Acidic soils that limit crop growth and productivity are often observed all over the world, especially in the tropics and subtropics. Approximately 30% of the world's ice-free land is acidic, and approximately 12% of crops are cultivated on acidic soils (von Uexküll and Mutert, 1995). What is worse, soil acidification is becoming an increasingly major problem due to the improper application of chemical fertilizers—particularly the overuse of nitrogen (N) fertilizers—alongside acid rain and intensive agriculture and monoculture (Wu et al., 2013; Yang et al., 2013). The effects of aluminum (Al)-toxicity—a major factor limiting crop productivity on acidic soils—on plants have drawn widespread attention, but few studies have investigated the damage to plants from low pH (Yang et al., 2015).

Poor crop growth and yield on acidic soils is usually due to the combination of toxicities of H⁺, Al, and manganese (Mn) and a lack of nutrients—namely phosphorus (P), calcium (Ca), magnesium (Mg), potassium (K), and molybdenum (Mo)— and a reduced uptake of water (von Uexküll and Mutert, 1995; Bian et al., 2013). In tropical America, over 70% of the acidic soils display Al-toxicity and Mg and Ca deficiencies, and almost all the acidic soils are P-deficient or have a high P-fixation capacity (George et al., 2012). For example, Zhang et al. (2014) showed that pH 3.0 decreased the uptake and utilization efficiency of P in Juglans regia seedlings. Forest ecosystems with acidic soils are often restricted by low Ca and Mg availability (St Clair and Lynch, 2005). Schubert et al. (1990) showed that transferring Vicia faba plants from pH 7 to pH 4 led to the reduced uptake of N, P, K, Ca, Mg, and sulfur (S). Malkanthi et al. (1995) observed that the levels of K, Ca, Mg, Mn, and Zn in the roots and tops of wheat, barley, and chili plants were lower at pH 3.8 than at pH 5.5. Similarly, the K, Ca, Mg, and Mn levels in Pinus pinaster roots and needles were lower at pH 3.5 than at pH 4.5, 5.5, and 6.5, whereas the levels of P and Fe were higher at pH 3.5 and 4.5 than at pH 5.5 and 6.5 (Arduini et al., 1998). However, Anugoolprasert et al. (2012) reported that the uptake of N, P, K, Ca, and Mg, and their concentration in roots, leaflets, petioles and whole plant, were not altered over the range of pH 3.6 to 5.7 for 4.5 months; this possibly explains the normal growth of sago palm seedlings at pH 3.6. Kidd and Proctor (2001) have suggested that the direct toxicity of H⁺ was the primary cause of the poor growth in H⁺-intolerant plants growing in very acidic soils.

Low pH can affects plant water uptake. Kamaluddin and Zwiazek (2004) observed that low pH caused a large and rapid decrease in both the water flow rate and the hydraulic conductivity in seedling roots of paper birch (Betula papyrifera). A pH 4.5 decreased the whole-root water conductivity in the H⁺-sensitive maize cultivar Adour 250, but it did not in the H⁺-tolerant maize cultivar BR 201 F (Gunsé et al., 1997). Tournaire-Roux et al. (2003) showed that the inhibition of water hydraulic conductivity (water uptake) in Arabidopsis roots by anoxia was primarily caused by cytosol acidosis, while changing the pH between 5.5 and 8.0 of a root-bathing solution did not affect the cytosol pH nor the root water hydraulic conductivity. Finally, Yang M. et al. (2011) observed that a low pH decreased the water content in Eucalyptus roots, stems, and leaves.

