Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2007 Jan 3;99(2):293–299. doi: 10.1093/aob/mcl257

Regulation of Summer Dormancy by Water Deficit and ABA in Poa bulbosa Ecotypes

Micha Ofir 1,*, Jaime Kigel 1
PMCID: PMC2802989  PMID: 17202183

Abstract

Background and Aims

Survival of many herbaceous species in Mediterranean habitats during the dry, hot summer depends on the induction of summer dormancy by changes in environmental conditions during the transition between the winter (growth) season to the summer (resting) season, i.e. longer days, increasing temperature and drought. In Poa bulbosa, a perennial geophytic grass, summer dormancy is induced by long days, and the induction is enhanced by high temperature. Here the induction of summer dormancy in a Mediterranean perennial grass by water deficit under non-inductive photoperiodic conditions is reported for the first time.

Methods

Plants grown under 22/16 °C and non-inductive short-day (9 h, SD) were subjected to water deficit (WD), applied as cycles of reduced irrigation, or sprayed with ABA solutions. They were compared with plants in which dormancy was induced by transfer from SD to inductive long-day (16 h, LD). Responses of two contrasting ecotypes, from arid and mesic habitats were compared. Dormancy relaxation in bulbs from these ecotypes and treatments was studied by comparing sprouting capacity in a wet substrate at 10 °C of freshly harvested bulbs to that of dry-stored bulbs at 40 °C. Endogenous ABA in the bulbs was determined by monoclonal immunoassay analysis.

Key Results

Dormancy was induced by WD and by ABA application in plants growing under non-inductive SD. Dormancy induction by WD was associated with increased levels of ABA. Bulbs were initially deeply dormant and their sprouting capacity was very low, as in plants in which dormancy was induced by LD. Dormancy was released after 2 months dry storage at 40 °C in all treatments. ABA levels were not affected by dormancy relaxation.

Conclusions

Summer dormancy in P. bulbosa can be induced by two alternative and probably additive pathways: (1) photoperiodic induction by long-days, and (2) water deficit. Increased levels of endogenous ABA are involved in both pathways.

Key words: Poa bulbosa, abscisic acid, water deficit, ecotype, geophyte, perennial grass, summer dormancy

INTRODUCTION

The phenological phase of summer dormancy occurs frequently in perennial herbaceous species and geophytes inhabiting regions with a bi-seasonal Mediterranean type of climate, i.e. a growth season during the mild, rainy winter and spring, followed by a resting season during the hot and dry summer. Induction of summer dormancy and the transition from active growth to the resting stage involve major developmental changes, such as increased storage in basal parts of the shoots and in the roots, including production of special storage organs in geophytes (e.g. bulbs, corms, tuberous roots), formation of dormant regeneration buds, arrest of meristematic activity leading to termination of leaf production, followed by senescence of above-ground parts, and re-translocation of metabolites to the storage organs (Rees, 1992). Initiation of the dormant stage and associated plant structures can be a programmed, autonomous developmental process (Naor and Kigel, 2002), or an environmentally induced process. In the case of induced summer dormancy, the inductive environmental conditions occur during the transition between the winter–spring growth season and the summer resting season, i.e. increasing day length and temperature (Poa scabrella: Laude, 1953; Hordeum bulbosum: Ofir and Koller, 1972; Poa bulbosa: Ofir and Kerem, 1982; Ranunculus sp.: Farina et al., 1985; Anemone coronaria: Ben-Hod et al., 1988). The response to these factors may be promoted by exposure to the winter conditions, i.e. short days and low temperature (H. bulbosum: Ofir and Koller, 1974; M. Ofir, unpubl. res.; P. bulbosa: Ofir and Kigel, 1999). In these species a stage of true, i.e. endo- summer dormancy (Lang et al., 1987) is induced, since growth is arrested even under favourable conditions, until dormancy is gradually released by the exposure to high temperatures during the summer (Ofir et al., 1967; Ofir, 1986).

