View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

South African Journal of Botany 2003, 69(4): 455–461 Copyright © NISC Pty Ltd Printed in South Africa — All rights reserved SOUTH AFRICAN JOURNAL OF BOTANY EISSN 1727–9321

Minireview

Thermoinhibition of seed germination

PN Hills and J van Staden*

Research Centre for Plant Growth and Development, School of Botany and Zoology, University of Natal Pietermaritzburg, Private Bag X01, Scottsville 3209, South Africa * Corresponding author, e-mail: [email protected]

Received 21 August 2003, accepted in revised form 25 August 2003

Thermoinhibition describes the inability of seeds to ger- perature induced changes to the structures enclosing minate at high temperatures, although germination pro- the embryo which prevent radicle emergence, to the ceeds immediately when the temperature is reduced interaction of a number of different factors, and probable below a certain threshold level. This phenomenon is expression of certain genes inhibitory to germination distinct from thermodormancy, where some form of dor- which may be temperature regulated. Thermoinhibition mancy-breaking treatment is required before germina- occurs in a large number of important crop species, so tion can proceed at the favourable temperature. Like that an understanding of this phenomenon is both of seed dormancy, thermoinhibition is manifested in a scientific interest and practical importance. number of different ways, ranging from simple high-tem-

Introduction

Temperature strongly influences both physiological and bio- Mayber 1975). These temperatures may be considerably chemical processes, including those occurring in germinat- modified by several other factors, such as exposure to light ing seeds. In the field, temperature acts to regulate germi- of different wavelengths or the application of different com- nation in three main ways (Bewley and Black 1986). Firstly, pounds, including phytohormones such as ethylene, it may be involved in the removal of either primary or sec- abscisic acid (ABA) and gibberellins. ondary dormancy. Secondly, temperatures outside of the Within the temperature range of a species, the rate of ger- normal limits for germination may cause the establishment mination usually increases as the temperature rises, of secondary dormancy in a seed when conditions are not although the final percentage germination may decline favourable for seedling establishment. Finally, the tempera- (Heydecker 1977). However, as the temperature approach- ture at which seeds are incubated determines their capacity es the maximum for germination, the germination rate for germination and the rate at which this occurs. begins to slow (Heydecker 1977). Variations in the rates of Each species has a range of temperatures over which ger- germination of different seeds ensure that the seeds in a mination may proceed and within which seedling establish- population germinate at different times, thereby leading to ment is possible (Bradbeer 1988). There are three cardinal the temporal dispersal of germination (Bewley and Black temperatures for germination, namely the maximum, mini- 1986). In wild populations this is a distinct advantage. mum and optimum germination temperatures (Bewley and However, it is disadvantageous in a field crop where unifor- Black 1982). The maximum and minimum temperatures rep- mity is an important consideration. resent the extremes of the range of temperatures over which In the environment, temperatures greater than the maxi- the seed is able to germinate. The optimum germination mum favourable temperature for germination will result in temperature may be defined as that temperature at which the suspension of germination. At these supra-optimal tem- the highest percentage germination may be obtained in the peratures, seeds may enter into a state of either thermodor- shortest possible period of time (Mayer and Poljakoff- mancy or thermoinhibition. It is important that a distinction Mayber 1975). These temperatures vary considerably be made between these two phenomena. In thermodormant between species, but may also vary between cultivars of the seeds, a state of secondary dormancy is induced by expo- same species (Bewley and Black 1982) and are thus deter- sure to the elevated temperature, which must then be mined by the source of the seeds, genetic differences with- released by some form of dormancy-breaking treatment in a species and the age of the seed (Mayer and Poljakoff- before the seeds are able to germinate again at their optimal 456 Hills and Van Staden temperature (Vidaver and Hsiao 1975). Seeds are said to be does exist between these different factors in the various thermoinhibited where they fail to germinate at a high tem- species or genotypes that eventually results in the seed perature, but where germination proceeds immediately upon becoming thermoinhibited, for the purposes of this review, transfer to a temperature suitable for germination of seeds of each potential factor which has been proposed as playing a that species (Horowitz and Taylorson 1983). In both ther- role in thermoinhibition will, as far as possible, be dealt with modormant and thermoinhibited seeds, if the temperature is separately. too high or is maintained for an extended period, thermal death will result (Horowitz and Taylorson 1983). In certain Involvement of the embryo coverings circumstances, however, this distinction is hard to make and terminology has been a problem in the study of thermoinhi- In