Meiwa kumquat (Fortunella crassifolia
Swingle) is an ever-flowering tree that usually blooms in spring (April), summer (June to August), and autumn (September) in Japan. The spring flowers appear on the previous year shoots, but are very few in number and not usually used for fruit production. By contrast, the summertime flowers appear on each axil of the current spring shoots that emerge around May and are the most useful for fruit production [1
]; therefore, the flower bud differentiation period of the summertime flowers is regarded as approximately 1 month from the termination of spring shoot elongation to flowering [2
]; however, the flowering of summertime flowers is divided into three to four separate time points, and when flowers do not bear fruitlets, new flowers appear at the same axil after 10–14 days. The resulting flowers are known as the first-, second-, and third-flush flowers in Japanese production areas [1
]. The fruits from first-flush flowers tend to grow well and mature early. The fruit quality also tends to be good in first-flush flowers. Since there are few first-flush flowers—the number of first-flush flowers also varies annually—the second-flush flowers are used for fruit production in general [1
Many previous studies on Meiwa kumquat have shown that the application of a soil water deficit (SWD) for 2–4 weeks during the flower bud differentiation period increases the number of first-flush flowers [4
]; however, the optimal duration of SWD treatment and its effectiveness varies between years. Iwasaki et al. [7
] found that severe drought stress that reduced the predawn leaf water potential to below −1.7 MPa or increased the leaf abscisic acid (ABA) content to above 5 nmol g−1
dry weight (DW) did not increase the number of first-flush flowers, suggesting that the leaf water potential needs to be maintained between 0 and −1.7 MPa; however, this condition is difficult to achieve because the leaf water potential fluctuates not only with soil water content but also with the relative humidity of the atmosphere [6
Iwasaki and Hiratsuka [8
] reported that the leaf water potential tends to be lower in Meiwa kumquat trees with larger trunk and scaffold branches under drought conditions. They also noted that transpiration from the trunk surface does not change despite rapid decreases in leaf transpiration under drought conditions, resulting in continuous water loss from the trunk at approximately one-sixth of the leaf transpiration rate. In some woody species, water stored in the trunk is transported to organs such as the leaves when required, but this process does not occur in isohydric plants [12
]. Isohydric plants maintain a constant midday leaf water potential under both abundant water and drought conditions by reducing stomatal conductance to limit transpiration when required [14
]. Thus, the kumquat tree is considered isohydric because SDW treatment rapidly decreases the transpiration rate in the leaves [9
], and the water stored in the trunk does not migrate to the leaves. Plants that have encountered drought actively accumulate soluble sugars and free amino acids to prevent cell dehydration, via a process known as osmoregulation [16
]. Thus the migration of water between the organs within a tree may be related to the amount of soluble sugars that have accumulated in that organ.
In the present study, the accumulation of soluble sugars in the organs of Meiwa kumquat trees under SWD treatment was investigated and considered the relationship between the sugar content of each organ and the number of first-flush flowers.
2. Materials and Methods
2.1. Plant Materials and Treatment
Six eighteen-year-old Meiwa kumquat trees that had been grafted onto trifoliate orange [Poncirus trifiliata (L.) Raf.] were used in the experiment, which was carried out during the growing season in 2016. The trees were planted in 29-L non-woven fabric pots containing commercially available humus-rich soil and each tree had a height and canopy diameter of approximately 1.2 and 1.5 m, respectively. Bud burst and elongation of spring shoots began around 25 April 2016, and were completed around 20 May 2016.
Three trees were allocated to each of the SWD treatment and non-treated control (CONT) groups on June 4 and maintained for 12 days in a greenhouse at Meiji University (Kawasaki, Kanagawa, Japan; 35°61′ N, 139°55′ E). For the SWD treatment, the soil water content was reduced by withholding irrigation, and a small volume of water (approximately 300 mL) was applied when the soil water content fell to <15%. During the treatment, the soil water content was monitored at a depth of approximately 10 cm between the trunk and the pot rim using an ECH2O soil moisture sensor (EC-5) with an Em50 digital data logger (METER Group, Inc., Pullman, WA, USA). The leaf water potential was determined on June 8, 11, and 14 using a psychrometer (WP4-T; METER Group, Inc., Pullman, WA, USA) to monitor the severity of water stress in the trees. One matured leaf in the middle of two current shoots per tree was collected at 9:00 a.m. on each measurement day.
