Protein reserves in the cereal endosperm are sequentially degraded to small peptides and amino acids during germination and these are translocated across the scutellum to support growth of the embryo. Peptide transport in the germinating barley grain is mediated by specific carriers localized to the plasma membrane of the scutellar epithelium. In isolated barley embryos peptide transport is rapidly inhibited by amino acid concentrations comparable with those found in the post-germination barley grain. However, this inhibition of HvPTR1 activity is not effected at either the transcriptional or translational level. The protein phosphatase inhibitor okadaic acid repressed transport of Ala-[14C]Phe, but not [14C]Ala, into the barley scutellar epithelium. In vivo [32P]orthophosphate labelling studies of barley scutellar tissue in combination with immunoprecipitation studies using antiserum raised to HvPTR1 showed that HvPTR1 (66 kDa) is phosphorylated in the presence of amino acids. Immunopurified HvPTR1 was further demonstrated to be phosphorylated on serine residues. Digestion with the N-glycosidase enzyme PNGase F results in a shift in the molecular mass of the protein by 10 kDa, indicating that HvPTR1 is an N-linked glycoprotein. These results provide strong circumstantial evidence that HvPTR1 peptide transport activity in the germinating barley grain is regulated at the post-translational level by phosphorylation in response to rising levels of amino acids emanating from the endosperm as a result of storage protein breakdown and mobilization. This is potentially an important element in balancing the flux of organic nitrogen and carbon from the endosperm to embryo during germination and seedling establishment.
This paper is available online free of all access charges (see http://jxb.oupjournals.org/open_access.html for further details)
The online version of this article has been published under an Open Access model. Users are entitled to use, reproduce, disseminate, or display the Open Access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and the Society for Experimental Biology are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact email@example.com
Peptide transport is important to the nutrition of bacteria, yeasts, and animals but to date has not been widely studied in plant systems (Saier, 2000; Stacey et al., 2002). The active transport of small peptides during nutrient reserve mobilization in barley grain germination is the best characterized function of peptide transport in plants at both the biochemical and molecular levels (Higgins and Payne, 1980; Waterworth et al., 2001). During cereal grain germination, endosperm protein reserves are hydrolysed by the concerted action of proteinases and peptidases to form a pool of small peptides and amino acids that are translocated into the embryo across the scutellum, a specialized absorptive tissue which separates the endosperm from the embryo structures. Peptide transport plays a vital role in cereal grain germination because peptides, rather than amino acids, form the initial products of endosperm protein breakdown, to provide nutrients for initiation of growth processes (Higgins and Payne, 1978a). In germinating cereal grains, peptide transport is solely responsible for the supply of organic nitrogen to the embryo during germination, whereas amino acid transport is essentially a post-germinative event which only becomes significant 48–72 h after the commencement of imbibition. The active transport of peptides is distinct and independent of amino acid transport and displays high stereoselectivity for α-peptide bonds and L-amino acid residues but low specificity with respect to amino acid side-chains (Waterworth et al., 2001). Peptide transport is H+-coupled (Higgins and Payne, 1977), specific for di- and tripeptides (and possibly tetra- and pentapeptides) (Higgins and Payne, 1978a, b, 1980; Sopanen et al., 1977), and stereospecific (Higgins and Payne, 1978c). Peptide transport shows broad specificity and is strongly pH-dependent with a markedly low pH optimum of pH 3.8–4.0 similar to that of the germinating barley endosperm (Higgins and Payne, 1978c). The barley scutellar peptide transporter (HvPTR1) has been cloned, characterized by functional expression in Xenopus laevis oocytes, and identified as a member of the PTR family of small peptide transporters (West et al., 1998). HvPTR1 protein is localized to the plasma membrane of scutellar epithelial cells (Waterworth et al., 2000a) and HvPTR1 expression is seed specific with transcripts detected in scutellar epithelial cells during germination and in barley grain from the earliest stages of grain development onwards (Waterworth et al., 2003). HvPTR1 is one of the earliest known genes expressed during barley grain germination, challenging the hypothesis that nutrient reserve mobilization is a post-germinative event (West et al., 1998).
