E ﬀ ect of Nutrient Solution Concentration on the Growth of Hydroponic Sweetpotato

: Nutrient solution concentration (NSC) is a critical factor a ﬀ ecting plant growth in hydroponics. Here, we investigated the e ﬀ ects of hydroponic NSC on the growth and yield of sweetpotato ( Ipomoea batatas (L.) Lam.) plants. First, sweetpotato cuttings were cultivated hydroponically in three di ﬀ erent NSCs with low, medium, or high electrical conductivity (EC; 0.8, 1.4, and 2.6 dS m − 1 , respectively). Shoot growth and storage root yield increased at 143 days after plantation (DAP), depending on the NSC. Next, we examined the e ﬀ ect of NSC changes at half of the cultivation period on the growth and yield, using high and low NSC conditions. In plants transferred from high to low EC (HL plants), the number of attached leaves increased toward the end of the ﬁrst half of the cultivation period (73 DAP), compared with plants transferred from low to high EC (LH plants). Additionally, the number of attached leaves decreased in HL plants from 73 DAP to the end of the cultivation period (155 DAP), whereas this value increased in LH plants. These changes occurred due to a high leaf abscission ratio in HL plants. The storage root yield showed no signiﬁcant di ﬀ erence between HL and LH plants. Our results suggest that the regulation of hydroponic NSC during the cultivation period a ﬀ ects the growth characteristics of sweetpotato.


Introduction
Sweetpotato (Ipomoea batatas) is an important root vegetable cultivated in temperate and tropical zones, especially in Asia and Africa [1]. Storage roots of sweetpotato contain relatively high amounts of carbohydrates that support the demand for food in developing countries [2,3]. In recent years, sweetpotato has been also evaluated as a candidate for biofuel production [4,5]. Sweetpotato could potentially be used as an alternative to corn-based ethanol production to reduce fertilizer, water, and pesticide inputs and to utilize its ability to fix relatively large amounts of solar energy into starch in storage roots [6,7]. Several efficient methods of extracting biofuels and their residues (hydrogen, ethanol, and methane) from sweetpotato have been reported to date [5,[8][9][10][11]. Because the demand for sweetpotato is gradually increasing worldwide [12], it is necessary to establish an efficient and cheap cultivation method with low fertilizer requirement.
Fertilizers are widely used in agriculture to increase crop production. In sweetpotato, soil amendment using manure and inorganic fertilizers has a significant impact on plant growth and storage root development [13]. Among chemical fertilizers, nitrogen (N), phosphorus (P), and potassium (K) are the major elements required for supporting shoot and root growth in sweetpotato [14,15]. Under N

Experimental Conditions
Sweetpotato (Ipomoea batatas) cultivar "Narutokintoki" was used in this study. The hydroponics system for sweetpotato was prepared as described previously [43]. Briefly, this system consisted of vermiculite-filled vinyl pots (4.5 L) placed in nutrient solution-filled containers (59 cm × 39 cm × 18 cm). Storage roots developed in vinyl pots, and fibrous roots of sweetpotato plants extended from the upper vinyl pots into the containers placed below. Plants could absorb the nutrient solution from the vermiculite, as the bottom of each pot was in direct contact with the water absorption sheet that extended into the nutrient solution below. Therefore, vermiculite in vinyl pots remained saturated with the nutrient solution throughout the cultivation period. The surface of pots and bottom containers was covered with insulation sheets to maximize the utilization of sunlight by reflection for photosynthesis. The nutrient solution was prepared by mixing OAT house 1: OAT house 2 (OAT Agrio Co., Ltd., Tokyo, Japan) at a ratio of 3:2, and the NSC was adjusted to obtain the target EC (described below). The mixed nutrient solution contains N, 260 mg L −1 , P 2 O 5 , 120 mg L −1 ; K 2 O, 405 mg L −1 ; CaO, 230 mg L −1 ; MgO, 60 mg L −1 ; MnO, 1.5 mg L −1 ; B 2 O 3 , 1.5 mg L −1 ; Fe, 2.7 mg L −1 ; Cu, 0.03 mg L −1 ; Zn, 0.09 mg L −1 and Mo, 0.03 mg L −1 . Reduction in the water level in the tank due to evaporation was compensated by adding more water to the maximum Agronomy 2020, 10, 1708 3 of 14 level. Therefore, the EC of the NSC in each tank gradually decreased over time (Supplementary Figure S1). The nutrient solution was renewed approximately once every 30 days.
Two separate experiments were conducted (experiment 1 in 2018, and experiment 2 in 2019) to examine the effect of NSC on the growth and yield of sweetpotato ( Figure 1). In experiment 1, three NSCs, each with low (0.8 dS m −1 ), medium (1.4 dS m −1 ), or high (2.6 dS m −1 ) EC, were used throughout the cultivation period, and the effect of each NSC on plant growth was observed. The initial pH of high, medium, and low NSCs were 6.14, 6.46, and 6.72, respectively. In experiment 2, effects of changes in NSC on plant growth were examined using nutrient solutions with low EC (0.8 dS m −1 ) and high EC (2.6 dS m −1 ). Four treatments were conducted in experiment 2: (1) LL, plants were grown in low EC nutrient solution throughout the cultivation period; (2) LH, plants were grown in low EC nutrient solution until the end of the first half of the cultivation period, and then transferred to high EC nutrient solution and maintained until the end of the cultivation period; (3) HL, plants grown in high EC nutrient solution were transferred to low EC nutrient solution at the end of the first half of the cultivation period and maintained thereafter; (4) HH, plants were grown in high EC nutrient solution throughout the cultivation period.
Agronomy 2020, 10, x FOR PEER REVIEW  3 of 14 high, medium, and low NSCs were 6.14, 6.46, and 6.72, respectively. In experiment 2, effects of changes in NSC on plant growth were examined using nutrient solutions with low EC (0.8 dS m −1 ) and high EC (2.6 dS m −1 ). Four treatments were conducted in experiment 2: (1) LL, plants were grown in low EC nutrient solution throughout the cultivation period; (2) LH, plants were grown in low EC nutrient solution until the end of the first half of the cultivation period, and then transferred to high EC nutrient solution and maintained until the end of the cultivation period; (3) HL, plants grown in high EC nutrient solution were transferred to low EC nutrient solution at the end of the first half of the cultivation period and maintained thereafter; (4) HH, plants were grown in high EC nutrient solution throughout the cultivation period.   [50]. The average relative humidity was 69%, 77%, 74%, 69%, 78%, and 67% in May, June, July, August, September, and October, respectively [50]. The experimental field is about 17 km away from the meteorological station in Wakayama City, and 90 m higher than the station. The nutrient solution was renewed on 26 June, 31 July, and 30 August. In experiment 2, stem cuttings were planted in pots and grown for 25 days under the same growth conditions as those used for experiment 1. Pots were then transferred to  .5 • C in May, June, July, August, September, and October, respectively, according to the website of Japan Meteorological Agency [50]. The average relative humidity was 69%, 77%, 74%, 69%, 78%, and 67% in May, June, July, August, September, and October, respectively [50]. The experimental field is about 17 km away from the meteorological station in Wakayama City, and 90 m higher than the station. The nutrient solution was renewed on 26 June, 31 July, and 30 August. In experiment 2, stem cuttings were planted in pots and grown for 25 days under the same growth conditions as those used for experiment 1. Pots were then transferred to the hydroponic system and cultivated for 155 days (from 18 May to 20 October in 2019) at the open experimental field of Kindai University. The average temperature of Wakayama city in 2019 was 20.1 • C, 23.5 • C, 26.3 • C, 28.5 • C, 26.4 • C, and 20.7 • C in May, June, July, August, September, and October, respectively [51]. The average relative humidity was 60%, 72%, 79%, 76%, 71%, and 73% in May, June, July, August, September, and October, respectively [51]. The nutrient solution was renewed on 22 June, 12 July, 3 August, 31 August, and 28 September. The EC of the nutrient solution was changed for HL and LH plants on 3 August 2019.

Measurement of Plant Growth and Yield
In experiments 1 and 2, shoot and storage root fresh weight (FW) and storage root number were measured at 143 and 155 days after planting (DAP), respectively. Enlarged roots weighing more than 20 g were included as storage roots. In experiment 2, the number of attached leaves and the maximum length of the stem were measured at 3, 73, and 155 DAP. Total leaf number, abscised leaf ratio, stem number, and stem diameter were measured at 155 DAP. The number of total leaves was counted by adding the number of attached leaves and leaf petioles (without leaves). The abscised leaf ratio was calculated by dividing the number of total leaves with the number of attached leaves. In experiments 1 and 2, all measurements were recorded as the average of eight plants.

Measurement of Chlorophyll Contents
Relative chlorophyll contents were measured via a nondestructive assay using the Soil and Plant Analyzer Development (SPAD) chlorophyll meter (SPAD-502; Konica Minolta, Tokyo, Japan). Measurements were conducted at 3, 73, and 153 DAP using the second young fully expanded leaf of each plant.

Data Analysis
Data were analyzed using the JMP statistical package (SAS Institute, Cary, NC, USA). Significant differences among treatments were determined by one-way analysis of variance, followed by the Tukey-Kramer honest significant difference test for pairwise comparisons at p < 0.05.

Effect of NSC on the Growth of Hydroponic Sweetpotato (Experiment One)
Experiment one was conducted to examine the effect of NSC on the growth of sweetpotato in a hydroponic system. At 143 DAP, the shoot FW was the highest in the nutrient solution with high EC, followed by medium EC, and low EC ( Figure 2A). The storage root FW showed the same trend as that described above ( Figure 2B). Additionally, the number of storage roots showed no significant difference among the three treatments ( Figure 2C).

Effect of NSC on the Growth of Hydroponic Sweetpotato (Experiment One)
Experiment one was conducted to examine the effect of NSC on the growth of sweetpotato in a hydroponic system. At 143 DAP, the shoot FW was the highest in the nutrient solution with high EC, followed by medium EC, and low EC (Figure 2A). The storage root FW showed the same trend as that described above ( Figure 2B). Additionally, the number of storage roots showed no significant difference among the three treatments ( Figure 2C).

Effect of Variation in NSC on the Growth of Hydroponic Sweetpotato (Experiment Two)
Next, we examined whether changes in NSC affect the growth of hydroponic sweetpotato. Plant shoot growth was measured at three time points: 3 DAP, 73 DAP (4 days before changing the NSC), and 155 DAP (harvest day). The leaf chlorophyll content increased from 3 to 73 DAP in all plants, reaching similar levels in all treatments ( Figure 3A). No significant differences were detected among treatments at each time point, although the leaf chlorophyll contents of HH and LH plants at 155 DAP

Effect of Variation in NSC on the Growth of Hydroponic Sweetpotato (Experiment Two)
Next, we examined whether changes in NSC affect the growth of hydroponic sweetpotato. Plant shoot growth was measured at three time points: 3 DAP, 73 DAP (4 days before changing the NSC), and 155 DAP (harvest day). The leaf chlorophyll content increased from 3 to 73 DAP in all plants, reaching similar levels in all treatments ( Figure 3A). No significant differences were detected among treatments at each time point, although the leaf chlorophyll contents of HH and LH plants at 155 DAP tended to be higher than that of HL and LL plants ( Figure 3A). The number of attached leaves was higher in HH and HL plants than in LH and LL plants at 73 (Figure 4). To examine leaf abscission, we counted the number of total leaves, including previously abscised leaves, at 155 DAP. HH plants showed the highest number of total leaves, followed LH, HL, and LL plants ( Table 1). The abscised leaf ratio was the highest in HL plants, followed by LL, HH, and LH plants (Table 1). At 73 DAP, the maximum shoot length was higher in HH and HL plants than in LH and LL plants ( Figure 3C). At 155 DAP, shoot length was the highest in HH plants and lowest in LL plants, while HL and LH plants showed similar intermediate shoot lengths (Figure 3C). The number of stems was significantly higher in HH plants compared with plants in the other treatments ( Figure 5A). Stem diameter was the highest in HH plants, followed by HL and LH plants, and the lowest in LL plants ( Figure 5B).
were rarely attached to the petiole at the bottom and middle sections of the stem (Figure 4). To examine leaf abscission, we counted the number of total leaves, including previously abscised leaves, at 155 DAP. HH plants showed the highest number of total leaves, followed LH, HL, and LL plants ( Table 1). The abscised leaf ratio was the highest in HL plants, followed by LL, HH, and LH plants (Table 1).
At 73 DAP, the maximum shoot length was higher in HH and HL plants than in LH and LL plants ( Figure 3C). At 155 DAP, shoot length was the highest in HH plants and lowest in LL plants, while HL and LH plants showed similar intermediate shoot lengths (Figure 3C). The number of stems was significantly higher in HH plants compared with plants in the other treatments ( Figure 5A). Stem diameter was the highest in HH plants, followed by HL and LH plants, and the lowest in LL plants ( Figure 5B). These parameters were examined after 3, 73, and 155 days after plantation. Vertical bars represent the means ± SE (n = 8). Different letters indicate significant differences among the treatments at p < 0.05 by Tukey-Kramer's test.   The biomass of shoots and storage roots was measured at 155 DAP. Shoot FW was the highest in HH plants, followed by LH, HL, and LL plants ( Figure 6A). Storage root FW was the highest in HH plants, followed by HL and LH plants, and the lowest in LL plants ( Figure 6B). The number of storage roots showed no significant difference among treatments ( Figure 6C). Storage roots developed within vinyl pots in all treatments. Storage roots were round in shape, with a short length and partially undeveloped parts (Figure 7), consistent with previous observations [43]. These morphological characteristics of storage roots exhibited no variation among the different treatments (Figure 7). The biomass of shoots and storage roots was measured at 155 DAP. Shoot FW was the highest in HH plants, followed by LH, HL, and LL plants ( Figure 6A). Storage root FW was the highest in HH plants, followed by HL and LH plants, and the lowest in LL plants ( Figure 6B). The number of storage roots showed no significant difference among treatments ( Figure 6C). Storage roots developed within vinyl pots Agronomy 2020, 10, 1708 8 of 14 in all treatments. Storage roots were round in shape, with a short length and partially undeveloped parts (Figure 7), consistent with previous observations [43]. These morphological characteristics of storage roots exhibited no variation among the different treatments (Figure 7).
The biomass of shoots and storage roots was measured at 155 DAP. Shoot FW was the highest in HH plants, followed by LH, HL, and LL plants ( Figure 6A). Storage root FW was the highest in HH plants, followed by HL and LH plants, and the lowest in LL plants ( Figure 6B). The number of storage roots showed no significant difference among treatments ( Figure 6C). Storage roots developed within vinyl pots in all treatments. Storage roots were round in shape, with a short length and partially undeveloped parts (Figure 7), consistent with previous observations [43]. These morphological characteristics of storage roots exhibited no variation among the different treatments ( Figure 7).

Discussion
In hydroponics, the optimal NSC varies among plant species, with EC ranging from 1.5 to 2.5 dS m −1 [52,53]. Several studies have shown that high NSCs reduce the growth and photosynthetic parameters of hydroponically-grown plants [30,32,35,[54][55][56]. High NSC-dependent growth restrictions are observed at EC > 1.8 dS m −1 in peace lily and at EC > 2.8 dS m −1 in peppermint and lettuce [37,54,55]. By contrast, hydroponically-grown bush snap beans can tolerate EC up to 3.6 dS m −1 [33]. In the current study, the growth of shoots and storage roots of hydroponic sweetpotatoes increased in an NSC-dependent manner up to an EC of 2.6 dS m −1 (Figure 2). Given that continuous growth of sweetpotato plants in a nutrient solution with an EC of 2.6 dS m −1 did not influence the leaf chlorophyll content (Figure 3A), this NSC appears to be more favorable for plant growth and storage root development rather than an osmotic stress condition that would deter growth and

Discussion
In hydroponics, the optimal NSC varies among plant species, with EC ranging from 1.5 to 2.5 dS m −1 [52,53]. Several studies have shown that high NSCs reduce the growth and photosynthetic parameters of hydroponically-grown plants [30,32,35,[54][55][56]. High NSC-dependent growth restrictions are observed at EC > 1.8 dS m −1 in peace lily and at EC > 2.8 dS m −1 in peppermint and lettuce [37,54,55]. By contrast, hydroponically-grown bush snap beans can tolerate EC up to 3.6 dS m −1 [33]. In the current Agronomy 2020, 10, 1708 9 of 14 study, the growth of shoots and storage roots of hydroponic sweetpotatoes increased in an NSC-dependent manner up to an EC of 2.6 dS m −1 (Figure 2). Given that continuous growth of sweetpotato plants in a nutrient solution with an EC of 2.6 dS m −1 did not influence the leaf chlorophyll content ( Figure 3A), this NSC appears to be more favorable for plant growth and storage root development rather than an osmotic stress condition that would deter growth and photosynthetic activity.
Plants sense the nutrient dose and alter the biomass partitioning accordingly [57]. In sweetpotato, the dose of N fertilizer alters the biomass partitioning of storage roots and shoots [58,59]. Increasing the N fertilizer dose from 0 to 1.2 g N per plant increases the biomass partitioning to storage roots [58]. However, at a higher N dose, the storage root biomass decreases, whereas the shoot biomass increases [58]. In experiment one, the growth of shoots and storage roots were enhanced as the NSC increased to an EC of 2.6 dS m −1 (Figure 2A,B). Considering the reports that hydroponic plants have different growth characteristics compared with soil grown plants [60,61], the responsiveness of hydroponically-grown sweetpotato to the nutrient dose might be different from that of soil grown sweetpotato plants.
In hydroponic potatoes, shoot growth was enhanced when NSC was increased up to EC 2.4 dS m −1 , whereas tuber biomass was not affected by the EC of the nutrient solution [56]. Therefore, NSC-dependent partitioning of biomass may differ among plant species in hydroponics. The number of storage roots tend to be higher in experiment two ( Figure 6C) compared to experiment one ( Figure 2C). This may be caused by the different cultivation periods between experiments. Because two experiments were conducted only one time, the data may be influenced by the environmental condition.
The nutritional requirements of plants vary with the developmental stage [62]. Several studies have shown that changes in NSC influence plant growth characteristics [29,[63][64][65][66]. In strawberry, restriction of N application at an early developmental stage increased the fruit biomass by enhancing reproductive growth [63]. In hydroponic tomato, increasing the NSC during fruit development reduces the fruit size and increases the sugar content [66]. Nutrient solution formulations have been developed for various growth stages in hydroponic tomato [67]. In sweetpotato, the timing of N fertilizer application influences plant growth and storage root yield [68][69][70][71]. Split application of N fertilizer could increase the storage root yield of sweetpotato by improving the efficiency of N uptake [69,70]. The timing of N fertilizer application is also important for increasing the marketable sweetpotato yield [71]. In experiment two, the storage root FW showed no significant difference between HL and LH plants ( Figure 6B). This suggests that the timing of high-dose N application is not important for storage root development in hydroponically-grown sweetpotato. On the other hand, the shoot biomass and total leaf number were higher in LH plants compared with HL plants ( Figure 6A, Table 1). These results implicate that higher dose of nutrient application at the storage root hypertrophic stage (the second half of growth stage) may enhance the development of shoots as well as storage roots. It should also take into account for the high abscised leaf ratio in HL plants (Table 1) because the abscised leaves, which did not contribute to shoot biomass, were partly responsible for the low shoot FW of HL plants. In general, at the late stage of sweetpotato cultivation, the storage root growth is enhanced, whereas shoot growth is retarded [59]. Thus, nutrient limiting condition at the hypertrophic stage of storage roots in HL plants may represent the field-grown sweetpotato characteristics in the shoot and root development. Because the amount of photosynthetic products translocated to storage roots partly depends on the shoot biomass, modifying the timing of NSC changes might improve the storage root yield.
Leaf abscission occurs during the senescence process and is induced by various stress responses [72]. Before the onset of leaf cell separation, the abscission zone encounters the repression of auxin biosynthesis and enhancement of ethylene production and sensitivity, resulting in the activation of cell wall degradation enzymes [73][74][75]. In experiment two, leaf abscission was induced at the late stage of cultivation in all NSC treatments (Table 1). This growth stage-dependent leaf abscission in sweetpotato has also been observed in open field conditions [76][77][78], suggesting a consistent senescence related phenomenon. N or P limitation is known to induce leaf abscission by enhancing ethylene production and sensitivity [79]. Sweetpotato leaves also abscise at the late growth stage under low N or P condition [80]. In HL plants, plant shoot biomass (the number of attached leaves and maximum shoot length) increased during the first half of the cultivation period in the nutrient solution with high EC (Figure 3B,C); however, these shoots grew in relatively poor nutrient conditions during the second half of the cultivation period. These nutrient poor conditions might trigger the high ratio of leaf abscission associated with N or P deficiency. Leaf senescence is accompanied by the breakdown of chlorophyll [81]. HL and LL plants showed a higher abscised leaf ratio and lower relative chlorophyll content ( Figure 3A, Table 1), suggesting accelerated leaf abscissions by the progression of senescence. N deficiency also causes oxidative stress to the leaf [82]. Given that oxidative stress could trigger leaf abscission [83][84][85][86], it is possible that HL plants exhibit leaf abscission during the second half of the cultivation period due to oxidative stress triggered by N deficiency. On the other hand, LH plants were relatively nutrient-rich condition at the late cultivation period. Therefore, leaf senescence and abscission were thought to be suppressed by relatively rich-N supplement.

Conclusions
Compared to traditional soil culture systems, sweetpotato hydroponics saves absorbent material (soil) and can be used anywhere exposed to sunlight. In addition, hydroponics can efficiently utilize nutrient components as supplied components are not dispersed to the soil. In fact, almost all nutrients were absorbed in plants grown on EC 0.8 and 1.4 in this study (Supplementary Figure S1). Here, we presented NSC-dependent storage root yield in hydroponic sweetpotato (experiment one). Although the timing of high and low NSC did not have a significant impact on the storage root yield, shoot growth was apparently increased by high NSC (experiment two). A more precise adjustment of the NSC may increase the yield of storage roots relative to the fertilizer input. Thus, given its flexibility in manipulating the nutrient status, hydroponics could be used as an efficient tool for sweetpotato production.