The Plant Cell, Vol. 5, 443-450, April 1993 O 1993 American Society of Plant Physiologists

The N- and C-Terminal Regions Regulate the Transport of Wheat y-Gliadin through the Endoplasmic Reticulum in Xenopus Oocytes

Yoram Altschuler, Nurit Rosenberg, Rotem Harel, and Gad Galili’ Department of Plant Genetics, The Weizmann lnstitute of Science, Rehovot 76100, Israel

Following sequestration into the endoplasmic reticulum (ER), wheat storage proteins are naturally either retained and packaged into protein bodies within this organelle or exported to the Golgi apparatus. To identify protein domains that control the sorting of wheat storage proteins within the ER, a wild-type y-gliadin storage protein as well as two of its deletion mutants, each bearing either of the two autonomous N- and C-terminal regions, were expressed In Xenopus oocytes. Our results demonstrated that the N-terminal region of the gliadin, which is composed of severa1 tandem repeats of the consensus sequence PQQPFPQ, was entirely retained within the ER and accumulated in dense protein bodies. In contrast, the C-terminal autonomous region was efficiently secreted to the medium. The wild-type y-gliadin, contain- ing both regions, was secreted at a lower rate and less efficiently than its C-terminal region. These results suggest that sorting of the wheat y-gliadin within the ER may be determined by a balance between two opposing signals: one func- tions in the retention and packaging of the storage protein within the ER, while the second renders the protein competent for export from this organelle to the Golgi apparatus.

INTRODUCTION

Sorting of secretory proteins within the endoplasmic reticu- lum (ER) to the Golgi apparatus is a highly regulated process (Pfeffer and Rothman, 1987; Rose and Doms, 1988). It is now generally agreed that, both in animal and plant cells, export of secretory proteins from the ER to the Golgi apparatus oc- curs by default, whereas retention within the ER requires positive signals. This has been recently demonstrated for a variety of ER-resident proteins that were shown to be retained within the ER by virtue of a specific signal, KIHDEL, located at their C termini (Pfeffer and Rothman, 1987, Rose apd Doms, 1988; Pelham, 1989; Rothman, 1989; Chrispeels, 1991). How- ever, other ER retention signals may also exist inasmuch as even proteins that lack the KlHDEL signal may be retained at least partially within the ER. It was also shown that even mutants of the ER-resident proteins that lack the KlHDEL sig- na1 are mostly retained within the ER (Munro and Pelham, 1987; Pelham, 1989,1990). Recent studies have also suggested that sorting of secretory proteins within the ER may be controlled by the leve1 of Ca2+ within the organelle (600th and Koch, 1989; Rudolph et al., 1989; Sambrook, 1990; Wadaet al., 1991).

Wheat prolamin storage proteins are a group of KIHDEL- less proteins whose transport via the ER appears to be com- plex. Electron microscopy and subcellular fractionation studies showed that following sequestration into the ER, these pro- teins are either retained and packaged into dense protein

To whom correspondence should be addressed.

bodies (PBs) inside this organelle or transported from the ER via the Golgi apparatus tovacuoles (Campbell et al., 1981; Miflin and Burgess, 1982; Parker, 1982; Parker and Hawes, 1982; Bechtel and Barnet, 1986; Kim et al., 1988; Levanony et al., 1992; Rubin et al., 1992). These processes may also be de- velopmentally regulated as at earlier stages of wheat grain development, a larger proportion of the storage proteins may be routed via the Golgi apparatus. Recently, we have próvided evidence suggesting that PBs formed within the ER of wheat endosperm cells may subsequently be internalized directly into vacuoles by an autophagy-like process (Levanony et al., 1992). These observations suggested that wheat prolamins possess specific structural properties that affect their retention or ex- port from the ER to the Golgi apparatus.

The majority of wheat storage proteins belong to a family of closely related sulfur (S)-rich gliadins. Each member of this family appears naturally as a chimeric protein containing two separately folded autonomous regions that evolved indepen- dently as separate proteins and were apparently fused during the evolution of gliadins (Kreis et al., 1985; Shewry and Tatham, 1990). The N-terminal region is composed of 7 to 16 tandem repeats rich in glutamine and proline, which are apparently arranged in p-turn configurations (Tatham and Shewry, 1985; Shewry and Tatham, 1990). The tandem repeats vary slightly between the different S-rich gliadins, but they are essentially based on either PQQPFPQ, PQQQPPFS, or PQQPQ con- sensus sequences (Shewry and Tatham, 1990). The C-terminal

444 The Plant Cell

pSP65yA

ATCjSpel

RepetitiveTGA

Sail Ball!; Unique ;;

BATG

rt RepetitivepSP65YA-N-Ter^-6Fi|-t5Pl l»»l»m*l

Ball

pSP65yA-C-Ter.