many species, seed dormancy is imposed by the struc- bition. Many authors, particularly in earlier papers, refer to tures surrounding the embryo, including the endosperm, thermodormant seeds, when the seeds in fact appear to be perisperm, pericarp and testa (Bewley and Black 1982). thermoinhibited. Vidaver and Hsiao (1975) stated: These structures may impose dormancy through a number ‘Secondary dormancy is a distinctly different condition from of mechanisms, including mechanical restraint on embryo so-called thermodormancy in these seeds. The term ther- extension, the establishment of permeability barriers inter- moinhibition should probably be substituted for thermodor- fering with water uptake, gaseous exchange or the outward mancy since seeds which have not yet become dormant will diffusion of endogenous germination inhibitors. Similar germinate merely by lowering the temperature.’ This is a mechanisms may also be associated with thermoinhibition in particular problem in lettuce seed studies. The threshold some species. temperature at which lettuce seeds fail to germinate varies In the legume Lathyrus sativus, supra-optimal tempera- between different cultivars, with some genotypes becoming tures increase hardseededness, preventing seed hydration inhibited at temperatures as low as 25°C (Nascimento et al. and germination (Agrawal et al. 1980). In spinach (Spinacia 2001). When this threshold temperature is exceeded during oleracea) seeds, germination is inhibited at temperatures in imbibition, seeds become thermoinhibited. However, if this excess of 30°C (Leskovar et al. 1999). At this temperature, supra-optimal temperature is maintained for more than the pericarp appears to act as a physical barrier to germina- approximately 72h, the seeds enter into a state of ther- tion as well as a source of inhibitory compounds (Leskovar modormancy, such that these seeds will then fail to germi- et al. 1999). Complete removal of the pericarp was able to nate even when the temperature is reduced below the overcome thermoinhibition at 30°C, however, placing the inhibitory threshold. A similar situation appears to occur in pericarp near the embryos had a slight inhibitory effect, celery seeds, in which seeds that do not germinate at high which was more pronounced if the embryos were re-insert- temperatures will germinate when transferred to a lower ed into the excised pericarps (Leskovar et al. 1999). temperature. The longer the seeds are incubated at the The integuments of apple seeds are rich in phenolic com- supra-optimal temperature, the longer they appear to take to pounds that fix oxygen through oxidation, thereby lowering germinate when the temperature is reduced (Biddington and the levels of oxygen available to the embryo (Côme and Thomas 1978). It is therefore important to examine the actu- Tissaoui 1973). When the seeds are imbibed at high tem- al germination curves of the various seed lots used in the lit- peratures the embryo’s requirement for oxygen is increased erature to determine whether the seeds used were in fact as a result of an increase in the metabolic rate. The amount thermodormant or thermoinhibited. of oxygen available to the embryo is, however, decreased as Observations of temperature effects on the germination of oxidation of phenolics rises and oxygen becomes less solu- seeds of a number of different species have led to the for- ble in the water in the seed coat and germination is conse- mulation of a number of hypotheses concerning the basis of quently inhibited (Côme and Tissaoui 1973). Hypoxic condi- thermoinhibition in those species. An analysis of the avail- tions imposed by the seed coat at supra-optimal tempera- able literature reveals that there appears to be a variety of tures may have a different effect in seeds that require ethyl- possible mechanisms resulting in the arrest of germination. ene for germination. Since the ethylene-forming enzyme The situation is made more confusing by the fact that there (ACC oxidase) is oxygen-dependant, the conversion of ACC appears to be a complex interplay between several different (1-aminocyclopropane-1-carboxylic acid) to ethylene may factors in a number of species. This is particularly evident in be inhibited, preventing germination (Yang 1985). lettuce seeds. In this species, not only do different factors Embryo coverings also play a role in the thermoinhibition appear to be involved in thermoinhibition in different culti- of lettuce seeds at supra-optimal temperatures (Drew and vars, but in certain cultivars, a number of factors seem to act Brocklehurst 1984). The involvement of different tissues sur- synergistically to prevent germination at supra-optimal tem- rounding the embryo in thermoinhibition appears to vary peratures. In the cultivar Grand Rapids for instance, ther- between seeds of different cultivars. In cv. Cobham Green moinhibition has been proposed to be due to restrictions on lettuce seeds, control over germination appears to be exert- radicle expansion by the endosperm or pericarp (Drew and ed by the pericarp, since relief of thermoinhibition at 35°C Brocklehurst 1984), by changes in endogenous cytokinin resulted when the pericarp was weakened by hypochlorite and gibberellin levels (Saini et al. 1986), by changes in phy- treatment (Drew and Brocklehurst 1984). Hypochlorite treat- tochrome in the seeds (Kristie and Fielding 1994), by ment of the pericarp also appeared to overcome thermoinhi- abscisic acid (ABA) (Yoshioka et al. 1998) and by anaerobic bition in cultivars Empire and Benita, whilst treatment of respiration and reduced ATP production (Small et al. 1993). seeds of cultivars Sabine and Bellona resulted in faster Whilst it is important to consider the complex interplay that recovery from thermoinhibition than non-treated seeds South African Journal of Botany 2003, 69: 455–461 457

(Drew and Brocklehurst 1987). In seeds of cv. Grand thermosensitive cultivar Dark Green Boston at 35°C, whilst Rapids, however, thermoinhibition was relieved by weaken- only 4% of non-primed seeds germinated at this tempera- ing of the endosperm (Drew and Brocklehurst 1984). Dutta ture (Nascimento et al. 2001). These results suggest that et al. (1997) also observed that even minor punctures to the enzymatic weakening of the endosperm may play an impor- endosperm resulted in complete alleviation of thermoinhibi- tant role in the germination of certain lettuce seed cultivars. tion in cv. Pacific lettuce seeds. Bradford and Somasco (1994) explained the role of Gibberellins and cytokinins embryo coverings in thermoinhibition in lettuce seeds by analysing water relations in lettuce cv. Empire seeds. This From their work on celery (Apium graveolens) cv. Lathom work did not attempt to distinguish between the effect of the Blanching, Biddington and Thomas (1978) reported that endosperm and the pericarp, but rather refers to the effects temperature may control germination in these seeds by of the embryo coverings as a whole. The ability of the affecting endogenous levels of cytokinins and gibberellins. embryo to absorb water from its surroundings and initiate Exogenous application of these hormones increased the growth is determined by the water potential (Ψ) of its cells. germination of celery seeds subjected to supra-optimal tem- Thermoinhibition is associated with an increase in the yield peratures in both the light and the dark. These two hor- threshold that must be exceeded in order for radicle emer- mones also appeared to act in a synergistic manner in the gence to take place (Karssen et al. 1989). This total yield dark (Biddington and Thomas 1978). A similar result has threshold is made up of a component from the radicle cell also been shown for the lettuce cv. Grand Rapids (Saini et walls (Yr) and a component from the tissues surrounding the al. 1986). Biddington and Thomas (1978) proposed that tem- embryo (Ye). As the incubation temperature increases, the perature may control germination in celery seeds by affect- osmotic potential of the embryo (Ψπ), the embryo turgor ing the levels of endogenous gibberellins or a gibberellic threshold for growth (Yr) and endosperm resistance (Ye) all acid/protein complex. They suggested that germination is increase relative to seeds incubated at optimal tempera- primarily induced by gibberellins, endogenous levels of tures, causing thermoinhibition (Bradford and Somasco which decline as temperatures increase and rise again when 1994). The removal of the structures enclosing the embryo the temperature is reduced. thus results in the alleviation of thermoinhibition by eliminat- Relief from thermoinhibition was enhanced by white or red ing the contribution of Ye to the total yield threshold light (Biddington and Thomas 1978). Quantitative and quali- (Bradford and Somasco 1994). This was also shown in the tative changes in the cytokinin contents of seeds have been lettuce cv. Pacific, where slitting the seeds to disrupt the reported in response to dormancy-breaking red light treat- integrity of the endosperm extended the temperature range ment of celery, lettuce (Van Staden 1973) and Rumex for germination and maintained lower base water potential in obtusifolius (Van Staden and Wareing 1972). These results the slit seeds as compared with intact seeds (Dutta and suggest that phytochrome-mediated control over the germi- Bradford 1994). The fact that removal or scarification of the nation process is expedited via endogenous cytokinins. endosperm does not affect differences in temperature sen- Biddington and Thomas (1978) therefore suggested that sitivity between seeds of different lettuce seed cultivars red-light may enhance the effect of gibberellins on the relief (Dunlap et al. 1990, Prusinski and Khan 1990), however, of thermoinhibition through increased cytokinin levels. indicates that the embryo itself also plays a key role in ther- Cytokinins may play a subsidiary role in alleviating ther- moinhibition (Bradford and Somasco 1994). moinhibition in lettuce seeds by increasing ethylene biosyn- In lettuce seeds, the protrusion of the radicle through the thesis or the sensitivity of the seeds to ethylene. micropylar endosperm signals the end of germination. Exogenously applied kinetin and ACC act synergistically to Weakening of the endosperm through cell wall autohydroly- increase pre-germination levels of ethylene and enhance sis thus may be important for germination in lettuce seeds germination at temperatures of 32°C and 35°C (Khan and (Dutta et al. 1994). Autolytic activity appeared to be an enzy- Prusinski 1989). Removing the restriction imposed by the matic process as it was dependant on temperature and pH tissues enclosing the embryo by cutting the testa allowed (Dutta et al. 1994). When isolated cell walls from thermo- ACC to be readily taken up and converted to ethylene, inhibited or ABA-treated seeds were examined, autohydrol- removing the synergistic effect of the added kinetin. This ysis was reduced by up to 25% compared with control result suggests that cytokinins have an important regulatory endosperms isolated from germinating seeds. Treating role for ethylene biosynthesis and hence for germination in seeds with kinetin and GA increased the rate of hydrolysis intact lettuce seeds at high temperatures, and that the seed by 20% to 30% compared with thermoinhibited controls coat may be an important factor in this regulation (Khan and (Dutta et al. 1994). In a later study (Dutta et al. 1997), it was Prusinski 1989). This may involve enhanced utilisation of reported that this autohydrolysis was related to a cell-wall ACC as a result of activation of the ethylene-forming bound endo-β-mannanase which was regulated by the same enzyme (EFE) or ACC oxidase (Huang and Khan 1992) and conditions as was germination, such that enzymic activity an interaction of the ACC-derived ethylene with the cytokinin was almost completely suppressed in thermoinhibited (Khan and Prusinski 1989). In lettuce cv. Grand Rapids, seeds. Endo-β-mannanase activity has since been shown to treatment with kinetin and oxygen also stimulated germina- be higher in thermotolerant than in thermosensitive geno- tion at thermoinhibitory temperatures, possibly either by types, especially at high temperatures (Nascimento et al. causing seeds to bypass their ethylene requirement or 2000, 2001). Priming of lettuce seeds increased endo-β- increasing their sensitivity to ethylene (Small et al. 1993). mannanase activity and allowed for 100% germination of the 458 Hills and Van Staden

Ethylene increase in the threshold concentrations of this hormone required to elicit germination. Prusinski and Khan (1990) In a number of species, seed germination is strongly related have also reported that seeds and seedlings have a greater to the biosynthesis and action of ethylene (Gallardo et al. sensitivity to ethylene when stressed than under non-stress- 1996). Huang and Khan (1992) found that pre-conditioning ful conditions, which may suggest a greater need for this seeds of the lettuce cv. Mesa 659 with the moist solid carri- phytohormone for growth processes under stress condi- er Micro-Cel E conferred greater thermotolerance on these tions. These authors found a positive correlation between seeds, alleviating thermoinhibition at 35°C. The ability of the the ethylene-producing capacity of seeds of a number of let- pre-conditioned seeds to rapidly synthesise ACC and to tuce cultivars and the ability of the seeds to germinate under respond to ACC and ethylene on exposure to temperatures conditions of salinity, osmotic and temperature stress. which were normally thermoinhibitory appeared to be the Nascimento et al. (2000) also observed a correlation critical factor for the alleviation of thermoinhibition. Exposing between thermosensitivity and ethylene production in lettuce seeds to aminoethoxyvinylglycine (AVG), an inhibitor of ACC at high temperatures, with thermotolerant cultivars such as synthesis, prevented pre-conditioned seeds from germinat- Everglades and PI251245 producing more ethylene than ing at 35°C. This inhibitory effect of AVG was reversed by thermosensitive cultivars such as Dark Green Boston and supplying the seeds with exogenous ACC, ethephon or eth- Valmaine at 35°C. ylene (Huang and Khan 1992). The use of ACC is enhanced The effect of ethylene in overcoming thermoinhibition by cytokinins under conditions of temperature stress (Khan appears to be linked fairly closely with other metabolic and Huang 1988), and the production of cytokinin or changes within lettuce seeds at supra-optimal temperatures, cytokinin-like factors may also play a regulatory role in eth- including changes in phytochrome (Saini et al. 1989). ylene biosynthesis during thermoinhibition (see above) Ethylene, GA3, kinetin, carbon dioxide (CO2) and light were (Huang and Khan 1992). all shown to have no effect on overcoming thermoinhibition Alleviation of thermoinhibition in chickpea (Cicer ariet- at 32°C, but that the action of any one of the hormones inum) may also be induced by the addition of exogenous required the presence of at least one other growth regulator, ethylene or by the stimulation of ethylene production within light or CO2 or a combination of these. To achieve 100% ger- the seeds (Gallardo et al. 1996). Thus the addition of ethrel mination, at least three stimuli were required (Saini et al. (chloroethylphosphonic acid), ACC or spermine to the ger- 1986). Nascimento et al. (2000) also observed an interaction mination medium or the inhibition of polyamine biosynthesis, between ethylene and light. This suggests that thermoinhibi- which competes with the ethylene biosynthetic pathway for tion in relation to ethylene may be regulated at a number of the same precursors, by CHA (cyclohexamine) or MGBG points. (methyl-glyoxal-bis-guanylhydrazone), could induce germi- The mechanism whereby ethylene is involved in the regu- nation of chickpea seeds at thermoinhibitory temperatures lation of germination appears therefore to differ in different (Muñoz De Rueda et al. 1994). species of plants. It should, however, be noted that the alle- Unlike lettuce seeds, where high temperatures inhibit ACC viation of thermoinhibition by ethylene is limited to species in production, supra-optimal temperatures appear to stimulate which seed germination is dependent on that phytohor- ACC synthase activity in chickpea seeds. Despite this stim- mone. ulation, ACC levels decrease in thermoinhibited seeds pos- sibly due to an increase in ACC conjugation (Gallardo et al. Abscisic acid 1991). With an increase in temperature, a decrease in the levels of ACC in the embryonic axis of chickpea seeds and Fluoridone (1-methyl-3-phenyl-5-[3-trifluorylmethyl-(phenyl)]- a concomitant increase in the levels of conjugated ACC in 4-(1H)-pyridinone) is an inhibitor of the carotenoid biosyn- the embryonic tissues was observed. Furthermore, malonyl thetic pathway enzyme phytoene desaturase. Since ACC-transferase, the enzyme responsible for the conjuga- carotenoids are the main precursors of abscisic acid (ABA) tion of ACC, has a very low Km value. It therefore appears in plants, inhibition of carotenogenesis also inhibits ABA that, in chickpea seeds, ACC malonisation and the conse- biosynthesis (Yoshioka et al. 1998). Application of this com- quent inhibition of ethylene synthesis is one of the principle pound to lettuce cv. Grand Rapids was able to overcome causes of thermoinhibition (Gallardo et al. 1996). The con- thermoinhibition up to 33°C, although at this temperature version of ACC to ethylene may also be affected at high tem- germination was markedly delayed as compared to 23°C peratures, possibly through some temperature effect on (Yoshioka et al. 1998). Fluoridone application also overcame ACC oxidase (Gallardo et al. 1996). In sunflower (Helianthus inhibition of germination at supra-optimal temperatures in annuus) too, thermoinhibition is associated with the loss of seeds of Chrysanthemum parthenim, Freesia hybrida, the seed’s ability to synthesise ethylene (Corbineau et al. Cerastium glomeratum, Stellaria neglecta, Agrostis alba, 1989). Conyza canadensis, Cryptotaenia japonica, Dactylis glomer- Besides changes in ethylene biosynthesis, ethylene and ata, Festuca rubra, Medicago sativa, Plantago lanceolata, ethylene action have also been implicated in the alleviation Trifolium repens and Vicia angustifolia, although the range of of stress on lettuce seed germination (Prusinski and Khan temperatures at which fluoridone promoted germination dif- 1990). Saini et al. (1989) have reported that in lettuce seeds, fered between species (Yoshioka et al. 1998). Fluoridone thermoinhibition does not appear to be related only to a had no effect on two gramineous weeds, Bromus catharticus reduction in the ability of the seeds to synthesise ethylene, and Lolium perenne, which were primarily dormant as but may rather be due to high temperatures causing an opposed to thermoinhibited. These authors therefore sug- South African Journal of Botany 2003, 69: 455–461 459 gested that ABA plays a decisive role in the regulation of may be a factor in thermoinhibition, Takeba and Matsubara seed germination at supra-optimal temperatures. (1976) reported that a thermo-labile process is involved in the germination of lettuce cv. New York seeds. Germination Repression of metabolic reactivation was completely inhibited at a temperature of 30°C in the dark, but was re-activated after incubation at 20°C. Although In Phacelia tanacetifolia seeds, germination is inhibited at Takeba and Matsubara did not establish the nature of the temperatures in excess of 26°C (Pirovano et al. 1997). This thermo-labile factor, they proposed that it was probably pro- inhibition is correlated with the lack of activation of metabol- tein-based. The longer the period that the seeds were pre- ic processes within the seed. In the first 24h of imbibition, an incubated at 30°C, the longer they required at the lower tem- increase in the levels of glucose-6-phosphate (Glc-6-P) was perature before germination occurred. A similar trend has inhibited first, followed by inhibition of increases in enzymic been observed in Tagetes minuta, where seeds incubated activity and then by inhibition of increases in levels of ATP, for extended periods of more than 20 days at 35°C begin to reducing sugars, RNA and DNA. The inhibition of radicle take longer to achieve germination (Forsyth and Van Staden emergence correlated most strongly with the levels of Glc-6- 1983). Takeba and Matsubara suggested that this thermo- P. After 24h, when germination was inhibited by 60%, only labile factor was thermoreversible and that the degree of its the level of Glc-6-P was still inhibited to the same extent re-activation upon return to favourable temperatures is (Pirovano et al. 1997). Appreciable qualitative differences in dependant on the extent of its inactivation. Three models transcriptional and translational activities were detected at were proposed to attempt to interpret these results in terms 16°C and 30°C. Both activities were higher in seeds incu- of the responses of the lettuce seeds to light and tempera- bated at 30°C than in seeds incubated at the lower temper- ture (Takeba and Matsubara 1976): ature after 9h, but with time the translational activities 1 Active phytochrome may promote germination by increas- became much lower at 30°C than at 16°C (Pirovano et al. ing the amount of the precursor of the reaction mediated 1997). Examination of protein patterns also revealed that a by the thermo-labile factor, number of polypeptides which disappeared during imbibition 2 The product(s) of the phytochrome system is(are) at 16°C did not disappear in temperature-inhibited seeds changed to the essential metabolite for germination by (Pirovano et al. 1997). combining with the product(s) of the reaction mediated by High temperatures may inhibit germination at two levels the thermo-labile factor, and (Pirovano et al. 1997). Firstly, the high temperature may 3 The phytochrome system increases the stability of the have a general effect on the rate of the reactivation process- thermo-labile factor by unknown mechanisms. es occurring within the imbibing seeds. This may be linked to the fact that cells of seeds incubated at 30°C also had Phytochrome lowered levels of mobile potassium (K+) ions and the plasma membranes showed greater permeability to K+, compared In lettuce, the promotion of seed germination through the with cells of seeds incubated at 16°C. The control of K+ con- red/far-red reversible or Low Fluence Response (LFR) is centrations in cells is crucial for the regulation of metabolic closely linked to temperature (Kristie and Fielding 1994). For activity and these authors suggested that reduced cellular the majority of lettuce cultivars, a pulse of red light is able to concentrations of K+ may inhibit the activation of K+-depen- promote germination within a certain range of temperatures. dant mechanisms and hence germination in seeds at 30°C. When this range is exceeded, the response to the red light Secondly, elevated temperatures may also work directly on pulse is diminished and seeds become thermoinhibited single reactions involved in the inhibitory mechanism of ger- (Vidaver and Hsiao 1975). In lettuce cv. Grand Rapids, dark mination in P. tanacetifolia seeds, either inducing a lack of germination at 20°C results in almost 100% germination, but function through denaturation of key components or by thermoinhibition prevents germination at 27°C and above increasing processes related to the synthesis of inhibitory (Saini et al. 1989). This has been attributed to the reversion substances. of Pfr to Pr before it has time to produce an effect, a hypoth- In lettuce cv. Grand Rapids, thermoinhibition at 38°C esis which is supported by the finding that repeated red light markedly reduced ATP and total adenylate content of the pulses raise the upper threshold temperature for germina- seeds, as well as adenylate energy charge (Small et al. tion by several degrees as compared to a single pulse of red 1993). Seeds had a high ethanol content at this temperature light (Saini et al. 1989). Kristie and Fielding (1994) suggest- and appeared to undergo ethanolic fermentation, probably ed that dark germination was governed by levels of Pfr and as a result of a reduction in oxygen solubility at this temper- that the mean level of Pfr required to induce dark germina- ature. Thermoinhibition was alleviated and seeds contained tion was increased as temperatures rose. It was proposed normal levels of ATP when treated with 100% oxygen and that this may be attributable to an increase in dark-reversion kinetin at 38°C (Small et al. 1993). A reduction in aerobic of Pfr to Pr at high temperatures (Kristie and Fielding 1994). respiration and consequently in ATP production may there- Preliminary Pfr requirement curves generated from a num- fore be a contributing factor in thermoinhibition (Small et al. ber of different lettuce cultivars suggest that the shape of the 1993). Pfr requirement curve may partially account for differences in the temperature response curves for dark germination The involvement of a thermo-labile factor that occur in the different cultivars (Kristie and Fielding 1994). Before Pirovano et al. (1997) proposed that denaturation The alleviatory effect of the phytochrome system at high 460 Hills and Van Staden temperatures may, as previously mentioned, be mediated expressed at any stage during germination of the achenes at through an effect on the levels of various phytohormones 25°C (Hills et al. 2001). These polypeptides all disappeared required for germination, including ethylene (Saini et al. rapidly when the temperature was reduced and the achenes 1986, 1989, Nascimento et al. 2000) and cytokinins allowed to germinate, in a manner which correlated tightly (Biddington and Thomas 1978). Nascimento et al. (2000) with an increase in the germinability of the achenes. also observed that endo-β-mannanase activity was eliminat- Concomitant with the decline of these thermoinhibition- ed when lettuce seeds were imbibed in the dark at 35°C, associated polypeptides, a number of other proteins which thus the activity of this enzyme may also in some manner be were not synthesised during thermoinhibition and which are mediated through the phytochrome system. presumably involved in germination and in post-germinative events were produced (Hills et al. 2001). This suggests that Gene expression thermoinhibition in this species may be imposed in a manner not previously described, possibly by an actively-imposed From the above, it would seem that in all cases described block on germination through gene expression. Thus, these thus far in the literature, thermoinhibition is a passive state, ten thermoinhibition-associated polypeptides may be imposed on seeds held under conditions of temperature involved in actively repressing germination at temperatures stress through heat-induced changes in their normal physio- which are not conducive to seedling establishment, in a logical functioning. Whilst most studies have focussed on manner similar to the inhibition of precocious germination or specific aspects of seed physiology, such as plant growth to genetically-imposed seed dormancy. regulators, enzymic reactions or phytochrome, it would appear that control of germination is levied at multiple sites Conclusions and that thermoinhibition is actually the result of a complex interaction of a number of factors in these species. Whilst this paper has dealt with each of the many factors In Tagetes minuta achenes, which have an optimum ger- which appear to cause thermoinhibition in different species mination temperature of 25°C, temperatures in excess of separately, it is important to remember that in many species 35°C result in the achenes becoming thermoinhibited. If the several different factors may be involved in causing the sus- temperature is reduced even slightly below 35°C, germina- pension of germination and that these factors may interact in tion proceeds rapidly (Drewes 1989). None of the factors exceedingly complex ways. To attempt to explore all these described in the literature appear to be involved in ther- interactions is beyond the scope of this review. moinhibition in this species. The achenes easily take up Thermoinhibition is a particularly interesting and important water when imbibed at 35°C. Whilst the seeds have a light part of seed biology. It differs from thermodormancy in that requirement for germination, seeds incubated at 35°C in no dormancy breaking treatment is required before germi- white light become thermoinhibited. Ethylene does not nation is able to continue. This occurs as soon as the tem- appear to be involved in thermoinhibition in T. minuta ach- perature at which the seeds are incubated has fallen to lev- enes. The application of ethrel to achenes imbibed at 35°C, els which favour germination for that particular species. either as short pulses or through continuous exposure, had Since many of the factors which have been proposed to no statistically significant alleviatory effect on thermoinhibi- cause thermoinhibition are the same or are similar to those tion of achenes of this species (Drewes 1989). When ethrel which are implicated in seed dormancy, thermoinhibition, was applied at 25°C, the germination rate was not affected, like dormancy provides an opportunity to examine process- indicating that T. minuta achenes are not dependant on eth- es in germinating seeds which are able to interrupt the ger- ylene for their germination (Drewes 1989). Indeed, ethrel mination process and thus to gain greater insight into germi- treatment of the achenes resulted in abnormal post-germi- nation itself. Unlike dormancy, however, thermoinhibition native development and both hypocotyl and radicle exten- allows the unique possibility of then allowing the seeds to sion were severely retarded. The cytokinins kinetin and ben- continue germinating without any need for treatments which zyladenine (BA) also had no effect on thermoinhibited ach- may in some way interfere with the experimental situation. enes (Drewes 1989). No alleviatory effect was observed when achenes were treated with a commercially available Acknowledgements — The authors wish to acknowledge the finan- cial support of the National Research Foundation of South Africa mixture of gibberellin GA4+7 and BA (Promalin), indicating that even a combination of light and multiple growth regula- and the University of Natal Pietermaritzburg. tors was unable to prevent thermoinhibition in this species. The application of gibberellins alone, however, did partially References alleviate thermoinhibition in this species, with GA4+7 being more effective than GA (Drewes 1989). Yet even GA appli- Agrawal PK, Sethi KL, Mehra RB (1980) Effect of temperature on 3 germination of Lathyrus sativus seeds: An analysis. Seed cation was unable to induce more than approximately 75% Research 8: 25–32 germination at 35°C and no major changes in endogenous Bewley JD, Black M (1982) Physiology and Biochemistry of Seeds gibberellin-like substances were detected in thermoinhibited in Relation to Germination, Vol 2. Springer-Verlag, Berlin, pp 297, achenes (Drewes 1989). 304 When protein expression in germinating and thermoinhib- Bewley JD, Black M (1986) Seeds: Physiology of Development and ited achenes was compared using two-dimensional poly- Germination, Vol 2. Springer-Verlag, Berlin, pp 297, 304 acrylamide gel electrophoresis, ten polypeptides were iden- Biddington NL, Thomas TH (1978) Thermodormancy in celery tified from the thermoinhibited achenes which were not seeds and its removal by cytokinins and gibberellins. Physiologia Plantarum 42: 401–405 South African Journal of Botany 2003, 69: 455–461 461

Bradbeer JW (1988) Seed Dormancy and Germination. Blackie, Khan AA, Prusinski J (1989) Kinetin enhanced 1-aminocyclo- London, pp 1–3, 6 propane-1 carboxylic acid utilization during alleviation of high Bradford KJ, Somasco OA (1994) Water relations of lettuce seed temperature stress in lettuce seeds. Plant Physiology 91: thermoinhibition. I. Priming and endosperm effects on base water 733–737 potential. Seed Science Research 4: 1–10 Kristie DN, Fielding A (1994) Influence of temperature on the Pfr Côme D, Tissaoui T (1973) Interrelated effects of imbibition, tem- level required for germination in lettuce cv. Grand Rapids. Seed perature and oxygen on seed germination. In: Heydecker W (ed) Science Research 4: 19–25 Seed Ecology. Butterworths, London, pp 157–168 Leskovar DI, Esensee V, Belenfant-Miller H (1999) Pericarp, Corbineau F, Rudnicki RM, Côme D (1989) ACC conversion to eth- leachate, and carbohydrate involvement in thermoinhibition of ylene by sunflower seeds in relation to maturation, germination germinating spinach seeds. Journal of the American Society for and thermodormancy. Plant Growth Regulation 8: 105–115 Horticultural Science 124: 301–306 Drew RLK, Brocklehurst PA (1984) Investigations on the control of Mayer AM, Poljakoff-Mayber A (1975) The Germination of Seeds, lettuce seed germination at high temperatures. Journal of Vol 2. Pergammon, Oxford, pp 32–35 Experimental Botany 35: 986–993 Muñoz De Rueda PM, Gallardo I, Sanchez-Calle P, Matilla AJ Drew RLK, Brocklehurst PA (1987) The effect of sodium hypochlo- (1994) Germination of chickpea seeds in relation to manipulation rite on thermo-inhibition of lettuce seed. Acta Horticulturae 215: of the ethylene pathway and polyamine biosynthesis by inhibitors. 