2.2. Observation of Flowering Behavior and Determination of Total Soluble Sugars and Starch
Ten current-year spring shoots of approximately 10 cm length were selected from each tree to observe flowering behavior. The number of flowers that opened on each shoot was recorded each day from 20 June to 31 July, and the number of first-flush flowers and the total number of flowers per 10 shoots per tree was counted. One scaffold branch per tree was harvested on 15 July at the end of SWD treatment, and the plant parts were categorized into current-year spring leaves (current leaves), current-year spring stems (current stems), previous-year leaves (previous leaves), previous-year stems (previous stems), and scaffold branch, which included branches >2 years old. The scaffold branch was then further divided into bark and wood (xylem tissue). The fresh weight of each organ was measured, following which the organs were stored at −70 °C for further analysis. To analyze the soluble sugar and starch contents of the organs, the samples were freeze-dried and ground to powder. Soluble sugars were then extracted by adding 0.1 g of dried sample to 50 mL of 80% ethanol and leaving for 24 h at room temperature, following which the ethanol was removed from the extract using a rotary evaporator (N-1000; Tokyo Rika Kikai Co., Tokyo, Japan) at 35 °C and the remaining aqueous phase was diluted to 50 mL. The total sugar content in 1 mL of the diluted sample was then determined using the anthrone method. The starch content of the organs was determined with the starch-iodine colorimetric method [17
] using the alcohol-insoluble solids (AIS) that were obtained during sugar extraction. The soluble sugar and starch contents of each plant part were expressed as mg g−1
dry weight (DW). The water content (WC) of each plant part was calculated as WC = (FW − DW)/FW where FW is the fresh weight.
2.3. Sugar Composition of the Xylem Tissue and Bark Extracts of the Scaffold Branch
Following the extraction of sugars from each dried sample as described above, the aqueous phase that remained after ethanol removal was lyophilized. The residue was then dissolved in 2 mL of ultrapure water to prepare an analytical sample, which was filtered using a 0.45 μm membrane filter before being injected into a high-performance liquid chromatography (HPLC) system (LC-2000Plus series, JASCO, Tokyo, Japan). Sugars were separated in a Shodex SC1011 column (Shodex SC1011; Showa Denko K.K., Tokyo) that was maintained at 80 °C using ultrapure water as the mobile phase with a flow rate of 1.0 mL min−1. Sucrose, glucose, and fructose were then identified and quantified by comparison with peaks of known standards using a refractive index detector (RI-2031; JASCO, Tokyo, Japan).
2.4. Statistical Analyses
Each treatment was performed on three trees (one tree per plot) and differences between the treatments were assessed using t-tests. The differences in sugar and starch content among the organs were compared using the Tukey HSD test; the coefficients of correlation and determination were obtained by regression analysis. Percentage values were arcsine transformed before analysis. Differences between the means were analyzed using the software KaleidaGraph v.4.5 (Synergy Software, Reading, PA, USA) when ANOVA was significant at p = 0.05 or p = 0.01, whereas BellCurve for Excel, v.2.00 (Social Survey Research Information Co., Ltd. Tokyo, Japan) was used for regression analysis.
In this study, it was found that SWD treatment reduced the frequency of sequential blooming of summertime flowers and increased the number of first-flush summertime flowers in Meiwa kumquat; however, the total number of flowers did not significantly change, indicating that SWD promoted flower bud development rather than differentiation. These results support the findings of previous studies [4
] and suggest that the application of SWD treatment approximately two weeks after the cessation of spring shoot elongation during flower bud differentiation would accomplish uniform fruit maturation, removing the need to evaluate fruit maturity at harvest and thus simplifying harvesting practices.
In many crops, deficit irrigation is known to increase fruit sugar content without decreasing the yield [18
], and nowadays, the water stress is often applied to produce high-quality fruits. We found that kumquat trees accumulated considerable amounts of sugars in their scaffold branches, particularly in the xylem tissue, under SWD conditions. Although the scaffold branches had a lower total soluble sugar content per gram dry weight than the leaves, they exhibited a greater proportional increase following SWD treatment compared with the CONT trees (3.1 fold increase in xylem tissue while a 1.4 fold increase in the leaves). In this study, the trunk was divided into bark and wood. The wood is the xylem tissue consists of vessels, tracheid, parenchymal cell, etc., but since the vessels and tracheid are dead cells, sugars cannot be actively absorbed into the cells [22
]; therefore, we concluded that most sugars are accumulated in the xylem parenchyma cells. In the xylem tissue, sucrose was detected alongside glucose and fructose, and the sucrose content also increased alongside glucose or fructose under SWD conditions. Since the water content of the xylem tissue did not change as a result of SWD treatment, the increase in sugar content might involve not only osmoregulation [16
] but also other factors. Trees growing in cold areas such as birch are known to accumulate sucrose, raffinose, and stachyose in xylem parenchyma cells during the cold season [25
]. These sugars are considered to not only increase the freeze-tolerance of cells but have a stabilizing effect of cell membrane under stress conditions such as freezing, desiccation and high temperature, and that the effect is higher for trisaccharides or disaccharides having a higher molecular weight than monosaccharides [26
It has been previously shown that the carbohydrate contents of the leaves or bark of shoots that produce the next flowers gradually increase during flower bud differentiation/development period in Satsuma mandarin (Citrus unshiu
) trees growing outdoors in Valencia, Spain [27
], or in early-heating plastic houses in Fukuoka, Japan [28
], although they sometimes decrease temporarily; however, none of these studies found a direct relationship between flower bud differentiation and the sugar or starch content. In the present study, the number of first-flush flowers was most closely related to the sugar content of the current shoots that set flowers for fruit production. Similarly, Niii and Okamoto [29
] reported that the amount of stored carbohydrates influenced the development of flower buds in Satsuma mandarin, with flower buds showing better development when old leaves existed during budburst. It has also been shown that buds of peach (Prunus persica
) actively absorbed carbohydrates during dormancy release, which are used for growth metabolism and thereby induce bud development [30
]. In addition, Nakajima et al. [31
] reported that water stress from early September through December increased the number of ‘Tosa Buntan’ pomelo (Citrus grandis
) flowers and caused them to bloom earlier the following spring. Consequently, the authors considered that water-stressed trees were forced into dormancy to promote flower induction, and released from dormancy by water stress in a similar way to the flower buds of coffee (Coffea arabica
]. SWD treatment may have caused flower bud dormancy to occur earlier in kumquat, although the dormant period, occurring immediately after the cessation of spring shoot elongation, may be short. In recent years, sugars such as sucrose have been reported to be important signals that regulate bud outgrowth and act before hormones in releasing apical dominance [33
], i.e., sucrose promotes auxin export from the bud, then the axillary buds have sustained outgrowth. Since the flower buds of kumquat are formed in the axils of shoots, the mechanism involved in their development may be similar to that of apical dominance; therefore, the sucrose accumulated in xylem, just as in glucose and fructose, may have promoted the development or budburst of floral buds that originally should have been the second- or third-flush flowers, and made them the first-flush flowers.
Consequently, such sugar accumulation may decrease the water potential of the xylem tissue, resulting in the active accumulation of water due to osmoregulation. Although the sugar content of the trunk was not investigated in this experiment, the sugar content of the scaffold branches is expected to reflect this. Both the trunk and scaffold branches can be regarded as water reservoirs in many tree species [12
]. Nevertheless, the rapid decrease in leaf water potential that was observed during SWD treatment in this experiment indicated that the water supply from the trunk and scaffold branches to the leaves was restricted. Iwasaki et al. [9
] reported that the water potential of the trunk xylem tissue was always lower than the leaf water potential in kumquat trees under SWD conditions, which may result from the large accumulation of sugars. It was also found that the relationship between the number of first-flush flowers or current leaf water potential and the sugar content of the current stems changed when the sugar content reached approximately 100 mg g−1
DW. Similarly Iwasaki et al. [7
] reported that severe water stress expressed by a leaf water potential below −1.7 MPa or a leaf ABA content above 5 nmol g−1
did not increase the number of first-flush flowers. Thus, factors related to the level of water stress, such as leaf water potential and sugar content of the plant tissues, appear to have a limited range of influence on the number of first-flush flowers. Although it is necessary to investigate the detailed mechanism of sugar accumulation under drought conditions, the xylem tissue of the trunk and scaffold branches seems to play an important role in the whole-tree water relations through active sugar accumulation.