Abstract
Seed germination is a key developmental process that has to be tightly controlled to avoid germination under unfavourable conditions. Abscisic acid (ABA) is an essential repressor of seed germination. In Arabidopsis, it has been shown that the endosperm, a single cell layer surrounding the embryo, synthesizes and continuously releases ABA towards the embryo. The mechanism of ABA transport from the endosperm to the embryo was hitherto unknown. Here we show that four AtABCG transporters act in concert to deliver ABA from the endosperm to the embryo: AtABCG25 and AtABCG31 export ABA from the endosperm, whereas AtABCG30 and AtABCG40 import ABA into the embryo. Thus, this work establishes that radicle extension and subsequent embryonic growth are suppressed by the coordinated activity of multiple ABA transporters expressed in different tissues.
Similar content being viewed by others
Introduction
Seed germination is tightly controlled in plants. Arabidopsis seeds exhibit primary dormancy, a trait whereby germination of freshly produced seeds is blocked even under favourable germination conditions1,2. Many studies have sought to decipher the complex regulatory network that controls seed germination, and it is apparent that abscisic acid (ABA) and gibberellins (GAs) antagonistically regulate seed germination in both dormant and non-dormant seeds3,4,5,6,7. Whereas ABA represses seed germination, GAs promote it2,8,9. In non-dormant seeds, environmental parameters, such as light quality, temperature or water potential, determine the endogenous production of ABA and GA10,11,12. Recent work has shown that GAs promote germination by negatively regulating ABA levels13, which stimulate the production of germination repressors, such as transcription factors ABI3 and ABI5 (refs 12, 13, 14, 15). GA promotes the proteasome-mediated destruction of DELLA factors that promote endogenous ABA accumulation, leading to seed germination12,13.
Dormant seeds typically block germination by maintaining high levels of ABA on imbibition16. Accordingly, mutant seeds unable to synthesize ABA lack dormancy; that is, they germinate precociously and may even germinate while still attached to the mother plant17. This phenomenon, termed vivipary, results in economic loss in crop species. Arabidopsis natural accessions display markedly different dormancy levels: the Cape Verde Island (Cvi) ecotype exhibits much stronger dormancy at harvest than the Columbia-0 (Col-0) or Landsberg erecta (Ler) ecotypes18,19. These ecotypes differ in ABA content in the dry seed and the speed of ABA degradation on imbibition20. Quantitative trait loci involved in the delay of germination have been identified by comparing the Cvi and Ler ecotypes21,22, and some of the loci have been suggested to be related to ABA signalling or GA sensitivity21,22,23.
One intriguing and poorly understood aspect of seed physiology is the spatial and temporal regulation of ABA transport on seed imbibition. The Arabidopsis seed consists of an embryo, an active endosperm layer surrounding the embryo, and an outer layer of dead tissue, the testa, of maternal origin24. A former study indicated that in dormant seeds, ABA is produced by both the embryo and endosperm25,26,27,28. More recent studies revealed that some inhibitory compounds produced by tissues surrounding the embryo are responsible for dormancy29,30,31. Elegant experiments conducted by Bethke et al.25 further showed that the endosperm layer is responsible for most, if not all, of the germination-repressive activity in Arabidopsis. Recently, Lee et al.32 developed a seed coat bedding assay (SCBA) that allows the monitoring of growth of dissected embryos cultured on a layer of seed coat bedding, which consists of testa and endosperm. This assay is useful to identify components active in either endosperm or embryo that control germination. Indeed, this assay provided direct evidence that ABA is released from the endosperm of dormant seeds on imbibition to inhibit embryo development. The SCBA was used to further show the importance of the endosperm in non-dormant seeds. An early far-red light pulse that blocks germination involves phytochrome B (phyB) inactivation in the endosperm, which blocks the expression in GA biosynthesis genes in the endosperm, and induces ABA release towards the embryo33. Furthermore, recent work suggested that in ga1 seeds, which are unable to synthesize GA, the endosperm also releases ABA to block germination34.
However, how ABA is transported between the endosperm and embryo in seeds remains poorly understood. Until recently, it was assumed that ABA, being a weak acid (pKa=4.9), passes through biological membranes via passive diffusion when protonated, and is then trapped in the dissociated form in the neutral pH environment of the cytosol35. However, auxin, another weak acid and a potent growth-promoting phytohormone, is not only transported by passive diffusion but also by different types of transporters, which enables the plant to respond accurately and rapidly to environmental changes by adjusting its growth36,37,38. Thus, it is tempting to speculate that a multitude of ABA transporters coordinately function to deliver ABA to specific tissues in response to environmental and developmental cues. Indeed, recent studies identifying several ABA transporters might be a prelude to the discovery of such a network of ABA transporters39,40,41,42,43,44. Two ABC transporters of the ABCG subfamily, AtABCG25 and AtABCG40, export ABA from the vascular bundle and import it into guard cells, respectively39,40. Furthermore, another ABC transporter, AtABCG22, was proposed to function as an ABA transporter41. In addition, several members of the ABA-importing transporter family function as ABA importers at the influorescence stem42,43. Recently, a MATE-type transporter has also been reported to act as an ABA efflux carrier in guard cells44.
We reasoned that ABA delivery from the endosperm to the embryo must be accurately controlled and thus might be mediated by multiple coordinated groups of transporters, rather than by diffusion alone. We searched for candidate transporters among the ABCG subfamily of ABC proteins, since this family mediates the efflux and influx of a wide variety of terpenoids45, including ABA.
Here we report that four AtABCG proteins function together to deliver ABA from the endosperm to the embryo in mature, imbibed seeds. AtABCG25 and AtABCG31 are mainly localized to the endosperm and export ABA from the endosperm to the embryo, whereas AtABCG30 and AtABCG40 are localized to the embryo and transport ABA into the embryo. ABA transport represents an important step in embryo germination and early seedling establishment. Thus, this work demonstrates that a developmental step critical for a plantâs survival is efficiently regulated by the coordinated activity of multiple ABA transporters localized in different tissues.