pSP65aA pSP65yA

ATG TGASpel

pSP65vA-FpSP54-1133A pSP65yA

Figure 1. Schematic Diagrams of the y-Gliadin and Its Deletion andFusion Constructs Cloned into pSP65A.(A) pSP65yA, encoding a wild-type y-gliadin.(B) pSP65yA-N-Ter, a truncated y-gliadin construct lacking a Ball-to-Ball DMA fragment encoding the last 120 amino acids of the C termi-nus and containing an additional three codons for Met, Lys, and Ala,followed by a termination codon.(C) pSP65yA-C-Ter, a truncated y-gliadin construct lacking the entireN-terminal repetitive region, which was obtained by fusion of the DMAencoding the C-terminal region of the y-gliadin with DMA encodingan ER signal peptide from an cc-gliadin. The sequence of the double-stranded synthetic DNA used to fuse the y-gliadin to the a-gliadin isshown above.(D) pSP65yA-F, a construct encoding a fusion y-gliadin containing alonger N-terminal repetitive region (24 repeats) than that of the wild-type protein (16 repeats). Each triangle in the repetitive regionsrepresents DNA encoding one repeated amino acid sequence of aconsensus PQQPFPQ (pSP65yA) and PQQQPPFS (pSP64-1133A).SP6, promoter for SP6 RNA polymerase; SP, DNA encoding an ERsignal peptide; ATG, initiation codon; Repetitive, DNA encoding theN-terminal repetitive region; Unique, DNA encoding the C-terminalunique sequence (nonrepetitive) region; TGA, termination codon.

The secreted proteins can be easily separated from those thataccumulate within the ER. Our study suggested that theN-terminal repetitive region of the y-gliadin was responsiblefor its retention and packaging into PBs within the ER, whereasthe C-terminal region was responsible for its export from theER to the Golgi apparatus. Sorting of the wild-type y-gliadinappears to be determined by a balance between these tworegions.

RESULTS

Localization of the y-Gliadin Inside Oocyles

We have previously shown that a wheat y-gliadin expressedin Xenopus oocytes was partly secreted via the Golgi appara-tus to the medium, and the rest was retained within the oocytecells following 24 hr of incubation of the oocytes with 35S-methionine (Simon et al., 1990). To study more accurately thetime course of this secretion, an mRNA coding for a wild-typey-gliadin was transcribed in vitro from the pSP65yA clone, asshown in Figure 1A, and was injected into oocytes; the cellswere then incubated with 35S-methionine for 24, 48, or 72 hrbefore harvesting. As shown in Figure 2, the y-gliadin wasgenerally detected in the medium after 24 hr, and it continuedto accumulate in the medium with increasing incubation time.However, its secretion was relatively slow, and a considerableamount of the y-gliadin was retained within the oocytes evenafter 72 hr. Based on nine different experiments, the secre-tion of ygliadin following 72 hr of incubation amounted to

oocyte mediumkD Mr

40

24 48 72 24 48 72' .Incubationtime (hr)

regions of the S-rich gliadins are apparently arranged predom-inantly in a-helix and (5-sheet configurations (Tatham andShewry, 1985; Shewry and Tatham, 1990). These regions alsocontain six to eight cysteine residues, which form three to fourintramolecular disulfide bonds (Shewry and Tatham, 1990).

In this report, we have studied the role of the N- andC-terminal regions of a wheat y-gliadin in the sorting of thisprotein within the ER. This was addressed by expressing wild-type y-gliadin in Xenopus oocytes as well as two deletionmutants containing only one of its autonomous regions. Xeno-pus oocytes are a very powerful system for this type of studyinasmuch as these cells lack vacuoles; hence, storage pro-teins exported from the ER to the Golgi apparatus are secretedinto the medium by the default pathway (Simon et al., 1990).

-y-gliadin

30

Figure 2. Expression of the Wheat y-Gliadin in the Oocytes.Oocytes were injected with a y-gliadin SP6 mRNA, transcribed frompSP6yA, and allowed to stand overnight. The oocytes were then in-cubated with 35S-methionine for increasing time periods andharvested. Alcohol extracts of homogenized oocytes (oocyte) or theincubation medium (medium) were electrophoresed on SDS-polyacryl-amide gels and fluorographed. The migration of protein markers andtheir molecular masses (in kilodaltons) are shown at left.

Subcellular Transport of Wheat Gliadins 445

uninjected oocytesinjected oocytes

ER Dense PB10 11 12 13

Top Botlon

Figure 3. Densities of PBs Obtained from Oocytes Expressing theWheat y-Gliadin and from Developing Wheat Grains.

Membrane homogenates from oocytes, either noninjected or injectedwith y-gliadin mRNA and incubated for 48 hr with 35S-methionine (Ato C), or from developing wheat grains at 16 days after anthesis (D)were fractionated on a metrizamide density gradient. In (B) to (D),alcohol-soluble proteins were prepared from equal aliquots of eachfraction, collected along the gradient (fractions 1 to 13), and subjectedto SDS-PAGE. Proteins were detected either by fluorography (B and C)or by Coomassie blue staining (D).(A) Fractions from oocyte membranes assayed for NADH cytochromec reductase activity. The locations of lipids (L), mitochondria (M), andyolk platelets (YP) are indicated (Wallace et al., 1988).(B) and (C) Oocytes with low and high efficiency of protein synthesis,respectively.(D) PBs from wheat grains.The position of the y-gliadin in (B) and (C) is indicated by arrows at right.The molecular masses (in kilodaltons) of protein markers (D) are shownat right.