Another important environmental factor associated with the oncoming of summer is water deficit, but its potential role in the induction of summer dormancy has not been studied yet. In some Mediterranean perennial grasses, a summer resting phase is enforced by drought, but growth is resumed as soon as water becomes available during the summer after summer storms or irrigation, i.e. enforced, eco-summer dormancy (Lang et al., 1987), as in Phalaris aquatica (Oram, 1983) and Dactylis glomerata (Volaire, 2002; Piano et al., 2004; Norton et al., 2006). Here induction of true summer dormancy by water deficit was studied in Poa bulbosa, a small perennial grass geophyte that thrives in the Mediterranean basin and in neighbouring arid regions (Heyn, 1962), and in which dormancy is induced by long days and high temperature (Ofir and Kerem, 1982; Ofir and Kigel, 1999). Poa bulbosa grows in a wide range of climatic and rainfall conditions (Heyn, 1962). In Israel ecotypic differentiation occurs along the rainfall gradient between the Mediterranean and desert regions, with earlier onset of summer dormancy, enhanced flowering and smaller leaves with higher reflectance typically occurring in the desert ecotypes (Ofir and Kigel, 2003). The frequency and amount of rain declines gradually during the transition between the winter and summer seasons, in regions with a Mediterranean climate, particularly in the drier eastern Mediterranean region (Bitan and Rubin, 1991) Thus, it was questioned whether exposures to increasing day length, temperature or drought are alternative pathways for dormancy induction in P. bulbosa. This possibility is particularly relevant for this species, since it inhabits open sites with shallow soils and, therefore, is frequently exposed to drought and high temperature stresses even late in winter, when days are still relatively short and non-inductive for dormancy (Ofir and Dorenfeld, 1992). Previously it had been found that in P. bulbosa inductive long days increase endogenous levels of ABA and that application of ABA induced summer dormancy (Ofir and Kigel, 1998). On the other hand, since it is well established that drought stress increases endogenous ABA (Volaire et al., 1998b; DaCosta et al., 2004; Wang et al., 2004; Davies et al., 2005; Zhang et al., 2006), it can be argued that water deficit during drought stress may be involved in the induction of summer dormancy in P. bulbosa.

Thus, the aims of the present research were: (a) to examine whether the stress resulting from water deficit can induce dormancy in P. bulbosa under non-inductive short days and moderate temperatures; (b) to compare the effects of water deficit on plant development and dormancy induction with those of exogenous ABA and long-day induction; (c) to asses the depth of dormancy induced by long-day, ABA and water-deficit treatments. Since propensity for summer dormancy increases in P. bulbosa populations across gradients of greater aridity, i.e. dormancy was imposed earlier in populations of more arid habitats (Ofir and Kigel, 2003), these questions were examined in two contrasting ecotypes, from a mesic and an arid site.

MATERIALS AND METHODS

Plant material and growth conditions

The following procedures were employed in all treatments: resting, dry bulbs of Poa bulbosa were collected at the end of the dormant phase, in September 2000, before the first autumn rains, from natural populations (‘ecotypes’) in two sites in Israel, differing in their aridity (Fig. 1): (1) a mesic ecotype from Tirat Shalom in the coastal plain, 32°N35°E, 50 m a.s.l., 530 mm rain year−1, with calcareous shallow degraded sandstone soil, and (2) an arid ecotype from Lehavim in the northern Negev, 31°N34°E, 280 m a.s.l., 310 mm rain year−1, with stony loess soil (loess derived Serozem). The bulbs were kept in paper bags at room temperature (20 ± 5 °C) until planting (22 February 2001) in the phytotron of the Faculty of Agriculture, Rehovot, Israel, in glass-covered growth rooms transmitting 80 % of the outside solar radiation (Ofir and Kigel, 2003, 2006). The bulbs were planted in a moist substrate (vermiculite–tuff gravel; 1 : 1, v/v), in drained trays, with short days at 12 °C until sprouting. After sprouting the plants grew at day/night air temperature and relative humidity of 22/16 °C ± 0·5 °C and 65/75 % ± 5 %, respectively. Day conditions were maintained between 0700 and 1600 h. Day/night temperature changes were gradual, spanning 3 h. Day-length conditions applied in the present study were short days (SD), 9 h (0700–1600 h) of natural light, and long days (LD), 16 h (0400–2000 h), attained by extending the natural day length with supplementary lighting (3–5 µmol m−2 s−1 PAR at plant level), using 75-W incandescent tungsten lamps (LM960 Osram GMBH, München, Germany).

Fig. 1.

Fig. 1.

Open in a new tab

Mean monthly rainfall in meteorological stations nearest to the two locations from which the bulbs of P. bulbosa were collected: the arid habitat at Lehavim and the mesic habitat at Tirat Shalom. Data for the latter are from Bet Dagan, 12 km north of Tirat Shalom. The data are from Bitan and Rubin (1991).

Dormancy induction treatments

Onset and release of dormancy were studied in plants subjected to water deficit and in ABA-treated plants, grown at SD and 22/16 °C, photo-thermal conditions in which summer dormancy is not induced (Ofir and Kigel, 1999), and compared with plants in which dormancy was induced by LD, or to plants growing continuously under non-inductive SD.

