Zinc and Selenium Biofortification Modulates Photosynthetic Performance: A Screening of Four Brassica Microgreens

Abstract

Microgreens, having short growth cycles and efficient nutrient uptake, are ideal candidates for biofortification. This study investigated the effects of selenium (Se) and zinc (Zn) on photosynthetic performance in four hydroponically grown Brassica microgreens (broccoli, pak choi, kohlrabi, and kale), using direct and modulated chlorophyll a fluorescence and chlorophyll-to-carotenoid ratios (Chl/Car). The plants were treated with Na₂SeO₄ at 0 (control), 2, 5, and 10 mg/L or ZnSO₄ × 7H₂O at 0 (control), 5, 10, and 20 mg/L. The results showed species-specific responses with Se or Zn uptake. Selenium enhanced photosynthetic efficiency in a dose-dependent manner for most species (8–26% on average compared to controls). It increased the plant performance index (PI_tot), particularly in pak choi (+62%), by improving both primary photochemistry and inter-photosystem energy transfer. Kale and kohlrabi exhibited high PSII-PSI connectivity for efficient energy distribution, with increased cyclic electron flow around PSI and reduced Chl/Car up to 8.5%, while broccoli was the least responsive. Zinc induced variable responses, reducing PI_tot at lower doses (19–23% average decline), with partial recovery at 20 mg/L (9% average reduction). Broccoli exhibited higher susceptibility, with inhibited Q_A re-oxidation, low electron turnover due to donor-side restrictions, and increased pigment ratio (+3.6%). Kohlrabi and pak choi tolerated moderate Zn levels by redirecting electron flow, but higher Zn levels impaired PSII and PSI function. Kale showed the highest tolerance, maintaining stable photochemical parameters and total electron flow, with increased pigment ratio (+4.5%) indicating better acclimation. These results highlight the beneficial stimulant role of Se and the dual essential/toxic nature of Zn, thus emphasizing genotype and dose-specific optimizations for effective biofortification.

1. Introduction

Microgreens are young, tender, edible seedlings of vegetables and herbs, typically harvested 7–21 days after seeding at the cotyledon or first true leaf stage. They have attracted increasing attention due to their rich nutritional profiles, characterized by high concentrations of health-promoting phytochemicals, vitamins, and minerals compared to their mature counterparts [1,2,3]. These characteristics make microgreens a functional food with potential health benefits and disease prevention properties. Additionally, their short cultivation cycle, minimal spatial requirements, and suitability for controlled-environment agriculture make them an ideal crop for urban farming, hydroponic setups, and sustainable production systems [4,5].

Among various microgreen vegetables, those from the Brassicaceae family, including broccoli, cauliflower, radish, cabbage, and kohlrabi, are the most widely consumed vegetables worldwide, the most commercially important, and therefore the most extensively studied, owing to their rapid growth, ease of cultivation, and high content of antioxidant compounds, such as ascorbic acid, carotenoids, tocopherols, phenolic compounds, and glucosinolates [2,6,7]. Recent studies have demonstrated that Brassica microgreens exhibit significantly higher nutrient quality scores than their mature counterparts, especially in vitamins A and E, calcium, and manganese [8]. Their biochemical composition can be optimized through controlled nutrient ratios and is also strongly influenced by environmental factors such as solar radiation. Variations in light exposure have been shown to affect not only growth traits but also the levels of glucosinolates, phenolics, flavonoids, and antioxidant capacity, highlighting the importance of genotype selection and cultivation conditions in maximizing their functional and nutritional value [9].

Biofortification, a sustainable approach to combat micronutrient malnutrition (“hidden hunger”), enhances the vitamin and mineral content of crops during growth through agronomic practices, conventional breeding, or biotechnological methods [10,11]. Agronomic biofortification—via soil, foliar, or hydroponic nutrient application—is particularly effective in leafy vegetables and microgreens due to their rapid growth and efficient nutrient uptake [12,13].

Selenium (Se) and zinc (Zn) are among the key trace elements commonly targeted in biofortification strategies. While Zn is essential for both plant and human physiology, Se, though non-essential for plants, significantly benefits human health and is readily absorbed and metabolized by plants [14,15,16]. Selenium is a non-metal trace element in living organisms with a significant role in antioxidant defense, and it is a structural component of selenoproteins and antioxidant enzymes such as glutathione peroxidases [17,18]. Although plants do not require Se for their growth, its uptake can confer beneficial effects, such as increased stress tolerance, enhanced antioxidant defense, and delayed senescence [15,19,20]. Zinc, on the other hand, is essential for plant growth and development, acting as a cofactor for over 300 enzymes involved in protein synthesis, membrane integrity, and regulation of gene expression [14,21,22]. In humans, zinc deficiency is associated with impaired immune function, stunted growth, and increased susceptibility to infectious diseases, particularly in developing countries [23].

The biofortification of microgreens with Se and Zn offers a dual advantage by enhancing the dietary intake of these essential nutrients and potentially improving the physiological performance of the plants themselves. Studies have demonstrated that Se biofortification stimulates the production of secondary plant metabolites, such as vitamin C, polyphenols, anthocyanins, and flavonoids, while also increasing overall antioxidant capacity [24,25]. Zinc biofortification has been shown to improve the nutritional profile of radish (Raphanus sativus L.), pea (Pisum sativum L.), and sunflower (Helianthus annuus L.) microgreens cultivated in soilless systems. Treated plants exhibited increased Zn accumulation alongside elevated levels of beneficial phytochemicals, including total polyphenols, flavonoids, and antioxidant capacity [26]. This suggests that Se- and Zn-enriched foods may benefit human health not only by increasing intake of these essential microelements but also by boosting uptake of various other phytochemicals. Additionally, plant-based Se-biofortified foods have an advantage over synthetic supplements, as inorganic Se absorbed by plants is converted into organic forms with higher bioavailability and absorption efficiency [27,28]. Se and sulfur (S) share similar chemical properties and common metabolic pathways in plants, which allows Se to substitute for S in physiological processes, such as amino acid and protein synthesis. This substitution potential is especially relevant in Brassicaceae species, which are naturally rich in sulfur-containing compounds such as glucosinolates. Se toxicity could be alleviated by S supplementation through reducing non-specific Se incorporation into proteins and modulating the redox system. Therefore, the interplay between Se and S availability can significantly influence the nutritional quality and positive effects on human health of edible Brassica crops [29,30,31].

Recent research has also shown that exogenous application of Se and Zn can influence key aspects of plant metabolism, including antioxidant enzyme activity, osmotic regulation, and stress response [32,33,34]. However, the efficacy and safety of such biofortification strategies depend on multiple factors, including the concentration and chemical form of the micronutrient, plant species, and developmental stage. Careful optimization is essential, as many micronutrients exhibit a narrow beneficial range; at elevated levels, they may become phytotoxic, causing oxidative stress, membrane damage, impaired photosynthetic function, and growth inhibition [35,36].

