2.2 Measurements of endogenous phytohormones

Leaf samples were collected from the first leaf matures stage, the second leaf sprouting stage (because the second leaf had just emerged, the first leaves were selected as materials for this period), the second leaf development stage (from this stage, the samples were all taken from the second leaves), the second leaf matures stage and the lodging stage. Corm samples were taken at all stages. During first leaf maturity and second leaf emergence, the mother corm gradually shrinks and the daughter corm starts to develop ( Figure 1 ), so the mother corms were selected at the first leaf maturity stage, and the daughter corms were sampled from the second leaf sprouting period to the later periods. Samples from three individual plants were mixed as one biological replicate and three biological replicates were used in the study.

Sample extraction was carried out following a high-throughput target detection method by Shanghai Biotree Biotechnology Co., Ltd. (Shanghai, China). In brief, samples were first ground in liquid nitrogen, and then 1000 μL of extract solution (50% acetonitrile in water, precooled at -40°C, containing isotopically labeled internal standard mixture) was added. Information about the isotopically labeled internal standard used in this study is shown in Supplemental Data 1 . After vortexing for 30 seconds and sonicating for 5 minutes in an ice bath, the samples were homogenized at 40 Hz for 4 minutes. Following centrifugation at 12000 rpm for 15 min at 4°C, 90 μL of 10% ACN/H₂O (v/v) was added. The samples were centrifuged once again, filtered through a 0.22 mm polytetrafluoroethylene membrane filter, and then subjected to ultra-high-performance liquid chromatography tandem mass spectrometry (UHPLC-MS/MS) analysis.

A Waters ACQUITY UPLC CSH C18 column (150 * 2.1 mm, 1.7 mm, Waters Corporation, MA, USA) was used for UHPLC separation. Mobile phase A was 0.01% formic acid in water, and mobile phase B was 0.01% formic acid in acetonitrile. The column temperature was set at 50°C. The autosampler temperature was set at 4°C. The injection volume was 5 μL. The typical ion source parameters were as follows: curtain gas = 40 psi, ion spray voltage = ± 4500 V, temperature = 475°C, ion source gas 1 = 30 psi, and ion source gas 2 = 30 psi.

Flow injection analysis was used to optimize the multiple reaction monitoring (MRM) parameters for each targeted analyte, which was injected into the API source of the mass spectrum using the standard solutions of each analyte. MRM scan mode was used to optimize collision energy for each Q1/Q3 pair using some of the most sensitive transitions. The Q1/Q3 pairs that showed the best sensitivity and selectivity were selected as ‘quantifiers’ for quantitative monitoring among the optimized MRM transitions per analyte. An additional transition served as a ‘qualifier’ for identifying the target analytes. SCIEX Analyst Work Station Software (Version 1.6.3) and Sciex MultiQuant™ 3.0.3 were used for MRM data acquisition and processing. The extracted ion chromatographs (EICs) from a standard solution and a sample of the targeted analytes under the optimal conditions are shown in Supplemental Figure 1 .

The calibration solutions were analyzed with UPLC-MRM-MS/MS using the methods described above. The y-axis represents the ratio of peak areas for analyte/IS, and the x- axis represents the concentration (nmol/L) for an analyte. The least-squares method was used for the regression fitting. For the curve fitting, 1/x weighting provided the highest accuracy and correlation coefficient (R²). Levels were excluded from calibration if their accuracy fell outside of 80%-120%.

The precision of the quantitation was measured by the relative standard deviation (RSD), determined by injecting analytical replicates of a QC sample. The accuracy of quantitation was measured by the analytical recovery of the QC sample. The percent recovery was calculated as [(mean observed concentration)/(spiked concentration)] × 100%.