Effects of water deficit under non-inductive SD (‘SD + WD’)

Sprouted bulbs were transplanted into 3-L drained pots filled with a measured amount of wet, grade 2 perlite, four to six plants per pot at SD, 22/16 °C. The perlite was covered by a weighed amount of coarse tuff gravel (>5 mm), to reduce substrate disturbance during irrigation, water-loss from the perlite surface, and algae growth. Weight of the empty pot, gravel and dry perlite (i.e. tare weight), and the maximal amount of water retained by the perlite after irrigation to saturation [‘pot capacity’ (PC)] were determined prior to planting, for each pot. The substrate was kept moist by irrigating to saturation every 3–5 d alternately with water or 50 % Hoagland's nutrient solution. Treatments were: (a) irrigated control (‘C’): irrigation to saturation every 2–3 d until the end of the experiment, 97 and 110 d after planting, for the arid and mesic ecotypes, respectively; (b) mild water deficit (‘mild WD’); and (c) severe water deficit (‘severe WD’). Treatments started 53 d after planting (last saturation irrigation). The number of replicate plants was 12, 55 and 60 for the C, mild WD and severe WD treatments, respectively.

Pots subjected to water deficits were weighed every 1–3 d, from the last saturation irrigation to the termination of the experiment. The fresh weight of the plants as measured at the start of the water-deficit treatments was added to the tare weight. The amount of water held in each pot was determined by subtracting the tare weight from the pot gross weight, and calculated as a percentage of pot capacity (% PC). When % PC reached a pre-set minimal value, the amount of water (or nutrient solution) required to restore a higher pre-set % PC was added to each pot (Fig. 2). Imposition of the water deficit treatments was gradual and in two steps: during the first 25 d of the treatments the level of moisture of the substrate at which plants were irrigated was gradually reduced from approx. 15 % PC to 5 % PC, and the pre-set levels after irrigation were 50 and 30 % PC for the mild and severe WD treatments, corresponding to three and five irrigation cycles, respectively. In the second step, the pre-set levels after irrigation were lowered to 30 and 15 % PC, until the end of the experiment (Fig. 2). WD treatments were carried on until all the plants became dormant.

Fig. 2.

Fig. 2.

Open in a new tab

Water-deficit (WD) treatments applied to P. bulbosa plants. The moisture` content of substrate (perlite) is expressed as % of the maximal amount of water-holding capacity of the substrate (pot capacity, PC). During the drying and partial re-watering cycles substrate moisture decreased by evapo-transpiration until a certain minimum level was reached. In the mild WD treatment, substrate moisture was partially restored by irrigations to 50 % and later to 30 % of the maximal PC. In the severe WD treatment irrigations restored 30 % and later 15 % of PC.

Dormancy induction by ABA at non-inductive SD (‘SD + ABA’) and by inductive LD

Sprouted bulbs were transplanted to 1-L drained pots, in a substrate of vermiculite–volcanic tuff gravel (1 : 1 v/v). After growing for 2 months in SD and 22/16 °C, plants were subjected to: (a) LD at 22/16 °C (‘SD → LD’), 38 and 63 plants for the arid and mesic ecotypes, respectively; (b) dormancy imposition by application of ABA The plants (20 per ecotype) continued to grow at SD and were sprayed every 2 d for 40 d, with increasing concentrations of ABA (approx. 2 mL per plant): 0·1 mm in the first 10 d, 0·2 mm in the next 10 d and 0·3 mm in the last 20 d. Stock solutions of 5 m ABA were prepared by dissolving 0·040 g of S-( ± )-ABA (Sigma) in 1 mL 1 m KOH, diluting with a surfactant solution of 0·05 % (v/v) Tween-20 in water, adjusting to pH 5·5 and storing at –10 °C. Dilutions to required concentrations were made before use with 0·05 % Tween in water.

Plant and bulb dormancy

A plant was considered dormant when most (>75 %) of its leaves were yellowing and drying and most of its tillers had bulbs (Ofir and Kigel, 1999). Dormant plants were lifted from the pot and the following data were recorded: age at dormancy onset, number of tillers, number and percentage of bulbing and dormant tillers. The depth of dormancy of the bulbs produced in the different treatments was assessed by comparing the sprouting capacity (% sprouting), in a moist vermiculite–tuff gravel substrate (1 : 1, v/v) at 10 °C, of bulbs planted immediately after lifting with that of bulbs from the corresponding treatments that were dry-stored at 40 °C for 2 months to reduce dormancy (Ofir, 1986), and planted under the same conditions (pre- and post-storage sprouting, respectively). Sprouting percentages were determined 40 d after bulb planting, when sprouting reached its full capacity (i.e. no more bulbs sprouted). Four replicates of 20 bulbs were used in the mild and severe WD treatments, and five replicates of ten bulbs in the SD → LD and SD + ABA treatments.

