Introduction

Sugarcane is one of the most important agricultural products of the Nansei Islands, which is located in southwest Japan. However, sugarcane production in this region remains sluggish, mainly due to a reduction in the harvested area and the number of growers. Therefore, technical improvement in both the yield and quality of sugarcane is needed. Fertilizer management practice is one of the factors that affect sugarcane quality (Meyer and Wood 2001). Watanabe et al. (2016a) reported that K+ and Cl− are the most abundant ions in sugarcane juice that are negatively correlated with juice sucrose concentration. They concluded that K+ and Cl− negatively affect sugarcane quality. Watanabe et al. (2016b) conducted pot experiments with variable potassium levels using potassium chloride (KCl) and potassium sulfate (K2SO4) to examine the effects of K+ independent of Cl−. They observed a significant reduction in juice sucrose concentrations only when plants were treated with higher doses of KCl, suggesting that Cl−, not K+, is more responsible for sucrose reduction. However, it is difficult to conclude whether the combination of K+ and Cl− or only Cl− is involved in this process since both the ions are closely associated, with Cl− considered to be the counterion of K+ (Laties et al. 1964; Stuart and Jones 1978; Schnabl and Raschke 1980). In order to improve the quality through fertilizer management practices, it is important to determine whether we need to lower both K+ and Cl− or Cl− alone. Some studies have revealed specific ion toxicities by separately using different types of salt, e.g., individual effects of Na+ and Cl− (Joshi and Nail 1980; Sümer et al. 2004; Tavakkoli et al. 2011; García and Medina 2013). Therefore, the use of different types of Cl− salts could help in examining the effects of only Cl−. Moreover, the process of sucrose reduction has not been fully characterized. Sucrose is one of the major products of photosynthesis (Fewkes et al. 1971), so sucrose reduction by KCl overdose may occur through changes in the physiological functions associated with photosynthesis, for example, the inhibition of CO2 assimilation by stomatal closure (Meinzer et al. 1994; Du et al. 1996; Gomathi et al. 2010; Vasantha et al. 2010; Ribeiro et al. 2013), metabolism impairment (Meinzer et al. 1994; Du et al. 1996; Gomathi et al. 2010; Markelz et al. 2011; Ribeiro et al. 2013), and accumulation of carbohydrates in leaves (Hartt and Burr 1967; Goldschmidt and Huber 1992; Paul and Pellny 2003). The objectives of this study were to determine the individual effects of Cl− on sugarcane quality and investigate a possible mechanism for sucrose reduction in terms of photosynthesis, using different kinds of K+ and Cl− salts.

Materials and Methods

A pot experiment was conducted under greenhouse conditions at the University of the Ryukyus, Okinawa, Japan (26°25′N, 127°77′E; 125 m a.s.l.) from April to November 2015. Seedlings of a commercial sugarcane cultivar (Saccharum spp. cv. NiF8) were collected from the fields at the Subtropical Field Science Center of the University of the Ryukyus. One-bud seedlings were immersed in a solution of Benlate-R (5 g L−1, Sumitomo Chemical) and in tap water for 24 h each for sterilization and to improve germination rate. These seedlings were planted and grown in containers from April 11 to May 13, 2015. After the first fully expanded leaves were confirmed, seedlings were transplanted into 1/2000a Wagner pots filled with soil mixture composed of three materials: dark red soil (Shimajiri mahji), sea sand, and peat moss (1:1:1, v v−1). Tillers were immediately removed after emergence. Irrigation was carefully performed with tap water (Na+, 19; K+, 1; Mg2+, 3; Ca2+, 9; Cl−, 29; SO4 2−, 8 mg L−1) through daily soil moisture evaluation to prevent water stress.

Fertilization was performed once a week after transplantation using a modified Hoagland solution containing 6 mM Ca(NO3)2, 4 mM KNO3, 2 mM KH2PO4, 2 mM MgSO4, 100 μM C10H12FeN2NaO8, 25 μM H3BO3, 10 μM MnSO4, 2 μM ZnSO4, 0.5 μM CuSO4, and 0.5 μM H2MoO4. The treatments were started 7 weeks after transplantation, and five treatments were established by dissolving different types of K+ and Cl− salts: 50 mM KCl, 25 mM K2SO4, a mixture of 12.5 mM magnesium chloride (MgCl2) and 12.5 mM calcium chloride (CaCl2), and 50 mM sodium chloride (NaCl) (the treatments as well as the plants treated with these solutions are referred to as KCl, K2SO4, MgCl2 + CaCl2, and NaCl, respectively) into the Hoagland solution. The Control plants were given only the Hoagland solution. The concentrations of these salts in the Hoagland solution were determined based on the results of a pretest from July 2014 to March 2015, in which we had applied 2, 10, and 50 mM of K+ either by KCl or K2SO4. We observed a significant reduction in sucrose following 50 mM KCl application, while K2SO4 application barely affected the sucrose concentration (Fig. 1). Hence, the salt treatments contained 50 mM K+ and/or 50 mM Cl− given in different forms. Eight pots for each of the treatments were prepared.

Fig. 1
figure 1

Effects of different levels of K+ (2, 10, and 50 mM) either by KCl or K2SO4 on sucrose concentration in juice of the pretest from July 2014 to March 2015. Values are means of each of the treatments and vertical bars indicate SD (n = 4). Different letters mean significant differences at the 5 % level by the Tukey test

Full size image

Stem height from the ground to the base of the top visible dewlap leaf, number of green leaves, and soil–plant analysis development (SPAD) value of the top visible dewlap leaf using a SPAD meter (SPAD-502, Minolta Camera) were measured monthly to evaluate the effects of the treatments on plant growth. Four plants were harvested on September 19 (T 1) and then on November 19 (T 2), and the weight of millable stalks and leaf area was measured. Based on their length, the stems were divided into three parts: upper, middle, and lower, and the internodes located at the center of each part were squeezed separately. Juice samples were diluted 50-fold with extra-pure water and passed through a membrane filter (diameter, 13 mm; pore size, 0.45 μm; Advantec) for ion and sugar analyses.

Photosynthesis and chlorophyll fluorescence measurements were performed on the second to the fourth visible dewlap leaves before and 1 month after the treatments started (July and August, respectively) and before T 1 (September) and T 2 (November), using an infrared open gas exchange system (Li-6400, LI-COR). Minimum fluorescence (F o) and maximum fluorescence (F m) were measured before dawn, and variable fluorescence (F v, F m − F o) and the maximum photochemical quantum yield of photosystem II (F v/F m) were calculated. The CO2 assimilation rate (A), stomatal conductance (g s), and intercellular CO2 concentration (C i) were measured at a photosynthetic photon flux density of 2000 μmol m−2 s−1, using a constant flow rate of 400 μmol s−1. The mean ± SD of leaf temperature and vapor pressure deficit based on leaf temperature were 35.2 ± 1.1 °C and 2.2 ± 0.5 kPa, respectively. The measurements were performed at 400 (A 400) and 2000 (A 2000) μmol mol−1 CO2 to estimate stomatal limitation (SL), calculated using the following formula by referring to Long and Bernacchi (2003) with a slight modification.

$$ {\text{SL}} = \left( {A_{2000} {-}A_{400} } \right)/A_{2000} $$

One day after the photosynthesis measurements in September and November, approximately 10 cm2 of green leaf pieces was taken from the measured leaves at 08:00, 14:00, and 20:00 h. This temporal sampling was necessary because sugar compositions in sugarcane leaves undergo diurnal changes (Du et al. 2000). Freeze-dried leaf samples were cut into small pieces, and 50 mg of a sample was extracted with 3 ml of boiled water for 10 min. After centrifugation at 3000 rpm for 10 min, the supernatant was collected. This cycle was repeated twice with 1 ml of boiled water, and then 5 ml of the combined supernatant was used for sucrose, glucose, and fructose analyses. The pellet was retained for starch analysis. Starch was extracted following Du et al. (1998), and the measurement was performed photometrically (UV-1800, Shimadzu) using a modified enzymatic method (Jones et al. 1977). Starch concentration was expressed as glucose weight. The extracts of leaf samples at 14:00 h were used for ion analysis.

The concentrations of Na+, K+, Mg2+, Ca2+, Cl−, and SO4 2− were determined by ion chromatographs (ICS-1600, Thermo Fisher Scientific). The columns and eluents used for cation analysis were Ion Pac CS12 and 20 mM methane sulfonic acid solution, respectively, and those for anion analysis were Ion Pac AS22 and a solution of 4.5 mM sodium carbonate and 1 mM sodium hydrogen carbonate, respectively. The concentrations of sucrose, glucose, and fructose were determined by high-pressure liquid chromatography (Shimadzu) with a liquid chromatograph (LC-20AD), a column oven (CTO-20A), a refractive index detector (RID-10A), and an autosampler (SIL-20A). The column and eluent used were SCR 101-H and degassed extra-pure water, respectively. Ion and sugar concentrations in juice were expressed as the averaged values of the three stem parts.

Statistical analysis was performed using the software R (R Core Team 2015). Data were subjected to one-way analysis of variance between the treatments. When significances were found, the Tukey test was conducted, and the significant differences were accepted based on a P value <0.05.

Results

Visual observation suggested that all the plants were growing well. Stem height gradually increased and finally reached 250–280 cm (Fig. 2). At T 2, the Control had a significantly higher stem height value than K2SO4 and NaCl. Stem weight at T 2 ranged from 1007 to 1152 g and was the highest in Control and lowest in NaCl, but there was no significant treatment effect (data not shown). The number of green leaves and SPAD value showed similar changes in all the treatments, except that the numbers of green leaves in KCl were significantly lower than those in Control and MgCl2 + CaCl2 at 14 weeks after transplantation and in NaCl at 18 weeks after transplantation. The leaf area was approximately 5000 cm2 and did not differ significantly between the treatments (data not shown).

Fig. 2
figure 2

Change of growth parameters during the experiment with different kinds of K+ and Cl− salts. Values are means of each of the treatments (n = 4). Asterisk means a significant difference at the 5 % level

Full size image

Juice ion compositions were similar at T 1 and T 2, except that K+ and Cl− concentrations increased in all the treatments from T 1 to T 2 (Table 1). KCl and K2SO4 accumulated a great deal of K+. Juice Cl− concentration was remarkably high only in KCl, and the other Cl− salt treatments, namely MgCl2 + CaCl2 and NaCl, had lower Cl− concentrations than K2SO4, although those were all significantly higher than that of Control. SO4 2− concentration was low in KCl and high in K2SO4, and the other treatments had similar values. We did not see the effects of MgCl2 + CaCl2 and NaCl as these salts barely affected Mg2+, Ca2+, and Na+ concentrations.

Table 1 Effects of different kinds of K+ and Cl− salts on ion composition in juice at T 1 and T 2
Full size table

Sucrose concentration ranged from 16.0 to 18.0 % at T 1 and from 17.0 to 20.6 % at T 2 (Fig. 3). At T 1, sucrose concentration in K2SO4 was the highest and was significantly higher than those in KCl and NaCl. Compared to the values at T 1, sucrose concentration increased in all the treatments at T 2; however, the increment was smallest in KCl. Sucrose concentrations in MgCl2 + CaCl2 and NaCl rose over 20 % and were significantly higher than in KCl. At both T 1 and T 2, sucrose concentration in KCl was the lowest. The concentration of reducing sugar, which was composed of glucose and fructose, decreased from 0.6–1.4 % at T 1 to 0.2–0.4 % at T 2. Reducing sugar concentration was significantly lower in KCl and K2SO4 at T 1, with KCl decreasing to 0.2 %, while K2SO4 stayed at the same level, resulting in significant differences between KCl and other treatments at T 2. Similar results to sucrose concentration were obtained for total sugar concentration: KCl had a significantly lower total sugar concentration than plants treated by other treatments except for Control at T 2.

Fig. 3
figure 3

Effects of different kinds of K+ and Cl− salts on sugar composition in juice at T 1 and T 2. Values are means of each of the treatments and vertical bars indicate SD (n = 4). Within each of the sampling times (T 1 and T 2), different letters mean significant differences at the 5 % level by the Tukey test

Full size image

The K+ and Cl− salt treatments tended to increase leaf K+ and Cl− concentrations, respectively; yet, the differences were less significant than juice ion composition (Table 2). MgCl2 + CaCl2 and NaCl also had some effects on leaf ion compositions, and the results were clearer at T 2 than at T 1: high leaf Mg2+ and Ca2+ concentrations in MgCl2 + CaCl2 and high Na+ concentration in NaCl, respectively.

Table 2 Effects of different kinds of K+ and Cl− salts on ion composition in leaf at T 1 and T 2
Full size table

In the leaf, sucrose concentration was low at 08:00 h, increased at 14:00 h, and decreased again at 20:00 h, whereas starch concentration constantly increased from morning to night (Fig. 4), which is consistent with Du et al. (2000). Reducing sugar remained relatively stable irrespective of the sampling time. All of the sugar concentrations, especially reducing sugar, tended to decrease from T 1 to T 2. KCl had the highest starch concentration at T 2, and only small differences between treatments were observed.

Fig. 4
figure 4

Effects of different kinds of K+ and Cl− salts on sugar composition in leaf at T 1 and T 2. Values are means of each of the treatments (n = 4). Asterisk means a significant difference at the 5 % level

Full size image

A linearly decreased as plants matured (Fig. 5). The differences between the treatments became greater, and A 400 ranged from 25.2 to 30.5 μmol m−2 s−1 and A 2000 from 29.4 to 35.4 μmol m−2 s−1 in November. All the treatments had similar changes in g s and C i at 400 μmol mol−1 of atmospheric CO2 concentration. C i varied depending on the treatments but with no clear pattern. The increasing CO2 level caused an increase in C i and reduction in g s, which resulted in higher A. Both A 400 and A 2000 were the lowest in KCl in September and November although the differences were not significant. SL increased from 0.06 in July to 0.08–0.16 in the later periods, and we observed a significant difference between KCl and NaCl only in August. F v/F m was barely changed by the treatments and was always 0.78–0.80 irrespective of the date and treatment, and no significances were found.

Fig. 5
figure 5

Effects of different kinds of K+ and Cl− salts on A, g s, and C i. Values are means of each of the treatments (n = 4). Note that the scales for y axis in C i are different

Full size image

Discussion

This study was conducted using different types of K+ and Cl− salts to examine the individual effects of Cl− on sugarcane. Many salinity stress studies, mainly using NaCl, have been conducted in sugarcane, and it is generally reported that NaCl stress induces inhibition of growth (e.g., Wahid 2004), yield (e.g., Akhtar et al. 2001), and photosynthesis (e.g., Medeiros et al. 2014), as well as quality deterioration (e.g., Gomathi and Thandapani 2005) with accumulation of Cl−. In the present study, however, visual observation and quantitative parameters revealed no obvious treatment effects on sugarcane growth (Fig. 1). Nevertheless, ion and sugar compositions in the juice were greatly affected, indicating that the treatment effects observed in this study differed from those induced by common salt stress.

KCl application significantly increased both K+ and Cl− concentrations in juice (Table 1) as previously reported (Watanabe et al. 2016b); yet the other Cl− salt treatments, MgCl2 + CaCl2 and NaCl, did not induce sugarcane to absorb as much Cl− as KCl (Table 1). Notably, the counter ions of Cl−—Mg2+, Ca2+, and Na+—were barely absorbed, but K+ concentrations in MgCl2 + CaCl2 and NaCl were higher than that in Control (Tables 1, 2). On the other hand, K+ accumulation was markedly enhanced by K2SO4 application (Table 1), which led to increasing Cl− concentration in accordance with Watanabe et al. (2016b). These results suggest that K+ can primarily move into sugarcane alone, but Cl− needs K+ to be efficiently absorbed and other cations, such as Mg2+, Ca2+, and Na+, hardly accompany Cl−. As a result, only the KCl treatment lowered sucrose concentration (Fig. 3) with an increase in Cl− in juice. Therefore, from the present study, we could not evaluate the effects of Cl− independent of K+ accumulation and conclude whether Cl− alone or both K+ and Cl− have adverse effects on sugarcane quality.

Although the differences were not significant, both A 400 and A 2000 were lower in KCl (Fig. 5). We observed the decline of neither stomatal conductance nor F v/F m (Fig. 6), but KCl showed a slightly higher leaf starch concentration in the later growth period, which may have resulted in the lower A 400and A 2000 readings. This indicates the possibility that the sucrose reduction was induced through feedback regulation of photosynthesis because starch accumulation in leaves is correlated with a decline in CO2 fixation (Hartt and Burr 1967; Goldschmidt and Huber 1992; Paul and Pellny 2003). However, the results are not strong enough to prove this hypothesis, and Lunn and Hatch (1995) found no evidence for feedback inhibition of photosynthesis as carbohydrate accumulated in leaves of various C4 plants; thus, a more detailed study is needed to confirm this hypothesis.

Fig. 6
figure 6

Effects of different kinds of K+ and Cl− salts on SL and F v/F m. Values are means of each of the treatments (n = 4). Asterisk means a significant difference at the 5 % level

Full size image

Furthermore, given that sucrose is not only one of the major products of photosynthesis (Fewkes et al. 1971) but also a translocated photosynthate and a soluble reserve substance (Stewart et al. 1973), the physiological functions related to translocation and distribution of sugar may have also been affected. The sucrose reduction, however, was not simply induced by the decomposition of sucrose to its components, glucose and fructose because the reducing sugar concentration in juice also markedly declined in KCl (Fig. 3). To further characterize this, studies assessing the effects of KCl application on sucrose accumulation are currently underway.

The Cl− salt treatments other than KCl, namely MgCl2 + CaCl2 and NaCl, showed lower juice Cl− concentrations than plants in K2SO4 (Table 1) that were given no Cl− through fertilization. Also, in these two treatments, concentrations of up to 50 mM Cl− did not adversely affect sugarcane quality, but sucrose concentration in juice declined with 50 mM KCl treatment (Fig. 3). Taken together, it appears that juice Cl− concentration is determined by available K+ rather than by available Cl− in soil and that Cl− could have a negative impact on sugarcane quality only when K+ is supplied above sufficiency levels. Therefore, we need to reduce K+ as well as Cl− concentrations to improve sugarcane quality through fertilizer management practices when both K+ and Cl− are excessively accumulated in the environment.