Results
Screening for candidate ABA transporters in seeds
To select candidate ABA transporters that regulate seed germination, we first performed an in silico analysis (Arabidopsis eFP browser; http://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi) of all 43 members of the ABCG subfamily of ABC transporters, and identified 10 genes that are highly expressed in mature seeds. To determine whether any of these candidates may function as ABA transporters within seeds, we screened the mutants for altered patterns of seed germination (Fig. 1). Six of these ten lines did not have an altered germination phenotype. In contrast, the mutant lines of two candidate genes, AtABCG30 and AtABCG31, exhibited clear differences in germination when compared with wild-type seeds. Seeds from freshly collected atabcg31 and atabcg30 mutants germinated earlier than the corresponding wild-type seeds on imbibition without stratification (Fig. 1a,c,e), whereas seed stratification abolished the difference between the mutants and wild type (Fig. 1f). To further investigate whether atabcg31 and atabcg30 mutant seeds exhibit deficiencies in ABA transport, we compared their responses with exogenously added ABA and found that germination of stratified atabcg31 and atabcg30 seeds was less inhibited by 0.1â1.0âμM ABA than was that of the corresponding wild-type seeds (Supplementary Fig. 1aâe). Similarly, we examined whether AtABCG25 and AtABCG40 function as ABA transporters in seeds. Indeed, atabcg40 seeds exhibited a higher germination rate than the wild type on medium containing 0.1âμM ABA (Supplementary Fig. 1b), suggesting that ABA uptake was reduced in these seeds, consistent with a previous report40. Furthermore, seeds from freshly collected, non-stratified atabcg40 mutants also germinated earlier than those from the corresponding wild type (Fig. 1d,e). In contrast, the germination of non-stratified, freshly collected atabcg25 seeds was similar to that of wild-type seeds (Fig. 1b,e), even when grown on medium containing exogenous ABA (Supplementary Fig. 1f). This finding contradicts a recent report that showed that the germination of atabcg25 seeds is delayed on medium containing ABA and that AtABCG25 acts as an ABA exporter39. This discrepancy may be explained by the ecotype difference and mild nature of the mutantâs phenotype, which may depend on plant growth conditions. Given the convincing data presented previously39, we continued to include atabcg25 in our studies. Taken together, these results strongly suggest that at least four AtABCG transporters are involved in controlling seed germination.
AtABCGs are expressed either in the embryo or endosperm
To further explore whether these AtABCG proteins indeed influence germination by regulating ABA transport within the seed on imbibition, we analysed the transcript levels of the corresponding genes in the endosperms and embryos of mature wild-type seeds. Tissue-specific quantitative reverse transcriptionâPCR (qRTâPCR) revealed that AtABCG31 was not expressed in dissected embryos, but was expressed at high levels in the endosperm (Fig. 2a). AtABCG25 was mainly expressed in the endosperm and, to a lesser extent, in the embryo (Fig. 2b). This result is consistent with previous in silico data31. In contrast, AtABCG30 and AtABCG40 transcripts were strongly expressed in dissected embryos, but only weakly in the endosperm (Fig. 2d,e). To validate our approach, we performed qRTâPCR with two genes known to be expressed either in the endosperm (AtEPR1, an extensin-like gene involved in seed germination)46 or the dissected embryo (AtABI4, encoding an ABA-response transcription factor)31. Consistent with previous reports, AtEPR1 transcripts were detected solely in the endosperm, whereas AtABI4 transcripts were detected nearly exclusively in the embryo (Fig. 2c,f). Our analysis of multiple lines of promoterâβ-glucuronidase-expressing plants (Fig. 2g and Supplementary Fig. 2a) confirmed the qRTâPCR results; the AtABCG31 promoter was active only in the endosperm layer, and β-glucuronidase activity driven by the AtABCG30 and AtABCG40 promoters was observed in dissected embryos, but not in the endosperm. Together, these results show that AtABCG31 and AtABCG25 expression is specific to the endosperm, where their protein products may act as ABA efflux transporters. In contrast, AtABCG30 and AtABCG40 expression is specific to the embryo, where their protein products may function as ABA influx transporters. To function as transporters that facilitate ABA flux from the endosperm to the embryo, these four transporters are expected to be localized to the plasma membrane. AtABCG25 and AtABCG40 were previously shown to localize to the plasma membrane39,40,47. Here we show that AtABCG31 and AtABCG30 are also localized to the plasma membrane of Arabidopsis mesophyll protoplasts and tobacco epidermis cells when transiently expressed under the control of the 35S promoter (Fig. 2h and Supplementary Fig. 2b). Co-expression of green fluorescent protein (GFP)-fused AtABCG31 or AtABCG30 with PM-rK (ref. 48), a plasma membrane marker protein (an mCherry fusion of the plasma membrane aquaporin, AtPIP2A), showed complete co-localization of the two ABCG proteins with PM-rK. Plasma membrane localization of AtABCG31 was recently also shown by Choi et al.49.
