PLS 622: Plant Physiology I

Wednesday, September 13, 2006

Section II: Embryo and Seed Development:

Lecture IX: Seed germination:

yeah…right!

Orthodox seeds undergo maturation desiccation and some remain on the mother plant for some time in a fruit or cone. Eventually, however, all seeds are released from the mother plant and dispersed by the elements or by other living things. It is in the guise of the seed that the plant is most resistant to external stress. For example, it is possible to take desiccated orthodox seeds of many species and immerse them in liquid nitrogen (-196°C), take them out after one minute or decade, warm them, water them and they complete germination normally! Their resistant nature permits seeds to withstand periods of unfavorable environmental perturbation and sustains life until conditions more favorable to germination and establishment arise. What is germination? Strictly speaking germination commences when a seed imbibes water and terminates as soon as some portion of the embryo (usually the radicle) emerges from the seed covers (e.g. testa, nucellus, perisperm, endosperm). The inappropriate use of the term to also describe seedling establishment has lead to considerable confusion and necessitates covering GERMINATION and POST-GERMINATION events separately for emphasis.

Anhydrobiosis: Mature, desiccated seeds can have a moisture content of less than 5% by weight and yet they are alive, they respire, although it is difficult to measure because it occurs at such a reduced rate. If seeds are capable of completing germination upon the provision of light, heat, water, and favorable atmospheric conditions, when desiccated they are said to be quiescent. If they are incapable of completing germination upon entering favorable conditions due to some internal impediment to radicle protrusion they are said to be dormant. We will discuss seed dormancy and its implications in the next lecture.

Germination events:

Water uptake: Seed water uptake can be divided into three stages (Fig. 1). Imbibition, the first stage, is usually rapid, commencing with the seed being placed on water where it quickly hydrates the cells and their constituents. This is followed by the lag phase of water uptake, where there is very little net gain of water. This is not to say that water is not taken up by the seed. In reality there is a steady state reached between the amount of water lost by the seed due to evaporation and that taken up by the seed from its surroundings. In natural conditions, the amount of water lost by a seed that is placed on the surface of the ground can actually exceed the amount it takes up from the soil and limits germination. Finally, the third phase of water uptake is one of rapid hydration to support the expansion of the embryo as it completes germination and emerges from the seed (Fig. 1). Due to the impermeability of the testa or other covers, some seeds cannot take up water freely when placed in contact with it but must wait for some physical abrasion or partial chemical breakdown to occur prior to imbibing.

The antagonistic role of ABA and GA in seed germination: Martin Koornneef and other geneticists have isolated many mutants involved in seed development and germination. Two such mutant families include a series of abscisic acid deficient (aba) or insensitive mutants (abi) and several mutants affected in their ability to synthesize (ga) or perceive (gai) gibberellic acid. These mutants have been instrumental in elucidating the fundamental role the interplay of these two potent plant hormones has on seed development and germination.

ABA is present in the seed during development, peaking in concentration during the period of maximum stored reserve accumulation. It is currently thought that ABA regulates when and to what extent storage proteins are synthesized in maize. This theory arises from the behavior of the ABA deficient (vp5) and ABA insensitive (vp1) mutants of maize which do not accumulate a specific globulin (storage protein). Neither the protein nor its mRNA is present in either mutant during development. Upon application of exogenous ABA, globulin production can be stimulated in the vp5 mutant but not the vp1. Strangely enough and probably partially based on differences in storage protein gene promoter characteristics, neither the ABA deficient arabidopsis (aba) nor tomato (sit^w) mutants have deficiencies in storage protein accumulation. However, the arabidopsis mutants, including those insensitive to ABA (abi1, 2, 3) are all somewhat leaky and the abi mutants all have higher than normal concentrations of ABA so a lack of abnormal storage protein accumulation may be due to the presence of some ABA and/or the partial sensitivity of the embryo to it. Recently, a new ABA insensitive mutant, abi3-3 has been isolated which is totally insensitive to ABA and this mutant does not accumulate either the 12S or2S (S = Svedberg Unit, the rate of sedimentation of a macromolecule in a centrifugal field) storage proteins. ABA is also known to prevent the precocious germination of embryos in developing seeds. When removed from the seed, many developing embryos that have attained maximal dry weight or close to it can complete germination on media if they have been desiccated slightly (5-10% loss of fresh weight). The ABA in the seed is thought to prevent this from occurring in vivo until the seed commences maturation desiccation, at which point there is insufficient water available to the seed to permit the completion of germination.

GA is a potent stimulatory hormone for seed germination. The fact that severe GA deficient mutants in arabidopsis (ga) and tomato (gib-1) will not complete germination without an exogenous supply of GA indicates how crucial this hormone is to seed germination. The current wisdom is that there is still sufficient ABA present in the mature, desiccated seed to inhibit germination. The newly synthesized GA made during the lag phase of water uptake overrides this ABA inhibition or hastens ABA metabolism to the biologically inactive metabolite phaseic acid, permitting radicle protrusion. However, treatments that prevent ABA metabolism such as illumination with far red light (730 nm) inhibits the completion of germination. Addition of exogenous ABA to seeds also prevents the completion of germination, seemingly overwhelming the stimulatory effect elicited by GA biosynthesis or preventing the synthesis of GA altogether.

This is not the whole story. Recent results from the lab of Peter McCourt have shown that both ethylene and brassinosteriod act to desensitize the embryo, at least, to ABA. Seeds producing ethylene required less GA to overcome the ABA inhibition of radicle protrusion than seeds that were not synthesizing ethylene. This is also the case for brassinosteriods. This fact helps to explain why GA deficient mutant seed of arabidopsis (ga) that usually require GA to complete germination, were able to do so without exogenous GA if they were exposed to ethylene. Using figure 2, can you understand why the GA-deficient seeds completed germination in ethylene?

Cellular events during seed germination…THE original (and persistent) “black box”: The answer to the children’s song above is still ‘NO!’. But we are making progress. Here is an overall synopsis of what must occur during the germination event (see also Figure 1 below). After phase I of water uptake, the cells within the seed must reconstitute their lipid membranes, repair damage to DNA and protein incurred during the quiescent phase, replace extensively damaged membranes, resume transcription, translation, and respiration, and commence instituting the sensory machinery to enable it to monitor its external environment to determine if conditions are favorable for establishment or not.

Figure 1: Time course of water uptake by an orthodox seed depicting the three stages of imbibition (phase 1), lag (phase 2), and completion of germination (phase 3) as well as the cellular events thought to occur in each phase (Redrawn and altered from Bewley 1997, The Plant Cell, 1055-1066).

Figure 2: The antagonistic relationship between GA and ABA during seed germination and subsequent radicle protrusion. This relationship is further complicated by the ability of both ethylene or brassinosteroid to desensitize the embryo to endogenous ABA.

Desiccated seeds contain an intracellular environment in which the cytoplasmic and organellar contents are extremely concentrated. High salt concentration, a long period in the desiccated state, and rapid imbibition, all exacerbates DNA breakage, loss of membrane integrity, and enzyme inactivation. During germination, a system capable of orchestrating considerable repair to a variety of macromolecules is invoked. The natural repair mechanism (NRM) includes both DNA and protein repair mechanisms. This repair must be effected at a point in the plant life cycle when energy, reductant, and structural carbon are poorly available since they are immobilized in storage materials. Additionally, the mitochondria are not efficient during the early part of germination since their membranes too have suffered compromise. All of these limitations necessitates the type of repair, if any, the NRM can conduct.

Despite all of the activity in repairing damage, and preparing the seed for the completion of germination, there is not a single visible exterior change discernable between the live and the dead seed, the desiccated and the imbibed! Even at the ultrastructural level, very few changes are detectable during seed germination and those only a few hours prior to the completion of germination. There are, however, three changes that can be discerned in many seeds that are about to complete germination but have not yet done so. They are: 1) the differentiation of the provascular cylinder of the embryo into xylem and phloem; 2) some very limited mobilization of stored reserves in the radicle tip and (endospermic seeds only) in the cells of the micropylar endosperm, and; 3) (again only in endospermic seeds) partial cell wall disassembly of the cells of the micropylar endosperm ahead of the radicle.

The differentiation of the provasculature into xylem and phloem is necessary in order to conduct metabolites from stored reserves located in the cotyledons or endosperm into the growing radicle and hypocotyl. Additionally, this permits the embryo to conduct water taken up from the soil immediately upon the completion of germination to support continued cell expansion throughout the embryo. Do not forget that, upon successful completion of germination in epigeal (in which the cotyledons are carried by the hypocotyl above the soil surface as opposed to hypogeal where the cotyledons stay beneath the soil surface) seedlings, the cotyledons are raised from the surface of the soil, losing any chance to absorb water by diffusion and therefore are dependent on functioning vasculature for water supply. The ABA insensitive mutant (abi 3-4) and embryogenesis mutation leafy cotyledon (lec 1-2) both prematurely differentiate their vasculature while the embryo is developing.

The limited hydrolysis of stored reserves in the radicle tip and micropylar endosperm adds energy and carbon and nitrogen building blocks to the actively growing radicle prior to and during the commencement of growth. Typically, this replaces the reserves of sucrose and raffinose family oligosaccharides (RFOs) that have been exhausted during the first stages of germination (Fig. 1). It must be emphasized that this mobilization of protein and carbohydrate/lipid is in a very limited part of the seed and the vast majority of stored reserves of protein and carbohydrate/lipid remain unmobilized until after the radicle has protruded from the seed (i.e. is a post-germinative event, see below)!

The partial disassembly of the cell wall of the region of the endosperm or other covers opposite the radicle has been studied extensively in endospermic seeds. Candidate enzymes that hydrolyse the hemicellulosic or pectin components of the cell wall that have been studied in this regard include endo-b-mannanase which randomly cleaves polymers of mannose anywhere along the backbone into shorter fragments, xylo-glucanases that randomly cleave xyloglucan polymers, expansins that simply disassociate hemicellulose polymers from para-crystalline cellulose microfibrils without cleavage, xylo-glucan endotransglycosylases that cleave and then re-polymerize xyloglucan polymers, and polygalacturonases that cleave single galacturonic acid moieties from the non-reducing ends of polymers of galacturonic acid (pectin). All of these hydrolases, transglycosylases, expansins probably play an interactive role in weakening the resistance of the endosperm or other covers to puncture by the radicle. Hence, it becomes increasingly easy for the embryo to push through or between the cells of the covers and emerge. This decrease in resistance to puncture has been demonstrated for many angiosperm and one conifer seed. Indeed, the requirement of endosperm weakening to permit radicle protrusion, in endospermic seeds, is one of the few facets of seed germination that is absolutely known through direct measurements of resistance to puncture using an instron (instrument capable of measuring extremely small forces). Additionally, the pleiotropic mutant deficient in gibberellic acid in both arabidopsis (ga) and tomato (gib-1) does not complete germination unless the seeds are provided with exogenous gibberellic acid or the testa and endosperm cap are removed surgically. The endosperm cap of the tomato gib-1 mutant does not weaken as the wild type does unless exogenous GA is applied.

Work on the poor germinating, GA-insensitive mutants sleepy and sneezy has uncovered a series of 5 proteins (GA-INSENSITIVE (GAI), REPRESSOR OF GA (RGA), REPRESSOR OF GA-LIKE1 (RGL1), RGL2 and RGL3). These proteins are collectively named ‘DELLA’ proteins because they all possess this highly conserved, pentameric amino acid combination in their amino-terminus. They are all inhibitors of GA action, including seed germination. In the presence of the DELLA proteins, GA fails to stimulate cellular metabolism leading to the completion of germination. These negative regulators of the GA response (DELLA proteins) must be actively tagged and degraded. This is accomplished in normal seeds by the SLEEPY and SNEEZY proteins. SLEEPY and SNEEZY are ‘F-box proteins’. F-box proteins can attach to target proteins, take them to the poly-ubiquitination complex and position them so that the ubiquitin ligases of the complex can polyubiquinate the target protein. This slates the target protein for destruction by the 26S proteasome. Without SLEEPY and/or SNEEZY the DELLA proteins persist in the cells of the seed, interfering with the ability of GA to stimulate germination.

Non-germination may be due simply to a lack of sufficient energy. Sugar is a potent signal for many developmental processes, including seed germination. The components of the sucrose non-fermenting (SNF) catabolic repression system (yeast) have been identified in plants and shown to interact with components of the yeast system in vitro. Rice SNF1 genes are transcribed during seed development and it is thought that the SNF system is conserved throughout the eukaryotes and therefore, likely to be functional in plants, controlling the switch between anabolism and catabolism. There is a fundamental switch in the primary metabolic processes occurring in seeds between development and germination. During development, the majority of the incoming assimilate is put into storage reserves (anabolism) so the seed primarily synthesizes compounds from smaller molecules. During germination, this process is reversed where the large storage polymers are catabolized in one part of the seed and used to make energy and structural molecules in others. The de-repression of catabolism in the germinating seed is absolutely necessary for radicle protrusion. Embryos excised from non-germinable barley seeds were still viable, capable of completing germination and forming autotrophic plants if they were given a metabolizable carbon source.

An example of how important the ability to mobilize stored reserves are to the completion of germination has been provided by the mutant aptly named ‘comatose’. The comatose (cts) mutant is unresponsive to GA, much like the sleepy and sneezy mutants mentioned above. The CTS gene encodes a transporter of acyl CoA situated in the peroxisome where it orchestrates the transport of acyl CoAs (produced in the lipid body through breakdown of storage lipid) into the peroxisome for energy and carbon-skeleton production. In the absence of a functional transporter, the lipid storing seed (Arabidopsis in this instance) literally does not have sufficient energy to germinate and hence ‘stalls’ prior to the completion of germination.

An alternative role for cell wall hydrolases (as opposed to the breakdown of the endosperm cell walls to permit the radicle to push through) is in protection of the seed from pathogen attack. Seed germination takes place in or on soil that is rife with microbes that can potentially infect and destroy the seed. This presents a quandary. While the seed is intact, the testa usually provides an effective barrier to infection. But the embryo itself must penetrate the seed covers from the inside in order to complete germination thus exposing the internal seed environment, moist and nutrient rich, to soil microbes. There are mechanisms that seeds invoke that prevent infection before and after radicle protrusion. Enzymes with a putative role in protection from infection (pathogenesis related, PR proteins) have been shown to accumulate in the cells of the endosperm cap of germinating seeds and some of these PR proteins are cell wall hydrolases. So, perhaps, cell wall hydrolases are not present in the seed cell walls to weaken these cell walls so much as they are there to inhibit microbial attack.

Shen-Miller, J., et al. 1995. Exceptional seed longevity and robust growth: Ancient Sacred Lotus from China. American Journal of Botany 82(November):1367-1380.