Endogenous ABA

ABA was determined only in the mesic ecotype. Bulbs from the plants in the different treatments were sampled for determination of endogenous ABA, before (i.e. freshly collected bulbs) and after dry storage at 40 °C for 2 months. Four replicates, of 25–30 bulbs each from SD + ABA and the mild WD treatments, and 8–12 bulbs from the SD → LD treatment were used (350–500 mg f. wt per replicate). The bulbs were cut into 0·5–1·0 mm pieces and stored at –20 °C in weighed 1·5-mL Eppendorf vials with 50 µL glass-distilled water. In the bulb-less plants from the SD treatment, the basal 7 mm of the tillers were sampled similarly, 20 tillers from ten plants, 500–600 mg per replicate.

ABA immunoassay

ABA extraction and determination were carried out as described by Ofir and Kigel (1998). Tissue sap was extracted by centrifugation in a microfuge for 10 min after thawing the frozen samples. After obtaining the sap, sample d. wt was determined by 24-h drying at 60 °C. Quantitative immunoassay analysis of free ABA was performed by ELISA, using a monoclonal antibody (DBPA1) raised against free (S)-ABA (Vernieri et al., 1989). Interference in ABA determination by non-specific substances present in the sap was negligible (Ofir and Kigel, 1998).

Statistical analyses

Experiments were arranged in a completely randomized design. Statistical analyses were carried out with JMPIN (version 4·04; SAS Institute Inc.). Means were compared using the LSMEANS test. ANOVA of data presented as proportions (% of sprouted bulbs) was carried out after arcsin transformations.

RESULTS

Imposition of dormancy

All the photo-induced plants (SD → LD), as well as the plants growing under non-inductive SD and subjected to water deficits (SD + WD) and ABA application (SD + ABA) became dormant (Table 1): bulbs developed at most (89–100 %) of the tiller bases, leaf production ceased and the mature leaves yellowed and dried up. In contrast, the control SD plants were non-dormant at the end of the experiment (age 90–110 d): most of their leaves (70–100 %) were green, with practically no dormant tillers (<2 %). Dormancy was imposed significantly earlier in the arid ecotype than in the mesic ecotype, by the SD → LD and SD + WD treatments (Table 1). Plants from the SD → LD treatment entered dormancy earlier than SD + ABA and SD + WD plants, in both ecotypes. Age of dormancy onset was similar in the mild and severe WD treatments.

Table 1.

Age at onset of dormancy, % bulbing and dormant tillers in arid and mesic ecotypes of Poa bulbosa as affected by mild or severe water deficit, and by ABA application in plants grown under 9 h SD (SD + WD and SD + ABA, respectively), compared with photo-inductive transfer from SD to 16 h LD (SD → LD)

Ecotype SD + WD
 SD + ABA SD → LD SD
Mild Severe
Age at onset of dormancy (days from sprouting)
 Arid 96·3 ± 0·5 98·2 ± 0·5 96·6 ± 1·0 77·7 ± 0·8 *
 Mesic 108·0 ± 0·4 109·0 ± 0·3 98·7 ± 0·9 89·9 ± 1·0 *
% bulbing tillers and % dormant tillers (in parentheses)
 Arid 89·3 ± 2·4 (88·3) 89·7 ± 1·7 (89·5) 100·0 (96·3) 94·6 ± 1·2 (91·9) 9·8 ± 5·7 (1·4)
 Mesic 100·0 ± 0·0 (99·3) 100·0 ± 0·0 (99·2) 100·0 (91·9) 100·0 (96·8) 0·0
ANOVA of age at onset of dormancy
SD + WD treatments
 SD + ABA and SD → LD
F P d.f.error F P d.f.error
Ecotype effect 660·0 0·0001 217 35·7 0·0001 136
Treatment effect
 In arid ecotype 7·0 0·009 102
 In mesic ecotype 3·3 0·070 115
 Both ecotypes 10·5 0·001 217 132·7 0·0001 136
 Ecotype × treatment 1·0 0·321 217 17·7 0·0001 136
Open in a new tab

Treatments started after 2 months of growth at SD. Growth temperatures were 22/16 °C (day/night). Water deficit (WD) was obtained by cycles of delayed irrigation, until the substrate was almost dry and re-watering to preset values that differed in the mild and severe WD treatments (cf. Fig. 1). In the SD + ABA treatment the plants were sprayed with increasing concentrations of ABA (0·1–0·3 mm). Control plants (continuous SD) did not become dormant until the end of the experiment. SD + WD plants grew in perlite in 3-L pots, while SD → LD and SD + ABA plants grew in vermiculite–tuff substrate in 1-L pots. Therefore the SD + WD treatments were subjected to a separate statistical analysis. Data are means and standard error; n = 10 plants for the SD controls; and 15–30 plants for other treatments.