69 ± 2.62% (SE). There was some variation between oocytesfrom different frogs in the efficiency of protein synthesis (bothendogenous proteins and the y-gliadin) as measured by in-corporation of the radioactive amino acids into trichloroaceticacid-precipitable counts (data not shown). Notwithstanding,the efficiency of y-gliadin secretion was not affected by theefficiency of protein synthesis.

Next we tested whether the y-gliadin retained within theoocytes accumulated within the ER. Membrane homogenateswere fractionated on a metrizamide gradient and separatedinto 13 fractions. Analysis of the activity of the ER markerenzyme NADH cytochrome c reductase showed that theER membranes sedimented mostly in fractions 3 to 5 in thegradient, as shown in Figure 3A. Lipids, mitochondria, andyolk platelets were visualized in this gradient as white opaquebands (Figure 3A)(Hurkmanetal., 1981;Wallaceetal., 1988).

Strikingly, when the distribution of the y-gliadin in these gra-dients was analyzed, there was a marked difference betweenvarious groups of oocytes. In oocytes with less efficient pro-tein synthesis, most of the y-gliadin sedimented in fractions3 to 5, identically to that of the ER marker NADH cytochromec reductase, indicating that the protein accumulated inside theER (Figure 3A and Figure 3B lanes 3 to 5). In oocytes withmore efficient protein synthesis, a considerable amount ofthe y-gliadin in the ER packaged to form dense PBs thatsedimented toward the bottom of the gradient in fractions 10to 13 (Figure 3C). These dense PBs had a very similar densityto that of natural PBs derived from developing wheat grains(Figure 3D). To confirm that the y-gliadin was localized insideER membranes, we tested its resistance to proteinase K diges-tion. The protein from both the ER and dense PB fractionswas resistant to proteinase K, unless Triton X-100 was presentin the incubation medium, as shown in Figure 4.

Transport and Packaging of y-Gliadin Deletion Mutantsin the Oocytes

The y-gliadin consists of two autonomous regions that appearto fold independently of each other: an N-terminal repetitiveregion containing 16 tandem amino acid repeats of a consensusPQQPFPQ sequence and aC-terminal region of unique (non-repetitive) amino acid sequence. To test which of these regionswas responsible for the retention or for the export of they-gliadin from the ER, two deletion mutants of the y-gliadingene were constructed. In the first, a deletion was performedat the 3'end of the N-terminal DNA coding sequence, yieldinga clone that codes for a truncated protein lacking most of theC-terminal region (pSP65yA-N-Ter) (Figure 1B). In the second,the DNA coding sequence of the entire C-terminal region of

Fractions 3-6 Fractions 10-13

proteinase K 0(u.g/mL)

Triton X-100 -

150 150 400 400 0 150 150 400 400

Figure 4. Treatment of Pooled Fractions from the Metrizamide Gra-dient with Proteinase K.

Fractions 3 to 6 and 10 to 13 from a metrizamide gradient, similar tothat shown in Figure 3, were pooled and each was divided into fiveequal samples. Samples were treated with zero (control) and 150 or400 ng/ml_ proteinase K in the presence (+) or absence (-) of TritonX-100. Following the proteinase K treatment, proteins were separatedon SDS-polyacrylamide gels and detected by fluorography.

446 The Plant Cell

Incubation oocyte mediumtime(hr)->- 24 48 72 24 48 72

Wild-type A

N-TerminalRegion

C-TerminalRegion

6

100

Wild-typeC-Terminal RegionN-Terminal Region

0 10 20 30 40 50 60 70% Ethanol

ER Dense PB

1 2 3 4 5 6 7 8 9 10 11 12 13

Top Bottom

Figure 5. Expression of the Truncated y-Gliadins in Oocytes.

Oocytes were injected with SP6 mRNAs transcribed from the constructspSP65yA (Wild-type), pSP65yA-N-Ter (N-terminal region), or pSP65yA-C-Ter (C-terminal region) and allowed to stand overnight.(A) Injected oocytes were incubated with a radioactive amino acid forincreasing time periods and harvested. Alcohol extracts of homogenizedoocytes (oocyte) or the incubation medium (medium) were electropho-resed on SDS-polyacrylamide gels and fluorographed.(B) Injected oocytes were incubated with a radioactive amino acid for48 hr and harvested. Gliadins were alcohol extracted, concentratedby evaporation, and resolubilized in increasing alcohol concentrations.Soluble and insoluble material from each extraction was separatedon SDS-polyacrylamide gels and detected by fluorography; gliadinbands were quantitated by densitometry. The percentage of proteinin the soluble fractions (average of three independent experiments)is plotted against the percentage of ethanol used for extraction.(C) Membrane homogenates from oocytes expressing the N-terminaldeletion mutant of the y-gliadin and labeled with 35S-methionine for48 hr were fractionated on a metrizamide density gradient. Individual