Low pH also inhibits CO₂ assimilation in some plant species, including J. regia (Zhang et al., 2014), Eucalyptus (Yang et al., 2015), sugar maple (Acer saccharum) and red maple (Acer rubrum) (Ellsworth and Liu, 1994; St Clair and Lynch, 2005). St Clair and Lynch (2005) also reported that the base cation stimulation of photosynthesis in sugar maple on acidic soils was correlated with its foliar nutrient status. Ellsworth and Liu (1994) had earlier suggested that photosynthesis in sugar maple on acidic soils might be co-limited by N and Ca, or by Ca × Mg interactions. Yang M. et al. (2011) observed that a low pH decreased the chlorophyll (Chl) level in Eucalyptus leaves. Yang et al. (2015) further investigated the effects of low pH on leaf gas exchange and Chl in four vegetatively-propagated Eucalyptus clones (G9, G12, G3, and G4); they found that pH 3.0 decreased leaf photosynthesis, transpiration, and Chl level in the four clones as well as the leaf water-use efficiency (WUE) in the G4 leaves, but pH 3.0 did not affect WUE in the G9, G12, and G3 leaves. Zhang et al. (2014) reported that pH 3.0 decreased the leaf net photosynthetic rate, transpiration rate, actual quantum yield of the photosystem II (PSII) electron transport (Φ_PSII), whereas it increased leaf non-photochemical quenching (NPQ); however, pH 3 had no effect upon leaf stomatal conductance, photochemical quenching (qP), and the maximum PSII efficiency of dark-adapted leaves (F_v/F_m), thus leading the authors to conclude that non-stomatal factors played a role in the low pH-induced inhibition of photosynthesis. Nonetheless, pH 4.0 did not influence spatial heterogeneity of Chl fluorescence, F_v/F_m, Φ_PSII, and quantum yields of regulated (Φ_NPQ) and nonregulated (Φ_NO) energy dissipation in the leaves of Plantago algarbiensis and P. almogravensis (Martins et al., 2013a,c). Altering the pH between 5.7 and 3.6 did not reduce the Chl concentration, photosynthetic rate, stomatal conductance, and transpiration rate in sago palm leaves (Anugoolprasert et al., 2012). However, to our best knowledge, little is still known about the effects of low pH on PSII photochemistry (i.e., absorption flux, trapped energy flux, electron flux, and dissipated energy flux) of leaves.

Low pH can induce oxidative stress and electrolyte leakage via the enhanced production of active oxygen species (ROS). Martins et al. (2013b) found that lipid peroxidation (malondialdehyde, MDA) was elevated in the pH 4.0-treated P. algarbiensis shoots, but not in the pH 4.0-treated P. almogravensis ones, and that the activities of antioxidant enzymes were enhanced or not affected in the shoots of the two Plantago species—suggesting that the higher antioxidant enzyme activities were insufficient to protect the low pH-treated P. algarbiensis shoots against oxidative damage. In another experiment, Martins et al. (2011) observed that pH 4.5 led to an increase in the MDA level in P. algarbiensis roots and shoots and P. almogravensis roots, but not in P. almogravensis shoots. Yang M. et al. (2011) reported that low pH increased membrane permeability in Eucalyptus leaves. Hydroponic experimentation showed that pH 3.5 led to an accumulation of H₂O₂ and severe lipid peroxidation that was accompanied by an increased activity of ascorbate peroxidase (APX) and decreased activities of superoxide dismutase (SOD) and catalase (CAT) in the roots of two rice cultivars (Zhang et al., 2015). Cucumber roots treated with pH 4.5 had a higher level of MDA and activities of monodehydroascorbate reductase (DHAR), guaiacol peroxidase (GPX), APX, and glutathione reductase (GR), but had lower activities of Cu/Zn-SOD, than did the pH 6.5-treated roots (Shi et al., 2006). However, pH 4.0 did not affect H₂O₂, MDA and the total soluble protein levels, electrolyte leakage, protein oxidation, and the SOD, CAT, APX, and GPX activities in the roots and leaves of P. algarbiensis and P. almogravensis (Martins et al., 2013c).

Citrus plants are considered insensitive to acidic soils (Yuda and Okamoto, 1965). Fang et al. (2011) used a solution culture approach to investigate the effects of pH 1.0, 2.0, 3.0, 4.0, 5.0, and 6.0 on several citrus rootstock seedlings. At pH 1.0, all seedlings died within 10 days after treatment, but the pH 4-treated seedlings showed normal growth except for a yellow tip that occurred in some leaves within 30 days. Using sand and solution cultures, Guest and Chapman (1944) found that Citrus sinensis seedlings died within a few days at pH 2.0, but they were not killed for months at pH 2.5 and 3.0 though their growth was limited or negligible. Nevertheless, citrus do not thrive in trongly acidic soils, because serious problems may arise when the soil pH is 5.0 or lower (Chapman, 1968). Citrus will often display poor growth and have a shortened lifespan when cultivated on soil with a low pH and high active Al (Lin and Myhre, 1990). In China, most of the citrus are grown in acidic and strongly acidic soils. Li et al. (2015) reported that the pH values of 319 soils sampled from pummelo (Citrus grandis) orchards in Pinghe, Zhangzhou, China had an average value of 4.34 and ranged from 3.26 to 6.22, with up to 90.0% of the orchard soils having a pH lower than 5.0. So far, however, only a handful of reports have empirically investigated the effects of low pH on citrus growth (Yuda and Okamoto, 1965), mineral nutrient uptake (Randhawa and Iwata, 1968; He et al., 1999; Li et al., 2015), and ROS metabolism alongside a few other physiological parameters (Fang, 2011). Randhawa and Iwata (1968) reported that the N, Ca, and Mg (Ca, Mg, and P) levels decreased in the leaves (roots), whereas the K level increased in the roots and leaves of Citrus natsudaidai seedlings, as the pH decreased from 7.0 to 4.0. He et al. (1999) observed that Fe, Zn, and Mn (Ca) in grapefruit (Citrus paradisi) leaves increased (decreased) with decreasing soil pH. The concentration of P and Ca in pummelo leaves decreased with decreasing soil pH (Li et al., 2015). Fang (2011) found that the activities of SOD, GPX, and CAT and the level of total soluble proteins displayed an upward trend, as a whole, as the pH decreased from 6.0 to 2.0; in contrast, the level of MDA decreased first to reach its lowest value at pH 4, but then increased as the pH decreased further.

The objectives of this work were (a) to determine how low pH affects gas exchange, related physiological parameters, and the mineral nutrient profiles in citrus seedlings; and (b) to understand the mechanisms by which low pH may lead to a decrease in leaf CO₂ assimilation.

Low pH very often affects the uptake of nutrients and water by plants (Kamaluddin and Zwiazek, 2004; Bian et al., 2013). As expected, pH 2.5 lowered the water uptake in citrus, as indicated by the reduced root and leaf RWC (Figures 9A,F). This result is supported by the finding that the water content was decreased in low pH-treated Eucalyptus roots, stems, and leaves (Yang M. et al., 2011). As shown in Figures 9–13, the uptakes of mineral nutrients were greatly altered at pH 2.5. Compared with pH 5, the pH 2.5 lowered N, P, K, Ca, and Mg uptake per plant or root DW, and the S uptake per plant. Low pH (4.0 relative to 7.0) induced decreases in the N, P, K, Ca, and Mg uptake per plant in V. faba (Schubert et al., 1990). Similarly, Malkanthi et al. (1995) observed that a pH 3.8 (relative to 5.5) decreased the K, Ca, and Mg uptake per plant in wheat, barley, and chili, and likewise in cowpea for Ca and Mg uptake per plant. However, the uptake of N, P, K, Ca, and Mg in sago palm seedlings was not changed in the range of pH 3.6–5.7 over a 4.5-month period (Anugoolprasert et al., 2012). Thus, it appears that the effects of low pH on macronutrient uptake per plant depend on both the plant species identity and the H⁺ strength (i.e., pH value).

On the whole, apart from a few exceptions, the concentrations of N, P, K, Ca, and Mg were decreased in the pH 2.5-treated C. grandis and C. sinensis roots, stems, and leaves (Figure 11). This agrees with the report that pH 3 decreased Ca and Mg levels in Picea abies roots and needles (George et al., 2012); that P, K, and Mg levels were lowered in the low pH-treated V. faba roots and shoots (Schubert et al., 1990); that the levels of K, Ca, and Mg in the roots and tops of wheat, barley, and chili were lower at pH 3.8 than at pH 5.7 (Malkanthi et al., 1995); and that P and Ca levels in pummelo leaves increased linearly with increasing soil pH (Li et al., 2015). However, the levels of N, P, K, Ca, and Mg in the roots, leaflets, petioles, and whole plant of sago palm seedlings did not differ among pH 3.6, 4.5, and 5.7 (Anugoolprasert et al., 2012). The concentrations of N, P, K, Ca, and Mg might have been reduced in sago palm seedlings if the pH was lower than pH 3.6, because the concentration of N, P, K, Ca, and Mg in citrus roots, stems, and leaves were greatly reduced at pH 2.5 but little affected at pH 4 relative to pH 5 (Figure 11). In contrast, the S level was increased in the low pH-treated C. grandis and C. sinensis roots, stems, and leaves (Figures 11F,L,R), which is consistent with the finding that the S concentration in the tops of ginger, maize, wheat, French bean, and tomato plants was higher at pH 3.3 than at pH 4.0 (Islam et al., 1980).

So far, however, there is little published information available on the effects of low pH on plant micronutrients. H⁺-toxicity is thought to inhibit the uptake of cations (George et al., 2012). However, treatment with pH 2.5 did not lower Fe, Cu, Mn, and Zn uptake per root DW in the two citrus species, or the Mn uptake per plant in C. sinensis, when compared with pH 5 (Figures 13N,R,S,U,V). This result may be related to the insensitivity of citrus plants to acidic soils, as reported previously by Yuda and Okamoto (1965). Interestingly, the B uptake per plant or per root DW was reduced by a low pH (Figures 13O,T). This result is supported by a work showing that B could alleviate low pH-induced damage in Arabidopsis roots (Koyama et al., 2001).

The Fe, Mn, Cu, and Zn concentrations in the C. grandis and C. sinensis roots, stems, and leaves were all higher at pH 2.5 than at pH 5, or they were similar between the two treatments, though there was a lower level of Fe detected in the C. grandis leaves at pH 2.5 than at pH 5 (Figures 12A,B,D–G,I–L,N–Q). The observed higher Fe, Mn, Cu, and Zn concentrations in the pH 2.5-treated C. grandis and C. sinensis roots, stems, and leaves might be associated with a reduced dilution due to decreased growth (Figure 1) and with higher uptake per root DW (Figures 13R,S,U,V). As shown in Figures 12A,D,F,I,K,N, the root Fe and Cu concentrations were higher at pH 2.5 than those at the other treatment levels of pH, while no such results were observed for the leaf and stem Fe and Cu concentrations; this may be explained by the increased Fe and Cu distributions in the roots, and the decreased or unchanged Fe and Cu distributions in the leaves and stems, at pH 2.5 (Figures S2A,D,F,I,K,N). By contrast, the B level was decreased in the pH 2.5-treated C. grandis and C. sinensis roots, stems, and leaves (Figures 12C,H,M) likely due to the decreased B uptake per plant or root DW (Figures 13O,T).

In this experiment, many of the fibrous roots became rotten and the living roots turned abnormally dark brown when exposed to pH 2.5 (Figures 2A,D). Thus, it is reasonable to presume that H⁺-toxicity may directly damage citrus roots, thus affecting the uptake of vital mineral nutrients and water.

Our results showed that pH 2.5 lowered the root, stem, leaf, and whole plant DW (Figures 1, 2). The low pH-induced poor growth of citrus seedlings may be due to the combined interplay of direct H⁺-toxicity—as shown by the damaged roots (Figures 2A,D)—deficiencies of macronutrients—as indicated by the decreased N, P, K, Ca, and Mg concentrations (Figures 11A–E)—and uptake per plant or root DW (Figures 13A–E,G–K), and the decreased water uptake—as indicated by the decreased root and leaf RWC (Figures 9A,F).

In spite of the reduced growth at pH 2.5, no seedling deaths occurred in the two citrus species at each given pH during the entire experiment. Similar results have been obtained for several citrus rootstocks (Fang, 2011; Fang et al., 2011), as well as for C. sinensis seedlings (Guest and Chapman, 1944). Based on the present results, we conclude that the two citrus species studied were insensitive to low pH. This above conclusion is supported by the fact that most of physiological parameters monitored in Figures 3, 4, 7, 9 were altered only at pH 2.5, and that pH 4 had almost no influence on these parameters and the OJIP transients (Figure 6).

As shown in Figure 2B, mottled bleached leaves were observed only in the pH 2.5-treated C. grandis seedlings (Figure 2B). Furthermore, the pH 2.5-induced alterations of many physiological parameters shown in Figures 3, 4, 7, 9, and of the JIP transients (Figure 6), were slightly greater in C. grandis than in C. sinensis leaves. Evidently, when the results are taken together, seedlings of C. sinensis had a slightly higher tolerance to a low pH than did those of C. grandis. However, the difference in low pH tolerance between the C. grandis and C. sinensis species is apparently lower than the difference between them in their Al-tolerance (Yang L. T. et al., 2011; Jiang et al., 2015; Li et al., 2016). This latter discrepancy is supported by a study showing that plant races were separately adapted to Al³⁺ or low pH- (H⁺-) toxicity (Kidd and Proctor, 2001).

We found that pH 2.5 greatly inhibited the CO₂ assimilation in C. grandis and C. sinensis leaves, and that this inhibition was slightly greater in C. grandis than in C. sinensis leaves (Figure 3A). The pH 2.5-induced decrease in leaf CO₂ assimilation could not be explained only by decreased stomatal conductance, because the intercellular CO₂ concentration increased and stayed unchanged in pH 2.5-treated C. granddis and C. sinensis leaves, respectively (Figure 3C), and because leaf CO₂ assimilation decreased with the increasing intercellular CO₂ concentration (Figure 5B). Thus, the pH 2.5-induced decrease in leaf CO₂ assimilation in citrus may be primarily driven by non-stomatal factors.

In this context, the pH 2.5-induced decreases in Chl a, Chl b, and Chl a+b were probably not the main factor inhibiting leaf CO₂ assimilation because their reductions were much lower than that for leaf CO₂ assimilation (Figures 3A, 4A–C). This conclusion is supported by our results showing that DI_o/RC, DI_o/ABS, NPQ, and qNP were all elevated in the pH 2.5-treated C. grandis and C. sinensis leaves (Figures 7F,G,N,O).

The observed positive ΔL-band at ca. 130 μs in the OJIP transients from the pH 2.5-treated leaves (Figures 6E,J) suggested that the grouping (stability) of the PSII units and the energy exchange between the independent PSII units were both reduced (Strasser et al., 2004; Liao et al., 2015). This interpretation is further supported by our result that P_2G was decreased in the pH 2.5-treated leaves (Figure 7K). The appearance of a positive ΔK-band at 300 μs in the OJIP transients from the pH 2.5-treated leaves (Figures 6C,H) indicated that the oxygen evolving complex (OEC) had been damaged (Srivastava et al., 1997). The observed positive ΔJ- and ΔI-bands at 2 ms and 30 ms, respectively, in the OJIP transients from the pH 2.5-treated leaves (Figures 6C,H) suggested that the reduction of the PSII acceptor side had been elevated (Strasser et al., 2004). The amount of electrons from the RCs at the acceptor side depends on both the capacity of electron donation to the RCs and the capacity of the electron transport chain from the RCs to the electron acceptors. Based on these results, we conclude that at pH 2.5, the PSII acceptor side was more severely damaged than was the PSII donor side. We observed that pH 2.5 led to increased DI_o/RC, decreased F_v/F_m and ET_o/ABS (Figures 7F,H,I), and altered the OJIP transients (Figure 6) in leaves, together indicating that photoinhibition damaged the PSII complexes in these citrus leaves (Maxwell and Johnson, 2000; Force et al., 2003). In the present study, the pH 2.5-induced decrease in F_v/F_m was caused by an increased F_o, since the F_m slightly increased with decreasing pH (Figures 7A,B). The observed higher F_o in the pH 2.5-treated leaves was likely associated with an increased inactivation of the PSII RCs, as increased by the decreased qP (Figure 7M), and with the enhanced damage to OEC, as indicated by the positive ΔK-band (Figures 6C,H). Furthermore, the higher F_o may have arisen from the pH 2.5-induced accumulation of reduced Q_A (Bukhov et al., 1990), as indicated by the increased M_o (Figure 7D). In addition, the pH 2.5-treated leaves displayed decreased RE_o/ABS, PI_{tot, abs}, F_m′/F_v′, Φ_PSII, and ETR (Figures 7J,L,P–R). Obviously, treatment with pH 2.5 impaired the whole electron transport chain from the donor side of PSII to the reduction of the PSII end acceptors, thus decreasing ETR. Regression analysis showed that leaf CO₂ assimilation increased with increasing F_v/F_m, ET_o/ABS, RE_o/ABS, P_2G, PI_{tot, abs}, qP, F_m′/F_v′, Φ_PSII, or ETR, (Figures 8H–M,P–R). Based on these results, we conclude that pH 2.5 damaged the whole photosynthetic electron transport chain, thus inhibiting leaf CO₂ assimilation in seedlings of these two citrus species.

Light-driven ROS production can cause oxidative damage to vital photosynthetic components and thereby inhibit photosynthesis (Foyer and Shigeoka, 2011). We observed that pH 2.5 greatly increased the H₂O₂ production and the electrolyte leakage in C. grandis and C. sinensis leaves, though more so in the C. grandis leaves (Figures 9G,H), and that leaf CO₂ assimilation decreased with increasing leaf H₂O₂ production and electrolyte leakage (Figures 10B,F). Hence, the observed higher ROS production may be responsible for the pH 2.5-induced inhibition of photosynthesis in citrus leaves.

The leaf photosynthetic rate decreases with decreasing leaf RWC (Lawlor, 2002). However, the relative importance of stomatal and non-stomatal limitations to photosynthesis under water stress is not yet fully understood. Typically, as the RWC decreases, the stomatal limitation of photosynthesis will also decrease and the metabolic limitation will increase (Lawlor, 2002; Zhou et al., 2007), which entails limitations to ribulose-1,5-disphosphate (RuBP) regeneration (Gunasekera and Berkowitz, 1993), photophosphorylation (Tezara et al., 1999), and Rubisco activity (Maroco et al., 2002; Parry et al., 2002). Zhou et al. (2007) observed that water stress decreased photosynthetic rate, Rubisco activity, the energy flux via linear electron transport, and increased ΔpH- and xanthophyll-mediated thermal dissipation. Our results showed that the pH 2.5-induced decrease in leaf CO₂ assimilation (Figure 3A) was accompanied by decreases in root and leaf RWC (Figures 9A,F), leaf Rubisco activity (Figure 3F) and ETR, and by an increase in NPQ (Figures 7O,R). Furthermore, leaf CO₂ assimilation decreased with decreasing root RWC, leaf RWC (Figures 10C,D), Rubisco activity (Figure 5C), or ETR (Figure 8R), and with increasing NPQ (Figure 8O); leaf Rubisco activity (y) increased with increasing leaf RWC (y = −61.1653 + 0.8351x, r² = 0.9174, P < 0.0001). Thus, it is reasonable to assume that a low pH lowered the water uptake and induced water stress, thus inhibiting photosynthesis in the C. grandis and C. sinensis leaves.

A study has shown that the base cation-induced increase in sugar maple photosynthesis on acid soils was associated with an improved foliar nutrient status (St Clair and Lynch, 2005). Ellsworth and Liu (1994) suggested that leaf photosynthesis of sugar maple on acidic soils was co-limited by N and Ca, or by interactions of Ca with other nutrients, such as Mg. We observed that leaf CO₂ assimilation decreased with increasing leaf N, P, Ca, or Mg concentrations (N, P, Ca, or Mg uptake per plant) (Figures 14A,B,D,E,G,H,J,K). Thus, the pH 2.5-induced decreases in these nutrients might be responsible for the observed lower leaf CO₂ assimilation.

Our results also showed that the growth of seedlings (Figures 1, 2) and the status of many of their physiological parameters (Figures 3, 4, 7, 9, 11, 12) reached their maximum at pH 5. This seems to contradict the early view that serious problems for citrus might arise when the soil pH was 5.0 or lower (Chapman, 1968). In our study, citrus seedlings were grown under favorable conditions of mineral nutrients and the direct toxicity of H⁺ might be the primary cause for the poor seedling growth. However, a significant difference might also occur when citrus are grown on acidic soils due to the increased solubility of Al and Mn, and/or decreased availability of P, Ca, Mg, and Mo (George et al., 2012; Kochian et al., 2015; Li et al., 2015). Thus, the optimum pH for citrus might be higher in a soil culture than when grown in solution or a sand culture (Yuda and Okamoto, 1965). These findings indicate that suitable fertilizers might alleviate the toxicity of acidic soils upon citrus. Adjusting the soil nutrients via careful fertilization should contribute to greater harvest yield and the sustainable management of citrus across a range of acidic soils.

Our results demonstrate that citrus seedlings were insensitive to low pH, and that C. sinensis is slightly more tolerant to this low pH than is C. grandis. H⁺-toxicity could directly damage the citrus roots, thus affecting their uptake of mineral nutrients and water. The results suggest that the low pH-induced inhibition of growth was caused by the combination of H⁺-toxicity, deficiencies of nutrients, and decreased water uptake. Here, only pH 2.5 noticeably inhibited leaf CO₂ assimilation, which was probably due to the combination of an impaired photosynthetic electron transport chain, increased ROS production, and decreased uptake of water and nutrients. In sum, these findings increase our understanding of the factors by which a low pH can decrease citrus growth, and of the mechanisms by which low pH inhibits leaf CO₂ assimilation.

AL performed most of the experiment and drafted the manuscript; JZ participated in the measurements of photosynthesis and fluorescence; LY participated in the direction of this study; XY and NL participated in the analysis of the nutrient elements; LT and DL participated in the cultivation of the experimental seedlings; LC designed and directed the study and also revised the manuscript. All authors have read and approved the final manuscript.

This work was financially supported by an earmarked fund for the China Agriculture Research System (No. CARS27).