Photosynthesis is a key physiological process driving plant growth and biomass production, and it is highly sensitive to micronutrient status. Both selenium (Se) and zinc (Zn) can modulate photosynthetic activity, depending on their concentration and context. Low Se levels have been shown to enhance photosynthesis by reducing oxidative stress, stabilizing membranes, and preserving chloroplast ultrastructure, as observed in sorrel and tomato seedlings under salt stress [37,38]. Conversely, excessive Se can decrease pigment content, disrupt chloroplast structure, and inhibit electron transport [39,40]. Zn deficiency impairs photosynthetic capacity by reducing the activity of key enzymes such as carbonic anhydrase and Rubisco, lowering CO₂ assimilation, stomatal conductance, and PSII efficiency, and decreasing chlorophyll content due to structural damage to chloroplasts [41,42].

Despite extensive research on selenium and zinc biofortification in mature plants, their short-term effects on microgreens—particularly regarding photosynthetic performance—remain poorly explored. Due to their rapid growth and sensitivity to nutrient levels, microgreens serve as a suitable model for studying nutrient–physiology interactions over short periods. This study aimed to assess how Se (Na₂SeO₄) and Zn (ZnSO₄ × 7H₂O) treatments affect photosynthetic responses in four Brassica microgreens—kale (Brassica oleracea L. var. acephala), kohlrabi (Brassica oleracea L. var. gongylodes), broccoli (Brassica oleracea L. var. italica), and pak choi (Brassica rapa L. subsp. chinensis). Species-specific responses to micronutrient treatments were evaluated using non-destructive techniques such as direct chlorophyll a fluorescence, modulated 820 nm reflection, measurements of root and hypocotyl length, pigment analysis, and the DW to FW ratio. The findings provide insight into light-driven photosynthetic reactions in hydroponically grown Brassica microgreens and contribute to the development of strategies for effective biofortification that enhance both nutritional value and physiological performance.

2. Materials and Methods

2.1. Experimental Setup

Ecologically certified seeds for microgreen cultivation of four species from the Brassicaceae family: broccoli (Brassica oleracea var. italica), pak choi (Brassica rapa var. chinensis), kohlrabi (Brassica oleracea var. gongylodes), and kale (Brassica napus var. pabularia) were purchased from a local supplier Lokvina d.o.o. (Savska Ves, Čakovec, Croatia), certified with an ecological certificate for the production of high-quality organic seeds.

Microgreens were cultivated hydroponically in plastic containers using tap water. Six-day-old microgreens were transferred to the aqueous solutions of sodium selenate (Na₂SeO₄) or zinc sulfate heptahydrate (ZnSO₄ × 7H₂O), respectively, at different concentrations. Concentrations of 2 mg/L (Se2), 5 mg/L (Se5), and 10 mg/L (Se10) were used for selenium treatments, while for zinc treatments, concentrations of 5 mg/L (Zn5), 10 mg/L (Zn10), and 20 mg/L (Zn20) were chosen based on the literature review to balance nutritional enhancement and physiological safety [26,43,44,45,46,47,48].

The electrical conductivity (EC) of the solutions was measured using a pH/conductivity meter (SevenCompact Duo; Mettler-Toledo GmbH, Greifensee, Switzerland) and ranged from 0.929 ± 0.008 dS/m in the control (tap water) to 0.943 ± 0.005 dS/m in aqueous Se and Zn solutions, indicating that no salinity stress affected results. The control plants continued to grow on tap water. The treatment was carried out for the following four days in a growth chamber (light intensity 320 µmol m⁻²s⁻¹; photoperiod 16/8 light/dark, temperature 20 ± 2 °C) under violet LED light (VGD Lumia, Mouans-Sartoux, France) with peak intensities at 450 and 630 nm. The experiment was repeated three times, and in each experiment, randomly chosen, dark-adapted (30 min) microgreen plantlets were used for in vivo measurements of fast chlorophyll a fluorescence (ChlF) and modulated 820 nm reflection (MR).

2.2. Chlorophyll a Fluorescence Measurements

Ten-day-old cotyledons were used for measurements of chlorophyll a fluorescence (Handy PEA, Hansatech Instruments Ltd., Norfolk, UK). ChlF transients were induced by applying a saturating pulse of red light (~3200 μmol m⁻² s⁻¹, maximum intensity at 650 nm), and the measured parameters were used for the calculation of selected JIP-test parameters (Table S2) [49,50,51].

Normalized OJIP transients represent the mean values of 30 measurements (n = 30) of each Se and Zn treatment, respectively, in relation to their corresponding control. Total photosynthetic electron transport, log Pitot, was presented as total driving forces (DF_total), which were obtained by summing up the corresponding partial driving forces: log γRC/(1 − γRC), log φP0/(1 − φP0), log ψE0/(1 − ψE0), and log δR0/(1 − δR0) [52,53]. The difference for the control and each Se and Zn treatment, respectively, ∆DF_total, was calculated as ∆DF = DF_treatment − DF_control [54].

2.3. Modulated 820 nm Reflection Measurements

For the MR measurements, M-PEA (Hansatech Instruments Ltd., Norfolk, UK) was used. A modulated 820 nm LED was used to induce a high-quality P700 reflectance signal. The reflected beam from the MR signal was used to calculate the MR/MR₀ ratio; the first reliable MR measurement (MR₀) was taken at 0.7 ms. The specific slopes corresponding to fast (oxidation) and slow (reduction) phases of reflection intensity were calculated from the gained MR signals [55,56,57,58,59].

2.4. Agronomic Traits and Photosynthetic Pigment Determination

The lengths of the root, hypocotyl, and total plant were measured in 15 randomly chosen plantlets from each species and Se/Zn concentration.

Total chlorophyll content (Chl a+b) and carotenoids (Car) were extracted with cold acetone for 24 h at −20 °C from grounded stems and cotyledons of the ten-day-old microgreens. Concentrations were determined using the extinction coefficients and the equations derived by Lictenthaler [60] using absorbances measured spectrophotometrically (Specord 40, Analytik Jena, Jena, Germany) at 470, 661.6, and 644.8 nm, respectively. Grounded tissue was weighed (fresh weight, FW) and then oven-dried at 105 °C for 24 h, after which the dry weight (DW) was measured. The DW/FW radio was calculated and expressed as %DW.

2.5. Selenium and Zinc Content Analysis

The stems and cotyledons of four Brassica microgreens were harvested after a 4-day treatment, washed in deionized water, and then dried to a constant weight at 60 °C. The dried samples were ground into a fine powder using a pestle and mortar. All plant samples for measuring Se and Zn concentrations were digested with 10 mL of a 5:1 mixture of HNO₃ and H₂O₂ at 180 °C for 60 min in microwave vessels (CEM Mars 6, Matthews, NC, USA). For the Se determination, 5 mL of concentrated HCl was added to the cooled digest to reduce Se6+ to Se4+, while the Zn concentrations were determined by inductively coupled plasma optical emission spectroscopy, ICP-OES (PerkinElmer Optima 2100 DV, Überlingen, Germany) using internal pooled plasma. The reference material was prepared in the same way as the microgreen samples [61].