*All the control SD plants were non-dormant at the termination of the experiment (age 90–110 d).

Bulb dormancy

Effects of the different dormancy-inductive treatments on bulb dormancy were examined by comparing the percentage of sprouted bulbs in a moist substrate, at 10 °C, with or without 2 months dry storage at 40 °C (i.e. pre- and post-storage % sprouting; Fig. 3). Dormancy of the unstored bulbs was deep: sprouting was very low and did not differ between treatments and ecotypes (P > 0·05) (Fig. 3 and Table 2). After 2 months of storage at 40 °C, sprouting percentages increased markedly and significantly in both ecotypes and in all treatments (Fig. 3 and Table 2). Ecotype differences in post-storage sprouting were not significant (Table 2). Highest sprouting percentages were attained in bulbs from the SD + WD plants, but differences between the two WD treatments were not significant. Sprouting of SD + ABA and SD → LD bulbs were significantly lower (P<0·006) compared with the WD bulbs. During storage, bulbs lost 40–50 % of their f. wt due to drying.

Fig. 3.

Fig. 3.

Open in a new tab

Effects of dormancy inducing treatments on the level of dormancy in bulbs of P. bulbosa. Treatments were: (1) exposure of plants growing under non-inductive SD (9 h) to water-deficit (WD) cycles: mild WD (SD + mWD) and severe WD (SD + sWD) treatments; (2) spraying plants in SD with 0·1, 0·2 and 0·3 mM ABA, 10, 20, and 30 d, respectively after start of treatment (SD + ABA); (3) transferring plants from SD to 16 h LD (SD → LD). All treatments started after 2 months of growth at SD. Plants grew continuously under 22/16 °C (day/night). Bulb dormancy, i.e. % sprouting of freshly harvested bulbs and of bulbs after 2 months dry storage at 40°C (pre- and post-storage, respectively), was determined 40 d after planting of the bulbs in a wet substrate at 10 °C. Data are means and s.e. of four replicates of 20 bulbs in the WD and in the SD + ABA treatments or five replicates of 10 bulbs in SD → LD.

Table 2.

ANOVA of the effects of ecotype (arid vs. mesic) and of dormancy-inducing treatments on the level of dormancy of bulbs produced in the different treatments

% sprouted bulbs
Pre-storage
 Post-storage
 Pre- vs. post-storage
F P F P F P
Ecotype effect 0·01 0·946 0·5 0·486
Treatment effect 2·0 0·142 8·0 0·0006
Storage effect 328·5 0·0001
Ecotype × treatment 1·5 0·226 0·8 0·528
Ecotype × storage 0·3 0·567
Treatment × storage 5·22 0·003
d.f.error 29 26 55
Open in a new tab

Dormancy was assessed by comparing the percentage sprouting of freshly harvested bulbs with that of bulbs dry-stored at 40 °C for 2 months (dormancy-releasing treatment). The bulbs were planted in a wet substrate at 10 °C. Inductive treatments were: transfer from SD to LD, application of ABA at SD and mild and severe water deficit at SD. The three treatments started after 2 months of growth at SD. Analysis of the data was carried out after arcsin transformation.

Endogenous ABA in the bulbs

Endogenous ABA levels were assayed only in bulbs of the mesic ecotype (Fig. 4). Two-way ANOVA showed that ABA levels were highest in the bulbs of the ABA- and WD-treated plants, followed by SD → LD bulbs, and lowest in the tiller bases of the non-dormant, still actively growing SD control plants (P = 0·0001, F = 12·9, d.f.=18). ABA levels in the SD control plants and SD → LD bulbs (approx. 60 and 240 ng g−1 d. wt, respectively) were comparable with the levels found previously in tiller bases and in bulbs from similar treatments (Ofir and Kigel, 1998, fig. 5C). Storage at 40 °C for 2 months had no significant effect on ABA levels (P = 0·23, F = 1·6, d.f.=18) and treatment × storage interactions were also not significant (P > 0·05). The increase of ABA observed in the SD + WD bulbs after storage was not significant (t = 1·85, d.f.=6, Pt=0·115).

Fig. 4.

Fig. 4.