the ygliadin was fused to a DMA encoding an ER signal pep-tide of another gliadin gene, yielding a protein lacking the entireN-terminal repetitive region, but containing a signal peptidefor insertion into the ER (pSP65yA-C-Ter) (Figure 1C). mRNAswere transcribed in vitro from the wild type and from the twodeletion forms of the Y-gliadin and injected into separate groupsof oocytes. While the truncated y-gliadin containing the N-ter-minal region was completely retained within the oocytes, thedeletion mutant containing only the C-terminal region was pre-dominantly secreted to the medium, as shown in Figure 5A.Secretion of this region was not only more efficient than thatof the wild-type y-gliadin, but also significantly faster, consider-ing that in all the experiments performed most if not all of theC-terminal region was already found in the medium after 24 hrof incubation with the radioactive amino acid.

Wheat gliadins are known to be poorly soluble in aqueoussolutions in vitro. To test for a possible correlation betweensecretion and solubility of the y-gliadin, the wild type and thetwo deletion mutants, expressed in oocytes, were purified byextraction in 70% ethanol. The ethanol was then removed byevaporation and the gliadin pellets were dissolved in increas-ing ethanol concentrations. Following an incubation of 1 hr inthese solutions, the proteins were precipitated and the rela-tive amounts found in the pellets (insoluble) and supernatant(soluble) were determined by SDS-PAGE analysis and den-sitometric tracing. Notably, as the alcohol concentration wasreduced, the N-terminal region, which was entirely retainedwithin the ER, was more soluble than either wild-type y-gliadinor the C-terminal region, both of which were secreted (Figure5B). This negates the supposition that the retention of thegliadin within the ER might stem from its low solubility withinthis organelle.

It was also of interest to test whether the y-gliadin deletionmutants could be packaged into dense PBs. To address thisissue, membrane homogenates from oocytes expressing theN-terminal region were fractionated on a metrizamide densitygradient, and alcohol-soluble proteins were subjected to SDS-PAGE. A considerable amount of the N-terminal region waspackaged in dense PBs that sedimented toward the bottomof the gradient (Figure 5C).

Transport of y-Gliadin Containing a Larger N-TerminalRepetitive Region

Based on our previous findings, we tested whether the reten-tion of the deletion mutant bearing the N-terminal region wasdetermined by the general conformation rather than by a spe-cific amino acid sequence in this region. For this purpose, wegenerated a fusion construct between the y-gliadin encoded by

fractions from the top (1) to the bottom (13) were obtained, and alcohol-soluble proteins were fractionated on SDS-polyacrylamide gels andfluorographed. The position of the N-terminal deletion mutant is indi-cated by an arrow at left.

Subcellular Transport of Wheat Gliadins 447

pSP65yA and a clone coding for another closely related S-richgliadin. The tandem repeats in this gliadin are of the PQQ-QPPFS sequence. The fusion construct, pSP64yA-F (Figure1D), encodes an identical protein to the y-gliadin encoded bypSP65yA, except for the presence of an additional eight PQQ-QPPFS consensus repeats in the N-terminal region. mRNAswere transcribed in vitro from pSP65yA and pSP65yA-F andinjected into separate groups of oocytes. As shown in Figures6A and 6B, secretion efficiency of the fusion protein wassignificantly lower than that of the wild-type y-gliadin. This sug-gested that the effectiveness of the N-terminal region in theretention of the y-gliadin within the ER was positively relatedto the number of repeats present in this region.

incubation oocyte mediumtime(hr)—*-24 48 72 24 48 72

Wild-type

Chimericprotein

B80

60 -

40 ^

Wild-type Chimericprotein

Figure 6. Expression of the Chimeric y-Gliadin Containing a LongerN-Terminal Region in the Oocytes.(A) Oocytes were injected with SP6 mRNAs transcribed from the con-structs pSP65yA (Wild-type) or pSP65yA-F (Chimeric protein) andallowed to stand overnight. The oocytes were then incubated with^S-methionine for increasing time periods and harvested. Alcohol ex-tracts of homogenized oocytes (oocyte) or the incubation medium(medium) were electrophoresed on SDS-polyacrylamide gels andfluorographed.(B) Quantitative analysis of the percentage of secretion of the wild-type y-gliadin (average of nine separate experiments) and the chimericprotein (average of six separate experiments). Secretion was quanti-tated by densitometric tracing, following 72 hr of incubation with^S-methionine. Bars on top of each histogram represent the standarderror. Histograms with different letters represent significant differencesat the 1% level as measured by a Duncan test.

DISCUSSION

Identification of Regions That Control the Retentionand Packaging of the y-Gliadin within the ER ofXenopus Oocytes

Previous studies (Campbell et al., 1981; Miflin et al., 1981;Parker, 1982; Parker and Hawes, 1982; Levanonyetal., 1992;Rubin et al., 1992) showed that wheat storage proteins mayeither be retained and packaged into PBs within the ER orbe exported to the Golgi apparatus. In the present report, wehave chosen Xenopus oocytes as a heterologous system toidentify regions in wheat y-gliadin storage protein that controlits sorting and packaging into PBs within the ER. Xenopusoocytes were chosen primarily because they lack vacuoles;hence, vacuolar proteins exported from the ER to the Golgiapparatus are eventually secreted into the medium by bulkflow (Pfeffer and Rothman, 1987; Rose and Doms, 1988;Pelham, 1989). Thus, the efficiency of secretion of storage pro-teins to the medium could be used to monitor their efficiencyof exit from the ER to the Golgi apparatus (Simon et al., 1990).

The results of this study showed that sorting of the y-gliadinwithin the ER of the oocytes was very similar to the sortingof storage proteins in wheat cells (Levanony et al., 1992), i.e.,part of the gliadin was retained within the ER, while the restwas exported to the Golgi apparatus. This fidelity of the oo-cyte system suggests that the ER elements that control thefolding and transport of secretory proteins via the ER wereconserved between wheat and Xenopus oocytes. In addition,the y-gliadin, when expressed on its own, was packaged withinthe ER of the oocytes into PBs having similar physical char-acteristics to the natural PBs of wheat endosperm. This impliesthat the y-gliadin most likely contains sufficient information toinitiate PB formation in the heterologous system of the oocytes.The mechanism controlling the retention and packaging ofwheat storage proteins within the ER is not known. Yet, wheatstorage proteins do not contain a K/HDEL sequence, suggest-ing that their retention and packaging into PBs within the ERoccur by virtue of other structural properties of these proteins.Expression of the deletion mutants of the y-gliadin in the oo-cytes showed that the N-terminal repetitive region of thisprotein, containing the 16 consensus PQQPFPQ repeats, wasresponsible for its retention and packaging into PBs within theER. In addition, when eight tandem repeats of a slightly al-tered consensus sequence (PQQQPPFS) were added to theN-terminal region of the y-gliadin, the efficiency of protein reten-tion was increased, suggesting that this process was relatedto the length (number of repeats) and to the putative p-turnconformation of this region rather than to a specific amino acidsequence. The minimal number of repeats that may functionas an ER retention element is not yet known. Such an analy-sis should be supported by structural data inasmuch as it isunknown whether isolated repeats can still maintain the sameputative 3-turn structure that they apparently possess as partsof multiple repeats.

448 The Plant Cell

Another possible explanation for the complete ER retention of the deletion mutant containing the N-terminal region is that, when expressed on its own, this region was not folded cor- rectly. However, this possibility seems unlikely because of the following arguments: (1) structural studies of wheat S-rich gliadins strongly suggest that their N- and C-terminal regions are autonomous and fold independently of each other (Shewry and Tatham, 1990); (2) the fact that the deletion mutant con- taining the C-terminal region was entirely secreted implies that this region, even when expressed on its own, was correctly folded; and (3) independent folding of the N- and C-terminal regions of the S-rich gliadins may be supported by the fact that native proteins containing regions analogous only to the N- or to the C-terminal regions of the S-rich gliadins appear naturally in plant seeds (Kreis et al., 1985; Shotwell and Larkins, 1988). One such group of proteins is the family of a-zein stor- age proteins of maize, whose major amino acid sequence is related to the N-terminal region of the y-gliadins. The a-zeins are also naturally packaged into PBs within the ER (Shotwell and Larkins, 1988).

Wheat gliadins are known to possess low solubility in aque- ous solutions in vitro. However, whether this phenomenon is responsible for their retention within the ER is not known. If this assumption is true, then the N-terminal repetitive region, which was entirely retained within the ER, was expected to be less soluble than the wild-type y-gliadin and the C-terminal region, both of which were secreted. However, based on the higher solubility of the N-terminal region in alcohol, this pos- sibility may be ruled out. On the other hand, the partia1 retention of the y-gliadin within the ER could have been dueto its capa- bility to initiate PB formation within the ER. This possibility seems likely in view of the fact that the N-terminal region of the y-gliadin was responsible both for the retention and for the packaging of this protein within the ER. Yet, it seems that the formation of dense PBs was not essential for the retention of the y-gliadin inasmuch as in oocytes with less efficient pro- tein synthesis, no packaging into dense PBs was evident, although a considerable amount of the y-gliadin was still re- tained within the ER. It is thus possible that the retention of y-gliadin within the ER is determined by specific interactions that initiate PB formation. Alternatively, retention of the y-gliadin within the ER may not be related to the process of PB forma- tion but to a specific recognition of the N-terminal repetitive region by other ER-resident proteins (Booth and Koch, 1989; Rudolph et al., 1989; Sambrook, 1990; Wadaet al., 1991). This is a subject of continuous interest in our laboratory.

Role of the C-Terminal Region in the Transport of the r-Gliadin via the ER

It has been clearly demonstrated that export of proteins from the ER occurs by default whereas retention within this organelle requires specific signals (Pfeffer and Rothman, 1987; Rose and Doms, 1988; Pelham, 1989, 1990; Chrispeels, 1991). Thus,

if the function of the N-terminal region in the retention and pack- aging of the y-gliadin within the ER was fully dominant, then the wild-type protein would have been retained, as was ob- served for the deletion mutant containing only the N-terminal region. However, the wild-type y-gliadin was secreted, although at a slower rate and less efficiently than the deletion mutant containing the C-terminal region. This demonstrated that the C-terminal region of the y-gliadin was not inert because it coun- terbalanced the effect of the N-terminal region in controlling the retention and packaging of the intact protein within the ER. Thus, our study suggests that sorting of the y-gliadin within the ER may be determined by a balance between two oppo- site signals: (1) the effectiveness of the N-terminal region in the retention and initiation of PB formation within the ER, and (2) the counteracting effect of the C-terminal region to avert packaging within the ER and to promote the exit of the protein to the Golgi apparatus.

METHODS

Plasmid Constructions and Site-Directed Mutagenesis

Cloning and sequence analysis of the y-gliadin gene, pW1621, as well as subcloning of its DNA coding sequence into pSP65A to construct pSP65yA (Figure 1A) were described previously (Sugiyamaet al., 1986; Simon et al., 1990). To delete the DNA sequence coding for most of the C-terminal uniquesequence region, a 360-bp Ball-te-Ball DNA frag- ment (Figure lA), encoding the last 120 amino acids of the y-gliadin, was deleted. As a result of a frame shift, three additional codons for Met, Lys, and Ala, which were followed by a stop codon, were added to the DNA coding sequence of the truncated construct forming the clone pSP65yA-NTer (Figure 16). To fuse the DNA coding sequence of the entire C-terminal region to a DNA encoding a signal peptide, we used a signal peptide from the a-gliadin gene pW8233 (Rafalski et al., 1984) because this gene contained a convenient Pstl site lo- cated at the 3’ end of the DNA encoding the signal peptide.

For site-directed mutagenesis, the following steps were performed. An Ndel site was installed 506 nucleotides downstream of the ATG codon by changing a CAACGG sequence to a CATATG sequence using site-directed mutagenesis (Kunkel et al., 1987). The Ndel site was placed in the DNA sequence linking the N- and C-terminal regions. A syn- thetic double-stranded DNA (5‘-GGTCG-3?3’-ACGTCCAGCAT-5’), containing a 5‘ Pstl site and a 3’ Ndel site, was used for an in-frame fusion of the DNA encoding the C-terminal region of the y-gliadin in pSP65yA (Simon et al., 1990) to the Pstl site of the a-gliadin gene, which was cloned in the plasmid pSP65aA (Simon et al., 1990). The resulting pSP65aAlpSP65yA fusion plasmid was termed pSP65yA- CTer (Figure 1C).

To construct a clone coding for a y-gliadin with a longer N-terminal repetitive region, pSP65yA was digested with Spel, which is located upstream of the N-terminal repetitive region, and fused in-frame to the Spel of pSP64-1133A to form pSP65yA-F (Figure 1D). pSP64-1133A contains another sulfur-rich prolamin cDNA, 8-1133 (Okita et al., 1985), cloned into pSP64A. This cDNA contains a Spel site at the end of the N-terminal repetitive region. Thus, pSP65yA-F encodes the same y-gliadin as pSP65yA, except for its longer N-terminal repetitive region

Subcellular Transport of Wheat Gliadins 449

bearing 24 repeats. The production of pSP64-1133A will be described elsewhere (Y. Altschuler and G. Galili, manuscript in preparation).

equally into five samples. Samples were treated with proteinase K (150 or 400 pglmL) for 30 min on ice in the presence or absence of 1% Triton X-100. The reactions were terminated by the addition of phenyl- methylsulfonyl fluoride to a final concentration of 5 mM.

In Vitro Transcription and Microinjection into Xenopus Oocytes

SDS-PAGE Analysis and Protein Detection In vitro transcription, using purified SP6 RNA polymerase, was per- formed as described previously (Krieg and Melton, 1984) using the cap analog G(5?ppp(5’)G. The reaction mix was incubated for 2 hr at 4OoC and then extracted with phenol/chloroform before ethanol precipitation.

Microinjection into oocytes (5 to 15 oocytes per individual treatment) and labeling with 35S-methionine were performed as described pre- viously (Simon et al., 1990). For detection of the repetitive region of the y-gliadin, which does not contain methionine residues, the oocytes were labeled with 3 mCilmL of 3H-leucine (57 Ci/mmol).

Solubility of y-Gliadin and Its Truncated Mutants in Different Alcohol Concentrations

Oocytes were injected and labeled as described above. After 48 hr of incubation with the radioactively labeled amino acid, the y-gliadin as well as its N- and C-terminal regions were purified as described pre- viously (Simon et al., 1990). An equal amount of radioactivity from each protein was placed in eight tubes followed by evaporation and resolu- bilized in increasing concentrations of alcohol (O to 70%). Samples were incubated at room temperature for 1 hr with frequent agitation followed by centrifugation for 30 min. The supernatants were removed to other tubes and concentrated by evaporation. Samples were elec- trophoresed on SDS-polyacrylamide gels, and the proteins were detected by fluorography (Simon et al., 1990) and quantitated using a densitometer (model3OOA; Molecular Dynamics, Sunnyvale, CA) at 633 nm.

Fractionation of Membrane Homogenates on Metrizamide Density Gradients

Developing wheat grains or oocyte cells injected with y-gliadin SP6 mRNA and incubated with 35S-methionine for 48 hr were homog- enized in a buffer containing 20 mM Tris-HCI, pH 7.6, 50 mM KCI, 10 mM MgClp, 0.3 M NaCI, 2 mM EDTA, and 10% sucrose. The homog- enates were then fractionated on a 10 to 50% metrizamide gradient as described previously(Wal1ace et al., 1988). Fractions of 0.4 mL were collected from the top of the gradient and brought to 70% ethanol plus 5O/O P-mercaptoethanol. Following incubation for 30 min at 6OoC and precipitation, the supernatant was collected and further precipitated by the addition of 2 volumes of 1.5 M NaCl and overnight incubation at 4OC. Extraction with alcohol plus P-mercaptoethanol and salt precipi- tation was repeated twice. Fractions were also tested for the activity of the ER marker enzyme NADH cytochrome c reductase, as previ- ously described (Hurkman et al., 1981).

Proteinase K Treatment

Following fractionation on the metrizamide gradient, fractions 3 to 6 and 10 to 13 (Figure 3) were pooled and each pooled fraction was divided

Gliadin was ethanol extracted from the oocytes as described previ- ously(Simon et al., 1990). These gliadins, as well as pellets of gliadin extracts obtained from the metrizamide gradients, were analyzed by SDS-PAGE as described previously (Laemmli, 1970). Gliadins from wheat grains were detected by Coomassie Brilliant Elue R 250 stain- ing (Galili and Feldman, 1983); labeled gliadins were detected by fluorography (Simon et al., 1990).

ACKNOWLEDGMENTS

We are greatly indebted to Regina Rubin for her assistance in the anal- ysis of the metrizamide gradients. We also thank Dr. Anthony Futerman, Dr. Shulamit Michaeli, and Yigal Avivi for critical reading of the manu- script; Professor Dieter Sol1 for pW8233 and pW1621; and Professor Thomas W. Okita for pB-1133. This work was supported by a grant from the Leo and Julia Forchheimer Center for Molecular Genetics and the Kimmelman Center for Biomolecular Assemblies at the Weiz- mann lnstitute of Science, Rehovot, Israel. G.G. is an incumbent of the Abraham and Jenny Fialkov Career Development Chair in Biology.

Received December 29, 1992; accepted February 19, 1993.

REFERENCES

Bechtel, D.B., and Barnet, B.D. (1986). A freeze fracture study on the storage protein accumulation in unfixed wheat starchy en- dosperm. Cereal chem. 63, 232-240.

Booth, C., and Koch, G.L.E. (1989). Perturbation of cellular calcium induces secretion of luminal ER proteins. Cell 59, 729-737.

Campbell, W.P., Lee, J.W., OBrien, T.P., and Smart, M.G. (1981). En- dosperm morphology and protein body formation in developing wheat grains. Aust. J. Plant Physiol. 8, 5-19.

Chrispeels, M.J. (1991). Sorting of proteins in the secretory system. Annu. Rev. Plant Physiol. Plant MOI. Biol. 42, 21-53.

Galili, G., and Feldman, M. (1983). Genetic control of endosperm pro- teins in wheat. l. The use of high resolution one-dimensional gel electrophoresis for the allocation of genes coding for endosperm protein subunits in the comon wheat cultivar Chinese Spring. Theor. Appl. Genet. 64, 97-101.

Hurkman, W.J., Smith, L.D., Richter, J., and Larkins, B.A. (1981). Subcellular compartmentalization of maize storage proteins in Xeno- pus oocytes with zein mRNAs. J. Cell Biol. 89, 292-299.

Kim, W.T., Franceschi, V.R., Krishnan, H.B., and Okita, T.W. (1988). Formation of wheat protein bodies: lnvolvement of the Golgi appara- tus in gliadin transport. Planta 176, 173-182.

450 The Plant Cell

Kreis, M., Shewry, P.R., Ford, B.G., Ford, J., and Miflin, B.J. (1985). Structure and evolution of seed storage proteins and their genes

- with particular reference to those of wheat, barley and rye. In Ox- ford Surveys of Plant Molecular and Cellular Biology, Vol. 2, B.J. Miflin, ed (New York: Oxford University Press), pp. 253-317.

Krieg, P.A., and Melton, D.A. (1984). Functional messenger RNAs are produced by SP6 in vitro transcription of cloned cDNAs. Nucl. Acids Res. 12, 7057-7070.

Kunkel, T.A., Roberts, J.D., and Zakour, R.D. (1987). Rapid and effi- cient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 154, 367-382.

Laemmli, U.K. (1970). Cleavage of structural proteins during the as- sembly of the head of bacteriophage T4. Nature 227, 680-685.