2.6. Statistical Analysis

2.6.1. Variable Selection and Multivariate Analysis

Data normality, outliers, and homoscedasticity were verified using the Shapiro–Wilk test on residuals, the Grubbs and Levene tests, respectively. Standardization (z-score normalization to place parameters on a common scale) and a heteroscedasticity-consistent correction were applied when these tests indicated violations. The following variable selection strategy was used to identify the smallest, most informative set of chlorophyll fluorescence parameters that were biologically meaningful, statistically non-redundant, and significantly responsive to Zn/Se treatment. A three-way MANOVA test (with Benjamini–Hochberg correction) was used to simultaneously test species effect (S), micronutrient treatment effect (T; Zn vs. Se), dose effect (D; nested within each micronutrient), and their interaction (S × T × D). The dose level was analyzed within micronutrients, as Se and Zn were added at different concentrations (Zn: 0, 5, 10, and 20 mg/L; Se: 0, 2, 5, and 10 mg/L), allowing each element to have its scale, avoiding an unbalanced design, which violates homogeneity assumptions. Multicollinearity tests (using the variance inflation factor, with a cutoff of 10) and correlations with the first canonical roots (r < 0.4 correlation threshold) were used to eliminate the redundant variables [62,63]. Because the data set had a high parameter-to-observation ratio (>60 variables, n ≈ 25 plants per dose), separate three-way MANOVAs were used on each functional group (Table S1) to identify the most informative variables for each level. Pillai’s trace was chosen for presenting the results as it is the most robust to unequal n, covariance matrix heterogeneity, and small-to-moderate sample sizes [64].

2.6.2. Canonical Discriminant Analysis

Canonical Discriminant Analysis (CDA) was applied to visualize how the selected variables combine to separate each Brassica microgreens and treatment (Zn effect from Se effect), as well as to identify key parameters that carry the most weight. Additionally, separate CDAs were conducted for each micronutrient (Zn, Se) to examine whether the four dose levels nested in species can be distinguished based on the selected parameters, to identify the variables that best follow the dose–response gradient, and to reveal species-specific dose patterns. Within each micronutrient, a two-way ANOVA (plant × dose) with Benjamini–Hochberg-adjusted pairwise comparisons was used to identify specific dose differences. Statistical analyses were performed using R statistical software [65], and graphical presentations were created with Microsoft Excel. Cross-validation for CDA stability was performed using the “leave-one-out” classification accuracy.

2.6.3. Post-Hoc and Correlation Analyses

Within each micronutrient, a two-way ANOVA (plant × dose) with Benjamini–Hochberg-adjusted pairwise comparisons was used to identify specific dose differences. To indicate the direction and strength of the relationship between growth traits (root and shoot growth, dry weight proportion as DW/FW), pigment composition, and OJIP parameters, the Pearson correlation coefficient was calculated. To account for inter-species variation, Brassica species (broccoli, pak choi, kohlrabi, and kale) were set as a weighting factor to reveal how parameters couple under Se vs. Zn (micronutrient-specific relationships), ensuring robust and treatment-driven relationships. Statistical analyses were performed using R statistical software, with the following packages: stats, p.adjust() with method “BH”, MANOVA.RM, car, mctest, and candisc in R 4.5.1 [61]. Graphical presentations were created with Microsoft Excel. For some ChlF and MR parameters, as well as agronomic traits and the Chl a+b/Car ratio, a one-way ANOVA followed by Tukey’s HSD post-hoc test was applied (Statistica v.13.1; Tibco Software Inc., Palo Alto, CA, USA). Differences between Brassica microgreens exposed to Se and Zn concentrations were considered significant at p ≤ 0.05. The results are presented as mean ± standard deviation (SD) of 30 replicates (n = 30) for ChlF, 10 replicates (n = 9) for MR measurements, 15 replicates (n = 15) for root, hypocotyl and whole-plant length and six (n = 6) for DW/FW and Chl a+b/Car ratios.

3. Results and Discussion

3.1. Variable Selection and Multivariate Analysis of Chlorophyll Fluorescence Parameters

The primary objective of this study was to evaluate the impact of potential biofortification with selenium or zinc on the photosynthetic performance of four Brassica microgreens (broccoli, pak choi, kohlrabi, and kale) under hydroponic conditions and to identify the best species-treatment combinations for achieving optimal physiological efficiency and yield. Multivariate analysis specified the primary drivers across all functional groups of chlorophyll fluorescence parameters, as shown in Supplementary Table S3.

Across all chlorophyll fluorescence functional groups of measured parameters, species was the dominant source of multivariate variance. Large Pillai’s trace values (ranging from 0.36 to 1.36) and eigenvalues (λ up to 98.9) indicated that each Brassica microgreen carries a distinctive fluorescence signal in every functional group of parameters. The strongest differences among microgreens were found in timing and area (V = 1.36, λ = 98.9, ρ₁² = 0.99), Q_A turnover set (V = 1.18, λ = 16.6, ρ₁² = 0.94), and wave-band integrals and connectivity (V = 0.93, λ = 1.08, ρ₁² = 0.52).

Treatment (Se vs. Zn), dose alone, and interaction of species, treatment, and dose were also significant for all nine functional groups of chlorophyll fluorescence parameters (p < 0.0001). Pillai’s trace values for treatment ranged from 0.13 to 0.60, indicating that Se vs. Zn explains 13 to 60% of the combined variance (Table S3). The discrimination between treatments was concentrated in timing and area (λ = 1.49, ρ₁² = 0.40) on a single canonical dimension. In OJIP amplitudes and energy fluxes, as well as performance and driving forces, approximately 25% of the multivariate variance was attributed to the treatment effect. Still, a lower eigenvalue from 0.31 to 0.37 (ρ₁² = 0.24–0.27) indicated that these chlorophyll fluorescence profiles still largely overlap, and that misclassification based on them can be high.

The dose effect tested within each micronutrient showed a consistent but smaller multivariate response in the majority of functional groups of fluorescence parameters (8% to 13%), with the strongest timing and area variables (λ = 8.0, ρ₁² = 0.89), followed by Q_A turnover and related number (λ = 1.25, ρ₁² = 0.56). As for the other groups, Pillai’s V showed a weak multivariate signal, indicating measurable but moderate scaling of fluorescence response with the added dose of micronutrient. Although a species × treatment × dose interaction was significant across all groups, except relative variable fluorescence, indicating that the exact shape of the dose–response curve differs between Se- and Zn-treated Brassica microgreens, its contribution to multivariate variance was weak in most groups, suggesting that the form of the dose–response curve was similar across different microgreens. The strongest response was observed in Q_A turnovers and related metrics, accounting for 34% of the variance (Table S3).

Overall, the MANOVA results indicate that Brassica identity dominated the chlorophyll fluorescence response. The type of micronutrient treatment matters, but primarily in terms of timing and area variables, rather than in energy fluxes and OJIP amplitudes. Increasing concentrations of micronutrients resulted in small but significant adjustments, particularly in fast electron transport kinetics.

The dominance of timing and area variable, Q_A turnover numbers, connectivity indices, energy flux parameters, performance and driving forces indices highlights the central role of reaction center integrity and excitation energy transfer in the early photochemical response to added micronutrients, suggesting that the main strategy for coping with extra micronutrients for these microgreens was to protect the integrity of PSII reaction centers and keep energy flow smoothly between PSII and PSI. Several authors support this relationship between micronutrients and chlorophyll fluorescence response [66,67,68].

3.2. Canonical Discriminant Analysis

Species-level CDA provided insights into the interspecific variation of four Brassica microgreens (Pillai’s trace = 1.293, F_33,2424 = 55.7, p < 0.0001), with three statistically significant axes accounting for 100% of the discriminatory power between species (87.4% on F1 and 7.1% on F2) (Figure 1a,b, Supplementary Table S4). F1 was the primary species discriminator, associated with the overall electron transport capacity and phenomenological energy fluxes (N[VG], ET₀/CS₀, RC/CS₀), as positive loadings and connectivity between PSII units and the primary electron acceptor pool (capacity (A to F_J) as negative loadings. F2 moderately correlated with dV/dt₀, φ_P0, and Fi on the positive side, while RE₀/CS₀ loaded negatively, indicating donor side acceleration and PSI drain. Cross-validated classification accuracy reached 98.8%, confirming the reliability and stability of the model.

Therefore, among four Brassica microgreens, two clusters were discriminated, with broccoli and pak choi grouped with positive canonical scores (high-energy flux species) vs. kale and kohlrabi, grouped with negative canonical scores (low-energy flux, acceptor-limited species, with a bigger acceptor pool that needs to be reduced or larger transient electron accumulation at the acceptor side). At the same time, the efficient energy flux (+F1) or limitation in the Q_A pool (-F1) in species discrimination can be detected by monitoring A to Fj, while the balance between RE₀/CS₀ and Fi reveals whether Brassica species respond by accelerating electron cycling through PSI or enhancing electron sinks beyond PSI [66,69].

Furthermore, the treatment-level (CDA) analysis (Figure 1c,d and Table S5) confirmed that chlorophyll fluorescence profiles discriminate among Zn and Se treatments (Pillai’s trace = 0.557, p < 0.0001), with three statistically significant canonical functions explaining approximately 91% of the discrimination. The first canonical function, F1 (55.49% of the variance), separated groups based on their electron transport efficiency (early rise amplitudes of the initial electron transport process vs. efficiency of electron transport downstream Q_A and the overall performance indices). All Se treatments were characterized by higher ET₀/RC, ψ_E0/(1 − ψ_E0), RE₀/RC, φ_R0, PI_total, and DF_total, indicating more efficient electron transport beyond Q_A and better overall photosynthetic performance; similar findings are supported in the literature [70,71,72]. In contrast, the Zn treatments were associated with lower performance indices and higher V_L and V_I, indicating impaired electron flow. Over-reduced Q_A pools and donor-side restrictions are consistent for Zn stress in the literature [73,74,75,76]. Since Zn is a cofactor of many antioxidant enzymes and supports the structure of glutathione reductase, its adequate content stabilizes the ratio of oxidized and reduced glutathione and sustains oxygen metabolism. Changes in Zn concentrations, which cause an imbalance, can disrupt these redox reactions [77]. The second canonical function, F2 (24.07% of the variance), contributed to additional discrimination based on early PSII electron transport dynamics and captured the dose-dependent response within each micronutrient, driven by an increase in A to Fj (size of Q_A pool) and a higher single turnover saturation constant, Ss[VG] (positive loadings), as Zn and Se concentrations increased. At the same time, the number of Q_A redox turnovers N[VG] declined (negative loadings), indicating slower Q_A turnover at higher concentrations (Figure 1a). Cross-validation using the “leave-one-out” method showed near-perfect class separation (training sample accuracy of 98.2%). Even when each observation was excluded from model estimation, the cross-validated classification accuracy reached 91.6%. All misclassifications occurred within the adjacent doses of the same treatment (e.g., between Se0 and Se2, 12% mistakes, or between Se2 and Se5, 14% mistakes), never across two elements, confirming the high confidence of the first canonical axis and indicating that the chlorophyll fluorescence signature provides a reliable fingerprint of both metal identity and its dosage.

3.3. Micronutrient Effect

Since the global model indicated overlapping effects of Se and Zn in different Brassica microgreens, separate CDAs were performed to visualize the complex responses of all four Brassica microgreens to varying doses of both metals individually (Figure 2). Their consistent separation in baseline measurements (Zn0 and Se0) most probably reflected inherent interspecies differences in fluorescence.

3.3.1. Zinc Effect

Dose–species discriminatory power was captured entirely on a single axis for both elements (F1 = 97.9%). In comparison, F2 explained 1.6–1.7% of the residual discriminatory power for Se and Zn, creating a slight vertical offset within each cluster (Figure 2a). Along the F1 axis, different Brassica species form four clusters, and within each, there is a shift with the dose increase (Figure 2b). For Zn, a single canonical dimension (F1) explained 97.9% of the species-dose separation (χ² = 3605, p < 0.001), and the second axis (F2) added only 1.6%, although it was statistically significant (χ² = 1221, p < 0.001).

Canonical F1 was dominated by A to F_J, φ_P0, and RC/CS₀, while canonical F2 was primarily influenced by the balance of Q_A turnovers N[RS], N[VG], as well as RC/CS₀ and φ_P0, explaining fine-scale separation between species or how they manage electron transport restrictions (Table S6). Since the size of the Q_A pool is one of the most significant drivers of species separation, kale and kohlrabi cluster toward high F1 (positive scores), while pak choi and broccoli cluster toward low (negative scores) F1. In all species, A to F_J increased with higher Zn doses, shifting samples in a positive F1 direction. The size of the shift is primarily influenced by the increase in A to F_J, counterbalanced by an increase/decrease in RC/CS₀ and φ_P0. In every species, φ_P0 decreases at 5–10 mg/L while more reaction centers close (RC/CS₀ increases), as shown in Figure 3 and in Tables S7 and S8.

Broccoli showed the most substantial relative vector shift in F1 (+51%), accompanied by a shift to positive F2. This large shift was driven by a decrease in N[VG] and a minor recovery in N[RS] and φ_P0, indicating a significant rearrangement of electron transport dynamics (Figure 3a). Kale exhibited the smallest percentage change in F1 (+16%) and minimal shift along the F2, similar to kohlrabi, making it the most Zn-tolerant of Brassica species in this experiment. Kale initially had the largest Q_A pool and more Q_A turnovers, which kept reaction centers open, suggesting a shift towards faster but less efficient electron transfer (Figure 3d). Although small N[VG] and N[RS] decreases would predict a positive shift along F2, a reduction in RC/CS₀ (with positive weight × negative change) drives the kale response towards negative F2 scores with increasing Zn. Kale also showed the strongest φ_P0 rebound at Zn20 (Figure 2 and Figure 3). Pak choi showed moderate sensitivity, characterized by a relatively stable F2 position with increasing Zn concentrations. Kohlrabi, similarly, exhibited modest adjustments, characterized by decreased N[VG], without corresponding changes in N[RS] or RC/CS₀, shifting samples toward lower F2 scores (Figure 2 and Figure 3c).

Overall, kale showed the lowest level of stress to Zn, with the most robust photochemistry (the smallest F1 rise and φ_P0 rebound were observed at the highest Zn concentration, above the ones at the control, Zn0). Pak choi was moderately sensitive to Zn treatment, showing stronger early stress signs but stabilizing later, since Zn does not overwhelm Q_A cycling pathways. On the other hand, kohlrabi was more tolerant than pak choi to Zn treatment, rerouting electron traffic rather than shutting down reaction centers, keeping photosynthesis in pace with the increase in Zn concentration.

3.3.2. Selenium Effect

For the Se treatments, CDA revealed similar dose–species separation (F1 explained 97.9%, χ² = 3371, p < 0.001, and F2 added only 1.6%, although significant discriminating power, χ² = 1197, p < 0.001). Standardized coefficients showed that F1 is driven by A to Fj and RC/CS₀, along with φ_P0 and N(RS), similar to the pattern observed in the Zn treatment, whereas F2 is defined by the balance between N[RS] and N[VG], RC/CS₀, φ_P0, and ET₀/CS₀. F1, again, measures the size of the primary electron transport limitation at Q_A as the dose increases. At the same time, F2 captures higher N[RS], RC/CS₀, and φ_P0 at the positive end, and higher N[VG] and ET₀/CS₀ at the negative end to reveal mechanistic diversity in how the Se treatment affects or fine-tunes the photosynthetic apparatus in different species (Table S6).

In all four species, increasing Se from 0 to 10 mg/L increased A to Fj (Figure 3), shifting the samples in the direction of F1 and towards a positive F2 (Figure 2). This time, pak choi had the highest relative F1 shift (+51%) and upward movement along F2, resulting from an increase in N[RS] and the RC/CS₀ strategy. Broccoli and kale showed lesser shifts, while kohlrabi maintained stable N[VG] values with enhanced electron transport (ET₀/CS₀).

3.4. PSII Primary Photochemistry and Electron Flow

The maximum quantum yield of primary photochemistry (φ_P0), with values ranging from 0.741 to 0.805, confirmed that PSII reaction centers were fully functional in all microgreens [69]. This is particularly significant because measurements were performed on cotyledons of immature plantlets, showing that they maintained intact charge separation after enrichment with either Se or Zn. Further, the probability that an electron is transported from reduced PQ to the acceptor side of PSI (δ_R0) increased significantly in broccoli and kohlrabi with Se10, suggesting that an accelerated intersystem electron flow and enhanced activity of PSI are driven by the Se treatment [78]. Nevertheless, a significant decrease in δ_R0 in kohlrabi and kale with Zn20, indicating a slower transport rate of electrons to the PSI acceptor side [79], coincided with a lower density of active RCs in photosystems (RC/SC₀), which was supported by unchanged ET₀/CS₀, suggesting that Zn primarily deactivates RCs rather than limits electron transport flux from Q_A to Q_B.

δ_R0 and ET₀/CS₀ are parameters used to quantify the I–P phase of the OJIP transient. Their values can be correlated with the appearance of L-band and the overall grouping probability of PSII (ω), reflecting changes in energetic cooperativity of PSII units [53,55,80]. According to Yusuf et al. [81], to achieve an efficient distribution of the absorbed energy to the primary acceptor, Q_A, there is a necessity for good grouping and connectivity between PSII reaction centers (RCs). It was suggested that closely grouped thylakoids are not likely to undergo structural changes, since they are not considered susceptible to stress [82]. The ω parameter indicated changes in downstream electron flow. In Se-treated plants, ω remained stable or increased, suggesting an oxidized PQ pool and efficient energy distributions among RCs [83]. In contrast, ω decreased under Zn, indicating reduced connectivity due to RC closure and PQ over-reduction. Thus, in addition to the L-band, the overall grouping probability of closed PSII RCs (ω, Table S1) provides a confident parameter to differentiate between Se and Zn effects on microgreens.

The area above the fluorescence rise to the J-step (A to Fj), as a measure of for the size of the total electron carrier pool between the PSII donor side and the PSI acceptor side [50,51,83], increased significantly in all microgreens after exposure to each Se and Zn concentration. A larger A to Fj implies more available carriers and more Q_A turnovers (N[RS]). However, except for pak choi at Se10, the results showed no significant change in N[RS], indicating that the additional carriers were not used for faster multiple turnovers but remained in a reduced state.

Moreover, the N[VG] parameter, which denotes the quantity of electron carriers in in vivo conditions, revealed a significant decrease after Se treatment in broccoli (Se10), pak choi (Se5, Se10), and kale (Se10) and after Zn treatment in broccoli and kale at all concentrations, in pak choi at Zn20 and in kohlrabi at Zn10 and Zn20. Such results indicate that Q_A could be reduced, but not re-oxidized efficiently, which means that PSII cannot perform multiple turnovers, a phenomenon reported in PSII lacking fully developed donor or acceptor sides [84]. Comparable limitations were described for developing spruce needles [85] and young fig leaves [86].

Therefore, the Se-treated microgreens appeared to compensate developmental immaturity by promoting cyclic electron flow around PSI, while excess Zn pushed the Q_A pool into an over-reduced state with restrictions from the PSII donor side (Figure 1 and Figure 2).

3.5. Performance Index and Driving Forces

To gain deeper mechanistic insight into Brassica’s strategies, the performance index (PI_total) and its component driving forces (DFs) were examined in detail (Figure 2a). Although a three-way ANOVA detected a statistically significant relationship (F_{31, 788} = 5.66, p < 0.0001), it explained only 18% of the variance and showed a large prediction error (MAPE = 38.44%). The treatment had a medium effect size (partial η² = 0.075, Cohen’s f = 0.285), similar to that of species (partial η² = 0.055, Cohen’s f = 0.240), while dose on its own had a negligible effect (partial η² = 0.004, Cohen’s f = 0.064), but the effect of dose depends on the treatment (treatment × dose interaction was significant, p < 0.0001, partial η² = 0.041, Cohen’s f = 0.206). In contrast, the three-way treatment × plant × dose effect was not significant (p = 0.198), so the results for the treatment by dose were analyzed separately for each species.

The PI_tot (Figure 4a) describes the functional activity of PSII and PSI, as well as the electron transport chain between these two photosystems [53,81,87]. Treatment with Se induced an increase in PI_total in pak choi, kohlrabi, and kale at Se5 and Se10 concentrations, while broccoli showed no changes regardless of applied concentration. On the other hand, Zn application resulted in a decrease in this parameter for all investigated microgreens. The analysis of driving forces (DF_total) revealed a detailed description and mechanisms underlying performance changes. As the overall DF_total was shown to be positive for Se treatments and negative for Zn treatments, a detailed analysis of partial DFs provided deeper insight into which part of the electron transport chain was the most sensitive to a specific element.

Moreover, relative differences in driving forces (ΔDFs) provided insight into specific differences for each treatment compared to the corresponding control (Figure 4b–e). The results revealed that PI_total in pak choi treated with Se increased mostly due to higher log ψ_E0/(1 − ψ_E0), with an additional increase in log φ_P0/(1 − φ_P0) at higher (Se5 and Se10) concentrations. Such results suggested that the potential for light reactions for primary photochemistry and the potential for energy conversion between the two photosystems due to dark reactions increased [53,54].

On the other hand, kohlrabi and kale showed that their increase in PI_tot was induced mainly by the positive difference in log δ_R0/(1 − δ_R0), suggesting the highly efficient PSI electron transport as a result of the positive Se stimulus [88]. The Zn treatment, however, showed a versatile negative effect on microgreens. There were two differential negative responses, however. The first one, the decrease in PI_tot due to negative log φ_P0/(1 − φ_P0) and log ψ_E0/(1 − ψ_E0), suggesting a decrease in energy conversion potential through both light and dark reactions, was observed in broccoli and kale at Zn10 and in pak choi and kohlrabi at Zn5. The second response was the diminished PI_total due to negative log δ_R0/(1 − δ_R0) compared to the corresponding control, which implies ineffective electron transport around PSI in all microgreens at the highest Zn concentration (Zn20), but also in pak choi and kohlrabi at Zn10 and in kale at Zn5.

Based on these observations, Se-treated microgreens demonstrated positive responses through highly regulated processes between primary photosynthetic reactions and the reactions leading to CO₂ assimilation. In contrast, the Zn-treated microgreens showed impaired photosynthetic performance through multiple mechanisms affecting different components of the electron transport chain.

3.6. Redox Dynamics of PSI—Modulated Reflectance at 820 nm

To further investigate the efficiency of electron transport dynamics at PSI, the differences in kinetics at 820 nm were analyzed (Figure 5). The modulated reflectance measurements (MR) revealed the redox states of PSI reaction center (P₇₀₀) and plastocyanin (PC) as MR/MR₀ ratios in Brassica microgreens exposed to different Se and Zn concentrations. The typical MR transients cover a fast-declining phase from MR₀ to MR_min, typically between 0.7 and 7 ms, and a slow-inclining phase from MR_min to MR_max, which occurs around 300 ms.

All four microgreens displayed these characteristic typical transients. However, the Se-treated microgreens showed higher MR_min signals compared to the controls and Zn-treated microgreens (Figure 5). Further, MR₈₂₀ signals additionally revealed specific oxidation (V_ox) and reduction (V_red) parameters (Table V_OX/RED). Se treatment induced a faster oxidation state of P₇₀₀ and PC in broccoli, pak choi, and kale, and the effect increased with higher Se concentrations, followed by a slower re-reduction state of P₇₀₀⁺ and PC⁺ in all investigated microgreens. Faster oxidation of P₇₀₀ and PC is usually associated with elevated photochemical activity at the acceptor side of PSI [55,58]. In Zn-treated plants, no changes were observed in Vox regardless of species, but a significant decrease in V_red in pak choi was noticed. Despite the slower re-reduction rate, the increased driving forces associated with the efficiency of PSI suggested that the Se-treated microgreens efficiently regulated the distribution of excitation energy between PSII and PSI [78]. These energy arrangements also indicate the generation of cyclic electron flow around PSI as an efficient adaptive mechanism for the protection of the PSI donor side [57,89,90]. Since the oxidation and re-oxidation of PC and P₇₀₀ depends on the available electron acceptor pool at the PSI acceptor side [55,59], our results suggest that the Zn treatment showed impaired photochemical activity of PSI, likely due to an over-reduced Q_A pool preventing efficient electron flow and restrictions on the PSII donor side limiting electron supply to PSI [78]. Thus, these results indicate that Zn treatment creates bottlenecks in the electron transport chain. This suggestion was supported by CDA, which showed treatment-dose separation (Figure 1, Table S3), since the changes in the MR/MR₀ transient correspond to the J–I and I–P phases of the ChlF transients and can therefore be linked to the changes in the (1–V_I) value [59,91].

3.7. Agronomic Traits and Pigment Ratio

Microgreens averaged from 6 to 9 cm (in Zn treatments) and from 4.3 to 6.8 cm (in Se treatments) of total plant length, with species differences detailed in

Table 1. Agronomic traits: lengths of the root, hypocotyl, and the whole plant (measured in cm), dry weight-to-fresh weight ratio (DW/FW, expressed as %) and Chl a+b/Car ratio determined in four Brassica microgreens exposed to Se (2, 5, and 10 mg/L) and Zn (5, 10, and 20 mg/L). The results are shown as the mean ± SD; different letters (lowercase for Se treatment and uppercase for Zn treatment) represent significant differences at p ≤ 0.05 (ANOVA, Tukey HSD).

		Root (cm)	Hypocotyl (cm)	Plant Total (cm)	DW/FW (%)	Chl a+b/Car
Treatment
Broccoli	Se0	6.933 ± 2.517 a	2.093 ± 0.675 a	9.027 ± 2.454 a	9.487 ± 0.446 d	4.138 ± 0.031 a
	Se2	6.740 ± 1.840 a	1.820 ± 0.699 ab	8.560 ± 2.150 a	12.215 ± 0.271 b	3.955 ± 0.044 b
	Se5	2.760 ± 0.702 b	1.427 ± 0.526 b	4.187 ± 0.959 b	13.722 ± 0.379 a	3.731 ± 0.062 c
	Se10	4.087 ± 1.037 b	1.473 ± 0.317 b	5.560 ± 1.037 b	11.109 ± 0.678 c	3.701 ± 0.037 c
	Zn0	3.713 ± 1.828 B	2.113 ± 0.867 B	5.827 ± 0.398 B	10.494 ± 0.301 B	4.106 ± 0.065 B
	Zn5	7.907 ± 2.970 A	3.367 ± 0.439 A	11.273 ± 2.865 A	10.952 ± 0.709 AB	4.055 ± 0.036 B
	Zn10	7.760 ± 2.611 A	2.867 ± 0.434 A	10.327 ± 2.749 A	11.412 ± 0.185 A	4.257 ± 0.027 A
	Zn20	6.167 ± 2.317 A	1.940 ± 0.575 B	8.107 ± 2.271 B	10.783 ± 0.767 AB	4.238 ± 0.063 A
Pak choi	Se0	3.853 ± 1.548 a	1.553 ± 0.334 a	5.407 ± 1.697 a	12.898 ± 0.395 b	4.155 ± 0.017 a
	Se2	8.040 ± 1.850 a	1.400 ± 0.511 a	4.400 ± 2.115 ab	16.239 ± 0.494 a	4.104 ± 0.095 a
	Se5	1.780 ± 0.592 b	1.493 ± 0.461 a	3.273 ± 0.631 b	15.530 ± 0.446 a	4.053 ± 0.215 a
	Se10	2.787 ± 0.680 ab	1.213 ± 0.236 a	4.000 ± 0.805 b	12.560 ± 1.482 b	4.165 ± 0.102 a
	Zn0	2.867 ± 1.313 B	2.200 ± 0.393 A	5.067 ± 1.337 B	13.957 ± 0.421 AB	4.209 ± 0.052 BC
	Zn5	5.533 ± 2.208 A	20227 ± 0.349 A	7.760 ± 2.277 A	13.572 ± 0.599 AB	4.239 ± 0.018 B
	Zn10	3.567 ± 2.191 A	2.147 ± 0.242 A	5.713 ± 2.232 B	12.595 ± 1.301 B	4.161 ± 0.039 C
	Zn20	3.393 ± 1.823 B	1.653 ± 0.414 B	5.047 ± 1.752 B	14.527 ± 0.792 A	4.346 ± 0.036 A
Kohlrabi	Se0	3.893 ± 0.869 a	1.980 ± 0.675 a	5.873 ± 1.297 a	13.859 ± 0.309 b	4.067 ± 0.025 a
	Se2	4.213 ± 0.966 a	1.853 ± 0.230 ab	6.067 ± 1.204 a	12.885 ± 1.809 b	4.003 ± 0.047 b
	Se5	2.100 ± 0.815 b	1.687 ± 0.444 ab	3.787 ± 0.823 b	16.151 ± 0.563 a	3.758 ± 0.044 c
	Se10	2.593 ± 1.024 b	1.535 ± 0.223 b	3.947 ± 1.026 b	14.371 ± 0.930 b	3.822 ± 0.054 c
	Zn0	2.980 ± 1.385 B	2.253 ± 0.529 A	5.233 ± 1.362 B	15.502 ± 0.704 B	4.248 ± 0.084 A
	Zn5	5.927 ± 2.514 A	2.307 ± 0.546 A	8.233 ± 2.661 A	15.022 ± 0.325 B	4.039 ± 0.041 C
	Zn10	5.120 ± 1.015 A	2.327 ± 0.343 A	7.447 ± 1.119 A	13.827 ± 1.063 A	4.130 ± 0.049 B
	Zn20	4.527 ± 2.708 AB	1.907 ± 0.471 A	6.433 ± 1.931 AB	18.530 ± 2.892 B	4.148 ± 0.036 B
Kale	Se0	4.080 ± 1.375 ab	1.608 ± 0.512 a	5.760 ± 1.314 a	12.023 ± 0.369 a	4.142 ± 0.035 a
	Se2	4.987 ± 1.529 a	1.533 ± 0.461 a	6.520 ± 1.500 a	12.056 ± 0.619 a	4.023 ± 0.018 b
	Se5	2.253 ± 0.601 c	1.500 ± 0.492 a	3.773 ± 0.824 b	11.186 ± 0.277 b	4.042 ± 0.028 b
	Se10	3.527 ± 1.102 b	1.013 ± 0.207 b	4.540 ± 1.079 b	12.578 ± 0.403 a	3.790 ± 0.039 c
	Zn0	5.573 ± 1.167 AB	2.040 ± 0.267 A	7.613 ± 1.157 AB	12.968 ± 0.499 BC	4.204 ± 0.022 B
	Zn5	6.907 ± 2.534 A	1.193 ± 0.294 B	8.100 ± 2.476 A	14.014 ± 0.604 AB	4.091 ± 0.105 BC
	Zn10	6.733 ± 1.758 A	1.953 ± 0.323 A	8.687 ± 1.722 A	14.267 ± 0.996 A	4.048 ± 0.043 C
	Zn20	4.253 ± 1.192 B	1.730 ± 0.561 A	5.973 ± 1.395 B	12.178 ± 0.388 C	4.396 ± 0.084 A