Open in a new tab

Effects of dormancy-inducing treatments on endogenous levels of ABA in bulbs of P. bulbosa. Treatments were: (1) exposure of plants growing under non-inductive SD (9 h) plants to water deficit (SD + WD); (2) spraying plants in SD with ABA (SD + ABA); (3) transferring plants from SD to 16 h LD (SD → LD); (4) SD controls, plants remained non-dormant until the end of the experiment and bulbs were not formed. All treatments started after 2 months growth at SD. Plants grew continuously under 22/16 °C (day/night). ABA assays were carried out by ELISA in freshly harvested bulbs (pre-storage) and in bulbs after 2 months of dry storage at 40 °C, i.e. dormancy releasing treatment (post-storage). Data are means and s.e., n = 4 replicates of 100–300 mg d. wt each.

DISCUSSION

Results clearly show that water deficit can induce summer dormancy in P. bulbosa independently, in the absence of inductive long days and at moderate temperatures (Table 1 and Fig. 3). This is the first report of induction of summer dormancy imposed by drought stress resulting from water deficit in Mediterranean perennial grasses. The effects of drought are characterized by the same developmental traits typical of those observed in this species after controlled photo-induction of summer dormancy, or under natural growth conditions: production of basal bulbs, arrest of leaf and tiller production, and leaf senescence (Ofir and Kerem, 1982; Ofir and Dorenfeld, 1992; Ofir and Kigel, 1999). This is a true, endo-dormancy since (a) it was induced by pulses of water deficit applied alternately with partial irrigation, particularly in the mild WD treatment, and (b) the freshly produced bulbs remained dormant when transferred to wet conditions and low temperature (Fig. 3). However, the fact that prolonged water deficit failed to induce dormancy under winter conditions in a population of P. bulbosa from the Mediterranean region in France (Volaire et al., 2001), suggests that genotypic differences among populations adapted to different climatic conditions may affect their responsiveness to dormancy induction by water deficit. Recently, true summer dormancy (i.e. growth arrest under summer irrigation) has been shown in a Mediterranean climate for D. glomerata ‘Kasbah’ (Norton et al., 2006), an early-flowering cultivar bred from an arid Moroccan population (Oram, 1990). In this cultivar summer dormancy was induced during late spring, probably by longer days, increasing temperature and drought, and was maintained even under irrigation during the warm and dry summer. However, it is not clear whether increasing drought by itself could induce dormancy under non-inductive photo-thermal conditions in this case, as found for P. bulbosa. Moreover, it is important to note a basic difference between these two species: while dormant plants of P. bulbosa are highly tolerant to desiccation (approx. 10 % water content in the bulbs; Ofir, 1986), in D. glomerata the lethal water content in the surviving leaf and meristematic tissues at the base of the tillers is higher (25–30 %; Volaire et al., 2001) and does not differ among populations with different drought tolerance (Volaire et al., 1998a; Volaire, 2002).

It is hypothesized that the stress resulting from water deficit induces summer dormancy in P. bulbosa through an increase in endogenous ABA. This is supported by the fact that ABA application under non-inductive short days resulted in a similar dormancy syndrome as induction by long-day (Table 1 and Fig. 3), and the finding that ABA levels increased in the leaves and tiller bases under inductive long-day conditions (Fig. 4; and Ofir and Kigel, 1998). ABA has been shown to be involved in dormancy development in other geophytes (Suh and Kwack, 1992; Yamazaki et al., 1999; Ii et al., 2002).

Dormancy of the newly produced bulbs was deep in all the treatments and in both ecotypes, e.g. only 0–25 % of the bulbs sprouted when planted in a moist substrate at 10 °C (Fig. 3). Dry storage at high temperature, simulating summer conditions, causes dormancy relaxation (Ofir, 1986). In the harvested bulbs, 2 months of dry storage at 40 °C removed dormancy almost completely (85–100 % sprouting) in bulbs from the plants that were subjected to water-deficit treatments, but only partially in bulbs from the ABA-treated plants or from the SD → LD plants (60–75 % sprouted bulbs; Fig. 3). Dormancy relaxation during dry storage was not associated with a corresponding change in the level of endogenous ABA in the bulbs (Fig. 4). Moreover, ABA levels were low in the more dormant SD → LD bulbs, compared with the less dormant bulbs from the WD treatment (Figs 3 and 4). These results indicate that while ABA was closely involved in the imposition of dormancy (Ofir and Kigel, 1998), relaxation of dormancy was apparently not associated with changes in endogenous ABA. In seeds, in contrast, after-ripening, stratification and other dormancy-releasing mechanisms lowered ABA content of imbibed dormant seeds (Gubler et al., 2005). Onset of dormancy was earlier in the arid ecotype than in the mesic ecotype in the photo-induced plants (cf also Ofir and Kigel, 2003), as well as in the drought-stressed plants under non-inductive SD (Table 1). These results indicate that the ‘arid’ plants are more sensitive than the ‘mesic’ plants to the dormancy-inducing signal, whether it originates in photoperiodic change or in drought stress.

It is concluded that summer dormancy can be induced in natural populations of P. bulbosa by two alternative and probably additive mechanisms: (1) photoperiodic induction by long days, a process enhanced by high temperature; (2) drought stress resulting from water deficit. The critical day-length for dormancy in P. bulbosa is approx. 11–12 h (Ofir and Kigel, 1999), reached by mid-February–March. Thus, the inductive drought stress pathway may allow earlier dormancy induction during the winter under non-inductive day length, in case of unusual dry winters. In contrast, in normal years higher water availability will postpone dormancy to a later date, as found for the more mesic ecotypes along a gradient of increasing rainfall (Ofir and Kigel, 2003). Co-occurrence of these two pathways for dormancy induction can be of critical importance for the survival of P. bulbosa, since it inhabits open sites exposed to direct solar irradiation, with a shallow soil with low water-storage capacity. Moreover, this physiological plasticity may increase the range of micro-habitats colonized by similar clones, in this highly apomictic species (Heyn, 1962).

LITERATURE CITED

  1. Ben-Hod G, Kigel J, Steinitz B. Dormancy and flowering in Anemone coronaria L. as affected by photoperiod and temperature. Annals of Botany. 1988;61:623–633. [Google Scholar]
  2. Bitan A, Rubin S. Climatic atlas of Israel for physical and environmental planning and design. Israel Meteorological Service, Beit Dagan: Ramot Publishing Co., Tel Aviv University; 1991. [Google Scholar]
  3. DaCosta M, Wang Z, Huang B. Physiological adaptation of Kentucky bluegrass to localized soil drying. Crop Science. 2004;44:1307–1314. [Google Scholar]
  4. Davies WJ, Kudoyarova G, Hartung W. Long-distance ABA signaling and its relation to other signaling pathways in the detection of soil drying and the mediation of the plant's response to drought. Journal of Plant Growth Regulation. 2005;24:285–295. [Google Scholar]
  5. Farina E, Paterniani T, Suffia G. Experimental trials on the reaction of hybrid ranunculus to photoperiod. Annali dell' Istituto Sperimentale per la Floricoltura. 1985;16:93–104. [Google Scholar]
  6. Gubler F, Millar AA, Jacobsen JV. Dormancy release, ABA and pre-harvest sprouting. Current Opinion in Plant Biology. 2005;8:183–187. doi: 10.1016/j.pbi.2005.01.011. [DOI] [PubMed] [Google Scholar]
  7. Heyn CC. Studies of bulbous Poa in Palestine. 1. The agamic complex of Poa bulbosa. Bulletin of the Research Council of Israel. 1962;11D:117–126. [Google Scholar]
  8. Lang GA, Early JD, Martin GC, Darnell RL. Endo-, para-, and eco-dormancy: physiological terminology and classification for dormancy research. HortScience. 1987;22:371–377. [Google Scholar]
  9. Ii K, Okubo H, Matsumoto T. Control of bulb dormancy in hyacinth – a molecular biological approach. Acta Horticulturae. 2002;570:241–246. [Google Scholar]
  10. Laude HM. The nature of summer dormancy in perennial grasses. Botanical Gazette. 1953;114:284–292. [Google Scholar]
  11. Naor V, Kigel J. Temperature affects plant development, flowering and tuber dormancy in calla lily (Zantedeschia) Journal of Horticultural Science and Biotechnology. 2002;77:170–176. [Google Scholar]
  12. Norton MR, Lelievre F, Volaire F. Summer dormancy in Dactylis glomerata L.: the influence of season of sowing and a simulated mid-summer storm on two contrasting cultivars. Australian Journal of Agricultural Research. 2006;57:565–575. [Google Scholar]
  13. Ofir M. Seasonal changes in the response to temperature of summer-dormant Poa bulbosa L. bulbs. Annals of Botany. 1986;58:81–89. [Google Scholar]
  14. Ofir M, Dorenfeld Y. Induction of summer dormancy in Poa bulbosa L. under natural environment and subsequent controlled photo-thermal conditions. Israel Journal of Botany. 1992;41:265–277. [Google Scholar]
  15. Ofir M, Kerem D. The effects of temperature and photoperiod on the onset of summer–dormancy in Poa bulbosa L. Annals of Botany. 1982;50:259–264. [Google Scholar]
  16. Ofir M, Kigel J. Abscisic acid involvement in the induction of summer-dormancy in Poa bulbosa, a grass geophyte. Physiologia Plantarum. 1998;102:163–170. [Google Scholar]
