PLS 622: Plant Physiology I, September 15, 2006

Section II: Embryo and Seed Development:

Lecture X: Seed Dormancy:

IMPORTANT TOPICS TO BE COVERED:

Seed dormancy:

We will learn about...

- why dormancy is a sound strategy in some instances.

- how the plant hormone ABA plays a major role in the maintenance of the dormant state.

- that the metabolism of the dormant seed can be different from that of a seed that has completed germination:

- about the role phytochrome and the quality of light impacting the seed has on the dormancy status of some seeds. We will also investigate how two different phytochromes can influence the dormancy status of arabidopsis seeds.

- We will briefly go over skotodormancy (a.k.a. secondary dormancy).

Seed dormancy:

First, dormancy is not a trait that is unique to seeds. Buds too can be dormant in plants that grow in climates that have seasonal variations in favorable growing conditions. These buds can be on tubers and bulbs, or be the apical meristems of shoots and roots. In a larger context, dormancy is also a trait of many life forms spanning diverse kingdoms, Monera, Protista, Fungi, Plantae, even Animalia.

Quiescent vs dormant: Live seeds in which none of the germination events are taking place, usually due to a low moisture content, are said to be quiescent. They are alive and have metabolism ongoing at a barely detectable rate but some environmental factor necessary for germination to commence (usually the addition of water) is lacking. Seeds that are in an environment optimal for germination, that is to say they are provided with ample water, heat, light, and oxygen and yet fail to complete germination are said to be dormant. The block to the completion of germination is an attribute of the seed itself and must be removed before the seed will be able to complete germination under favorable conditions.

Why dormancy?: Because the purpose of the seed is to complete germination and produce the next generation of plant why some seeds should be shed with an inherent block to germination is not immediately obvious. It can be argued that the developing embryo of all seeds goes though a period of dormancy because, upon removal from the seed, slight desiccation, and germination on nutrient media, many embryos can complete germination whereas they cannot do so in the seed. This inability to precociously complete germination has been found to be largely due to elevated ABA amounts in the seed and tissue surrounding it during development on the mother plant. The advantages of this imposed block to germination during the development of a seed are easily understandable. But upon dissemination from the mother plant, why would mature seeds be incapable of completing germination without first undergoing some environmental stimulus and what evolutionary advantage could this hold? Particularly for plants growing in climates that vary considerably in their favorability for plant establishment (e.g. seasonal cold, dry, or flooding variations) immediate germination is not a desirable attribute. There must be some way of delaying the completion of germination until conditions favorable for the germination process and seedling establishment are, again, prevalent. Ideally, the environmental variations that occur are seasonal in nature, with conditions favorable and unfavorable for germination and establishment cycling yearly (e.g. cold in temperate or boreal climates). In this example, the switch from the conditions that make it difficult or impossible to complete germination and establish successfully (winter; freezing temperatures and lack of liquid water) to conditions conducive to completion of germination and establishment (spring; warmth, rain, and melting ice and snow) can be used as a cue to alleviate dormancy and permit the seed to complete germination at the start of an optimal growing period. Seeds from plants of desert annuals do not have the luxury of such a predictable cycle between a period conducive to establishment and one that is not and must be capable of reacting quickly to the availability of water at whatever period they get it. In such unpredictable climates, it is common to have a large variation in the degree of dormancy that individual seeds from the same plant experience. Hence, some seeds from the same maternal plant can be released with practically no dormancy, while others are deeply dormant. This spreads the number of seeds from a particular plant that are completing germination out over time and maximizes the possibility that some, at least, will establish successfully.

The role of ABA in inducing/maintaining seed dormancy: There have been innumerable studies of the types of dormancy exhibited by seeds and the covers or inhibitors responsible for dormancy imposition. In some cases, the distinction is not clear, as in arabidopsis. Wild type arabidopsis seeds, at least those of the land race (ecotype) Landsberg erecta (Ler) have coat imposed dormancy. This dormancy is initiated by an increase in ABA content occurring about halfway through normal seed development. ABA concentration then declines slightly during the latter phases of development until seed maturity. Hence, both ABA and the testa impose dormancy on the embryo. Additionally, there have been testa mutants isolated that show a reduced amount of dormancy. These so-called aberrant testa shape (ats) mutants do not need a period of afterripening to be able to complete germination to a high percentage. It is thought that the mutation, which alters the shape of the testa, also weakens it physically eliminating it as a barrier to radicle protrusion. However, work with the ABA deficient mutants of arabidopsis has shown that without ABA, testa imposed dormancy is not initiated and the embryos complete germination readily! So some aspect of the presence of ABA during development must trigger changes in the arabidopsis testa and/or embryo that result in dormancy imposition by this seed part in the mature seed. Elegant work using an ABA deficient mutant was performed that elucidates where the ABA that is responsible for establishing dormancy arises. Is the ABA coming from the mother plant that translocates the hormone to the developing seeds or is it synthesized by the seeds themselves as they develop on the plant? Homozygous mutants (aba) of a recessive gene imparting ABA deficiency were crossed with wild type (ABA) plants. The resulting progeny were all heterozygous ABA/aba and were backcrossed to the mutant (aba) (Fig.1). This resulted in seeds of two genotypes developing on a heterozygous, ♀ plant, a plant that could produce ABA. The seeds produced by the backcross however were either heterozygotes themselves (ABA/aba) and capable of producing ABA or homozygous for the recessive mutation (aba/aba) and ABA deficient. An analysis of seed dormancy showed that dormancy was a trait of only those seeds that were heterozygous, the homozygotes being without dormancy. If the mother plant was providing the ABA to the seeds that was responsible for the imposition of dormancy, all seeds from the back cross would be dormant, regardless of their genotype. Since this was not the case, the ABA responsible for dormancy imposition in arabidopsis must be seed derived (Fig. 1).

ABA plays a major part in initiating dormancy in seeds, but it is not alone in this process. Mutants have been isolated that complete germination without a requirement for GA. The seeds completed germination on media containing inhibitors of GA biosynthesis. Of these mutants, some were ABA deficient or insensitive but one, spindly, was hypersensitive to GA. This signifies that the antagonism mentioned in the context of seed germination between these two potent plant growth regulators is also functional in determining whether a seed is dormant or not. Large amounts of GA or hypersensitivity to it, lead to a non-dormant phenotype as does the lack of ABA or insensitivity to it. Based on the phenotype of ga mutants deficient in GA, lack of GA leads to extreme dormancy that cannot be alleviated without exogenous application of this hormone. Additionally, the reduced dormancy (rdo) mutants of arabidopsis are not deficient, insensitive, or hypersensitive to either ABA or GA. Their testa color is normal as is testa shape, unlike the transparent testa (tt) and ats mutants, respectively both of which can display reduced dormancy. The rdo mutations are further divorced from mutations affecting the testa by the observation that the F1 progeny of a cross between the homozygotic mutant ♀ and wild-type ♂ produced a wild type phenotype indicating the mutations affect the embryo and not maternal tissue (testa).

Figure 1: Seed dormancy in arabidopsis is installed by seed produced ABA during seed development. Maternal ABA is not responsible for initiating dormancy in seeds produced on the plant.

Metabolism of dormant seeds: There have been many different metabolic pathways hypothesized to play a role in alleviating dormancy in seeds over the years that this phenomenon has been studied. The synthesis of nucleic acids or proteins has been implicated as has a switch in metabolism which alters gas exchange, adenylate charge and/or abundance or uses an alternative pathway to produce energy and reducing power such as the pentose phosphate pathway (PPP), reverse glycolysis and cyanide-insensitive pathway. Some of the evidence accumulated has debunked several of these metabolic changes from having any influence on dormancy alleviation. Others, such as the PPP, are still under investigation. Support for the PPP being involved in dormancy alleviation comes from observations that the dormancy of some species seeds can be alleviated by the application of inhibitors of respiration. Substances that inhibit terminal oxidation and the tricarboxylic acid pathway are effective in alleviating dormancy in these seeds as are inhibitors of glycolysis. Additionally, electron acceptors can alleviate dormancy. These inhibitors are thought to function by diverting cellular oxygen from regular respiration to the PPP, where it is used to oxidize NADPH to NADP. The electron acceptors replace oxygen in this capacity and again result in elevated amounts of NADP. However, studies have shown that the NADP/NADPH ratios are completely unrelated to the dormancy status of the seeds. Whatever is occurring to alleviate dormancy upon application of inhibitors of respiration or electron acceptors, it does not appear to be due to elevated NADP amounts. On the other hand, Bob Buchanan and co-workers are continuing to elucidate the role thioredoxin plays in permitting the reduction of disulfide bonds in storage proteins and inhibitors of alpha-amylase thereby permitting storage protein utilization in the first instance and inhibitor inactivation in the second. Thus, the PPP pathway, the only known source of NADP/NADPH in the seed, is the sole source of reductant to reduce thioredoxin, via the enzyme NADPH thioredoxin reductase, back to a form capable of reducing disulfide bridges since NAD/NADH is not capable of this reaction (Fig. 2).

One of the problems defining metabolic limitations that impose seed dormancy is that different species seeds behave differently and may alleviate dormancy through a different metabolic switch than others. Another is accurately determining the dormancy imposing tissue. For instance, lettuce embryos appear to be constrained by the endosperm and it is this tissue that imposes dormancy upon them in the absence of light. Illuminate the seed and you alleviate dormancy. This could occur through either cell wall weakening of the endosperm or

increased embryo thrust permitting them to push through the endosperm opposing their expansion. Excised embryos complete germination in darkness with no apparent dormancy. However, if they are placed under water stress, the excised embryos germinated in the dark fail to elongate at much less sever water deficits than embryos germinated in the light suggesting that light enables the lettuce embryo to generate more thrust than is possible in the dark. So, the embryos in light have a lower water potential than embryos germinated in darkness. What is different between the two? Are the light-germinated-embryos more osmotically active or do they have more extendable cell walls resulting in lower turgor pressure? Measurements of osmotic potential failed to reveal differences between light-germinated and dark-germinated embryos. However, a decrease in turgor pressure has been documented in response to light. This means that the cell walls of the embryo, when it is illuminated, weaken, allowing the cells to elongate more easily than if the embryo was held in darkness. The mechanism though which light acts appears to be in a pH decrease in the apoplast. There is a vacuolar proton ATPase that is upregulated during germination in tomato, but to date, no report of any of the number of plasmamembrane proton ATPases known to exist being likewise regulated has been documented. However, isolated lettuce embryos do seem capable of decreasing the pH of the media they are in if illuminated, providing strong evidence that such a proton pump does exist.

Figure 2: The necessity of the Pentose phosphate pathway to provide a source of NADPH to the cells of the germinating seed. Thioredoxin h activity permitting catabolic utilization of stored reserves depends on the recycling of thioredoxin h to the reduced form via NADPH.

Phytochrome and Dormancy: One of the more spectacular discoveries in plant physiology involved the control light quality has on the ability of many species seeds to complete germination. There are no fewer than 5 phytochromes present in arabidopsis. These chromophores were first discovered and investigated due to the marked effect phytochrome B and possibly others, has on lettuce seed germination. When imbibed seeds of lettuce were illuminated with a period of far-red light, the percentage of seeds that subsequently completed germination in the dark was very low. If the period of far-red illumination was followed by a period of red light illumination, the seeds subsequently completed germination to almost 100%. If however, the second, red light illumination was followed by another period of far-red illumination, seed germination was again drastically inhibited in the dark. This cycle of germination inhibition and stimulation can continue ad infinitum until, at some point much advanced in seed germination, the seeds “escape” from phytochrome control and complete germination in the dark regardless of the illumination they perceived last.

The antagonistic nature of red/far red illumination on seed dormancy has led to intensive

investigation of phytochrome states after illumination from light of different spectral qualities.

As mentioned previously, there are at least 5 different phytochromes present in arabidopsis. Of the phytochromes, A and B are by far the best studied, having been cloned and for which null mutants are available. Studies of the phytochrome A deficient, phytochrome B deficient, and double mutant, have led to the conclusion that both phytochrome A and B play a role in seed germination. Phytochrome A, upon absorbing far red light is actually stimulatory for germination while phytochrome B is the phytochrome that is purportedly responsible for the photoreversible effect on seed dormancy. Recently, however, the arabidopsis double phytochrome AB null mutant was used to show that seed germination is still under phytochrome control, signifying that at least a third phytochrome is involved in determining light regulated seed dormancy in arabidopsis.

Studies with microbeams of intense light at different wavelengths has demonstrated that phytochrome is present in greatest amounts in the axis of embryos and very little is present in the cotyledons. This is significant because it is the elongation of the radicle that leads to the completion of germination and the phytochrome control of dormancy probably should be localized to this portion of the embryo rather than the cotyledons. No mention was made of whether phytochrome was also found in the endosperm. This is not crucial to our interpretation below that, in some seeds at least, phytochrome mediated endosperm wall weakening might control dormancy and germination. It is fully possible that, even if phytochrome is present only in the axis, its photoconversion to the active form Pfr elicits a response from the radicle cells that includes the production and transport of a signal to the cells of the endosperm in the immediate proximity of the radicle i.e. the micropylar endosperm. This signal would then dictate that the endosperm cells should produce enzymes that would weaken their cell walls.

Partial cell wall disassembly and dormancy: Recently more interest has been directed to determining whether or not the cell walls of the endosperm cap region in endospermic seeds weakens during dormancy alleviation. If the micropylar endosperm does not weaken, as we saw in the gib-1 mutant of tomato and ga mutant in arabidopsis, the embryo will not complete germination. There does appear to be some evidence suggesting that endosperm weakening does have to occur in several species and that this weakening might be prevented in seeds that are dormant. The enzymes that have been investigated to date all hydrolyze hemicellulose present in the cell walls of the micropylar endosperm. In Datura ferrox endo-b-mannanase and a cellulase, both of whose transcription is under phytochrome control, accumulate prior to radicle protrusion and endosperm cap weakening. In lettuce, endo-b-mannanase also appears to be under phytochrome control and may accumulate prior to radicle protrusion, although this is contentious. In tomato several forms of endo-b-mannanase exist, one of which is apparently specific to the endosperm cap the gene of which is downregulated by ABA and upregulated by GA prior to the completion of germination.

Secondary (skoto) dormancy: Secondary dormancy is brought about when fully hydrated, mature seeds experience gravely sub-optimal germination conditions. Some of the conditions that invoke secondary dormancy include, anoxia, darkness, excessively strong light, higher (lower) than maximal (minimal) temperatures for the completion of germination, and water stress. These conditions do not universally elicit secondary dormancy for all species, some species are affected and other are not. Whatever the mechanism whereby skotodormancy is initiated, ABA does not appear to be involved, at least not in the response of arabidopsis to higher than maximal temperatures for germination. Even ABA deficient mutant seeds are capable of skotodormancy. Secondary dormancy plays a large part in the maintenance of a persistent seed bank of numerous species. Those seeds that fail to complete germination, for whatever reason, during the favorable portion of the year for them are prevented from radicle protrusion later due to unfavorable conditions. Prolonged exposure to these unfavorable conditions elicits secondary dormancy that persists until the seeds undergo a dormancy alleviating treatment. This treatment is often overwintering for summer annuals or afterripening during the next summer for winter annuals. Return to favorable climatic conditions gives the now non-dormant seeds a second chance to complete germination. Often, secondary dormancy decreases the seeds sensitivity to chemicals known to promote germination of non-dormant seeds. Such is the case with the small annual weed, Sisymbrium officinale which decreases its sensitivity to both light and nitrate upon entering secondary dormancy.