, as well as in Table S9. The differences between the four species under control conditions (0 mg/L Se or Zn) are as follows: The broccoli seedlings were the longest (5.8–9 cm), followed by kale (5.8–7.6 cm), kohlrabi (5.2–5.9 cm), and pak choi (5.1–5.4 cm). Zn increased the total plant length in broccoli (by 39–93%), kohlrabi (by 23–57%), and pak choi (0–53%), while in kale, it increased length by up to 14% at Zn10 but decreased it by 22% at Zn20. Se, on the other hand, reduced total plant length by 54% in broccoli, 39% in pak choi, 36% in kohlrabi, and 34% in kale—all at Se5. The size of Brassica microgreens (total plant length, cm) was dominated by root growth in both treatments (r = 0.97, p < 0.001) and moderately influenced by hypocotyl growth (r = 0.47 in Se and r = 0.39 in Zn; p < 0.001 for both; see Table S9). In the early stages of plant development, root growth is a primary indicator of microgreens’ vigor and health, enabling water and nutrient absorption for rapid and uniform growth [43]. The microgreens with higher dry weight content allocated more biomass above ground in hypocotyl and cotyledons, as indicated by positive but moderate-to-weak correlation coefficients with shoot-to-root ratio (r = 0.33, p < 0.001 in Se-treated plants, and r = 0.14, p < 0.05 in Zn-treated plants). On the other hand, moderate negative correlations of DW/FW with root growth (r = −0.41 for Se, and r = −0.25 for Zn) are characteristic of turgor-driven rapid elongation of microgreens in this early phase of their development (Table S9).

In Se treatments, Brassica microgreens with more chlorophyll than carotenoids (Se 0 mg/L or control) had longer roots, as indicated by a positive, although weak, correlation with Chl a+b/Car (r = 0.17, p < 0.01). Root and shoot growth were reduced with the increase in Se concentration in almost all microgreens (Table 1). At the same time, all Brassica microgreens exposed to Zn showed increased root elongation, and broccoli showed increased aboveground growth, as opposed to other species that had slightly decreased growth at the highest Zn concentration level (Table 1). The growth of microgreens in the Zn treatment was negatively correlated with the Chl a+b/Car ratio (r = −0.16 for shoot, r = −0.36 for root), suggesting that, across the species-weighted dataset, higher growth was associated with a lower pigment ratio, with a potential shift in pigment composition towards carotenoids as Zn increases, mitigating Zn-induced stress [92]. However, Chl a+b/Car increased with Zn concentration in most species (Table 1), implying a greater proportion of chlorophylls, which could enhance light absorption. This result could have been a consequence of species-specific variations and non-linear effects at high doses (Zn20), combined with species-weighting in the correlation analysis.

Redox potentials in chloroplast thylakoids are one of the most important factors influencing carotenoid biosynthesis [93,94]; therefore, any changes in redox state can influence carotenoid accumulation. Moreover, carotenoids are recognized as signaling molecules in plant development and response to stressful conditions [83,93,95] and proven to have a protective function [96]. Therefore, a decrease in the total chlorophyll-to-carotenoids ratio (Figure 6) in broccoli, kohlrabi, and kale exposed to Se biofortification suggests that more carotenoids are accumulated in order to protect the photosynthetic apparatus, and in pak choi, this suggests better Se tolerance. It was recently reported [95] that Se treatment reduced total chlorophyll content in broccoli sprouts but, at the same time, increased carotenoid content, which ultimately significantly enhanced the nutritional quality of the broccoli sprouts.

On the other hand, Zn treatment induced an increase in the Chl a+b/Car ratio in broccoli, pak choi, and kale and a decrease in kohlrabi. Even though [88] reported inhibition of both Chls and Cars after Zn treatment in wheat, several researchers [97,98] revealed that Zn does not cause a decrease in Car; however, it can cause changes in chlorophyll content. It was suggested that in this case, carotenoids can control the redox potential, leading to acclimatization to stressful conditions induced by Zn biofortification.

The weak correlations observed between microgreens’ growth traits (plant size and dry weight content) and OJIP parameters (r was predominantly < 0.3) suggested that photosynthetic performance was not limited, but implied possible photosynthetic acclimation response to Se and Zn (Table S10). However, weak correlations could also reflect biological variability, addressed through prior multivariate analysis, which revealed structured variability, validating the weak correlation coefficients as reflective of the complex acclimation patterns. Thus, due to the high sample size, the weak correlations were significant, but the variance explained was low, suggesting potential minor associations.

In the Se-fortified Brassica microgreens, shoot growth was slightly reduced with an increase in Se dose, probably due to mild Se stress affecting cell elongation [99,100]. The weak negative correlations between shoot growth and overall photochemical performance (r = –0.16, p < 0.01 for PI_total and r = −0.18, p < 0.01 for DF_total) potentially indicate minor photosynthetic acclimation by enhancing end-electron transport (correlations with δ_R0 and φ_R0, RE₀/RC and RE₀/CS₀), while increased dry weight proportion, possibly due to enhanced Se accumulation (Figure 6), reinforces this acclimation process to Se. A decrease in the pigment ratio with increasing Se (Table 1) due to increased carotenoid content, which enhanced photoprotection, was coupled with higher A to FJ (r = −0.4, p < 0.001, moderate effects) and slower Q_A turnover (r = 0.36, p < 0.001, moderate effects). This positive correlation between Chl a+b/Car and N[VG] suggests that Q_A reduction is related to photosynthetic pigment balance; thus, a decrease in N[VG] may favor carotenoids for photoprotection and increased cyclic electron flow (enhanced end electron transport), possibly compensating for Se-induced photosynthetic stress [55,78].

In the Zn-treated microgreens, total plant growth increased with Zn concentration; shoot growth was not affected (increased only in broccoli), was slightly decreased only at the highest Zn concentration (20 mg/L), and was negatively correlated with photosynthetic performance parameters (Table S10), reflecting reduced photosynthetic efficiency. Pigment ratio showed no significant correlations with most OJIP parameters, suggesting that photosynthesis or photoprotection pigments were regulated independently of fluorescence measured under Zn.

3.8. Selenium and Zinc Content

Preliminary analysis of Se and Zn content (Figure 6) in four Brassica microgreens revealed substantial increase in the accumulation of each element, corresponding to an increase in applied concentration. In the Se-treated broccoli, the Se content at Se2 was 10 times higher compared to the control (4.41 μg/g DW), while Se5 and Se10 showed almost 25- and 30-fold increases, respectively. Pak choi showed the highest accumulation compared to other microgreens; there were 55, 88, and 152 times higher concentrations in Se2, Se5, and Se10 compared to Se0 (1.95 μg/g DW). Further, kohlrabi showed the highest accumulation at Se0 (15.27 μg/g DW), in contrast to other microgreens, but with 5.3-, 10-, and 23-fold increases compared to Se0, indicating high accumulation potential. A similar capacity for Se accumulation was also observed in kale microgreens, which revealed 42, 66, and 97 times higher concentrations after Se treatment compared to Se0 (3.78 μg/g DW). The latest research on three microgreens: kohlrabi, kale, and wheat, showed that kohlrabi was the best Se accumulator among the investigated species [101], which is in accordance with our findings. Further, a recent investigation on three different hydroponically grown microgreens showed that increasing the availability of Se in the medium increased its accumulation in specific microgreens depending on the species [102], thus indicating that certain species could be cultivated with a specific Se content to achieve desirable accumulation. However, it has been suggested that several other factors, such as seeding density, growth conditions, or harvest stage could also play a significant role in the accumulation of a specific element. Despite that, several microgreens, such as basil, arugula, mizuna, or cress, were shown to be excellent candidates for Se biofortification, since they showed good potential for Se accumulation even at low concentrations (2 mg/L) [43]. The recommended daily allowance (RDA) of Se for adults was estimated to 55 μg to maintain proper metabolism function [103]; therefore, our results suggested that Se biofortification in hydroponic systems can be an appropriate method for increasing Se content in microgreens.

On the other hand, the Zn treatment did not show the same accumulation potential in four Brassica microgreens as Se did. However, a gradual increase in Zn accumulation was observed with increasing applied Zn concentrations. In contrast to the treatment with Se, higher initial values of Zn concentrations were observed: 78.55 μg/g DW in broccoli, 146.5 μg/g DW in pak choi, 121.6 μg/g DW in kohlrabi, and 108.8 μg/g DW in kale. Zn5 and Zn10 induced similar increases in all microgreens, ranging from 1.4- to 1.8-fold compared to the corresponding controls, while Zn20 induced 2- to 2.9-fold increases when compared to Zn0. It has been recently shown that the potential for Zn biofortification is highly influenced by the cultivar in hydroponically grown basil [97], as well as by Brassica species [46], thus highlighting the crucial role of selecting good target species for specific treatments. Additionally, it was reported that the short growth cycle of microgreens favors absorption of Zn from the roots to the upper parts of the sprout at early growth stages [104]. The recommended daily intake of Zn for an adult person is 8–11 mg [105,106]; therefore, hydroponic biofortification strategies that enhance Zn accumulation could serve as an effective method for delivering adequate micronutrient levels to meet the RDA for this element.

4. Conclusions

Based on the obtained results, it can be concluded that adding selenium and zinc provoked distinct responses in the four investigated Brassica microgreens. While selenium generally enhances photosynthetic performance and tolerance to the treatments, zinc compromises photosynthetic reactions and induces stress response, especially at higher concentrations. Selenium addition enhanced photosynthetic efficiency most in pak choi by improving both primary photochemistry and energy conversion between photosystems. In addition, kale and kohlrabi also revealed well-regulated energy distribution due to good connectivity between photosystems, suggesting dynamic photochemistry upon selenium treatment. Faster oxidation and slower re-reduction of P700 and PC, suggesting building of the cyclic electron flow around PSI, thus contributing to the high overall selenium tolerance. Moreover, selenium slightly reduced shoot growth, likely due to mild stress, and also promoted acclimation responses such as increased dry weight, while increasing the chlorophyll-to-carotenoid ratio in broccoli, kale, and kohlrabi, implying increased photoprotection and enhanced carotenoid-based photoprotection. On the other hand, zinc provoked a more versatile physiological response in the four investigated microgreens. Broccoli showed impaired photochemical regulation reflected in inhibited Q_A re-oxidation and low turnover ability, suggesting broad disruption across light and dark reactions. Kohlrabi and pak choi were characterized by moderate sensitivity to zinc, since they were able to reroute electron flow rather than inactivate reaction centers. However, higher Zn concentrations compromised PSII turnover and decreased re-oxidation, suggesting reduced energy conversion potential and reduced PSI efficiency. Kale showed the highest overall tolerance to zinc by maintaining efficient photochemistry and a stable overall electron flux, while a decrease in the Chl a+b/Car ratio, suggests increased carotenoid accumulation, possibly as a protective mechanism. Zinc treatment increased total plant growth in all investigated microgreens, while changes in pigment levels were regulated independently of fluorescence-based photosynthetic parameters. Therefore, to provoke effective biofortification without compromising photosynthetic performance, it is necessary to modify the specific dosage for individual microgreens. Future research should focus on analyses of antioxidant status to better link Se and Zn accumulation with oxidative stress defense mechanisms, which may help explain the observed species-specific responses. Tailoring biofortification approaches by integrating physiological screening with dose-response optimization might ensure both functional quality and environmental sustainability.