQ redox status across treatments( Figure 6 ). The kinetics of P700 dark reversion accelerated progressively with longer high-light exposure in GX7; under T30, however, ND12 exhibited a 38.7% faster reversion rate, indicative of more efficient electron turnover.

Both cultivars responded to extended high-light exposure with elevated thylakoid energization ( Figure 6 ). Prolonged treatments increased ΔΨ and ΔpH by 34.8% and 27.3%, respectively, but ND12 consistently showed higher energization under T30, with ΔΨ and ΔpH values 52.4% and 47.1% greater than GX7 (P < 0.01). Together, these results demonstrate a core set of photoprotective responses shared between genotypes, while highlighting contrasting strategies in energy partitioning and electron transport regulation under varying durations of intense midday light.

The photosynthetic induction responses of both soybean cultivars to low-high-low light transitions revealed significant treatment-dependent patterns ( Figure 7 ). Under T200 and T150 treatments, ND12 and GX7 exhibited comparable kinetics in Y(I) and Y(II) fluctuations, characterized by initial increases during low light, subsequent decreases under high light, and recovery upon return to low light. However, under T30 treatment, GX7 showed significantly lower Y(I) and Y(II) increases than ND12 during the initial low-light phase (P<0.05). Notably, upon sudden light intensification, GX7 demonstrated dramatic reductions in Y(I) and Y(II) to 0.2 and 0.06, respectively, while ND12 maintained significantly higher values (P<0.01). PSI donor-side limitation analysis through Y(ND) revealed that all treatments showed gradual increases during low light, rapid elevation to 0.7 under high light, and sharp declines to approximately 0.1 upon light reduction, with GX7 under T30 exhibiting significantly higher Y(ND) than ND12 during high-light exposure.

Dark-adapted initial Y(NA) values (at 10s) followed the pattern T30 > T150 > T200, with GX7 under T30 maintaining significantly higher Y(NA) than ND12 throughout the light transitions. Following light intensification, Y(NA) peaked around 50s before gradually decreasing to 0.1, with subsequent slow declines during the final low-light phase. Prolonged midday illumination significantly enhanced GX7’s NPQ capacity while reducing its Y(NO), as evidenced by ND12’s consistently lower Y(NO) (non-regulated energy dissipation at PSII) and higher NPQ (regulated energy dissipation) across all light phases under T30 (P<0.05). Furthermore, midday light duration treatments improved GX7’s photochemical quenching capacity (QP and QL), though ND12 maintained superior QP and QL values than GX7 under T30 throughout the light transitions.

Prolonged exposure to intense midday light acts as a crucial environmental signal driving convergent photoprotective responses in soybean. Under treatments with extended midday irradiance duration (row spacings corresponding to T150–T200), the two cultivars examined, consistently exhibited enhanced regulated energy dissipation, as reflected by significant increases in DIo/CSm (+18.7–22.3%) ( Figure 4 ). This conserved adjustment highlights the central role of dynamic non-photochemical quenching (NPQ) in safeguarding the photosynthetic apparatus under prolonged high-light conditions, in line with previous observations across diverse species (Goss and Lepetit, 2015; Lazzarin et al., 2024). The robust induction of NPQ suggests that soybean cultivars share a common regulatory framework aimed at rapidly dissipating excess excitation energy and minimizing photoinhibition during midday peaks (Yang et al., 2015; Liu and Yang, 2024).

In addition to enhanced energy dissipation, both ND12 and GX7 displayed coordinated modifications in electron transport processes, indicative of regulation of the proton motive force (pmf) and cyclic electron flow (CEF) (Takahashi and Badger, 2011; Huang et al., 2019). These adjustments likely optimize ATP/NADPH balance and maintain thylakoid membrane stability under sustained light stress. Such photoprotective coordination reflects a generalized adaptive strategy that enhances resilience to temporally heterogeneous irradiance, particularly relevant for improving light capture and use efficiency in intercropping systems.

Despite these shared responses, cultivar-specific differences were observed in the modulation of electron transport. ND12 maintained a more stable ETR(II) and exhibited faster ΔpH formation under prolonged light, suggesting a constitutive activation of protective pathways advantageous for environments dominated by fluctuating light, such as shaded understories. In contrast, GX7 showed a marked increase in CEF with prolonged light exposure, indicative of an inducible photoprotective strategy that may confer benefits under more predictable light regimes. While these differences underline the existence of genotype-specific nuances, the overall findings emphasize a broadly conserved photoprotective framework among soybean cultivars in response to midday irradiance (Dang et al., 2023; Wu et al., 2023).

The patterns of proton motive force (pmf) regulation observed under prolonged midday light suggest a shared adaptive mechanism among soybean cultivars, centered on the dynamic adjustment of thylakoid redox states (Ho et al., 2022). Both ND12 and GX7 exhibited flexible modulation of ΔpH and ΔΨ components in response to increasing light duration, indicating that fine-tuning the balance between electric potential and proton gradient is a critical strategy for optimizing photosynthetic efficiency under fluctuating irradiance. Such regulation is essential for maintaining ATP production and preventing over-reduction of the electron transport chain, thereby stabilizing photosystem performance during intense midday light a feature highly relevant to relay intercropping systems where light environments are temporally heterogeneous (Wang et al., 2022b).

In addition, both cultivars demonstrated the capacity to adjust electron transport dynamics, including mechanisms that stabilize PSI redox poise and mitigate donor-side limitations under high light stress (Wang et al., 2022a; Wang et al., 2022b). These findings underscore that dynamic control of electron flow and pmf partitioning constitutes a fundamental photoprotective response across soybean genotypes exposed to extended midday irradiance (Yamori, 2016).

Despite this shared regulatory framework, subtle differences were observed between cultivars. ND12 showed a greater reliance on ΔΨ formation relative to ΔpH, along with faster P700 re-reduction kinetics in darkness, suggesting a strategy that emphasizes rapid ATP synthase activation and cyclic electron flow to enhance recovery under transient shading (Tanaka et al., 2019). In contrast, GX7 exhibited an expansion of the plastoquinone (PQ) pool under prolonged light exposure, coupled with delayed relaxation of donor-side PSI limitation, indicative of a potential “electron capacitor” mechanism to buffer excess reducing equivalents during high-light phases. These cultivar-specific nuances reflect alternative adaptations to midday light stress, although the overarching photoprotective strategies remain largely conserved.

These findings demonstrate that the duration of midday high-light exposure acts as a key driver modulating photoprotective plasticity in soybean, underscoring the critical role of temporal light dynamics beyond traditional intensity-focused frameworks (Ho et al., 2022). Both ND12 and GX7 exhibited dynamic adjustments in energy dissipation and electron transport processes, indicating a common capacity among soybean cultivars to optimize photoprotective responses according to the temporal characteristics of the light environment. Such flexibility has important implications for designing intercropping systems that account for not only light quantity but also its temporal distribution.

Although shared photoprotective strategies were evident, subtle cultivar-specific differences emerged. ND12 predominantly utilized rapid ΔpH-mediated NPQ induction, while GX7 favored CEF-dependent electron flow adjustment under prolonged illumination (SI.fig1), providing mechanistic insights into genotype-specific light adaptation. The observed expansion of the plastoquinone pool in GX7 under T200 conditions (21.96%) further suggests a potential role for redox-mediated retrograde signaling in fine-tuning duration-dependent photoprotection. Future studies incorporating time-resolved transcriptomic analyses and multi-environment field trials will be essential to unravel the molecular underpinnings of these responses and to validate their agronomic relevance under diverse cropping conditions.