179–184 Plant Science 97: 31–37 Drewes FE (1989) An Examination of the Thermoinhibition of Nascimento WM, Cantliffe DJ, Huber DJ (2000) Thermotolerance in Tagetes minuta L. Achenes. MSc Thesis, University of Natal lettuce seeds: Association with ethylene and endo-β-mannanase. Pietermaritzburg, Pietermaritzburg, South Africa Journal of the American Society for Horticultural Science 125: Dunlap JR, Scully BT, Reyes DM (1990) Seed coat-mediation of let- 518–524 tuce germination responses to heat and sodium chloride. Journal Nascimento WM, Cantliffe DJ, Huber DJ (2001) Endo-β-mannanase of the Rio Grande Valley Horticultural Society 43: 55–61 activity and seed germination of thermosensitive and thermotol- Dutta S, Bradford KJ (1994) Water relations of lettuce seed ther- erant lettuce genotypes in response to seed priming. Seed moinhibition. II. Ethylene and endosperm effects on base water Science Research 11: 255–264 potential. Seed Science Research 4: 11–18 Pirovano L, Morgutti S, Espen L, Cocucci SM (1997) Differences in Dutta S, Bradford KJ, Nevins DJ (1994) Cell-wall autohydrolysis in the reactivation process in thermosensitive seeds of Phacelia isolated endosperms of lettuce (Lactuca sativa L.). Plant tanacetifolia with germination inhibited by high temperature Physiology 104: 623–628 (30°C). Physiologia Plantarum 99: 211–220 Dutta S, Bradford KJ, Nevins DJ (1997) Endo-β-mannanase activi- Prusinski J, Khan AA (1990) Relationship of ethylene production to ty present in cell wall extracts of lettuce endosperm prior to radi- stress alleviation in seeds of lettuce cultivars. Journal of the cle emergence. Plant Physiology 113: 155–161 American Society of Horticultural Science 115: 294–298 Forsyth C, Van Staden J (1983) Germination of Tagetes minuta L. I. Saini HS, Bassi PK, Consolacion ED, Spencer MS (1986) Temperature effects. Annals of Botany 52: 659–666 Interactions among plant hormones, carbon dioxide, and light in Gallardo M, Del Mar Delgardo M, Sanchez-Calle I, Matilla AJ (1991) the relief of thermoinhibition of lettuce seed germination: Studies Ethylene production and 1-aminocyclopropane-1-carboxylic acid in a flow-through gaseous system. Canadian Journal of Botany conjugation in thermoinhibited Cicer arietinum L. seeds. Plant 64: 2322–2326 Physiology 97: 122–127 Saini HS, Consolacion ED, Bassi PK, Spencer MS (1989) Control Gallardo M, Sanchez-Calle I, Muñoz De Rueda P, Matilla AJ (1996) processes in the induction and relief of thermoinhibition in lettuce Alleviation of thermoinhibition in chickpea seeds by putrescine seed germination. Plant Physiology 90: 311–315 involves the ethylene pathway. Australian Journal of Plant Small JGC, Schultz C, Cronje E (1993) Relief of thermoinhibition in Physiology 23: 479–487 Grand Rapids lettuce seeds by oxygen plus kinetin and their Heydecker W (1977) Stress and seed germination: An agronomic effects on respiration, content of ethanol and ATP and synthesis view. In: Khan AA (ed) The Physiology and Biochemistry of Seed of ethylene. Seed Science Research 3: 129–135 Dormancy and Germination. North Holland, Amsterdam, pp Takeba G, Matsubara S (1976) Analysis of temperature effects on 237–282 the germination of New York lettuce seeds. Plant and Cell Hills PN, Viljoen CD, Van Staden J (2001) Differences in polypeptide Physiology 17: 91–101 expression in thermoinhibited and germinating achenes of Van Staden J (1973) Changes in endogenous cytokinins of lettuce Tagetes minuta L. Plant Growth Regulation 34: 187–194 seed during germination. Physiologia Plantarum 28: 222–227 Horowitz M, Taylorson RB (1983) Effect of high temperatures on Van Staden J, Wareing PF (1972) The effect of light on endogenous imbibition, germination, and thermal death of velvetleaf (Abutilon cytokinin levels in seeds of Rumex obtisifolius. Planta 104: theophrasti) seeds. Canadian Journal of Botany 61: 2269–2276 126–133 Huang XL, Khan AA (1992) Alleviation of thermoinhibition in pre- Vidaver W, Hsiao AI (1975) Secondary dormancy in light-sensitive conditioned lettuce seeds involves ethylene, not polyamine lettuce seeds incubated anaerobically or at elevated temperature. biosynthesis. Journal of the American Society of Horticultural Canadian Journal of Botany 53: 2557–2560 Science 117: 841–845 Yang SF (1985) Biosynthesis and action of ethylene. HortScience Karssen CM, Haigh A, Van der Toorn P, Weges R (1989) 20: 41–45 Physiological mechanisms involved in seed priming. In: Taylorson Yoshioka T, Endo T, Satoh S (1998) Restoration of seed germination RB (ed) Recent Advances in the Development and Germination at supraoptimal temperatures by fluoridone, an inhibitor of of Seeds. Plenum Press, New York, pp 269–280 abscisic acid biosynthesis. Plant Cell Physiology 39: 307–312 Khan AA, Huang XL (1988) Synergistic enhancement of ethylene production and germination with kinetin and 1-aminocyclo- propane-1-carboxylic acid in lettuce seeds exposed to salinity stress. Plant Physiology 87: 847–852

Edited by TH Thomas