  17. Ofir M, Kigel J. Photothermal control of the imposition of summer-dormancy in Poa bulbosa, a perennial grass geophyte. Physiologia Plantarum. 1999;105:633–640. [Google Scholar]
  18. Ofir M, Kigel J. Variation in onset of summer dormancy and flowering capacity along an aridity gradient in Poa bulbosa L., a geophytic perennial grass. Annals of Botany. 2003;91:391–400. doi: 10.1093/aob/mcg026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ofir M, Kigel J. Opposite effects of daylength and temperature on flowering and summer dormancy of Poa bulbosa. Annals of Botany. 2006;97:659–666. doi: 10.1093/aob/mcl021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ofir M, Koller D. A kinetic analysis of the relationships between flowering and the dormant state in Hordeum bulbosum L. – a perennial grass. Israel Journal of Botany. 1972;21:21–34. [Google Scholar]
  21. Ofir M, Koller D. Relationship between thermoinduction and photoinduction of flowering and dormancy in Hordeum bulbosum L., a perennial grass. Australian Journal of Plant Physiology. 1974;1:259–270. [Google Scholar]
  22. Ofir M, Koller D, Negbi M. Studies on the physiology of regeneration buds of Hordeum bulbosum. Botanical Gazette. 1967;128:25–34. [Google Scholar]
  23. Oram RN. Ecotypic differentiation for dormancy levels in over summering buds of Phalaris aquatica. Botanical Gazette. 1983;144:544–551. [Google Scholar]
  24. Oram RN. Register of Australian herbage plant cultivars. 3rd edn. East Melbourne: CSIRO; 1990. [Google Scholar]
  25. Piano E, Pecetti L, Annicchiarico P, Carroni AM, Fornasier F, Romani M. Combining drought tolerance and responsiveness to summer moisture availability in cocksfoot (Dactylis glomerata L.) germplasm grown in Mediterranean environments. Australian Journal of Agricultural Research. 2004;55:1197–1204. [Google Scholar]
  26. Rees AR. Ornamental bulbs, corms and tubers. Wallingford: CAB International; 1992. [Google Scholar]
  27. Suh JK, Kwack BH. Physiological changes in the course of bulbing and dormancy-breaking of gladiolus (Gladiolus gandavensis L.) Journal of the Korean Society of Horticultural Science. 1992;33:466–470. [Google Scholar]
  28. Vernieri P, Perata P, Armellini D, Bugnoli M, Presentini R, Lorenzi R, et al. Solid phase radioimmunoassay for the quantitation of abscisic acid in plant crude extracts using a new monoclonal antibody. Journal of Plant Physiology. 1989;134:441–446. [Google Scholar]
  29. Volaire F. Drought survival, summer dormancy and dehydrin accumulation in contrasting cultivars of Dactylis glomerata. Physiologia Plantarum. 2002;116:42–51. doi: 10.1034/j.1399-3054.2002.1160106.x. [DOI] [PubMed] [Google Scholar]
  30. Volaire F, Thomas H, Bertagne N, Bourgeois E, Gautier MF, Lelievre F. Survival and recovery of perennial forage grasses under prolonged Mediterranean drought: water status, solute accumulation, abscisic acid concentration and accumulation of dehydrin transcripts in bases of immature leaves. New Phytologist. 1998a;140:451–460. doi: 10.1111/j.1469-8137.1998.00287.x. [DOI] [PubMed] [Google Scholar]
  31. Volaire F, Thomas H, Lelievre F. Survival and recovery of perennial forage grasses under prolonged Mediterranean drought: growth, death, water relations and solute content in herbage and stubble. New Phytologist. 1998b;140:439–449. doi: 10.1111/j.1469-8137.1998.00288.x. [DOI] [PubMed] [Google Scholar]
  32. Volaire F, Conejero G, Lelievre F. Drought survival and dehydration tolerance in Dactylis glomerata and Poa bulbosa. Australian Journal of Plant Physiology. 2001;28:743–754. [Google Scholar]
  33. Wang Z, Huang B, Bonos SA, Meyer WA. Abscisic acid accumulation in relation to drought tolerance in Kentucky bluegrass. HortScience. 2004;39:1133–1137. [Google Scholar]
  34. Yamazaki H, Nishijima T, Yamato Y, Koshioka M, Miura H. Involvement of abscisic acid (ABA) in bulb dormancy of Allium wakegi Araki. I. Endogenous levels of ABA in relation to bulb dormancy and effects of exogenous ABA and fluridone. Plant Growth Regulation. 1999;29:189–194. [Google Scholar]
  35. Zhang JH, Jia WS, Yang JC, Ismail AM. Role of ABA in integrating plant responses to drought and salt stresses. Field Crops Research. 2006;97:111–119. [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES