Autotetraploid Plumbago auriculata Lam. (2n = 24) was induced by colchicine treatment of stem segments from diploid P. auriculata (2n = 12) in the laboratory of the Landscape Architecture Institute at Sichuan Agricultural University (Jiang et al., 2020). Both diploid and tetraploid cytotypes were obtained by tissue culture and were subcultured for 25 d. After root formation, all seedlings were transferred to a climate-controlled growth chamber. A total of 72 plants (36 diploids and 36 tetraploids) of similar size were grown for 3 months in a nutrient substrate composed of soil, vermiculite, and perlite in a 1:1:1 ratio (by volume). The plants were placed in an intelligent biochemical incubator (Ningbo Jiangnan Instrument Factory, Zhejiang, China) under controlled conditions: 25 °C/20 °C (day/night) temperature, 12-h day/12-h night photoperiod, 6,600 Lx light intensity, and 70% relative humidity. Observations on the structure and secretory components of the salt glands of diploids and tetraploids under normal growth conditions prior to salt stress provide a basic reference for analyzing ion secretion from diploid and tetraploid salt glands The sixth mature leaf, counted from the top, was selected from diploid and tetraploid plants for structural observation and analysis of salt gland secretory components.

To determine the effect of salt gland Ca²⁺ content on Na⁺ and K⁺ secretion in diploid and tetraploid plants under salt stress, plants were transferred to a hydroponic system containing 1/6th-strength Hoagland’s solution (pH 6.0 ± 0.2), with the nutrient solution refreshed at 5-d intervals. After 15 d of cultivation in the hydroponic system, diploid and tetraploid plants with similar growth were selected for six different treatments (Table 1). 300 mM NaCl was selected for salt stress treatment based on our preliminary experiments that it is the highest concentration at which both diploids and tetraploids will be stressed, but diploid salt gland secretion will not be inhibited because the stress is too severe. Treatments included LaCl₃, used to inhibit Ca²⁺ channels; Na₃VO₄, used to inhibit H⁺-ATPase activity; and Diphenyleneiodonium chloride (DPI), used to inhibit plasma membrane NADPH oxidase. The CaCl₂ and the inhibitors LaCl₃, DPI, and Na₃VO₄ were pre-treated 20 min before the salt stress treatment. Salt stress treatment duration was 2 d, with six plants per treatment for each ploidy level.

Leaf samples were prepared following the method described by Ding et al. (2010) with slight modifications. Fresh leaves of diploid and tetraploid P. auriculata were cut into small pieces of 0.5 cm × 0.5 cm from the leaf margin to the middle position of the leaf vein using a razor blade. Leaf pieces were fixed in 2.5% glutaraldehyde solution at room temperature for 24 h. After fixation, leaf samples were washed with 0.05 mM HNO₃ to remove salt crystals on leaf surfaces, followed by rinsing with distilled water. Leaves were gently dried with absorbent paper and then dehydrated in a graded ethanol series (30%, 50%, 70%, 80%, 90%, and 95% ethanol). Dehydration time for the 30% to 90% ethanol series was 20 min, whereas it was 1 h for 95% ethanol. After dehydration, samples were dried using a critical point dryer (Leica, Nussloch, Germany). Dried samples were observed and photographed using a scanning electron microscope (SEM; Carl Zeiss, Oberkochen, Germany), and the area of salt glands was measured using ImageJ software. Twenty salt glands from six individual plants were measured for both diploid and tetraploid P. auriculata.

Leaf samples were prepared following the method described by Yuan et al. (2013). Fresh leaves of diploid and tetraploid P. auriculata were cut into small pieces as previously described. Leaf pieces (0.5 cm × 0.5 cm) were fixed in Carnoy’s fixative (ethanol:acetone = 3:1) and then washed with 0.05 mM HNO₃ to remove salt crystals on leaf surfaces. Leaves were rinsed with distilled water, gently dried with absorbent paper, and then decolorized in 70% ethanol. Decolorized leaves were mounted with Hoyer’s solution, making temporary slides. To measure density and area of salt glands, slides were observed and photographed using a fluorescence microscope (Olympus, Tokyo, Japan) under UV excitation light (330–380 nm) at 20× magnification. Twenty different fields of view from six leaves were measured for both diploid and tetraploid P. auriculata.

Fresh leaves of diploid and tetraploid P. auriculata were cross-sectioned in the middle using a sharp blade. Leaf pieces (1.0 cm × 1.0 cm) were quickly fixed in formalin-aceto-alcohol (FAA) fixative at room temperature for 24 h. After fixation, leaf tissues were rinsed with distilled water for 10 min, dried with absorbent paper, and then sequentially dehydrated in a gradient series of ethanol (70%, 80%, 90%, 95%, 100%, and 100% ethanol for two times). After dehydration, tissues were cleared with xylene and then embedded in paraffin in a constant-temperature oven at 60 °C. Leaf-containing paraffin blocks were sectioned using a microtome (Leica, Nussloch, Germany), producing 4 μm-thick sections. Sections were placed on glass slides coated with glycerol mounting medium and dried in a constant-temperature oven at 50 °C. Following dewaxing with xylene, rehydration with a gradient series of ethanol, and staining with SafraninO/Fast green, sections were dehydrated with ethanol and cleared with xylene. Last, sections were mounted with Canada balsam and air-dried at low temperature. Salt gland anatomical structures were observed and photographed using a fluorescence microscope (Olympus), and the external cuticle layer of salt glands was observed under UV excitation light (330–380 nm) using fluorescence mode. Twenty salt glands from six individual plants were observed for both diploid and tetraploid P. auriculata.

Salt gland secretory crystals were observed and photographed using an SEM (Carl Zeiss) equipped with an energy dispersive X-ray (EDX) system. Cross sections of mature leaf portions (0.5 cm × 0.5 cm) from the lower part of P. auriculata diploid and tetraploid leaves were cut with a sharp blade. Leaf sections were placed in a vacuum dryer for 1 h, and then, sections were attached to conductive adhesive tape on the SEM stage for EDX analysis of crystals secreted on leaf surfaces. Twenty salt glands from six leaves were analyzed for both diploid and tetraploid P. auriculata.

After 2-d treatment of diploid and tetraploid P. auriculata, mature leaves from the same position were collected and washed with 0.05 mM HNO₃ to remove salt crystals on leaf surfaces. Leaves were rinsed with double-distilled water and quickly dried with absorbent paper to remove surface water. The lower epidermis of leaves was carefully torn off and placed in 0.4 mM Floura-8 Ca fluorescence probe dye solution, with addition of Pluronic F-127 to enhance fluorescence. To ensure sufficient contact between dye and leaf epidermis, a vacuum pump was used for 10 min, followed by incubation at room temperature for 40 min (avoiding light). After incubation, leaf epidermis tissues were thoroughly rinsed with double-distilled water to remove residual dye. Rinsed epidermis tissue was mounted with a fluorescence decay-resistant medium, making temporary slides. The fluorescence of Ca²⁺ in salt glands was observed with a laser scanning confocal microscope (Olympus LSCM) in the FITC channel, and fluorescence intensity was analyzed using ImageJ software. Twenty salt glands from six plants were observed for each treatment.

Instantaneous secretion rates of Na⁺ and K⁺ from individual salt glands on the abaxial side surface were measured using an NMT system (NMT-100SIM-YG; Younger USA LLC, Amherst, MA, USA) as described by Chen Z. et al. (2010). Leaves from the lower part of the same position of both diploid and tetraploid P. auriculata were collected following 2 d of salt stress treatment. The abaxial side epidermis of leaves was gently peeled off using forceps and fixed in a culture dish with the outer surface facing up. Test solutions (Na⁺: 1.0 mM NaCl, 0.1 mM KCl, 0.2 mM MES, pH 5.8; K⁺: 30 mM NaCl, 0.5 mM KCl, 0.2 mM MES, pH 5.8) were added to the culture dish, and samples were left undisturbed for 30 min before measurement. Under a microscope, a target salt gland was located, and a selective microelectrode (Na⁺: XY-SJ-Na; K⁺: XY-SJ-K; Xuyue, Beijing, China) was positioned approximately 5 μm above the outer surface of the salt gland without touching it. Each sample was measured for 5 min. Each group included nine replicates. The Na⁺ and K⁺ flux data were read directly and outputted using imFluxes v3.0 software (Xuyue, Beijing, China). Positive values represented Na⁺ and K⁺ secretion from a salt gland to the external environment, whereas negative values represented absorption from the external environment into a salt gland.

Data were analyzed using SPSS 19.0 (SPSS Inc., Chicago, IL, USA) for Windows, and all values are reported as the mean ± standard deviation (SD). Two-way ANOVA and Student’s t-test were used to compare means of different treatments for each data set at the significance level of P < 0.05.

Polyploidization significantly increased salt gland area of P. auriculata, with the area of tetraploid salt glands 41.1% larger than that of diploid salt glands. However, salt gland density decreased significantly in tetraploid plants (Figure 1). There were no significant differences in salt gland structure between diploid and tetraploid P. auriculata (Figures 2, 3). Salt glands in both cytotypes consisted of 16 salt gland cells, with four collecting cells, four accessory cells, four cup cells, and four secretory cells, each with a secretion pore in the center.

In SEM images of secreted salt gland crystals, the volume of salt crystals secreted on leaf surfaces was significantly larger in tetraploids than in diploids, indicating higher secretion capacity of individual salt glands in tetraploids under normal growth conditions. According to the EDX analysis of secreted crystals in diploids and tetraploids, the main ion secreted was Ca²⁺, followed by Mg²⁺, whereas the secretion of Na⁺ and Cl^– was minimal (Figure 4). This result suggested that polyploidization of P. auriculata did not significantly affect composition of salt gland secretions.

Under control conditions, Ca²⁺ content in salt glands of diploids was significantly higher than that in salt glands of tetraploids, with content in diploids 4.5 times higher (Figure 5). Compared with the control, NaCl stress significantly increased Ca²⁺ accumulation in the salt glands of tetraploid plants, whereas it did not significantly affect Ca²⁺ accumulation in diploid salt glands. After adding CaCl₂, Ca²⁺ content in the salt glands of diploids under NaCl stress increased significantly compared with that in the NaCl treatment alone, whereas no significant change was observed in tetraploids, and then, there was no significant difference in Ca²⁺ content between diploid and tetraploid salt glands. The addition of the Ca²⁺ inhibitor LaCl₃ suppressed NaCl-induced accumulation of Ca²⁺ in tetraploid salt glands, resulting in no significant difference in Ca²⁺ accumulation compared with the control. Notably, Ca2+ content in the salt glands of diploids increased significantly with LaCl₃ treatment, surpassing that of both the NaCl treatment alone and tetraploid salt glands treated with LaCl₃. When the H₂O₂ inhibitor DPI or the H⁺-ATPase inhibitor Na₃VO₄ was added, Ca²⁺ content in the salt glands of both diploids and tetraploids decreased significantly compared with that in the NaCl treatment alone.

Salt stress induces strong and stable Na⁺ secretion. After salt stress, Na⁺ secretion rate increased significantly in both diploids and tetraploids and was 212.11 and 50.92 times higher, respectively, than that in the control. The Na⁺ secretion rate was 5.28 times higher in tetraploids than in diploids (Figures 6A, B). Thus, salt stress activated Na⁺ secretion in P. auriculata, and Na⁺ secretion capacity in individual salt glands under salt stress was greater in tetraploids than in diploids. Addition of CaCl₂ significantly increased Na⁺ secretion in diploid salt glands under salt stress, with secretion 1.8 times higher than that in the NaCl treatment alone. However, CaCl₂ addition significantly inhibited Na⁺ secretion in tetraploid salt glands, reaching 67.56% of the NaCl treatment alone. Nevertheless, the Na⁺ secretion rate in tetraploid salt glands remained significantly higher than that in diploid salt glands (Figure 6C). After addition of the Ca²⁺ channel inhibitor LaCl₃, Na⁺ secretion in tetraploids was severely inhibited, reaching only 6.5% and 9.68% of that in NaCl and CaCl₂ treatments, respectively, and with secretion only 3.26 times higher than that in the control. By contrast, the Na⁺ secretion rate in diploid salt glands increased significantly and was 2.60 and 1.44 times higher than that in NaCl and CaCl₂ treatments, respectively, and was 7.25 times higher than that in tetraploid salt glands under the same conditions (Figure 6D). Pretreatment with the H⁺-ATPase inhibitor (Na₃VO₄) or the H₂O₂ inhibitor (DPI) significantly decreased Na⁺ secretion rate under salt stress in both diploids and tetraploids, indicating the important roles of endogenous H⁺-ATPase and H₂O₂ in increasing Na⁺ secretion rate under salt stress (Figures 6E, F). Overall, the Na⁺ secretion rate in tetraploids was lower than that in diploids only in the LaCl₃+NaCl treatment, whereas in other treatments, the Na⁺ secretion rate was higher in tetraploids than in diploids (Figure 6G).

Salt stress induced a significant increase in K⁺ secretion rate in individual salt glands of both diploids and tetraploids, with the K⁺ secretion rate in tetraploids significantly higher than that in diploids by 3.20 times. However, the Na⁺/K⁺ secretion rate ratio was also higher in tetraploids, indicating stronger selective secretion of Na⁺ in tetraploids than in diploids under salt stress (Figures 7A, B). After addition of CaCl₂, K⁺ secretion in tetraploid salt glands under salt stress was inhibited, reaching 60.52% of that in the NaCl treatment, whereas K⁺ secretion in diploid salt glands increased significantly and was 3.51 times higher than that in the NaCl treatment (Figure 7C).

The K⁺ secretion rate was inhibited in tetraploid plants by the addition of LaCl₃ compared with NaCl alone due to the inhibitory effect of LaCl₃ on K⁺ channels (Wegner et al., 1994). However, it increased in diploid plants. This may be due to the fact that in addition to the inhibitory effect of LaCl₃ on K⁺ channels the presence of large amounts of retained Ca²⁺ in the salt glands increased the overall secretion level including the K⁺ secretion rate (Figure 7D). Notably, the addition of the H⁺-ATPase inhibitor (Na₃VO₄) or the H₂O₂ inhibitor (DPI) inhibited K⁺ secretion in tetraploid salt glands under salt stress, whereas in diploid salt glands, secretion increased significantly (Figures 7E, F). Overall, the K⁺ secretion rate in tetraploid salt glands was higher than that in diploid salt glands only in the NaCl treatment, and in other treatments, the rate was lower in tetraploids than in diploids (Figure 7G).

There may be a positive correlation between salt gland size and secretion rate (Feng et al., 2014; Mi et al., 2021). Polyploidization in P. auriculata significantly increased the size of salt glands, and under normal growth conditions, the volume of secreted crystals also increased significantly. It is hypothesized that polyploidization enhances the secretory capacity of individual salt glands of P. auriculata under normal growth conditions.

Under control conditions, Ca²⁺ content in tetraploid salt glands was only 22.22% of that in diploid glands, indicating that tetraploids decreased the allocation of Ca²⁺ to salt glands and increased that to leaf mesophyll cells for plant growth and metabolism (Dayod et al., 2010; Gilliham et al., 2011; Nomura and Shiina, 2014). In a previous study, under control conditions, P. auriculata had significantly lower Ca²⁺ secretion rates in tetraploid leaves than in diploid leaves, only reaching 33% of the diploid secretion (Duan et al., 2023). Therefore, the decrease in Ca²⁺ allocation to salt glands might be an important factor contributing to the decrease in Ca²⁺ secretion rate in tetraploids compared with that in diploids under control conditions. Salt stress stimulates Ca²⁺ influx into cells (Dong et al., 2022) but also inhibits root Ca²⁺ uptake (Guo et al., 2022), leading to a decrease in overall Ca²⁺ content in plants (Guo et al., 2015). However, although Ca²⁺ content showed no significant change in diploid salt glands under salt stress, it increased significantly in tetraploid salt glands, indicating that tetraploids significantly increased Ca²⁺ transport to salt glands.

With loss of Ca²⁺, selective transport of Ca²⁺ to salt glands requires increases in energy consumption (Dassanayake and Larkin, 2017). Under control conditions, tetraploids decreased the allocation of Ca²⁺ to salt glands, thereby reducing the energy expenditure for transport and retaining more Ca²⁺ in leaf mesophyll cells as essential nutrients and signaling molecules for normal plant growth and metabolism (Dayod et al., 2010). Under salt stress, Ca²⁺ promotes Na⁺ secretion (Mahajan et al., 2008; Guo et al., 2009), and increased allocation of Ca²⁺ to salt glands might be the reason for higher Na⁺ secretion rate in tetraploids (Duan et al., 2023).

Ca²⁺ content in salt glands is affected by both Ca²⁺ transport from the chloroplasts and Ca²⁺ secretion from the salt glands. No direct evidence has been obtained to confirm the pathway of salt transport into the salt gland, but ion carriers or channels, plasmodesmata and vesicles may play an important role in transporting (Yuan et al., 2016). Ca²⁺ influx in plant cells is predominantly mediated by NSCC, and it has been shown that Ca²⁺ channel inhibitors do not only block Ca²⁺ influx but also Ca²⁺ efflux (Zherelova et al., 1994), and La³⁺ can also block outward current (Terry et al., 1992). The diploid salt glands secreted large amounts of Ca²⁺ under control conditions, and the addition of LaCl₃ resulted in a significant increase in the Ca²⁺ content of the salt glands, probably due to the inhibition of Ca²⁺ secretion leading to its accumulation in the salt glands. At this time, we found that the Ca²⁺ content in diploid salt glands was significantly higher than that in tetraploids, and the rate of Na⁺ secretion from salt glands was also significantly higher than that in tetraploids, so we hypothesized that the Ca²⁺ content of salt glands was one of the important factors affecting the rate of Na⁺ secretion.

It was shown that Ca²⁺ promotes the secretion of Na⁺ from salt glands of recretohalophytes (Ding et al., 2010; Maeda, 2019). In our previous study on whole-leaf secretion rates in P. auriculata under salt stress, Ca²⁺ content in diploid leaves was higher than that in tetraploid leaves, but the Na⁺ secretion rate was lower in diploids than in tetraploids (Duan et al., 2023). Besides, it has been suggested in the other research that K⁺ in the salt gland promotes salt gland NaCl secretion (Feng et al., 2015). After salt stress, Ca²⁺ content in diploid salt glands was significantly lower than that in tetraploid salt glands, and the Na⁺ secretion rate in individual diploid salt glands was also significantly lower. Therefore, it was hypothesized that Ca²⁺ content in salt glands is a key factor determining salt gland Na⁺ secretion rate, not the overall Ca²⁺ content in leaves. Besides, increasing Na⁺ secrete rate is companied with increased Ca²⁺ content in salt gland of tetraploid. The plasma membrane Na⁺/H⁺ antiporter (SOS1) been shown to be involved in Na⁺ secretion in recretohalophyte with SOS1 genes up-regulated after salt treatment and Na⁺ secretion rate was decreased when Silencing SOS1 (Guo et al., 2020; Zhang et al., 2017). The higher secretion rate in tetraploid in salt glands under salt stress might suggested higher SOS1gene expression, and the molecular mechanisms require further research.

Notably, although addition of CaCl₂ did not significantly affect Ca²⁺ content in tetraploid salt glands, the Na⁺ secretion rate in tetraploids was significantly lower than that in the NaCl treatment alone. It was presumed that because Ca²⁺ content of salt glands did not increase, the decrease in Na⁺ secretion from salt glands was due to CaCl₂ reducing the activation of Na⁺ secretion by NaCl stress (Mulaudzi et al., 2020).

The Ca²⁺ ion is an important secondary messenger that regulates H⁺-ATPase activity and H₂O₂ generation (Demidchik and Shabala, 2018; Shabala, 2019). In the present study, we found that inhibition of H⁺-ATPase activity or H₂O₂ production inhibited both diploid and tetraploid Na⁺ secretion rates in P. auriculata. Ca²⁺ may also indirectly regulate Na⁺ secretion through H⁺-ATPase and H₂O₂ (Kong et al., 2016; Wang et al., 2020), and the mechanisms need to be further investigated.

Although the mechanism of salt secretion in recretohalophytes remains not fully understood, salt secretion is an active transport process that requires H⁺-ATPase to provide energy (Chen JA et al., 2010). Therefore, inhibiting H⁺-ATPase activity significantly suppresses ion secretion in salt glands. In this study, inhibiting H⁺-ATPase significantly inhibited Na⁺ secretion in both diploids and tetraploids. Notably, under salt stress, inhibiting H⁺-ATPase activity suppressed K⁺ secretion in tetraploids but significantly increased it in diploids. Therefore, it was hypothesized that the K⁺ efflux channels DA-KORCs and DA-NSCCs, activated by plasma membrane depolarization, were involved in K⁺ secretion in P. auriculata salt glands (Shabala et al., 2006; Sun et al., 2009). Therefore, inhibiting H⁺-ATPase activity induced severe depolarization of the plasma membrane in diploid salt gland cells, activating DA-KORCs and DA-NSCCs efflux channels (Goncalves et al., 2000) and increasing the rate of K⁺ secretion. Additionally, Ca²⁺ inhibits the activation of the K⁺ efflux channels DA-KORCs and DA-NSCCs by depolarization, and the decrease in Ca²⁺ content in salt glands after H⁺-ATPase inhibition contributed to the increase in K⁺ efflux and increased K⁺ secretion. Hydrogen peroxide is a signaling molecule that increases H⁺-ATPase activity (Lu et al., 2013), and the addition of the NADPH oxidase inhibitor DPI, which inhibits H₂O₂ production, increases the loss of K⁺ caused by depolarization of the plasma membrane (Ma et al., 2012). By contrast, the secretion of K⁺ was inhibited in tetraploids following inhibition of H⁺-ATPase or H₂O₂, suggesting that the effect of depolarization-induced K⁺ efflux on tetraploids was relatively minor than diploids.