ABCG31 and ABCG25 repress embryonic growth
We next sought to determine whether the embryo or endosperm is responsible for the altered germination phenotype of mutant seeds by performing SCBAs as described by Lee et al.32,33. SALK insertion Arabidopsis lines are generated in the Col-0 ecotype, which displays relatively weak and short-term dormancy compared with ecotypes such as Ler or Cvi, for which the SCBA was initially used. As a result, a SCBA using freshly collected Col-0 material may result in less robust embryonic growth arrest results. To better test the role of AtABCG31 and AtABCG25 in controlling embryonic growth, we performed SCBAs in presence of paclobutrazol (PAC), a GA synthesis inhibitor. Lack of GA synthesis promotes ABA accumulation and efficiently blocks embryonic growth in SCBAs13,34,50. Indeed, in presence of PAC, seed coats of Col-0 prevented embryonic growth (Supplementary Fig. 3a), and expressed the germination repressor AtABI5 (Supplementary Fig. 3b)15, as effectively as those of Cvi. However, consistent with previous results, PAC did not affect the embryonic growth of dissected embryos or their ability to establish seedlings9 (Supplementary Fig. 3c). In this assay, we also used dissected embryos of the aba2-1 mutant, which is defective in ABA biosynthesis and consequently germinates precociously51. For quantitative analysis, we measured the angle between cotyledons and radicles as well as radicle length (Fig. 3a,b), as indicators of early embryonic growth. As shown in Fig. 3, seed coats (testa and endosperm) of the atabcg31 or atabcg25 single knockouts did not inhibit the growth or early seedling establishment of the aba2-1 embryo as effectively as did those of the wild type in the SCBA. Our findings for the atabcg25 seed coat are inconsistent with our observation that the germination rate of atabcg25 whole seeds is similar to that of the wild type (Fig. 1b,e and Supplementary Fig. 4a). This inconsistency may be due to a defect in atabcg25 embryos, which exhibit retarded embryonic growth even when dissected out from the surrounding tissues (Supplementary Fig. 4b). The retarded growth of embryos seems to cancel out the weak inhibitory effect of the endosperm on germination; thus, no aberrations in whole-seed germination are apparent. Reciprocal crossing of atabcg25 with atabcg31 yielded g31/g25-1 (atabcg31 pistil; atabcg25 pollen) and g31/g25-2 (atabcg31 pollen; atabcg25 pistil) double mutants, which lack both ABC transporters. The seed coats and endosperms of the double knockouts (g31/g25-1 or g31/g25-2) inhibited the growth and seedling establishment of aba2-1 embryos much less effectively than did those of the single knockouts or the wild type (Fig. 3câe and Supplementary Fig. 7). These results suggest that atabcg31 and atabcg25 endosperms release reduced amounts of germination inhibitor. In contrast, the seed coat bedding of the other candidate ABC transporter mutants, atabcg30 and atabcg40, did not differ from those of the wild type in terms of their effect on the development of aba2-1 mutant embryos (Supplementary Fig. 5a, and Fig. 3c,d), indicating that the amount of inhibitor molecules released from the seed coat beddings of wild-type, atabcg30 and atabcg40 plants was similar.
atabcg30 and atabcg40 embryos are impaired in ABA uptake
We next sought to determine whether AtABCG30 and AtABCG40 are necessary for embryonic uptake of endospermic ABA. For this purpose, we tested whether atabcg30 and atabcg40 mutants are defective in the embryonic uptake of ABA. We evaluated the growth of atabcg30 (g30), atabcg40 (g40) and g30/g40 embryos in SCBAs using seed coats isolated from the highly dormant Cvi ecotype, which were previously shown to efficiently block wild-type embryonic growth in SCBAs32. While most embryos of Col-0 wild-type seeds did not grow until 85âh after imbibition (Supplementary Fig. 5a), the embryos dissected from the seeds of atabcg30 and atabcg40 and the corresponding double-knockout plants (g30/g40) grew and developed to plantlets (Supplementary Fig. 5a and Fig. 4aâc). Thus, these observations indicate that the mutations in AtABCG30 and AtABCG40 genes could not efficiently uptake endospermic ABA released by Cvi seed coats relative to wild-type embryos. We further investigated this possibility by monitoring the development of dissected atabcg30 and atabcg40 embryos in the absence or presence of exogenous ABA. Under control conditions, no difference in the development of embryos was observed between the wild-type and mutant embryos (Supplementary Fig. 5b); however, in the presence of ABA, the development of embryos dissected from atabcg30, atabcg40 and g30/g40 double-mutant seeds was inhibited to a lesser extent than those dissected from the wild type (Fig. 4dâf). These observations further indicate that atabcg30 and atabcg40 embryos do not take up the ABA released from the seed coat bedding of the Cvi ecotype as effectively as does the wild type.
Four AtABCGs mediate ABA transport in seeds
To directly test whether AtABCG31 and AtABCG25 release ABA from the tissues surrounding the embryo, we removed embryos from the seeds and compared the remaining tissues for secreted and retained endogenous ABA levels. As shown in Fig. 5a,b (with PAC) and Supplementary Fig. 6c,d (without PAC), the tissues surrounding the embryos of atabcg31 and atabcg25 single and double mutants secreted less and retained more ABA. On the basis of these results and previous studies that found that endospermic ABA inhibits germination25,29,30,31,32,33, we conclude that AtABCG31 and AtABCG25 contribute to the secretion of ABA from the endosperm. It is interesting that the total ABA content did not differ between the wild type and atabcg31 or atabcg25 (Supplementary Fig. 6e), further indicating the importance of endospermic ABA release to control embryonic growth.
Next, we incubated embryos dissected from the wild type, the atabcg30 and atabcg40 single mutants, and the corresponding double mutants with radiolabelled ABA and determined the amount of radioactivity taken up after 30âmin at room temperature. Both mutants took up significantly less ABA than the wild type, and the double-knockout mutants took up less than the single knockouts (Fig. 5c). In contrast, embryos incubated in the presence of radiolabelled ABA on ice did not differ in their ABA contents (Supplementary Fig. 6a). Therefore, the difference in ABA uptake shown in Fig. 5c was mediated by an active transport mechanism and not by simple diffusion. Taken together, these observations further strengthen and lend direct support to the notion that AtABCG30 and AtABCG40 function as endospermic ABA importers. While ABA transport activity has been reported for AtABCG25 and AtABCG40 in heterologous systems39,40, it has not been reported for AtABCG30 and AtABCG31. Therefore, to obtain additional evidence that these transporters also catalyse ABA fluxes, we expressed these AtABCGs in yeast and performed ABA transport experiments using the yeast system. Yeast expressing AtABCG31 took up lower amounts of ABA than yeast transformed with an empty vector, while yeast expressing AtABCG30, which was localized to the plasma membrane (Supplementary Fig. 6f), accumulated far more radiolabelled ABA (Fig. 5d). These results are in line with those obtained with embryos and seed coats from the corresponding mutants, indicating that AtABCG31 acts as an exporter, while AtABCG30 catalyses the import of ABA.
Discussion
In this study, we characterize two novel ABA transporter candidates, AtABCG31 and AtABCG30 (Fig. 5d), and provide various lines of evidence that the ABC transporters AtABCG31 and AtABCG25 mediate ABA secretion from the endosperm, and that AtABCG30 and AtABCG40 mediate ABA uptake into the embryo. Interestingly, although all these AtABCG transporters exhibit a common substrate transport activity, they are structurally different and not closely related in phylogenetic tree. AtABCG25 is a half-size ABC transporter, whereas AtABCG31, AtAtABCG30 and AtABCG40 are full-size ABC proteins52. Moreover, the three full-size transporters belong to different phylogenetic clusters52.
We found that AtABCG31 transcripts were present in the endosperm (Fig. 2a), which is consistent with a recent report53. Freshly collected intact atabcg31 seeds exhibited an enhanced germination phenotype in the presence (Supplementary Fig. 1c,d) or absence (Fig. 1a) of exogenous ABA. Thus, AtABCG31 appears to function as an efflux transporter that transports ABA from the endosperm to the embryo to prevent germination under unfavourable environmental conditions. AtABCG31 was previously reported to be expressed specifically in guard cells54 and the tapetum49. ABA plays an important role in both cell types, functioning in stomatal closure and pollen maturation, respectively; however, no direct link to ABA transport was demonstrated.
AtABCG25 was previously reported to function as an ABA exporter that was active mainly in ABA-synthesizing tissues39. Here we show that this transporter also exports ABA from the endosperm (Figs 2 and 5a). Unlike AtABCG31, AtABCG25 seemed to be expressed in the embryo too (Fig. 2b). However, the expression levels of AtABCG25 in seed tissues were much lower than those of AtABCG31, and the embryonic expression level of AtABCG25 was barely above the background level. Therefore, this result either indicates that AtABCG25 is not expressed in the embryo or that it is only expressed at low levels. It is possible that low levels of AtABCG25 contribute to the proper distribution of ABA within the embryonic tissues. However, the major site of AtABCG25 function in seeds is most likely the endosperm, where it is expressed at significant levels. The observation that the double knockouts of AtABCG25 and AtABCG31 exhibited a stronger phenotype than any of the single knockouts indicates that both transporters are needed to efficiently block germination when GA levels are low and that both transporters exhibit only a limited ABA efflux capacity. In addition, the double knockouts exhibited lower PAC-induced ABA accumulation than the wild type (Supplementary Fig. 6b), suggesting that a defect in ABA secretion may have inhibited ABA biosynthesis by a negative feedback mechanism.
It is interesting to note that AtABCG40, which is broadly expressed in plants, regulates many different physiological properties, including lateral root number, main root growth, seed germination and stomatal movement40. These properties can all be explained by the transport of a single substrate, ABA38. In contrast, AtABCG30 is specifically expressed in imbibed seeds, and thus its role seems to be restricted to the inhibition of germination. Our data indicate that both AtABCG40 and AtABCG30 contribute to ABA uptake into the embryo and are required to maintain ABA fluxes. This is supported by former data showing ABA transport activity in heterologous systems40 and our data demonstrating that AtABCG30-expressing yeast take up far more ABA than the corresponding empty vector control (Fig. 5d). When expressed in yeast, AtABCG40 and AtABCG30 exhibited similar ABA transporting activities (Supplementary Fig. 6f). Since these two genes exhibit different expression patterns in planta, the relative contribution of each transporter to overall ABA flux changes depending on the particular tissue and developmental stage. Thus, the transporters involved in ABA release and uptake are not completely redundant, and all four ABC transporters are required to efficiently suppress seed germination. Recently, ABA-importing transporter 1 (AIT1), which is expressed mainly in the vasculature and also in the embryo, was reported to act as an ABA importer for exogenously supplied ABA42, raising the question of whether this ABA importer also suppresses embryo growth.
The four atabcg mutants studied here are in the Col-0 background, which is among the weakest in dormancy among Arabidopsis thaliana ecotypes16,19,55. As a result, we evaluated the roles of the transporters in the absence of GA synthesis, which leads to endogenous ABA accumulation13. This is because when seeds are highly dormant, GA is unable to efficiently downregulate DELLA factor accumulation, and the endosperm releases ABA in a DELLA-dependent manner to block embryonic growth32. Therefore, it is likely that the AtABCG transporters identified in this study also serve as ABA efflux and influx transporters to block the germination of naturally dormant seeds. However, this conclusion will require further evaluation using atabcg31, atabcg25, atabcg30 and atabcg40 mutant alleles isolated in highly dormant ecotypes.
Most studies on ABA transport have sought to elucidate aspects of the plantâs response to environmental stress, such as guard cell closure or long-term adaptation to drought or salinity. Thus, previous discussions on ABA transporters have focused mainly on long-distance transport from the drought-stressed root to the shoot56. However, here we have identified four ABA transporters that act in concert to regulate the developmental process of seed germination in adjacent tissues within the seed. These transporters transfer ABA from the site of ABA synthesis (the endosperm) to its target (the embryo) to prevent germination (Fig. 5e). It is plausible that other stress- and development-related responses also require the coordination of short- and long-distance transporters of ABA, in a manner similar to the numerous auxin transporters that collaborate at short and long distances to control plant development.
Methods
Plant material and growth conditions
All Arabidopsis wild-type plants used in this study were of the Columbia-0 ecotype. Plants as well as seed coats and embryos were incubated on ½ MurashigeâSkoog (MS) agar (0.8%) media (22/18â°C; 16/8âh day/night; 40âμmolâmâ2âsâ1 light). Experiments were performed with wild-type and mutant plants grown in parallel and seeds were collected at the same time point. Seeds were sown on ½ MS agar (0.8%) media and cultured for 10 days (22/18â°C; 16/8âh day/night; 40âμmolâmâ2âsâ1 light). Seedlings were transferred to the soil and cultured in a greenhouse (22/18â°C; 16/8âh day/night). Fully developed and ripened brown siliques were collected.
Seed germination assay
For the germination assay, 12 seeds were placed on a 0.45-μm pore-size nylon membrane on ½ MS medium and incubated at a light intensity of 40âμmolâmâ2âsâ1 (16/8âh day/night) at 22/18â°C. For the quantification, the seeds with (cold) or without (no cold) pretreatment in the dark at 4â°C for 2 days were sowed on ½ MS medium in the absence or presence of 0.5â1.0âμM ABA. The seeds were incubated under 16-h light and 8-h dark conditions at 22â°C.
Quantitative real time RTâPCR
Total RNA was extracted from dissected embryos or endosperms, or whole seeds of the Col-0 wild type using the RNeasy Plant Mini Kit (Qiagen). One microgram of total RNA was treated with DNAseI (Roche) to remove any residual genomic DNA. cDNA was synthesized with GoScript reverse transcriptase (Promega). qRTâPCR was performed with the SYBR Kit (TAKARA) to detect AtABCG25, AtABCG30, AtABCG31, AtABCG40, AtABI4, AtEPA1 and AtUbiqutin11 transcript levels. Amplified samples were normalized against AtUbiqutin11 levels. Primer sequences were as follows: for AtABCG25 (At1g71960), forward (5â²-GAGACGCCATGGCTTACTTTGA-3â²) and reverse (5â²-AATACATGTTGTTATTCCACCGCC-3â²); for AtABCG30 (At4g15230), forward (5â²-CATGGCAGTGACTCTACGCTCATCT-3â²) and reverse (5â²-TTGCCACGATAATGGGATAAGCG-3â²); for AtABCG31 (At2g29940), forward (5â²-TCCAAAAGCCTTTGATTCCAGT-3â²) and reverse (5â²-TCCCACAGGACTTTTTTTATCTTCTC-3â²); for AtABCG40 (At1g15520), forward (5â²-CTGAGCGAAAGCATAAGGCATG-3â²) and reverse (5â²-TGGAGAAATCCTCCTTACACAGCC-3â²); for AtABI4 (At2g40220), forward (5â²-ATGGACCCTTTAGCTTCCCA-3â²) and reverse (5â²-CCACTTGCGAGTGCGCTTAC-3â²); for AtEPR1 (At2g27380), forward (5â²-CCACCTGTCAAGCCACCGCCAAT-3â²) and reverse (5â²-GTAAGTTGGTGTCGGAGGCTTATGCA-3â²)46; for AtUbiqutin11 (At4g05050), forward (5â²-GAACCAAGTTCATGTATCGT-3â²) and reverse (5â²-ACACTCATCAAACTAAGC AC-3â²) (ref. 57).
Promoter constructs
The AtABCG30 promoter::uidA reporter gene construct was generated by PCR amplification of a 2.0-kb promoter region of AtABCG30 from genomic DNA using primers containing ClaI and XbaI restriction sites (5â²-ATCGATTCCACTGGCCAAATTTCTCTATACA-3â² and 5â²-TCTAGATTTCAGTTAAAATGCCAACGGAA-3â²) and ligation of the PCR product into pBI101.2 (Clontech). The AtABCG31 promoter::uidA reporter gene construct was generated by PCR amplification of a 2.5-kb promoter region of AtABCG31 from genomic DNA using primers (5â²-CTCATGTGGTAGACTGCTAA-3â² and 5â²-CTCCATACCAACTCTACGAA-3â²) and recombination of the PCR product into the pMDC163 using the Gateway System. All constructs were verified by sequencing.
Intracellular localization of AtABCG31 and AtABCG30
The AtABCG30 construct was generated by PCR amplification of a 6.2-kb genomic DNA region of AtABCG30 from Arabidopsis genomic DNA using primers containing KpnI restriction sites (5â²-GGGGTACCATGATCCAAACAGGTGAAGAAGATG-3â² and 5â²-GGTACCTTTCTTTTGGAAACTGAGTT-3â²). The resulting construct was ligated into the pART7 vector58. In the case of AtABCG31, a previously reported 35Spro::sGFP::AtABCG31 construct48 was used. Each construct was introduced into Arabidopsis mesophyll protoplasts using the polyethylene glycol method59,60 along with a plasma membrane marker PM-rK (CD3â1007, ABRE). Protoplasts were isolated from 6-week-old wild-type Arabidopsis leaves by enzyme treatments. For each transfection, 200âμl cells (2 à 105 protoplasts per ml) were incubated with 20âμg total DNA and 220âμl 40% (w/v) PEG4000 solution. Seventy hours after incubation under dim-light conditions, green (GFP) and red (mCherry) fluorescence signals were observed and captured using a confocal microscope (Leica TCS SP5).
Seed coat bedding assay
Detailed protocols on how to assemble SCBA can be found in Lee et al.33,34. For the SCBA, which tests the repressive activity of the endosperm, embryos were dissected from aba2-1 mutant seeds 4âh after imbibition33. Twenty dissected embryos were placed on a layer of 80 seed coat beddings dissected from freshly collected mutant seeds on the surface of 10âμM PAC containing ½ MS agar medium and incubated under 40âμmolâmâ2âsâ1 light (continuous light) at 22â°C.
To evaluate the embryoâs capacity to uptake germination inhibitor from the endosperm, 20 dissected embryos were placed on a layer of 80 seed coat bedding dissected from dormant Cvi seeds placed on ½ MS agar medium and incubated at a light intensity of 40âμmolâmâ2âsâ1 (continuous light) at 22â°C.
For the embryonic growth assay of dissected embryos, 12 embryos dissected from wild-type (Col-0), atabcg mutant seeds or whole seeds were placed on a 0.45-μm pore-size nylon membrane on ½ MS medium in the absence or presence of 0.1âμM ABA. In this experiment, we used a lower ABA concentration, since exogenously added ABA inhibits the growth of dissected, coatless embryos much more effectively than it inhibits the growth of embryos in whole seeds.
Assay of secreted and remaining ABA levels
Endosperms and testa were dissected from 500 seeds of the wild type and atabcg25 and atabcg31 single-knockout and atabcg25 atabcg31 double-knockout mutants imbibed for 24âh on ½ MS agar media complemented with 10âμM PAC or for 1âh on the same medium without PAC. To collect excreted ABA, the tissues were subsequently incubated in 100âμl ½ MS liquid medium without PAC for 24âh. After incubation, the remaining tissues were washed with ice-cold ½ MS medium. ABA was extracted with 80% methanol solution, vacuum dried and the pellet was resuspended with Tris-buffered saline buffer. ABA levels were determined using the ELISA Kit (Agdia)61.
3H-ABA accumulation assay
Fifty dissected embryos were incubated in 100âμl of a bathing solution (pH 6.5) containing 12.5ânM 3H-ABA (1.63âTbqâmmolâ1) and 17.5âpM 14C-glycerol (5.40âGBqâmmolâ1) for 30âmin at room temperature. After incubation, embryos were washed three times with 250âμl ice-cold bathing solution, and their radioactivity was determined by liquid scintillation counting.
Heterologous expression of AtABCGs and ABA loading assays
AtABCG31 cDNA was cloned into the NotI and XbaI sites of pYES2NT/C (Invitrogen). AtABCG30 cDNA was cloned into the KpnI and XbaI sites of pYES2NT/C. All constructs were verified by sequencing. AtABCGs expressing yeast lines were generated using the electroporation method and selected on uracil-minus minimal media. 3H-ABA-loading tests were performed as described previously with minor modifications40. Cells were cultured in minimum saltâgalactose medium in the absence of uracil (SG-URA), supplemented with 1.0 % raffinose, at pH 6.5, and collected by centrifugation at the mid-log phase (up to OD600=1.0). Yeast cells were washed two times using SG-URA medium and subsequently resuspended in reaction buffer (1âmM KCl, 1âmM MES (2-(N-morpholino) ethanesulfonic acid), 20âmM glucose, at pH 6.5) at an OD600=10. 3H-ABA (50ânM, 7.4âkBq, 1.63âTBqâmmolâ1) was then added to the cell suspension and gently mixed. At the times indicated, 100âμl of cell suspension was filtered through nitrocellulose membranes, and the cells remaining on the filter were washed two times with 2âml of ice-cold water. The radioactivity on the filter was determined using a liquid scintillation counter.
Additional information
How to cite this article: Kang, J. et al. Abscisic acid transporters cooperate to control seed germination. Nat. Commun. 6:8113 doi: 10.1038/ncomms9113 (2015).
References
Hilhorst, H. W. M. A critical update on seed dormancy. I. Primary dormancy. Seed Sci. Res. 5, 61â73 (1995).
Bewley, J. D. Seed germination and dormancy. Plant Cell 9, 1055â1066 (1997).
Kucera, B., Cohn, M. A. & Leubner-Metzger, G. Plant hormone interactions during seed dormancy release and germination. Seed Sci. Res. 15, 281â307 (2005).
Holdsworth, H. J., Bentsink, J. & Soppe, W. J. J. Molecular networks regulating Arabidopsis seed maturation, after-ripening, dormancy and germination. New Phytol. 179, 33â54 (2008).
Seo, M. et al. Interaction of light and hormone signals in germinating seeds. Plant Mol. Biol. 69, 463â472 (2009).
Graeber, K. et al. Molecular mechanisms of seed dormancy. Plant Cell Environ. 35, 1769â1786 (2012).
Nonogaki, H. Seed dormancy and germination-emerging mechanisms and new hypotheses. Front. Plant Sci. 5, 233 (2014).
Hilhorst, H. W. M. & Karssen, C. M. Seed dormancy and germination: the role of abscisic acid and gibberellins and the importance of hormone mutants. Plant Growth Regul. 11, 225â238 (1992).
Debeaujon, I. & Koornneef, M. Gibberellin requirement for Arabidopsis thaliana seed germination is determined both by testa characteristics and embryonic ABA. Plant Physiol. 122, 415â424 (2000).
Toh, S. et al. High temperature-induced abscisic acid biosynthesis and its role in the inhibition of gibberellin action in Arabidopsis seeds. Plant Physiol. 146, 1368â1385 (2008).
Verslues, P. E. & Bray, E. A. Role of abscisic acid (ABA) and Arabidopsis thaliana ABA-insensitive loci in low water potential-induced ABA and proline accumulation. J. Exp. Bot. 57, 201â212 (2006).
Piskurewicz, U., Turecková, V., Lacombe, E. & Lopez-Molina, L. Far-red light inhibits germination through DELLA-dependent stimulation of ABA synthesis and ABI3 activity. EMBO J. 28, 2259â2271 (2009).
Piskurewicz, U. et al. The gibberellic acid signaling repressor RGL2 inhibits Arabidopsis seed germination by stimulating abscisic acid synthesis and ABI5 activity. Plant Cell 20, 2729â2745 (2008).
Finch-Savage, W. E. & Leubner-Metzger, G. Seed dormancy and the control of germination. New Phytol. 171, 501â523 (2006).
Lopez-Molina, L., Mongrand, S., MeLachlin, D. T., Chait, B. T. & Chua, N. ABI5 acts downstream of ABI3 to execute an ABA-dependent growth arrest during germination. Plant J. 32, 317â328 (2002).
Ali-Rachedi, S. et al. Changes in endogenous abscisic acid levels during dormancy release and maintenance of mature seeds: studies with the Cape Verde Islands ecotype, the dormant model of Arabidopsis thaliana. Planta 219, 479â488 (2004).
Robertson, D. S. The genetics of vivipary in maize. Genetics 40, 745â760 (1955).
Van der Schaar, W. et al. QTL analysis of seed dormancy in Arabidopsis using recombinant inbred lines and MQM mapping. Heredity 79, 190â200 (1997).
Koornneef, M., Bentsink, L. & Hilhorst, H. W. M. Seed dormancy and germination. Curr. Opin. Plant Biol. 5, 33â36 (2005).
Preston, J. et al. Temporal expression patterns of hormone metabolism genes during imbibition of Arabidopsis thaliana seeds: a comparative study on dormant and non-dormant accessions. Plant Cell Physiol. 50, 1786â1800 (2009).
Alonso-Blanco, C. et al. Analysis of natural allelic variation at seed dormancy loci of Arabidopsis thaliana. Genetics 164, 711â729 (2003).
Bentsink, L. et al. Natural variation for seed dormancy in Arabidopsis is regulated by additive genetic and molecular pathways. Proc. Natl Acad. Sci. USA 107, 4264â4269 (2010).
Teng, S., Rognoni, S., Bentsink, L. & Smeekens, S. The Arabidopsis GSQ5/DOG1 Cvi allele is induced by the ABA-mediated sugar signaling pathway, and enhances sugar sensitivity by stimulating ABI4 expression. Plant J. 55, 372â381 (2008).
Muller, K., Tintelnot, S. & Leubner-Metzger, G. Endosperm-limited Brassicaceae seed germination: abscisic acid inhibits embryo-induced endosperm weakening of Lepidiumsativum (cress) and endosperm rupture of Cress and Arabidopsis thaliana. Plant Cell Physiol. 47, 864â877 (2006).
Bethke, P. C. et al. The Arabidopsis aleurone layer responds to nitric oxide, gibberellin, and abscisic acid and is sufficient and necessary for seed dormancy. Plant Physiol. 143, 1173â1188 (2007).
Kelly, K. M. et al. Seed coat structure and dormancy. Plant Growth Regul. 11, 201â209 (1992).
Karssen, C. M. et al. Induction of dormancy during seed development by endogenous abscisic acid: studies on abscisic acid deficient genotypes of Arabidopssi thaliana (L.) Heynh. Planta 157, 158â165 (1983).
Koornneef, M. et al. In vivo inhibition of seed development and reserve protein accumulation in recombinants of abscisicacid biosynthesis and responsiveness mutants in Arabidopsis thaliana. Plant Physiol. 90, 463â469 (1989).
Barrero, J. M. et al. Gene expression profiling identifies two regulatory genes controlling dormancy and ABA sensitivity in Arabidopsis seeds. Plant J. 61, 611â622 (2010).
Bassel, G. W., Mullen, R. T. & Bewley, J. D. ABI3 expression ceases following, but not during, germination of tomato and Arabidopsis seeds. J. Exp. Bot. 57, 1291â1297 (2006).
Penfield, S., Li, Y., Gilday, A. D., Graham, S. & Graham, I. A. Arabidopsis ABA insensitive 4 regulates lipid mobilization in the embryo and reveals repression of seed germination by the endosperm. Plant Cell 18, 1887â1899 (2006).
Lee, K. P., Piskurewicz, U., Tureckova, V., Strnad, N. & Lopez-Molina, L. A seed coat bedding assay shows that RGL2-dependent release of abscisic acid by the endosperm controls embryo growth in Arabidopsis dormant seeds. Proc. Natl Acad. Sci. USA 107, 19108â19113 (2010).
Lee, K. P. et al. Spatially and genetically distinct control of seed germination by phytochromes A and B. Genes Dev. 26, 1984â1996 (2012).
Lee, K. P. & Lopez-Molina, L. A seed coat bedding assay to genetically explore in vitro how the endosperm controls seed germination in Arabidopsis thaliana. J. Vis. Exp. 81, e50732 (2013).
Salisbury, F. B. Plant Physiology 3rd edn. 312 (1985).
Marchant, A. et al. AUX1 regulates root gravitropism in Arabidopsis by facilitating auxin uptake within root apical tissues. EMBO J. 18, 2066â2073 (1999).
Vieten, A., Sauer, M., Brewer, P. B. & Friml, J. Molecular and cellular aspects of auxin-transport-mediated development. Trends Plant Sci. 12, 160â168 (2007).
Swarup, K. et al. The auxin influx carrier LAX3 promotes lateral root emergence. Nat. Cell Biol. 10, 946â954 (2008).
Kuromori, T. et al. ABC transporter AtABCG25 is involved in abscisic acid transport and responses. Proc. Natl Acad. Sci. USA 107, 2361â2366 (2010).
Kang, J. et al. PDR-type ABC transporter mediates cellular uptake of the phytohormone abscisic acid. Proc. Natl Acad. Sci. USA 107, 2355â2360 (2010).
Kuromori, T. et al. Arabidopsis mutants of AtABCG22, an ABC transporter gene, increase water transpiration and drought susceptibility. Plant J. 67, 885â894 (2011).
Kanno, Y. et al. Identification of an abscisic acid transporter by functional screening using the receptor complex as a sensor. Proc. Natl Acad. Sci. USA 109, 9653â9658 (2012).
Leran, S. et al. A unified nomenclature of NITRATE TRANSPORTER 1/PEPTIDE TRANSPORTER family members in plants. Trends Plant Sci. 19, 5â9 (2014).
Zhang, H. et al. A DTX/MATE-type transporter facilitates abscisic acid efflux and modulates ABA sensitivity and drought tolerance in Arabidopsis. Mol. Plant 7, 1522â1532 (2014).
Jasinski, M. et al. A plant plasma membrane ATP binding cassette-type transporter is involved in antifungal terpinoid secretion. Plant Cell 13, 1095â1107 (2001).
Dubreucq, B. et al. The Arabidpsis AtEPR1 extnesin-like gene is specifically expressed in endosperm during seed germination. Plant J. 23, 643â652 (2000).
Lee, M. et al. AtPDR12 contributes to lead resistance in Arabidopsis. Plant Physiol. 138, 827â836 (2005).
Nelson, B. K., Cai, X. & Nebenfuhr, A. A multicolored set of in vivo organelle markers for co-localization studies in Arabidopsis and other plants. Plant J. 51, 1126â1136 (2007).
Choi, H. et al. The role of Arabidopsis ABCG9 and ABCG31 ATP binding cassette transporters in pollen fitness and the deposition of steryl glycosides on the pollen coat. Plant Cell 26, 310â324 (2014).
Lee, K. P. & Lopez-Molina, L. Control of seed germination in the shade. Cell Cycle. 11, 4489â4490 (2012).
Leon-Kloosteriel, K. M. et al. Isolation and characterization of abscisic acid-deficient Arabidopsis mutants at two new loci. Plant J. 10, 655â667 (1996).
Andolfo, G. et al. Genetic variability and evolutionary diversification of membrane ABC transporters in plants. BMC Plant Biol. 15, 51 (2015).
Dekkers, B. J. W. et al. Transcription dynamics of two seed compartments with opposing roles in Arabidopsis seed germination. Plant Physiol. 163, 205â215 (2013).
Galbiati, M. et al. Gene trap lines identify Arabidopsis genes expressed in stomatal guard cells. Plant J. 53, 750â762 (2008).
Koornneef, M. et al. in Dormancy in Plants eds Viémont J.-D, Crabbé J. Ch. 5, 365â373Oxon:CABI publishing (2000).
Jiang, F. & Hartung, W. Long-distance signaling of abscisic acid (ABA): the factors regulating the intensity of the ABA signal. J. Exp. Bot. 59, 37â43 (2007).
Gutierrez, L. et al. The lack of a systematic validation of reference genes; a serious pitfall undervalued in reverse transcription polymerase chain reaction (RT-PCR) analysis in plants. Plant J. 6, 609â618 (2008).
Gleave, A. P. A versatile binary vector system with a T-DNA organizational structure conducive to efficient integration of cloned DNA into the plant genome. Plant Mol. Biol. 20, 1203â1207 (1992).
Abel, S. & Theologi, A. Transient gene expression in protoplasts of Arabidopsis thaliana. Methods Mol Biol. 82, 209â217 (1998).
Jin, J. et al. A new dynamin-like protein, ADL6, is involved in trafficking from the trans-Golgi network to the central vacuole in Arabidopsis. Plant Cell 13, 1511â1525 (2001).
Jiang, T. et al. Root morphology and gene expression analysis in response to drought stress in maize (Zea mays). Plant Mol. Biol. Rep. 30, 360â369 (2012).
Acknowledgements
This work was supported by the National Research Foundation (NRF) grants funded by the Ministry of Science, ICT (Information and Communication Technology) and Future Planning of Korea (K20607000006, 2015R1A2A1A01004294, and BK21 Plus;10Z20130012243). The work in the laboratory of E.M. was partially supported by the Swiss National Science Foundation and the Forschungskredit from Universität zürich. L.L.-M. is supported by grants from the Swiss National Science Foundation and by the State of Geneva.
Author information
Authors and Affiliations
Contributions
J.K., L.L.-M. and Y.L. designed the experiments; J.K. and S.Y. performed most of the experiments; H.C. and A.K. generated some of the constructs used and transgenic plants; K.P.L., E.M. and L.L.-M. provided technical support and conceptual advice; J.K., E.M., L.L.-M. and Y.L. wrote the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Supplementary information
Supplementary Information
Supplementary Figures 1-7 and Supplementary References (PDF 1275 kb)
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articleâs Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Kang, J., Yim, S., Choi, H. et al. Abscisic acid transporters cooperate to control seed germination. Nat Commun 6, 8113 (2015). https://doi.org/10.1038/ncomms9113
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/ncomms9113
This article is cited by
-
Comparative Transcriptome Analysis of Hypocotyls During the Developmental Transition of C3 Cotyledons to C4 Leaves in Halimocnemis mollissima Bunge
Journal of Plant Growth Regulation (2024)
-
Genome-wide association study reveals the genetic basis of cold tolerance in soybean
Euphytica (2024)
-
Structural basis for abscisic acid efflux mediated by ABCG25 in Arabidopsis thaliana
Nature Plants (2023)
-
A VEL3 histone deacetylase complex establishes a maternal epigenetic state controlling progeny seed dormancy
Nature Communications (2023)
-
Identification and validation of seed dormancy loci and candidate genes and construction of regulatory networks by WGCNA in maize introgression lines
Theoretical and Applied Genetics (2023)