The aim of this study was to identify the precise time at which MT promotes AR formation in tissue-culture plantlets of Malus prunifolia apple rootstocks. The study consisted of five different treatment groups: MT, MT0–2, MT2–5, MT5–20 and one control group (Fig. 1A). No morphological changes were evident in any of the groups up to 5 d; however, ARs emerged from basal stem parts at 10 d (Fig. 1B). At 20 d, the greatest number of ARs was observed in the MT0–2 group, and groups MT2–5 and MT5–20 produced more ARs than the control and MT group at 20 d, but no significant difference was observed in the number of ARs among these groups (Fig. 1B). To observe the anatomy of stems at different stages of AR formation, sections were made from paraffin-embedded samples and were viewed using light microscopy. On day 0, cross-sections of the samples revealed the existence of competent cells, but mitotic cambial cell division was observed at 5 d and cell divisions were visible in the compactly arranged cells, and AR appeared in sections from stem bases cultured in medium for 10 days (Fig. 1C). No AR formation was observed in MP treated for 20 d with the auxin inhibitors N-1-naphthylphthalamic acid (NPA) or triiodobenoic acid (TIBA), but ARs were observed following MT treatment. All phenotypes are summarised in Figure S1.
We also measured the root rate, number of AR, crossings, root length, root surface area and root volume in all five treatment groups, and the data were consistent with the phenotypes of AR formation. All measured parameters were higher in the MT0–2 group than in other groups, and the minimum number of ARs and values for other root parameters were observed in the control group (Fig. 2). The results showed that MT mainly promoted AR formation at 0–2 d, during the AR induction stage.
On the basis of their diameter, ARs were classified into three groups: 0–2.0 mm, 2.0–5.0 mm and >5.0 mm. According to AR number, length and surface area, the 0–2.0 mm category contained the greatest percentage of the total; however, for root volume, the 2.0–5.0 mm category was the largest for all groups (Fig. 3). The MT0–2 group contained the greatest number of ARs in the 0–2.0 mm class, which was twice as many as in the control group (Fig. 3). We conclude that most ARs were fine roots (0–2.0 mm).
The levels of the hormones MT, IAA, ZR, GA1+3 and ABA were analysed in MP tissue culture plantlets after treatment with MT. The MT content was higher in the MT0–2 group than in other treatment groups during the early AR developmental stage, and reached a peak at 5 d in the MT0–2 group. In the MT and MT0–2 treatments, the levels of IAA, ZR and GA1+3 were higher than those in the control group during AR induction at 1 d and 2 d, but the levels were lower in the MT0–2 group than other groups at 10 d. The level of ABA responded opposite to that of IAA, ZR and GA1+3 (Fig. 4) in the treatments.
The expression of MdTDC1, MdHIOMT2, MdASMT1 and MdASMT2 genes, which are involved in MT and IAA signal transduction, was higher in plants in the MT0–2 groups than in other groups, except at 0 and 20 d, and expression of MdSNAT and MdHIOMT1 was higher in MT0–2 than in other groups at 1 d, 5 d and 20 d (Fig. 5).
To determine whether an interactive effect between MT and IAA existed, we measured the expression of genes related to IAA biosynthesis and signal transduction. The expression of MdYUCCA2, MdYUCCA10, MdARF7 and MdARF19 was higher in plants in the MT0–2 treatment than that in other groups at 1 d, 2 d and 3 d. The expression level of the IAA transport genes MdAUX1, MdPIN1, and MdPIN3 was also higher in the MT0-2 than in other groups at 2 d; however, expression of the IAA signal transduction gene MdIAA5 was lower following MT0–2 treatment than in other treatments during the AR induction stage (Fig. 6).
To investigate whether MT affects cell division, the expression of the cell-cycle genes MdCYCD1;1 and MdCYCD3;1 was analysed, and these genes were more highly expressed in MT0–2 at 3 d and 10 d. Therefore, we conclude that the application of MT promoted AR formation in apple rootstock, and RT-qPCR analysis showed that the expression of genes related to root development was higher at most time points in response to MT treatment. We observed that among all root-related genes, the expression of MdWOX11 following MT treatment was 5.6 times higher than that in control plants at 2 d (Fig. 7). This suggests that MdWOX11 probably plays an important role in AR induction in response to MT treatment.
The expression of MdWOX11 was induced by IBA treatment (Fig. 7). We generated the overexpression (OE) transgenic lines MdWOX11-OE15#, 16# and 20# in ‘GL3’, and confirmed the level of overexpression MdWOX11 transgenic lines (Figure S2). To confirm whether MdWOX11 transgenic lines exhibited an enhanced response to MT signalling, wild-type and transgenic apple plantlets growing in tissue culture were either treated with 3.45 μM IBA as a control, or with MT for 0–2 d (MT0-2). More ARs were observed in the MT0-2 group than in the control group and the overexpressing MdWOX11 transgenic lines and ‘GL3’, and the MdWOX11 overexpressing lines produced more ARs than ‘GL3’ (Fig. 8A). Overexpression of MdWOX11 also caused an increase in the rate of ARs (Fig. 8B). Furthermore, MdWOX11 overexpressing plants were more sensitive to exogenous MT treatment than wild type (Fig. 8A-C), which indicates that MdWOX11 induced AR formation in response to MT treatment.