Résumé
Biennial bearing habits are annual cyclical changes in cropping characterized by “on” and “off” years with “heavy” and “light” fruit loads, respectively. Within the domesticated apple, there is a large variation among cultivars in the amplitude of cycling and tendency to fruitfulness and regular cropping behavior, suggesting genetic control of this phenomenon. Based on previous investigations (Guitton et al., 2011), we focused on 3 floral genes, GA 20‐oxidase (MdGA20ox1a), TERMINAL FLOWER1 (MdTFL1) and SOC1 (MdSOC1), assumed to have a major role in the control of flower induction (FI) of apical shoot meristems in apple. GA20ox1 is involved in the later steps of the gibberellin biosynthetic pathway. MdTFL1 is assumed to repress flowering, thus maintaining the apical meristem vegetative. SOC1 integrates multiple flowering signals derived from photoperiod, temperature, hormone, and age‐related signals. The aim of this study was to reveal the temporal and spatial expression patterns of these 3 genes during floral bud induction and initiation by in‐situ hybridization. Apical meristems from 2‐year‐old spurs of Royal Gala/ M.9 “on” and “off” trees were collected between 19 April to 14 June 2012, at weekly intervals, as well as on 28 June and on 30 July. Single‐stranded antisense and sense RNA probes of the genes were transcribed with T7 polymerase and labelled with digoxiginin (DIG). At sampling, meristems were immediately fixed. Prior to in‐situ hybridization, samples were washed, dehydrated, embedded in paraffin, and cut in 8 μm sections. In‐situ hybridization was performed at 55°C (for genes GA20OX1a and SOC1) and at 57.6°C for TFL1a overnight with 0.2 μg ml‐1 of the digoxigenin‐labelled RNA probe. Several post‐hybridisation washes were then carried out before signal detection using an antiDIG antibody coupled with a phosphatise alkaline and VectorBlue as its substrate. Distinct temporal and spatial changes were revealed in the expression patterns of the 3 studied floral genes. All genes showed initially no or little expr ession in meristems. Thereafter, their expression level drastically increased over time, and each gene exhibited a characteristic spatial distribution, irrespective of subsequent induction or repression of flowering. Even though differences between on and off trees could be visually distinguished in mid and late June samples, these results must be further confirmed by quantitative analyses, such as qRT‐PCR.