Knowledge of the mechanisms which regulate and co-ordinate the expression and activity of plant solute transporters is now emerging. Regulation of transporter protein activity may occur at several levels, including transcriptional control of gene expression, post-transcriptional regulation of mRNA stability (turnover), mRNA translation, or post-translational control effected by a mechanism such as protein phosphorylation. Modulations in activity of the H+-ATPase, and hence the proton motive force across the plasma membrane, could also regulate activity of H+- coupled-symporters (Delrot et al., 2000). Surprisingly, the post-translational control of plant solute transporter activities by phosphorylation has been the subject of few published reports, despite being a universal and widespread mechanism through which the regulation of intracellular events can be effected. One exception is the control of sucrose transport in sugar beet leaves by phosphorylation (Roblin et al., 1998). It is reported here that amino acids, the products of hydrolysis of endosperm protein reserves, inhibit peptide transport in the germinating barley grain. Compelling evidence is also provided that this regulation is mediated in vivo at the post-translational level by phosphorylation of the plasma membrane-localized barley scutellar peptide transporter HvPTR1.
Materials and methods
Plant material and growth conditions
Barley grain (H. vulgare L. cv. Maris Otter 1993 harvest) was germinated at 23 °C in the dark on moist filter paper. In studies using isolated embryos, the embryos plus scutellum were excised from barley grain imbibed for the appropriate period of time and transferred to agar (1.2% w/v) containing uptake buffer (50 mM sodium phosphate–citrate buffer pH 3.8) and further additions as indicated.
Alanyl-[U14C]phenylalanine (Ala-[14C]Phe) was synthesized as described by West et al. (1998).
Solute transport by barley scutella
Transport of Ala-[U14C]Phe was monitored by uptake of radiolabel into barley scutellar tissue as described previously (West et al., 1998).
SDS-PAGE and western blotting
Membrane proteins were extracted from barley scutellar tissue as described in Waterworth et al. (2000a). Protein concentrations were determined by the Bio-Rad protein assay (Bio-Rad Laboratories, Hemel Hempstead, UK) using bovine serum albumin (BSA) as a standard. Protein samples were separated by SDS–PAGE and transferred to PVDF membrane (Bio-Rad) for 1 h at 100 V. The blots were sequentially probed with a 1/5000 dilution of peptide antiserum raised to HvPTR1 (Waterworth et al., 2000a) and a 1/30 000 dilution of anti-sheep coupled alkaline phosphatase (Sigma) before development with BCIP/NBT reagent (Sigma,). Phosphoserine, phosphothreonine, and phosphotyrosine antibodies were purchased from Sigma and used at 1/20 000, 1/10 000, and 1/4000 dilutions, respectively, followed by a 1/10 000 dilution of anti-mouse coupled alkaline phosphatase. SDS–PAGE gels were either Coomassie blue-stained or dried after treatment in 20% TCA overnight and radiolabelled proteins analysed by autoradiography or using a PhosphorImager (Fuji Photo Film Co., Tokyo, Japan).
In vivo [32P]orthophosphate labelling of barley scutellar tissue
Twenty epithelial scrapes were isolated from barley embryos imbibed for the appropriate length of time and incubated in 1.5 ml Eppendorf caps containing 150 μl 50 mM sodium phosphate–citrate buffer, pH 6.8 with 100 μCi [32P]orthophosphate for 3.5 h. Scutella were then washed in 3×20 ml distilled water to remove unincorporated label and membrane proteins were extracted from barley scutellar tissue as described by Waterworth et al. (2000a). Membrane proteins were resuspended in 1 ml TTBS containing 1 μl HvPTR1 antiserum and 0.5% (w/v) Triton X-100, incubated at 4 °C for 1 h on a rotator, and immunocomplexes recovered by centrifugation at 16 000 g for 15 min. The pellets were washed by resuspension in 500 μl TTBS and centrifuged at 16 000 g to recover the pellets. In larger scale preparations (0.5–1.0 g scutellar tissue) of membrane proteins a TL100 centrifuge (Beckman) was used to pellet proteins and immunoprecipitations scaled up as appropriate.
Purification of HvPTR1
Resuspended scutellar membrane proteins prepared as above were added to anti-HvPTR1 coupled to CNBR-Agarose beads (Autogen Bioclear, Calne, UK) and left on a rotator for 1 h at 4 °C. Immunoprecipitated protein was recovered by centrifugation at 16 000 g for 15 min and eluted from beads using 50 mM glycine (pH 2.8). Membrane fractions were concentrated using Amicon microconcentrators and preparations pooled for subsequent analysis.
PNGase F digestion of HvPTR1
HvPTR1 was purified by immunoprecipitation as described previously and separated from the antiserum by washing in 20 mM glycine pH 2.8. HvPTR1 protein was resuspended in 20 mM TRIS pH 7.5, 50 mM NaCl, and 1 mM EDTA. After the addition of a denaturing solution (2% w/v SDS, 1 M β-mercaptoethanol), the protein was heated at 100 °C for 5 min, cooled on ice, and 2.5 μl NP-40 and peptide N-glycosidase F (5 U) (PNGase F) were added. Protein was digested for 2 h at 37 °C and the reaction terminated by addition of Laemmli buffer (Sambrook et al., 1989).
Amino acids inhibit peptide transport across the barley scutellum
The effects of endosperm nutrient reserve hydrolysis products on peptide uptake in the barley scutellum were investigated. The germinating barley grain is an excellent model system with which to analyse the mechanisms which regulate nutrient transport because the barley embryo with attached scutellar tissue can easily be removed from the softened endosperm after several hours imbibition. Peptide transport activities comparable to those in the intact grain are retained for several hours after isolation onto agar (Sopanen, 1979; West et al., 1998). Barley embryos with scutellum attached were isolated from 20 h imbibed barley grains and placed scutellum surface down onto agar containing additives at concentrations representative of those present in the germinating barley grain for a further 4 h prior to assaying for peptide transport activity (Table 1).
Supplement to agar
Peptide transport activity (% of control values)
|5 mM glucose||165±23|
|5 mM amino acids||42±13|
|5 mM sorbitol||90±13|
Supplement to agar
Peptide transport activity (% of control values)
|5 mM glucose||165±23|
|5 mM amino acids||42±13|
|5 mM sorbitol||90±13|
Scutella were isolated from barley grain imbibed at 23 °C and incubated for 4 h on 1.2% (w/v) agar buffered with 50 mM sodium phosphate–citrate pH 3.8 in the presence of either no additives, amino acids (5 mM), glucose (5 mM, or sorbitol (5 mM). Transport of Ala-[14C]Phe was then determined. The 100% value for peptide uptake is 124 nmol g−1 FW min−1. Values are the mean ±standard error of three independent experiments.
The presence of a comprehensive equimolar mixture of amino acids (5 mM total) in the agar was found to inhibit peptide transport across the scutellum by 70% compared with peptide transport rates in barley embryos placed on agar containing no additives. Glucose, the major hydrolytic product of starch degradation in the endosperm and the form in which most carbohydrate reserves are transported across the scutellum (Bewley and Black, 1994), was shown to have a stimulatory effect on peptide transport with a 65% increase in transport activity after a 4 h incubation period relative to controls (Table 1). The possibility that osmotic effects accounted for these differences was eliminated by the use of sorbitol (5 mM) in place of amino acids or glucose in the agar. The presence of sorbitol had no significant effects on peptide transport across the barley scutellum. Competition for the peptide transport system by free amino acids could not account for the observed effects since amino acids are not substrates for peptide transporters in the germinating barley grain (Higgins and Payne, 1978a), and were not present in the uptake assay medium. Inhibition of peptide transport by amino acids displayed a concentration-dependent relationship between 0.1 mM and 5 mM total concentration of amino acids, indicating that the biologically effective regulation lies between 1–5 mM (Fig. 1A). This is comparable with the concentration of free amino acids (2–4 mM) determined to be present in the endosperm of the germinating barley grain after 3 d germination when amino acid transport begins to account for a substantial flux of nitrogen into the barley embryo (Higgins and Payne, 1981). A gradual decrease in peptide transport capacity is observed when isolated barley embryos are placed on agar for extended periods of time (24 h), but this is not significant over short periods of time (West et al., 1998). However, substantial inhibition (50%) of peptide transport by amino acids (1 mM) was observed within 4 h, indicating that the effects of amino acids on peptide transport are mediated relatively rapidly (Fig. 1B).
Are all amino acids effective in the down-regulation of peptide transport?
The effects of individual amino acids (1 mM final concentration) on peptide transport in the barley scutellum were examined (Fig. 2). Differential inhibition of peptide transport by individual amino acids was observed, indicating that only certain amino acids are effective in the down-regulation of peptide transport. Most marked repression of peptide transport was observed with tyrosine, asparagine, and isoleucine (85% inhibition relative to no amino acid control), whilst several amino acids including histidine, glutamate, lysine, alanine, aspartate, leucine, valine, tryptophan, and serine inhibited peptide transport by around 70%. The levels of these particular amino acids may represent the metabolite signal which is sensed and translated into regulation of peptide transport by an unknown signal transduction pathway. There is no obvious correlation between amino acids which inhibited peptide transport activity and their abundance in barley endosperm storage proteins or with the size or properties of their side-chain.
The mechanism by which down-regulation of peptide transport by amino acids occurs was investigated. Western analysis using peptide antiserum raised to the C-terminal and internal regions of HvPTR1 (Waterworth et al., 2000a) demonstrated that HvPTR1 protein levels did not change in amino acid-treated embryos compared with control embryos incubated in the absence of amino acids (Fig. 3A). Northern analysis also showed that mRNA transcript levels of the barley scutellar peptide transporter HvPTR1 also remained the same in amino acid-treated embryos (data not presented). These observations suggest that control of peptide transport activity in the scutellum of germinating barley is mediated not via transcriptional or translational control of HvPTR1, but at the post-translational level, and this was investigated further.
Okadaic acid inhibits peptide transport
Okadaic acid, an inhibitor of protein phosphatases 1 and 2a, was shown to inhibit peptide transport by over 90% in barley embryos after a 1 h preincubation (Fig. 3B).
This result supports a role for phosphorylation in the regulation of peptide transport activity across the barley scutellum. However, it is plausible that the observed effects of okadaic acid on peptide transport are attributable to the secondary effects of this phosphatase inhibitor on cell function., such as disruption of the proton gradient across the plasma membrane which energizes peptide transport. The plant plasma membrane H+-ATPase is known to be regulated by phosphorylation (Vera-Estrella et al., 1994). Amino acid transporters in the barley scutellum are H+-coupled symporters (Hardy and Payne, 1992), and therefore dependent on the maintenance of the proton gradient across the plasma membrane. However, okadaic acid did not inhibit amino acid transport in the barley scutellum over a 4 h incubation period (data not presented), providing supporting evidence that the inhibition of peptide transport across the barley scutellum by okadaic acid is a specific effect of okadaic acid on peptide transport per se. Immunoblot analysis of membrane proteins isolated from okadaic acid-treated and untreated (control) barley scutellum with HvPTR1 antiserum showed no differences in peptide transporter protein levels in scutellar tissue (Fig. 3C), supporting the possibility that inhibition of peptide transport activity is indeed mediated by the phosphorylation state of HvPTR1.
HvPTR1 is regulated by phosphorylation
The barley scutellar peptide transporter protein has been extensively studied by approaches based on peptide protectable thiol-affinity labelling (Payne and Walker-Smith, 1987; Hardy and Payne, 1991; Waterworth et al., 2000a). Xenopus oocytes expressing HvPTR1 cRNA, but not water-injected controls, express a 66 kDa protein which can be radiolabelled with [14C]NEM (Waterworth et al., 2001). A band of identical mass cross-reacts with peptide antiserum raised to C-terminal and internal epitopes of HvPTR1 (Waterworth et al., 2000a). Barley scutellar epithelial scrapes were prepared from 48 h-imbibed barley grain and incubated with [32P]orthophosphoric acid for a further 3 h in either the presence or absence of 5 mM total amino acids. Membrane proteins were isolated from epithelial tissue and phosphorylated proteins analysed by SDS–PAGE and autoradiography. Western blotting showed no change in HvPTR1 protein levels in the presence of amino acids (Fig. 4A). A protein identical in mass to HvPTR1 was heavily phosphorylated in the presence of amino acids (Fig. 4B). Confirmation that the phosphorylated protein was indeed HvPTR1 was obtained by immunoprecipitation of in vivo phosphorylated membrane proteins isolated from scutella labelled in the presence of amino acids with peptide antiserum raised to HvPTR1 (Fig. 4C). SDS-PAGE and autoradiography showed that a single heavily phosphorylated band of 66 kDa was immunoprecipitated using anti-HvPTR1 antiserum, but was absent from controls using preimmune serum. The identity of the protein as HvPTR1 was further confirmed by immunopurification of HvPTR1 from crude membrane fractions isolated from barley scutellar tissue using HvPTR1 antibodies immobilized onto Sepharose 4B (Fig. 4C). Purified HvPTR1 was analysed by SDS-PAGE and silver staining, demonstrating purification to near homogeneity (Fig. 4C, lane 1V).
Phosphoamino acid analysis of HvPTR1
When purified HvPTR1 was subject to phosphoamino acid analysis, a phosphoserine monoclonal antiserum cross-reacted with a 66 kDa protein, but no such cross-reaction was observed when phosphothreonine or phosphotyrosine antiserum were used (Fig. 5). HvPTR1 antiserum cross-reacted with a protein identical in molecular mass (66 kDa), confirming the identity of this protein as the barley scutellar peptide transporter. This strongly indicates that HvPTR1 becomes phosphorylated on serine residues in response to the presence of amino acids and that this phosphorylation results in a decrease in the ability of HvPTR1 to transport peptides. There are several putative phosphorylation sites in the predicted amino acid sequence of HvPTR1 but the specific site of the phosphorylated serine residue remains to be determined.
HvPTR1 is an N-linked glycoprotein
Plant plasma membrane proteins are often heavily glycosylated. The predicted protein sequence of HvPTR1 contains only one consensus N-glycosylation site at Asn72. The glycosylation status of HvPTR1 was investigated by incubating purified HvPTR1 protein with the N-glycosidase enzyme PNGase F. Analysis of the digestion products using SDS-PAGE followed by immunoblot analysis using HvPTR1 antiserum showed that PNGase F-treated HvPTR1 protein displayed a significantly lower molecular mass than untreated HvPTR1 protein (Fig. 6). The possibility that the observed shift in molecular mass with PNGase F was an artefact was addressed by performing an identical experiment with a protein known to be non-glycosylated (purified recombinant Arabidopsis thaliana DNA ligase 1 protein). No shift in molecular mass was obtained on PNGase F treatment of Arabidopsis DNA ligase 1 protein (Fig. 6).
Phosphorylation of proteins on the amino acid residues serine, threonine, and tyrosine is a fundamental mechanism through which the regulation of intracellular events may be effected. However, there are few previous reports concerning the post-translational regulation of plant solute transporter proteins by phosphorylation. Transcriptional regulation of plant transporter proteins is evidently a very common control mechanism, and there are numerous reports of plant sugar transporters regulated at this level (reviewed by Delrot et al., 2000; Williams et al., 2000). A small number of plant plasma membrane proteins have been shown to be regulated by phosphorylation including the H+-ATPase (Vera-Estrella et al., 1994), aquaporins (Johansson et al., 1996, 1998), and an anion channel in tobacco suspension culture cells (Zimmerman et al., 1994). Liu and Tsay (2003) reported that phosphorylation of the dual-affinity nitrate transporter CHL1 controls the switch between low and high affinity anion uptake. Okadaic acid was also shown to inhibit sucrose transport in plasma membrane vesicles isolated from sugar beet leaves, suggesting that phosphorylation of this sucrose transporter decreases activity, although the physiological significance of this remains to be determined (Roblin et al., 1998). It is reported here that amino acids at concentrations equivalent to those present in the endosperm of the germinated barley grain inhibit peptide transport in the barley embryo. Furthermore, strong evidence is provided that this regulation is mediated in vivo at the post-translational level by phosphorylation of the barley scutellar peptide transporter protein HvPTR1. The PEPT2 PTR family peptide transporter protein is down-regulated in the mammalian intestine in response to amino acids (Ogihara et al., 1999).
The regulation of nitrogen reserve mobilization in cereal grain germination has been poorly characterized and the physiological significance of HvPTR1 regulation by amino acids during germination remains to be determined. There is now an increasing awareness that amino acids translocated to the embryo are not merely used as building blocks for protein synthesis but may undergo significant metabolic interconversion (Limami et al., 2002). A number of studies demonstrated that substantial interconversion of amino acids occurs in germination, and catabolism of small peptides in early germination may possibly supply both energy and nutrients to the developing embryo (Lea and Joy, 1983; Limami et al., 2002). The germinating cereal grain may also need to balance the flux of nitrogen and carbon flowing from the site of reserve hydrolysis (the endosperm) across the scutellum and a means of achieving this regulation would be to maintain rates of glucose, peptide and amino acid transport in balance. Thus, the regulation of HvPTR1 peptide transport activity in response to the increasing levels of amino acids and glucose in the endosperm, as shown by a decrease in transport activity 2–3 d into germination, is potentially an important element in balancing the flux of organic nitrogen (amino acids and small peptides) and carbon from the endosperm to the embryo during seedling establishment following completion of germination (radicle emergence).
Nutrient reserve mobilization in the cereal endosperm has generally been considered to be a post-germinative event (Bewley and Black, 1994). However, both HvPTR1 expression and peptide transport activity are detectable very early during germination (from 6 h after imbibition commences) in the barley grain (West et al., 1998). Peptide transport plays a key role in the acquisition of organic nitrogen by the embryo from the endosperm in germinating barley grains. Not only is peptide transport responsible for approximately 50% of the total nitrogen flux from the endosperm to the embryo during germination and seedling establishment, but it is also almost solely responsible for nitrogen import into the embryo during germination and the first few days of seedling growth (Higgins and Payne, 1981). There is a significant pool of free peptides present in the endosperm of the quiescent barley grain and this peptide pool appears available for transport to the embryo during the initial water imbibition stages of germination (Higgins and Payne, 1981). Peptide transport capacity has also been shown to correlate closely with barley grain viability (Waterworth et al., 2000b). Recently, glucose was shown to overcome ABA inhibition of germination in Arabidopsis seeds (Garciarrubio et al., 1997), providing evidence for the importance of the provision of metabolizable substrates arising from reserve mobilization in the control of growth processes leading to successful seed germination.
HvPTR1 is one of the few genes known to be expressed so early in cereal grain germination, implicating nutrient supply to the embryo to be a critical process for the initiation of germination and commencement of growth processes. The importance of the provision of metabolizable substrates to the embryo of a quiescent seed to break dormancy and initiate growth processes associated with successful seed germination has recently been emphasized in seminal studies on the Arabidopsis comatose mutation (Footitt et al., 2002) and ABA (abi) mutants (Garciarrubio et al., 1997). The ped− (peroxisome-deficient) mutants of Arabidopsis (Hayashi et al., 1998) are defective in glyoxysomal fatty acid β-oxidation which plays a key role in producing sucrose from storage lipids. Seeds from these ped mutants only exhibit post-germinative growth when supplied with exogenous sucrose. These observations lend further support to the hypothesis that regulation of energy availability to the embryo is a major factor controlling dormancy, germination, and early seedling growth. The regulation of plant solute transporter activities by phosphorylation may be a key element in the steps leading to successful germination.
Understanding the regulation of seed germination has important and far-reaching implications for the improvement of seed quality through the ability to control both the timing and rate, or uniformity of seed germination. Through such control it should be possible to prevent mistimed germination such as seen during preharvest sprouting (Holdsworth et al., 2001) and to promote uniform rapid germination as seen after seed-priming treatments (Bray, 1995), both these traits being highly desirable in many areas of agriculture and horticultural practice. Elucidation of the mechanisms controlling transport of metabolizable substrates during germination may also have potential applications for seed quality in terms of an ability to control the onset of germination and maximize utilization of storage reserves for seedling establishment (seed quality and vigour), and also in malting processes where endosperm reserve degradation needs to be maximized whilst mobilization to the embryo needs to be minimized.
The financial support of the UK BBSRC through the award of a research grant and a postgraduate research grant is gratefully acknowleged.