Expressional regulation of PpDAM5 and PpDAM6, peach (Prunus persica) dormancy-associated MADS-box genes, by low temperature and dormancy-breaking reagent treatment.

The present study investigated the expressional regulation of PpDAM5 and PpDAM6, two of the six peach (Prunus persica) dormancy-associated MADS-box genes, in relation to lateral bud endodormancy. PpDAM5 and PpDAM6 were originally identified as homologues of Arabidopsis SHORT VEGETATIVE PHASE/AGAMOUS-LIKE 24 identified in the EVERGROWING locus of peach. Furthermore, PpDAM5 and PpDAM6 have recently been suggested to be involved in terminal bud dormancy. In this study, seasonal expression analyses using leaves, stems, and lateral buds of high-chill and low-chill peaches in field conditions indicated that both genes were up-regulated during the endodormancy period and down-regulated with endodormancy release. Controlled environment experiments showed that the expression of both PpDAM5 and PpDAM6 were up-regulated by ambient cool temperatures in autumn, while they were down-regulated by the prolonged period of cold temperatures in winter. A negative correlation between expression levels of PpDAM5 and PpDAM6 and bud burst percentage was found in the prolonged cold temperature treatment. Application of the dormancy-breaking reagent cyanamide to endo/ecodormant lateral buds induced early bud break and down-regulation of PpDAM5 and PpDAM6 expression at the same time. These results collectively suggest that PpDAM5 and PpDAM6 may function in the chilling requirement of peach lateral buds through growth-inhibiting functions for bud break.

Introduction

Bud dormancy allows perennial plants of temperate and boreal zones to survive low winter temperatures. Lateral buds are formed in early summer and enter a paradormant state that is mainly caused by apical dominance. By mid- to late autumn, the timing varies depending on species, the inhibitory control of bud growth shifts to the bud itself; these buds are referred to as being endodormant (Lang, 1987). Plants are incapable of emerging from this type of dormancy despite removal of terminal buds or defoliation. Endodormant buds require a certain amount of chilling to make the transition to an ecodormant state, in which buds are capable of resuming growth in favourable environments (Crabbe and Barnola, 1996; Faust ). In contrast to endodormancy, ecodormancy is imposed by external environmental factors such as cold or drought stress, which induce critical signals that prevent bud growth (Crabbe and Barnola, 1996; Lang, 1987; Horvath ).

Endodormancy is the result of physiological changes in response to an internal signal that inhibits the continuous rapid growth towards bud break. Certain signals that mediate the induction of endodormancy have been identified and characterized. The role of plant hormones such as abscisic acid, gibberellic acid, and ethylene in endodormancy is well established (Rohde ; Rohde and Bhalerao, 2007). Although a complex set of overlapping hormonal signals is known to be responsive to various environmental and physiological conditions related to dormancy (Horvath ; Rohde and Bhalerao, 2007), the manner in which environmental and hormonal signals are involved in dormancy remains to be elucidated.

Studies using both genetic and molecular approaches to examine the mechanisms that regulate bud endodormancy and dormancy release have recently been conducted for various plant species (for reviews, see Rohde and Bhalerao, 2007; Horvath, 2009). Endogenous processes that induce growth cessation and terminal bud set have been partly characterized. Böhlenius et al. (2006) suggested that homologues of CONSTANS and FLOWERING LOCUS T, which are flowering regulators in Arabidopsis thaliana, are involved in short-day-induced seasonal growth cessation of Populus trichocarpa. Recently, CENTRORADIALIS (CEN)/TERMINAL FLOWER 1 was also shown to be involved in the endodormancy release of Populus (Mohamed ). However, genes that are involved in lateral bud endodormancy induction and release in temperate fruit tree species are poorly understood.

Suppression subtractive hybridization supplemented with mirror orientation selection (SSH/MOS) and differential screening were performed to identify genes that are preferentially expressed in endodormant buds of the Japanese apricot (Prunus mume Sieb. et Zucc.), a temperate fruit tree species (Yamane ). A MADS-box gene that shows endodormancy-associated expression was obtained. Seasonal expression analysis suggested that the MADS-box gene was up-regulated during endodormancy induction and down-regulated during endodormancy release. Sequence and phylogenic analyses of the full-length cDNA clone revealed a similarity between this gene and StMADS11-clade MADS-box genes such as SHORT VEGETATIVE PHASE and AGAMOUS-LIKE 24 of Arabidopsis (Yamane ). In addition, Bielenberg identified six StMADS11-clade MADS-box genes as candidate genes associated with terminal bud formation in peach [Prunus persica (L.) Batsch] and named them DORMANCY ASSOCIATED MADS (DAM) genes. Six PpDAM genes showed distinct seasonal expression patterns in the terminal buds of peach, and PpDAM1, PpDAM2, and PpDAM4 were suggested to be closely associated with terminal bud formation (Li ). Horvath also indicated a possible involvement of the DAM homologues EeDAM1 and EeDAM2 with endodormancy induction in leafy spurge (Euphorbia esula). Recently, Fan showed the localization of a strong quantitative trait locus associated with the chilling requirement (CR) and a blooming date of peach PpDAM6 locus. Moreover, Jimenez showed that expression levels of PpDAM5 and PpDAM6 in shoot tips were negatively correlated with the time required for bud break in forcing conditions. These results suggested that StMADS11-clade MADS-box genes are candidates for internal factors controlling the endodormancy of perennial plants. However, no direct evidence has been presented for the involvement of DAM in bud dormancy through its growth inhibitory effects.

Currently, uniform flowering, which is indispensable for stable fruit production, is difficult to achieve in either protected or orchard cultivation in the southern areas of Japan. This situation is caused by irregular lateral bud endodormancy release, possibly due to global warming (Sugiura ). Therefore, the genetic factors that control endodormancy were investigated in order to understand the molecular basis of endodormancy regulation in temperate fruit tree species, which could lead to the artificial control of endodormancy through cultural practices and/or rapid breeding such as marker-assisted selection. As a first step, the investigation focused on PpDAM5 and PpDAM6, the most likely candidate genes for peach lateral bud endodormancy release (Fan ; Jimenez ; R. Sassaki, H. Yamane, T. Ooka, H. Jotatsu, T. Akagi, and R. Tao, unpublished data).

In this study, the different seasonal expression patterns of PpDAM5 and PpDAM6 were demonstrated in peach cultivars with different CR for bud break. Subsequently, temperature-regulated changes in the expression of PpDAM5 and PpDAM6 were shown, suggesting their association with temperature-mediated dormancy regulation. Finally, the effects of dormancy-breaking stimuli, such as a prolonged period of low temperature and the effect of the dormancy-breaking chemical cyanamide, on dormancy status and expression of PpDAM5 and PpDAM6, were investigated. Based on the results obtained, the biological functions of PpDAM5 and PpDAM6 are discussed.

Materials and methods

Plant materials

The peach cultivars Akatsuki, Shimizu Hakuto, and Tsukuba Ichigo were used in this study. Shimizu Hakuto and Akatsuki, grafted on Ohatsumomo rootstock, were obtained from a commercial nursery and planted at the Kyoto University experimental farm in 2002. Cuttings of Tsukuba Ichigo were kindly provided by Dr Kataoka from Kagawa University in 2002, and these cuttings were planted at the Kyoto University experimental farm in 2005. Tsukuba Ichigo is the F2 offspring of Akame (high-chill)×Okinawa (Tsukuba strain). Okinawa is a rootstock cultivar with a low CR for bud break (low-chill) and Tsukuba Ichigo was bred as a low-chill rootstock cultivar at the National Institute of Fruit Tree Science, Tsukuba, Japan. The chilling requirements for bud break of Akatsuki, Shimizu Hakuto, and Tsukuba Ichigo are ∼900 h CR, 1000 h CR, and 400–450 h CR, respectively. Akatsuki and Shimizu Hakuto are categorized as high-chill cultivars whereas Tsukuba Ichigo is a low-chill cultivar.

Long 1-year-old branches (current season's growth) of Shimizu Hakuto and Tsukuba Ichigo were collected monthly from June to March in 2007 or from June to February in 2008. Shimizu Hakuto and Tsukuba Ichigo trees were aged 7 and 6 years, respectively, and were considered to be in the adult phase. Air temperature was recorded at 15-min intervals and chilling accumulation was calculated according to the ‘chill hours’ (<7.2 °C). Pot-grown adult trees of Akatsuki were used for the ambient cool temperature (15–18 °C) treatment experiment. For the cold temperature (6–12 °C) treatment experiment, long 1-year-old branches of Akatsuki collected on 10 December 2008 were used. For the cyanamide treatment experiment, long 1-year-old branches of adult trees of Shimizu Hakuto collected on 25 December 2009 were used.

Seasonal endodormancy status and expression analysis

The timing of endodormancy induction and endodormancy release of lateral vegetative buds of Shimizu Hakuto and Tsukuba Ichigo were investigated by two different methods. From June to March, three 1-year-old branches of ∼50 cm length were cut from trees and the basal parts were placed in water containing Misakifarm™ (Otsuka Kagaku, Tokushima, Japan), a commercial preparation containing nutrients and fungicides for the prolonged life of flower cuttings. At the same time, the basal parts of 10 single-node cuttings obtained from the middle portions of each of three branches were placed in water containing Misakifarm™. The branches and single node cuttings were maintained at 22 °C under cool white fluorescent light with a 16-h light/8-h dark photoperiod. The water containing Misakifarm™ was replaced every 2 weeks. Lateral vegetative buds showing leaves were considered to have burst.

Vegetative buds of Shimizu Hakuto and Tsukuba Ichigo and leaves and stems of Shimizu Hakuto were collected monthly and immediately frozen in liquid nitrogen and stored at −80 °C until use. Total RNA was isolated from the collected tissues as described in Yamane . After DNase I treatment (Takara Bio, Inc., Ohtsu, Japan), 1 μg of total RNA was used for cDNA synthesis with SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Based on the genomic DNA sequence for PpDAM5 and PpDAM6 from the NCBI nucleotide database, gene-specific TaqMan® probes and primers for specifically detecting transcripts of PpDAM5 and PpDAM6, were designed and synthesized. The primer pairs DAM5F and DAM5R and PpDAM6F and PpDAM6R were used with the TaqMan® probes DAM5-T and DAM6-T, respectively, to detect transcripts of PpDAM5 and PpDAM6, respectively (Table 1). Real-time PCR analysis with TaqMan® probes was performed using LightCycler 480 (F. Hoffmann-La Roche AG, Basel, Switzerland) and probe master mix (F. Hoffmann-La Roche AG). The reaction mixture consists of 1× probe master mix, 500 nM each of forward and reverse primers, 200 nM TaqMan® probe, and cDNA equivalent to the amount synthesized from 4 ng of total RNA in a 20-μl reaction volume. As a control, UBIQUITIN (UBQ) transcript accumulation was monitored by real-time PCR using SYBR Green Master Mix (Roche). UBQ was selected as a reference because the transcript level of UBQ appeared to be most stable across tissue types and sampling dates among the transcript levels of three genes tested, UBQ, ACTIN, and Elongation Factor 1 (EF-1), although expression of both UBQ and ACTIN was far more stable than EF-1. A primer pair PaubiF and PaubiR was used for UBQ-specific real-time PCR (Table 1). PCR was performed using a programme of 45 cycles at 95 °C for 10 s, 56 °C for 20 s, and 72 °C for 1 s with an initial heating at 95 °C for 5 min. For UBQ-specific real-time PCR, dissociation curve analysis was performed to confirm that the fluorescence was only derived from the gene-specific amplification. Three technical replicates were performed for each gene and mesurement. The quantities of the transcripts of PpDAM5 and PpDAM6 in each sample were normalized using the UBQ transcripts.

Table 1.

Sequences of primers and TaqMan® probes used in this study

Primer or TaqMan® probe	Sequence (fluorescent probe)	Tm (°C)
DAM5F	5′-ATCTCCACCACCTGCAACAGT-3′	61.9
DAM5R	5′-CTTCTTAACGCCCCAGTTTGAG-3′	62.3
PpDAM6F	5′-TAATGTTGGAGGTGGAGGAGAA-3′	60.8
PpDAM6R	5′-GGGAAGCCCCAGTTTGAGA-3′	62.5
DAM5-T (probe)	5′-(VIC)TCTTGAAGATGACTCCTCCGA(MGB)-3′	69.3
DAM6-T (probe)	5′-(FAM)TGAAGATGACTGCTCCGATGT CACTTTATC-(TAMRA)-3′	69.0
PaubiF	5′-CGAACCCTAGCCGATTACAA-3′	56.5
PaubiR	5′-AGTGGTTCGCCATGAAAGTC-3′	56.3

Ambient cool temperature and prolonged low-temperature treatments in controlled environment conditions

The experiment was conducted on 8 September 2008; 10 pot-grown Akatsuki trees with five 1-year-old branches were used. Four pots each were placed in two different growth chambers with different growth conditions. The two growth conditions used were under a 12-h light/12-h dark photoperiod (300 μmol/m2/s) at 18 °C light/15 °C dark [cool(+)] and at 28 °C light/23 °C dark [cool(−)]. Buds consisting of both flower and vegetative buds were excised from the middle portions of branches at 0, 10, and 20 d after treatment and immediately frozen in liquid nitrogen and kept at −80 °C until use. Total RNA was isolated and used for the expression analysis of PpDAM5 and PpDAM6, as described above.

In prolonged low-temperature treatment, 1-year-old branches were cut from Akatsuki trees on 10 December 2008, and placed in a growth chamber under a 9-h light/15-h dark photoperiod cycle at 12 °C light/6 °C dark [cold(+)] or 24 °C light/22 °C dark [cold(−)]. To investigate the endodormancy status changes by treatment, three branches were transferred to forcing conditions (9-h light/15-h dark photoperiod cycle at 24 °C light/22 °C dark) at 0 d and 60 d after treatment and their bud burst was recorded for 60 d. Vegetative buds were excised from the middle portions of branches at 0 d and 60 d after treatment and immediately frozen in liquid nitrogen and stored at −80 °C until use. Total RNA was isolated from the buds stored at −80 °C and used for expression analysis of PpDAM5 and PpDAM6, as described above.

Dormancy-breaking reagent (cyanamide) treatment

On 25 December 2009, long 1-year-old branches were cut from Shimizu Hakuto trees. Their basal parts were placed in water containing Misakifarm™ and incubated in a greenhouse for a natural day length and under controlled temperature (18–22 °C). Subsequently, diluted CX-10™ (Nippon Carbide industries Co., Inc., Tokyo, Japan) solution (containing 0.6% effective cyanamide) with 0.05% Tween 20 (Wako Pure Chemical Industries, Ltd, Osaka, Japan) added as a surfactant was sprayed onto the branches. In the control treatment, 0.05% Tween 20 was used as a spray. Vegetative buds were excised from the middle portions of branches at 0, 2, and 4 weeks after treatment and immediately frozen in liquid nitrogen and stored at −80 °C until use. Total RNA was isolated and used for the expression analysis of PpDAM5 and PpDAM6, as described above. The percentages of the branches with bud break were recorded using 13–15 branches 5 weeks after treatment.

Results

Seasonal endodormancy status and expression patterns of PpDAM5 and PpDAM6 in lateral buds

Table 2 shows the seasonal changes in the percentage bud break of vegetative buds of the high-chill cultivar Shimizu Hakuto and the low-chill cultivar Tsukuba Ichigo in forcing conditions. In Shimizu Hakuto, no bud burst was observed in the branches sampled in July [chilling hours (CH), 0 h], August (CH, 0 h), September (CH, 0 h), October (CH, 2 h), November (CH, 64 h), and December (CH, 414 h). In January (CH, 1053 h), 32.7% of the vegetative buds had opened. In February (CH, 1831 h), >60% of the vegetative buds had burst. No bud burst was observed in Tsukuba Ichigo from September to November. In December, 31% of vegetative buds of Tsukuba Ichigo had burst. In January and February, the number of Tsukuba Ichigo buds that opened was higher than those of high-chill Shimizu Hakuto. In the case when single node cuttings were used instead of whole branches to examine dormancy status, Tsukuba Ichigo buds opened in all the months tested, while no Shimizu Hakuto buds opened from August to December (Table 2).

Table 2.

Seasonal endodormancy depth of lateral vegetative buds of two peach cultivars using whole branches (A) or branch parts (B)

A. Whole branches

Bud burst percentage by month

Cultivar

June

July

Aug.

Sept.

Oct.

Nov.

Dec.

Jan.

Feb.

Mar.

Shimizu Hakuto

73.3

0

0

0

0

0

0

32.7

62.7

100

Tsukuba Ichigo

86.7

31.7

3.3

0

0

0

31

54

95.5

NT

B. Single node cuttings

Days to bud burst by montha

Cultivar

June

July

Aug.

Sept.

Oct.

Nov.

Dec.

Jan.

Feb.

Mar.

Shimizu Hakuto

++

++

−

−

−

−

−

+

++

NT

Tsukuba Ichigo

++

++

++

++

+

+

++

++

++

NT

First bud burst was observed within 15 d (++), 30 d (+), or not observed within 30 d (−).

NT, not tested.

The seasonal expression changes of PpDAM5 and PpDAM6 in the lateral vegetative buds were monitored by real-time RT-PCR analysis using gene-specific TaqMan® probe and primers. From June to November, the expression levels of PpDAM5 and PpDAM6 in high-chill Shimizu Hakuto steadily increased. The highest transcript levels of PpDAM5 and PpDAM6 were observed in November. Subsequently, their transcript levels in vegetative buds steadily decreased towards spring (Fig. 1). From June to September, the expression levels of PpDAM5 and PpDAM6 in low-chill Tsukuba Ichigo were similar to or little bit lower than those in high-chill Shimizu Hakuto. However, their expression levels showed distinct patterns from October, with the transcript levels in low-chill Tsukuba Ichigo being much lower than those in Shimizu Hakuto. Transcript levels of PpDAM5 and PpDAM6 in Tsukuba Ichigo in January and February and those in Shimizu Hakuto in March were lowest (Fig. 1).

Fig. 1.

Seasonal expression changes of PpDAM5 and PpDAM6 in lateral vegetative buds of two peach cultivars, high-chill cultivar Shimizu Hakuto and low-chill cultivar Tsukuba Ichigo. Each data point represents the mean value of three technical replicates.

Seasonal expression changes of PpDAM5 and PpDAM6 in leaves and stems

In leaves of Shimizu Hakuto, the seasonal expression patterns of PpDAM5 and PpDAM6 steadily increased from early summer to autumn until the peak in October just before defoliation in the field (Fig. 2). Both PpDAM5 and PpDAM6 expression in stems peaked in November and December and drastically decreased in January and continued decreasing until spring. After December, the PpDAM5 and PpDAM6 expression patterns in stems were similar to those in lateral vegetative buds (Fig. 1).

Fig. 2.

Seasonal expression changes of PpDAM5 and PpDAM6 in leaves (A) and stems (B) of peach (cultivar Shimizu Hakuto). Each data point represents the mean value of three technical replicates.

Up-regulation of PpDAM5 and PpDAM6 in lateral buds by autumn ambient cool temperature (15–18 °C)

Seasonal expression analysis demonstrated that the transcript levels of both PpDAM5 and PpDAM6 increased in autumn when the ambient temperature decreased. It was assumed that changes in their expression level were controlled by alterations in the ambient temperature. To test this hypothesis, the effect of ambient cool temperature on the expression of PpDAM5 and PpDAM6 was investigated in a controlled environment experiment. The expression of both PpDAM5 and PpDAM6 increased in the cool(+) treatment, but remained constant in the cool(−) treatment (Fig. 3).

Fig. 3.

Effects of ambient cool temperature (15–18 °C) in autumn (September) on the expression of PpDAM5 (left) and PpDAM6 (right) in lateral vegetative buds of peach (cultivar Akatsuki). Each data point represents the mean value of three technical replicates.

Down-regulation of PpDAM5 and PpDAM6 through exposure to prolonged winter cold temperature (6–12 °C) treatment

Seasonal expression analysis (Fig. 1) demonstrated that the transcript level of both PpDAM5 and PpDAM6 steadily decreased from November to March when the plants were exposed to a prolonged cold temperature period over autumn and winter, and subsequently to warm temperatures in early spring. To examine the changes in expression of PpDAM5 and PpDAM6 in response to the prolonged cold temperature in early winter, branches excised in December were subjected to cold(+) or cold(−) temperature treatment, and dormancy status and PpDAM5 and PpDAM6 transcript levels in lateral vegetative buds were investigated. Cold temperature treatment for 60 d enhanced bud break. In forcing conditions, no Akatsuki buds opened after the cold(−) treatment, while most buds opened within 9 d after the cold(+) treatment. PpDAM5 and PpDAM6 transcript levels were constant after the cold(−) treatment, while they decreased after the cold(+) treatment. As a result, PpDAM5 and PpDAM6 transcript levels were lower in the cold(+) treatment than in the cold(−) treatment (Fig. 4).

Fig. 4.

Effects of prolonged low temperature (6–12 °C) in winter (December) on the expression of PpDAM5 (left) and PpDAM6 (right) in lateral vegetative buds of peach (cultivar Akatsuki). Each data point represents the mean value of three technical replicates.

Down-regulation of PpDAM5 and PpDAM6 by cyanamide treatment

Cyanamide, a dormancy-breaking reagent, treatment in late December (at ∼500 CH) effectively promoted Shimizu Hakuto bud break. Five weeks after treatment, at least one bud on eight of 13 (61.5%) cyanamide-treated branches had burst, whereas none of the buds on 13 untreated branches burst. Although PpDAM5 and PpDAM6 transcript levels decreased in both cyanamide-treated and -untreated branches, they were much lower in treated branches than in untreated ones (Fig. 5). Changes in the expression patterns of PpDAM5 and PpDAM6 differed, PpDAM6 showed a greater difference in expression level between treated and untreated branches than that observed for PpDAM5.

Fig. 5.

The effects of cyanamide treatment on the expression of PpDAM5 (left) and PpDAM6 (right) in lateral vegetative buds of peach (cultivar Shimizu Hakuto). Each data point represents the mean value of three technical replicates.

Discussion

It was previously reported that the putative homologue of PpDAM6 in Japanese apricot, PmDAM6, showed endodormancy-associated expression in lateral buds (Yamane ). Further comprehensive expression and transgenic analyses suggested an association between PmDAM6 with lateral bud endodormancy induction and release in Japanese apricot through its putative growth inhibitory functions against bud break and shoot growth (R. Sassaki, H. Yamane, T. Ooka, H. Jotatsu, T. Akagi, and R. Tao, unpublished data).

PpDAM genes were initially identified as candidate genes for growth cessation and terminal bud formation in peach using the peach evergrowing mutant (Bielenberg ). However, Li suggested that three peach PpDAM genes, namely PpDAM1, PpDAM2, and PpDAM4, were likely candidates to regulate terminal bud formation. Later, Jimenez showed that PpDAM5 and PpDAM6 had functions other than terminal bud formation. Namely, they demonstrated negative correlation of the expression of PpDAM5 and PpDAM6 with peach terminal bud break. Furthermore, Horvath reported the association of putative homologues of PpDAMs, EeDAM1, and EeDAM2 with endodormancy induction in leafy spurge crown buds. With regard to lateral buds, PpDAM5 and PpDAM6 have been shown to be co-localized with a quantitative trait locus associated with the chilling requirement and bloom date of peach (Fan ). Although this suggests that PpDAM5 and PpDAM6 are responsible for regulating the chilling requirement and/or bloom date in peach lateral buds, comprehensive expressional analyses of PpDAMs with lateral bud endodormancy have not been conducted. This study focused on the relationship between the expression of PpDAMs and lateral bud endodormancy status with different environmental and chemical stimuli.

The effects of environmental signals on the expression of DAMs have been studied in perennial plants. A reduction in photoperiod and a short period of cold exposure induces the expression of peach PpDAMs (Li ) and leafy spurge EeDAMs (Horvath ), respectively. The effects of prolonged cold exposure on PpDAMs (Jimenez ) and EeDAMs (Horvath ) were also investigated. Although down-regulation of DAMs was recorded following prolonged cold exposure in both plants, lack of a control treatment made it difficult to interpret the results conclusively.

In this study, the association of PpDAM5 and PpDAM6 with phenological changes in the endodormancy status of lateral vegetative buds of two peach cultivars with different chilling requirements for bud break were first validated. The effects of temperature and dormancy-breaking reagent stimuli on the expression of PpDAM5 and PpDAM6 were then characterized, with viable control treatments. The results further substantiated previous studies that have demonstrated PpDAM5 and PpDAM6 to have significant roles in endodormancy regulation in peach.

The low-chill cultivar entered endodormancy later than the high-chill cultivar, probably because of the prolonged period of shoot growth in low-chill cultivars in summer. Based on the bud burst percentage, the endodormancy of the high-chill cultivar appeared to be released after January, whereas that of the low-chill cultivar was released after November/December. Thus, the endodormancy period of the low-chill cultivar was shorter than that of the high-chill cultivar. Endodormancy was induced in July/August in Shimizu Hakuto and in September/October in Tsukuba Ichigo. However, prominent increases in the expression of PpDAM5 and PpDAM6 were not found in the respective period in either cultivar. This observation indicates that PpDAM5 and PpDAM6 are unlikely to trigger lateral bud endodormancy, although they seemed to be involved in endodormancy maintenance and release in late fall to early spring because their expression was subsequently up-regulated and down-regulated in coordination with endodormancy progression and release in the lateral buds of both cultivars. The same trends in dormancy-coordinated expression pattern were observed in PpDAM5 and PpDAM6 in stems and leaves. Dormancy-associated expression patterns of PpDAM5 and PpDAM6 were also found in peach terminal buds (Li ; Jimenez ). These observations collectively indicate that the expression of PpDAM5 and PpDAM6 is regulated in an endodormancy-coordinated manner in whole shoots of peach, contrasting with the dormancy-associated, but bud-specific, expression of EeDAM1 and EeDAM2 (Horvath ).

Lateral bud dormancy status is conceptually divided into the following five phases: paradormant, endodormant, ecodormant, and the two transition phases para–endo and endo–eco phases (Lang, 1987). Based on the present authors’ survey of seasonal endodormancy status, Shimizu Hakuto peach was paradormant in June, in the para–endo transition phase in July, endodormant from August to December, in the endo–eco transition phase in January, and ecodormant in February and March. Transcript levels of paradormant buds were similar to those of endodormant buds. In addition, when the bud burst percentage in forcing conditions and transcripts levels of PpDAM5 and PpDAM6 between February and March were compared, a higher bud burst percentage and lower transcript levels were found in March than in February. These results suggested that PpDAM5 and PpDAM6 could function as a dose-dependent growth inhibitor not only in endodormant buds but also in paradormant and ecodormant buds.

Li demonstrated that PpDAM5 and PpDAM6 were up-regulated by a reduction in photoperiod in controlled environment experiments. As the shortening of photoperiod in autumn coincides with a fall in temperature, the effects of temperature changes on PpDAM5 and PpDAM6 expression under a 12-h light/12-h dark photoperiod cycle in September were tested. Expression of both PpDAM5 and PpDAM6 was up-regulated only by cool(+) treatment, indicating that ambient cool temperature in September stimulates PpDAM5 and PpDAM6 expression. In July, similar experiments were conducted under a 14-h light/10-h dark photoperiod cycle with the same cool temperature regime (data not shown); however, this trend was only observed with PpDAM5 and not with PpDAM6. This suggests that the up-regulation of PpDAM5 and PpDAM6 expression by ambient cool temperature depends on the bud stage and/or photoperiod condition. Horvath found that EeDAM1 was cold stress (11 °C) responsive and contains putative C-repeat/DRE-binding factor (CBF) sites, which are known to be cis-regulating motifs targeted by the cold/drought stress CBF regulon, within the first 2000-base upstream region from the EeDAM1 translation initiation codon. Since CBF sites were found in both PpDAM5 and PpDAM6 (data not shown), it is possible that their cold-responsive expression was controlled by CBF protein as suggested for cold-responsive EeDAM1 expression.

The controlled low-temperature treatment experiment indicates that a prolonged period of low temperature (6–12 °C) is required for both the steady down-regulation of PpDAM5 and PpDAM6 and the increase in bud burst percentage; thus, further substantiating the observations made in peach shoot terminals by Jimenez . Horvath suggested that environmentally regulated down-regulation of DAM1 appeared to be controlled, at least in part, by chromatin remodelling. Since FLOWERING LOCUS C (FLC), a MADS-box gene involved in vernalization of Arabidopsis, was also repressed by chromatin remodelling triggered by a prolonged period of low temperature (Amasino, 2010), it would be intriguing to see whether the down-regulation of PpDAM5 and PpDAM6 is mediated by the same mechanisms.

Various chemicals and culture methods have been tested for their dormancy-breaking effects in several fruit tree species (Chang and Lin, 1989; Nee, 1991; Bound and Jones, 2004). One of the best known dormancy-breaking chemicals is cyanamide (hydrogen cyanamide), which has been used for dormancy studies and has been incorporated into cultural practices (Or , 2002; Walton ). Application of cyanamide at the end of the endodormant or endo–ecodormant stage in late winter/early spring is known to induce synchronized bud burst in grapevines. However, cyanamide is ineffective in the middle of an endodormant stage. Although the effects of cyanamide on peach dormancy have been poorly studied, probably because of its potential toxicity to peach buds, these application methods successfully enhanced Shimizu Hakuto peach bud burst with simultaneous down-regulation of PpDAM5 and PpDAM6. This result strongly supported the idea that PpDAM5 and PpDAM6 could act as a dose-dependent inhibitor to peach bud break not only at the endodormant stage but also at the ecodormant stage.

Endodormant buds differ from other dormant buds (paradormant and ecodormant buds), in their inability to resume growth under favourable conditions. This suggests that internal factors inhibiting bud burst and subsequent shoot growth should be present in endodormant buds (Faust ). Although PpDAM5 and PpDAM6 could be good candidates for this function in peach endodormant buds, the present study indicates that they act not only in endodormant buds but also in ecodormant buds. Therefore, PpDAM5 and PpDAM6 are different from the previously presumed internal inhibitors that are assumed to be specifically present in endodormant buds. Transgenic approaches are currently underway to completely elucidate the functions of PpDAMs.