The study of the characteristics of the secondary flowering of Cerasus subhirtella ‘Autumnalis’

Phenological observation

From June 2019 to April 2021, the phenological observations and recordings were made based on the date of appearance of C. subhirtella ‘Autumnalis’ and C. yedoensis ‘Somei Yoshino’. Observations were made every 3–5 days. When the phenological periods of these plant species overlapped and the plants grew rapidly, observations were made every day. We stopped monitoring the plants after they stopped growing in the winter. The specific phenological observation methods were developed based on the methods adopted by Fadón, Herrero & Rodrigo (2015) and Wan & Liu (1979). The timing of the growth and plant phenology of each plant was recorded using the plant BBCH coding system. In this study, based on the BBCH coding system, the coding method for the secondary flowering of cherry blossoms was developed with repetitive phenological events in plants.

Assessment of the chilling requirement

The branches of C. subhirtella ‘Autumnalis’ and C. yedoensis ‘Somei Yoshino’ were collected in mid-October 2019. Each sampling set contained five long branches (≥15 cm) and five short branches (≤10 cm). The length and width of the buds on the branches were recorded every six days. The collected branches were maintained in an artificial climate room for subsequent cultivation. The artificial climate room was maintained at 24 °C/18 °C (day/night) and 62% relative humidity. The bottom of the branches was cut off obliquely. The water was changed every two days, and the base of the branches was cut 3∼4 mm to ensure complete water uptake. The data on field temperature and humidity were recorded with the COS-03 automatic temperature and humidity recorder, and the instrument was set to record data once every 10 min.

The bud-break data was statistically analyzed based on the method developed by Wanglirong et al. (Wang et al., 2003). For the bud-break rate, the grading of flower bud-break status was as follows: Level 1, no bud-break; Level 2, budding; Grade 3, bud tip exposed green; Grade 4, showing petals; Level 5, flowers open. The bud-break rate is the weighted average value of flower buds at all levels, and the weight is the stage of the flower bud-break state. The grading of the chilling requirement was as follows: a bud-break rate ≥ 2.5 indicated that the low temperature was passed; for 2.5 <bud-break rate ≤ 3.0, the cooling demand was calculated based on the accumulated low-temperature hours in the field in this sampling period; for 3.0 <germination rate ≤ 3.5, the chilling requirement was calculated based on the arithmetic mean of the field accumulated low-temperature hours in this period and the last sampling period; for bud-break rate >3.5, the chilling requirement of the last sampling period was considered to prevail. The 0 ∼7.2 °C model and the Utah model were used as the estimation model of the chilling requirement, and the formula to calculate the chilling requirement is as follows:

${\displaystyle CH={\sum }_{i=1}^{t}T,T=\left\{\begin{array}{c}1\left(0°C\le T\le 7.2°C\right)\hfill \\ 0\left(else\right)\hfill \end{array}\right.}$ ${\displaystyle U\mathrm{tah}={\sum }_{i=1}^t=\left\{\begin{array}{c}T\le 1.40\hfill \\ 1.4<T\le 2.40.5\hfill \\ 2.4<T\le 9.11\hfill \\ 9.1<T\le 12.40.5\hfill \\ 12.4<T\le 15.90\hfill \\ 15.9<T\le 18.0-0.5\hfill \\ T\ge 18.0-1.\hfill \end{array}\right.}$

Extraction and identification of the DAM gene

Sampling was performed in mid-September 2020 when the flower bud differentiation of C. subhirtella ‘Autumnalis’ and C. yedoensis ‘Somei Yoshino’ was almost completed. Flower buds were collected every 15 days, 11 samples in total, and the branches that extended toward the sun were selected. The collected branches were placed in an ice box and transported to the laboratory. The collected flower buds were wrapped in tin foil, quickly frozen with liquid nitrogen, and stored in a refrigerator at −80 °C until analysis. The genes related to cooling requirements were searched in the NCBI (National Center for Biotechnology Information) database, and real-time quantitative PCR primers were designed based on the gene sequences reported in other studies (Qiu et al., 2020; Chen, 2017). The names of the synthetic primers and the upstream and downstream sequences used are presented in Table 1.

The total RNA of the samples was extracted using the Tiangen Polysaccharide Polyphenol Plant Total RNA extraction kit (DP441), and the obtained RNA solution was stored in a refrigerator at −80 °C to prevent degradation. The quality of the extracted RNA was visually assessed by agarose gel electrophoresis, and cDNA was synthesized using the PrimeScript™ 1st Strand cDNA Synthesis Kit (TaKaRa Bio Inc.). RSP3 was used as the internal reference gene in the real-time quantitative PCR (qPCR) technique, and the relative expression of the DAM4, DAM5, and DAM6 genes in cherry blossoms was determined using the MA-6000 real-time quantitative PCR system.

Changes in the relative expression of DAM4

The DAM gene triggers dormancy and inhibits flower bud-break in plants. In this study, after the plant entered dormancy, the expression of the DAM gene was upregulated rapidly, and then, the expression of the DAM gene was downregulated with storage at a low temperature. The relative expression of the DAM4 gene firstly increased and then decreased at the dormant stage of C. subhirtella ‘Autumnalis’ and C. yedoensis ‘Somei Yoshino’ (Fig. 3), but the timing and amplitude of the relative expression of the DAM4 gene in these two varieties differed significantly. Moreover, the relative expression of the DAM4 gene in the flower buds of C. yedoensis’ Somei Yoshino’ and C. subhirtella ‘Autumnalis’ was extremely low in September. After the flower bud differentiation was completed in mid-October, the relative expression of the DAM4 gene in the flower buds of C. yedoensis ‘Somei Yoshino’ increased till December. The relative expression of DAM4 reached the highest value in mid-December. In early January of the following year, the flower buds of C. yedoensis ‘Somei Yoshino’ accumulated cold energy. The relative expression of the DAM4 gene decreased rapidly to the same level as that before entering dormancy by the end of January. The relative expression of the DAM4 gene in the flower buds of C. subhirtella ‘Autumnalis’, however, increased slightly in mid-October and decreased to the same level as that before entering inner dormancy during autumn-flowering. The relative expression of the DAM4 gene was significantly upregulated in mid-December at the end of the autumn-flowering period. In C. subhirtella ‘Autumnalis’, the upregulation occurred within a shorter period and at a lower amplitude than that in C. yedoensis ‘Somei Yoshino’.

Changes in the Relative Expression of DAM5

The functional analysis of MADS-box genes related to flower bud dormancy in the Chinese cherry in other studies showed that the DAM4, DAM5, and DAM6 genes might be involved in flower bud dormancy and dormancy release (Shao, 2016). The relative expression of the DAM5 gene in the flower buds of C. yedoensis ‘Somei Yoshino’ first increased and then decreased during dormancy (Fig. 4). However, the relative expression of the DAM5 gene in the flower buds of C. subhirtella ‘Autumnalis’ showed a fluctuating trend, i.e., it was first upregulated and then downregulated, followed by the same trend. After the differentiation of the flower bud of C. yedoensis ‘Somei Yoshino’ was completed in mid-September, the relative expression of the DAM5 gene increased rapidly, and in C. subhirtella ‘Autumnalis’, it increased at a lower rate compared to that in C. yedoensis ‘Somei Yoshino’ in the same period. From October to November, the relative expression of DAM5 in C. yedoensis ‘Somei Yoshino’ was high, while in C. subhirtella ‘Autumnalis’, it was first upregulated and then downregulated. The low-temperature storage caused the inner dormancy to break, and the relative expression of DAM5 was the lowest during the autumn-flowering period. In mid-December, when C. subhirtella ‘Autumnalis’ entered dormancy, the relative expression of the DAM5 gene in flower buds increased, and the ungerminated flower buds could not continue to germinate. From mid-December to early January, the relative expression of the DAM5 gene in the flower buds of C. subhirtella ‘Autumnalis’ was maintained at a high level, and then, it decreased till the end of January. However, in the flower buds of C. yedoensis ‘Somei Yoshino’, the relative expression of the DAM5 gene remained at a high level from mid-December to early January. It decreased gradually to the lowest level at this time, and the flower buds of C. yedoensis ‘Somei Yoshino’ germinated and opened. These plants generally could not open because the external environmental conditions were not suitable for their growth. At the end of February, when the flower buds began germinating, the relative expression of DAM5 in C. yedoensis ‘Somei Yoshino’ and C. subhirtella ‘Autumnalis’ reached the lowest levels.

Changes in the relative expression of DAM6

The relative expression of the DAM6 gene in the flower buds of C. yedoensis ‘Somei Yoshino’ was first upregulated and then downregulated during dormancy (Fig. 5). It first decreased, then increased, and finally, decreased again. During low-temperature storage, the relative expression of the DAM6 gene was rapidly downregulated and was the lowest during the autumn-flowering period of C. subhirtella ‘Autumnalis’; however, it increased after the autumn-flowering period was over. In contrast, in the flower buds of C. yedoensis ‘Somei Yoshino’, the relative expression of the DAM6 gene remained at a high level during the same period.