Levanony, H., Rubin, R., Altschuler, Y., and Galili, G. (1992). Evi- dente for a nove1 route of wheat storage proteins to vacuoles. J. Cell Biol. 119, 1117-1128.

Miflin, B.J., and Burgess, S.R. (1982). Protein bodies from develop- ing seeds of barley, maize, wheat and peas: The effect of protease treatment. J. Exp. Bot. 33, 251-260.

Miflin, B.J., Burgess, S.R., and Shewry, P.R. (1981). The develop- ment of protein bodies in the storage tissues of seeds. J. Exp. Bot.

Munro, S., and Pelham, H;R.B. (1987). A C-terminal signal prevents secretion of luminal ER proteins. Cell 48, 899-907.

Okita, T., Cheesbrough, V., and Reeves, C.D. (1985). Evolution and heterogeneity of the a-/btype gliadin DNAsequences. J. Biol. Chem.

Parker, M.L. (1982). Protein accumulation in the developing endosperm of a high-protein line of Trircum dicoccoides. Plant Cell Environ. 5, 37-43.

Parker, M.L., and Hawes, C.R. (1982). The Golgi apparatus in develop- ing endosperm of wheat (Triicum aestivum L.). Planta 154,277-283.

Pelham, H.R.B. (1989). Control of protein exit from the endoplasmic reticulum. Annu. Rev. Cell Biol. 5, 1-23.

Pelham, H.R.B. (1990). The retention signal for soluble proteins of the endoplasmic reticulum. Trends Biochem. Sci. 15, 483-486.

Pfeffer, S.R., and Rothman, J.E. (1987). Biosynthetic protein trans- port and sorting by the endoplasmic reticulum and Golgi. Annu. Rev. Biochem. 56, 829-852.

32, 199-219.

260, 8203-8213.

Rafalski, A., Scheets, K., Metzler, M., Peterson, J.M., Hedgcoth, C., and Soll, D. (1984). Developmentally regulated plant genes: The nucleotide sequence of a wheat gliadin genomic clone. EM60 J.

Rose, J.K., and Doms, R.W. (1988). Regulation of protein export from the endoplasmic reticulum. Annu. Rev. Cell Biol. 4, 257-288.

Rothman, J.E. (1989). Polypeptide chain binding proteins: Catalysts of protein folding and related processes in cells. Cell 59, 591-601.

Rubin, R., Levanony, H., and Galili, G. (1992). Characterization of two types of protein bodies in developing wheat endosperm. Plant Physiol. 99, 718-724.

Rudolph, H.K., Antebi, A., Fink, G.R., Buckley, C.M., Dorman, T.E., LeVitre, J., Davidow, J.-I.M., and Moir, D.T. (1989). The yeast secre- tory pathway is perturbed by mutations in PMR7, a member of a Ca2+ ATPase family. Cell 58, 133-145.

Sambrook, J.F. (1990). The involvement of calcium in transport of secre- tory proteins from the endoplasmic reticulum. Cell 61, 197-199.

Shewry, P.R., and Tatham, A.S. (1990). The prolamin storage pro- teins of cereal seeds: Structure and evolution. Biochem. J. 267,l-12.

Shotwell, M.A., and Larkins, B.A. (1988). The biochemistry and mo- lecular biology of seed storage proteins. In The Biochemistry of Plants, Vol. 15, B.J. Miflin, ed (New York: Academic Press), pp.

3, 1409-1415.

297-345. Simon, R., Altschuler, Y., Rubin, R., and Galili, G. (1,990). Two closely

related wheat storage proteins follow a markedly different subcellu- lar route in Xenopus laevis oocytes. Plant Cell 2, 941-950.

Sugiyama, T., Rafalski, A., and Soll, D. (1986). The nucleotide se- quence of a wheat y-gliadin genomic clone. Plant Sci. 44,205-209.

Tatham, A.S., and Shewry, P.R. (1985). The conformation of wheat gluten proteins. The secondary structures and thermal stabilities of a, p, y and o-gliadins. J. Cereal Sci. 3, 103-113.

Wada, I., Rindress, D., Cameron, P.H., Ou, W.J., Doherty, J:J., II, Louvard, D., Bell, A.W., Dignard, D., Thomas, D.Y., and BeGe- ron, J.J.M. (1991). SSRa and associated calnexin are major calcium binding proteins of the endoplasmic reticulum membrane. J. Biol. Chem. 266, 19599-19610.

Wallace, J.C., Galili, G., Kawata, E.E., Cuellar, RdE:, Shotwell, M.A., and Larkins, B.A. (1988). Aggregation of lysine-containing zeins into protein bodies in Xenopus oocytes. Science 240, 662-664..

\ .

DOI 10.1105/tpc.5.4.443 1993;5;443-450Plant Cell

Y Altschuler, N Rosenberg, R Harel and G Galiliendoplasmic reticulum in Xenopus oocytes.

The N- and C-terminal regions regulate the transport of wheat gamma-gliadin through the

 This information is current as of January 8, 2021

 

Permissions 8X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw153229

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists