Stable expression of - Springer Link

0 downloads 0 Views 199KB Size Report
subunit genes in transgenic rye drastically increases the polymeric glutelin ... Rye flour has poor bread-making quality, despite the extensive sequence and ...
Plant Molecular Biology 54: 783–792, 2004.  2004 Kluwer Academic Publishers. Printed in the Netherlands.

783

Stable expression of 1Dx5 and 1Dy10 high-molecular-weight glutenin subunit genes in transgenic rye drastically increases the polymeric glutelin fraction in rye flourw Fredy Altpeter1,*, Juan Carlos Popelka3 and Herbert Wieser2 1

University of Florida – IFAS, Agronomy Department, Laboratory of Molecular Plant Physiology, 2191 McCarty Hall, P.O. Box 110300, Gainesville FL 32611-0300, USA (*author for correspondence, e-mail [email protected]fl.edu); 2Deutsche Forschungsanstalt fuer Lebensmittelchemie (DFA), Lichtenbergstr. 4, 85748 Garching, Germany; 3Present address: CSIRO Plant Industry, GPO Box 1600, Canberra, ACT, 2601, Australia Received 22 December 2003; accepted in revised form 6 April 2004

Key words: biolistic gene transfer, bread-making quality, high molecular weight glutenins, polymerized glutelins, transgenic rye Abstract We generated and characterized transgenic rye synthesizing substantial amounts of high-molecular-weight glutenin subunits (HMW-GS) from wheat. The unique bread-making characteristic of wheat flour is closely related to the elasticity and extensibility of the gluten proteins stored in the starchy endosperm, particularly the HMW-GS. Rye flour has poor bread-making quality, despite the extensive sequence and structure similarities of wheat and rye HMW-GS. The HMW-GS 1Dx5 and 1Dy10 genes from wheat, known to be associated with good bread-making quality were introduced into a homozygous rye inbred line by the biolistic gene transfer. The transgenic plants, regenerated from immature embryo derived callus cultures were normal, fertile, and transmitted the transgenes stably to the sexual progeny, as shown by Southern blot and SDSPAGE analysis. Flour proteins were extracted by means of a modified Osborne fractionation from wildtype (L22) as well as transgenic rye expressing 1Dy10 (L26) or 1Dx5 and 1Dy10 (L8) and were quantified by RPHPLC and GP-HPLC. The amount of transgenic HMW-GS in homozygous rye seeds represented 5.1% (L26) or 16.3% (L8) of the total extracted protein and 17% (L26) or 29% (L8) of the extracted glutelin fraction. The amount of polymerized glutelins was significantly increased in transgenic rye (L26) and more than tripled in transgenic rye (L8) compared to wildtype (L22). Gel permeation HPLC of the un-polymerized fractions revealed that the transgenic rye flours contained a significantly lower proportion of alcohol-soluble oligomeric proteins compared with the non-transgenic flour. The quantitative data indicate that the expression of wheat HMW-GS in rye leads to a high degree of polymerization of transgenic and native storage proteins, probably by formation of intermolecular disulfide bonds. Even c-40k secalins, which occur in non-transgenic rye as monomers, are incorporated into these polymeric structures. The combination 1Dx5 + 1Dy10 showed stronger effects than 1Dy10 alone. Our results are the first example of genetic engineering to significantly alter the polymerization and composition of storage proteins in rye. This may be an important step towards improving bread-making properties of rye whilst conserving its superior stress resistance. Introduction Rye has a superior biotic and abiotic stress tolerances compared to related small grain cereals w

Florida Agric. Exp. Stn. J. Series No. R-10176.

(Bushuk, 2001). Furthermore, rye represents the only hybrid crop among the small grain cereals and could serve as a model for breeding hybrids in other small grains in the future. Genetic transformation has a high potential to comple-

784 ment traditional rye breeding programs. Rye is known as one of the most recalcitrant species for regeneration from tissue cultures and genetic transformation (Castillo et al., 1994). Rye is the only cross-pollinated species among the small grain cereals. Selfing is naturally prevented by an effective gametophytic self-incompatibility mechanism. Self-fertile forms have been found in several populations and are used for developing inbred lines. The identification of inbred lines displaying a good regeneration response from tissue cultures (Popelka and Altpeter, 2001), and the optimization of gene transfer and selection parameters, significantly increased the reproducibility and efficiency of biolistic and Agrobacterium-mediated transformation experiments in rye (Popelka and Altpeter, 2003a, b). These genetic transformation protocols will support the development of genetically improved cultivars of rye in the near future. Worldwide rye is second to wheat in bread-making (Bushuk, 2001). Nevertheless bread-making quality in rye is very poor compared to wheat. The high-molecular-weight glutenin subunits (HMW-GS) of wheat are the major determinants of the gluten extensibility and elasticity (Shewry et al., 1992). HMW-GS form high molecular polymers together with the low-molecular-weight glutenin subunits (LMW-GS), giving wheat dough its superiority for making bread, pasta and other food products. While HMW-GS comprise only 5– 10% of the total wheat seed protein, differences in allelic pairs can account for more than 50% of the variation in bread-making quality of wheat (Payne et al., 1987; Lukow et al., 1989). Studies of 84 British wheat cultivars revealed that HMW-GS 1Dx5 and 1Dy10 are associated with good bread-making quality (Payne et al., 1987). An analysis of international cultivars confirmed that subunit 1Dx5, which has an additional cysteine residue, is one of the most important components of high-quality wheat (Wieser and Zimmermann, 2000). Transformation into wheat led to stable expression of transgenic HMW-GS (Altpeter et al., 1996; Blechl and Anderson, 1996) and improved functional properties under greenhouse (Barro et al., 1997) and field conditions (Vasil et al., 2001). Expression of transgenic HMW-GS in Triticum turgidum L. var. durum (He et al., 1999) and tritordeum (Rooke et al., 2000) also resulted in significant improvement in dough strength and stability. Rye is unable to form

cohesive viscoelastic gluten when flour is mixed with water, despite the extensive sequence and structure similarities of wheat and rye HMW-GS (Shewry et al., 1984; De Bustos and Jouve, 2003; Gellrich et al., 2003). Nevertheless, structural and quantitative differences between HMW secalins and HMW-GS have been proposed to be reasons for the lack of gluten formation of rye flour (Kipp et al., 1996, Ko¨hler and Wieser, 2000). Differences in functionality were clearly demonstrated by studies on gluten extensibility after the addition of reoxidized HMW-GS and HMW secalins to standard wheat flour (Kipp et al., 1996). The stable expression of superior wheat HMWGS genes 1Dx5 and 1Dy10 in transgenic rye has enabled us to investigate their interaction with different rye protein fractions and will allow analyzing quantitative and qualitative effects on functional properties of rye dough.

Results Rye inbred line L22 was chosen for transformation experiments due to its homozygous nature and superior in vitro response. Four to five days after pre-culture of immature explants, calli were bombarded with micro-particles coated with a constitutive nptII expression cassette and plasmids encoding the high-molecular-weight glutenin subunits (HMW-GS) 1Dx5 (p1Dx5) and/or 1Dy10 (p1Dy10) under control of their endogenous promoters. After two weeks of callus culture without selective agent, embryogenic calli were transferred to culture media without hormones and paromomycin sulphate selection was used for the identification and regeneration of transgenic plants. Stable expression of the selectable NPT II protein was confirmed by ELISA (data not shown). Stable expression of 1Dy10 (L26) or 1Dx5 and 1Dy10 (L8) were confirmed by SDS-PAGE of flour proteins (Figure 1A and B). While 1Dy10 is visualized on SDS page gels as an independently migrating band, 1Dx5 co-migrates with the HMW secalins and its expression is identified by the relative intensity of the band (Figure 1A). Generative progenies were produced by self fertilization of the rye inbred lines and progenies homozygous for the HMW-GS were identified by SDS-PAGE analysis of the resulting T2 sub lines. In sublines considered

785

Figure 1. SDS-PAGE analysis of flour protein composition of transgenic rye plants, stably transformed with wheat HMW-GS genes 1Dy10 and /or 1Dx5. SDS-PAGE of proteins from single seeds of rye inbred line L22 transformed with p1Dy10 (L26) or p1Dy10 and p1Dx5 (L8). Seed extracts from primary transformants (T0) (A), a segregating progeny (T1) (B, center) and a homozygous progeny (T2) (B, right) are shown in comparison to wildtype rye (L22) and wildtype wheat (Aspirant) (A, left). The sizes of protein standards are shown (A, right). Homozygous transgenic rye L8 (C) stably expressing 1Dx5 and 1Dy10 propagated in the greenhouse.

as homozygous, the presence of HMW-GS was detected in all analyzed 30 single seed extracts per subline. Also an elevated level of HMW-GS expression was observed in homozygous sublines, compared to segregating progenies (Figure 1B). Homozygous line 8 showed co-segregation of the subunits 1Dx5 and 1Dy10, while homozygous line 26 expressed just 1Dy10. Southern blot analysis was complicated by cross-hybridizing bands in wildtype rye, most probably due to the sequence similarity between wheat HMW-GS and HMW secalins. Nevertheless the individual banding pattern after hybridization with a probe from the 1Dx5 or 1Dy10 coding region confirmed the stable integration of 1Dy10 in L26 and 1Dy10 and 1Dx5 in line L8 (data not shown). Homozygous transgenic and wildtype ryes were simultaneously propagated (Figure 1C) in separate green houses under identical conditions to produce sufficient material for subsequent analysis.

The thousand kernel weight did not differ significantly between wildtype and transgenic rye expressing one (L26) or two HWG-GS (L8) (data not shown). Flour protein fractions were extracted by means of a modified Osborne fractionation and were quantified by RP-HPLC and GP-HPLC. The expression of transgenic 1Dy10 in flour of L26 represented 5.1% and the combination of transgenic 1Dy10 and 1Dx5 (L8) contributed to 16.3% of the total extracted protein. Wildtype and transgenic rye expressing one or two HMW-GS did not differ significantly in their albumin and globulin fractions. However drastic differences were observed in the ethanol-soluble prolamin and the polymerized, ethanol-insoluble glutelin fractions when wildtype and transgenic rye were compared (Table 1, Figure 2A and B). The amount of transgenic HMW-GS in homozygous rye seeds represented 17% (L26) or 29% (L8) of the extracted polymerized glutelin fraction. Compared to wildtype rye (L22; 22.4%, Table 1) the expression of

786 Table 1. Storage protein types in transgenic rye expressing 1Dy10 (L26) or 1Dy10 and 1Dx5 (L8) compared to wildtype rye and their relative distribution in the ethanol-soluble prolamin and ethanol-insoluble, polymerized glutelin fraction. Proteins/lines

Wildtype

L26 (1Dy10)

L8 (1Dy10 + 1Dx5)

Ethanol-insoluble glutelins HMW-wheat HMW-rye x 75k 40k Residue

22.4

37.4

68.1

0.0 4.0 0.0 9.8 3.6 5.0

6.2 7.4 0.0 15.6 3.6 4.6

19.6 6.3 0.0 20.8 11.6 9.8

Ethanol-soluble prolamins HMW-wheat HMW-rye x 75k 40k Residue S

77.6

62.6

31.9

0.0 4.9 13.4 29.3 25.2 4.8 100.0

0.0 1.9 14.6 19.9 22.0 4.2 100.0

0.0 0.8 8.2 6.6 13.6 2.7 100.0

(HMW-Rye = high molecular weight secalins, HMWWheat = high molecular weight glutelins, x = x-secalins, 40k = c-40k secalins, 75k = c-75k secalins, Residue = unidentified proteins).

1Dy10 (L26; 37.4%, Table 1) was associated with a 67% elevation of the proteins in the glutelin faction. The combined expression of 1Dy10 and 1Dx5 (L8;

68.1%, Table 1) resulted in a 304% elevation of the proteins in the glutelin faction compared to wildtype (L22; 22.4%, Table 1). The drastic increase of the polymerized glutelin fraction in transgenic rye was a consequence of the high-level expression of the transgenic 1Dy10 (L26) and 1Dx5 (L8), but surprisingly also HMW secalins and c-75k secalins were increased in the polymeric glutelin fraction of both transgenic rye lines (Table 1). In addition to these changes, line L8 showed 322% more c-40k secalins in the polymerized glutelin fraction than wildtype (Table 1; 11.6% versus 3.6%). The drastic increase of the polymerized glutelin fraction in transgenic rye (Figure 2A) is accompanied by a drastic reduction of the ethanol-soluble prolamin fraction (Figure 2B) compared to wildtype rye. In both transgenic rye lines, the proportions of HMW secalins, c-75k secalins c-40k secalins in the ethanol-soluble prolamin fraction were reduced. This is more pronounced in L8 expressing 1Dx5 and 1Dy10 than in L26 expressing 1Dy10. In addition, L8 shows a reduction of the x secalins in the ethanol-soluble prolamin fraction (Table 1). The quantitative changes of oligomers in the un-reduced, ethanol-soluble prolamin fractions were studied by GP-HPLC on Superdex 200. The proportions of oligomers and monomers were derived from the absorbance areas of peaks 1 and 2, respectively (Figure 3). The proportions of oligomers decrease from 39% (wildtype) to 28.6% (L26)

Figure 2. HPLC analysis of flour protein composition of transgenic rye plants, stably transformed with wheat HMW-GS genes 1Dy10 and /or 1Dx5. Flour proteins were extracted by means of a modified Osborne fractionation from wildtype and transgenic rye expressing one HMW-GS (L26) or two HMW-GS (L8). Protein fractions were quantified by RP-HPLC, and HPLC diagrams of the polymerized, ethanol-insoluble glutelin (A) and the ethanol-soluble prolamin fraction (B) are shown. (R ¼ unidentified proteins, HMW R ¼ high molecular weight secalin subunits, HMW W ¼ high molecular weight glutelin subunits, 40k ¼ c-40k secalins, 75k ¼ c-75k secalins, x ¼ x-secalins).

787

Figure 3. GP-HPLC analysis of the unreduced prolamin fractions on Superdex 200 (1 ¼ oligomers, 2 ¼ monomers).

and 16.1% (L8), and the amounts (AU/mg flour) of oligomers decreased from 321 (wildtype) to 200 (L26) and to 54 (L8). This indicates that ethanolsoluble oligomers were partially converted to insoluble glutelin aggregates by the expression of 1Dy10 and/or 1Dx5.

Discussion Rye (Secale cereale L.) is second only to wheat among the grains most commonly used in the production of bread (Bushuk, 2001). Nevertheless bread-making quality in rye is very poor compared

788 to wheat. Rye HMW-GS known as HMW secalins lack the ability to form gluten, despite their similarity to wheat HMW-GS. HMW-GS of wheat and rye share approximately 80% peptide similarity (Kipp and Wieser, 1999) and a typical structure with N- and C-terminal domains flanking the central repetitive region with tri-, hexa- and nonapeptides (De Bustos and Jouve, 2003). Differences in the number and position of cysteine residues allowed identification of rye HMW-GS orthologs of wheat HMW-GS (De Bustos et al., 2001; De Bustos and Jouve, 2003). However, both aggregative secalin types, HMW and c-75k secalins, do not support the formation of large aggregates. Potentially, intra-molecular disulphide bonds might terminate the polymerization of rye subunits (Ko¨hler and Wieser, 2000), and therefore be responsible for the poor dough-processing characteristics of rye flour. However, no conclusive relationship has been found between extra cysteines and/or other structural features and quality (Shewry et al., 2001). In this paper we therefore describe the stable expression of important components of wheat gluten quality, HMWGS 1Dx5 and 1Dy10 (Payne et al., 1987), in transgenic rye and their interaction with different rye protein fractions. Due to its superior in vitro response (Popelka and Altpeter, 2001) and homozygous nature, rye inbred line L22 was chosen for transformation experiments. The total period from culture initiation of the explants until transfer of transgenic plants to soil was less than three month (Popelka and Altpeter, 2003a); while in an earlier report more then six months were required to generate transgenic rye plants (Castillo et al., 1994). Probably as a consequence of the reduced time in tissue culture, transgenic rye plants were normal and fertile and developed like wildtype control plants and the thousand kernel weight (data not shown) did not differ between transgenic and wildtype plants. The estimation of transgene copy numbers by Southern blot analysis was complicated by cross hybridization between the wheat HMW-GS probes and wildtype rye genomic DNA, confirming the close relationship between sequences coding for wheat and rye HMW-GS. RP-HPLC revealed that homozygous transgenic rye line L26 expressed 1Dy10 at a level amounting to 5.1% of the total extracted protein, which is similar to the HMW-GS content in wheat, where HMW-GS

contribute to 5–10% of the total flour proteins (Blechl et al., 1998) (corresponding to 0.5–1.5% of the flour dry weight). Co-expression of transgenic 1Dx5 and 1Dy10 in T2 homozygous rye (L8) contributed to 16.3% of the total extracted grain protein, equivalent to 1.7% of the flour dry weight. This HMW-GS content exceeds the levels typically found in wheat. These results indicate that it is feasible to generate transgenic rye lines with medium to very high levels of HMW-GS. This also confirms that the wheat 1Dx5 and 1Dy10 regulatory and flanking sequences function very effectively as transcriptional control sequences in the heterologous genomic context of rye. In earlier reports, the stable expression of transgenic HMW-GS in wheat (Altpeter et al., 1996; Blechl and Anderson, 1996) had resulted in improved functional properties under greenhouse (Barro et al., 1997) and field conditions (Vasil et al., 2001). Expression of transgenic HMW-GS in Triticum turgidum L. var. durum (He et al., 1999) and tritordeum (Rooke et al., 2000) also caused significant improvement in dough strength and stability. Here we report for the first time the quantitative effect of transgenic HMW-GS expression on the composition of different rye protein fractions. The stable expression of 1Dy10 (L26) or 1Dy10 and 1Dx5 (L8) in transgenic rye flour resulted in a drastic increase of the polymerized glutelin fractions at the expense of the monomeric and oligomeric prolamins (Table 1, Figure 2A and B), while the proportion of the albumin/globulin fraction was not affected. Expression of 1Dy10 or the combination of 1Dy10 and 1Dx5 was associated with a 67% (L26) or a 304% (L8) elevation of the proteins in the glutelin fraction respectively (Table 1). Surprisingly the amounts of HMW secalins and c-75k secalins in the polymerized fraction were also increased in both transgenic rye lines (Table 1). Whereas their amounts were reduced in the ethanol-soluble prolamin fraction (Table 1). The quantitative changes of oligomers in the un-reduced, ethanol-soluble prolamin fractions were studied by GP-HPLC (Figure 3). Not only the proportions, but also the amounts (AU/mg flour) decreased from 321 (wildtype L22) to 200 (L26) and to 54 (L8). The amount and proportion of oligomers of wildtype rye L22 was similar to those of the rye cultivars Danko and Halo (Gellrich et al., 2003). In contrast, the oligomers of L8 were even lower than

789 oligomers in the unreduced prolamin fraction of wheat (Gellrich et al., 2003). This suggests that a large fraction of the alcohol soluble oligomers are converted to insoluble aggregates as a consequence of HMW-GS 1Dx5 and /or 1Dy10 expression. Surprisingly line L8 showed 322% more c-40k secalins in the polymerized glutelin fraction than wildtype (Table 1), associated with a reduction of the c-40k secalins in the ethanol-soluble prolamin fraction. This suggests that even c-40k secalins, which occur in nontransgenic rye as monomers, are incorporated into polymeric structures, probably by restructuring of intra-molecular disulfide bonds. The presented data indicate that the presence of wheat HMW-glutenin proteins promotes, either directly or indirectly, the aggregation or polymerization of rye proteins. The generation of transgenic rye differing in the type, number and composition of HMW-GS might allow tailoring rye flour functionality for different end uses. Several questions are raised by this report: Are the quantitative differences between the protein fractions of L8 and L26 a consequence of the amount or the type of HMW-GS expression, how are the wheat proteins affecting rye protein aggregation/polymerization and what effect will this have on the bread-making properties of rye flours?

Materials and methods Plants and explants Plants were grown under controlled environment conditions in a greenhouse at 15–20 C. Spikes were isolated with cellophane bags before anthesis to prevent cross-pollination. Surface sterilization of immature caryopses and preparation of immature embryos were previously described (Popelka and Altpeter, 2001). Culture media, callus culture and selection of transgenic events The callus induction medium consisted of MS salts (Murashige and Skoog, 1962), 30 g/l sucrose, 100 mg/l casein hydrolysate, 500 mg/l glutamine, 2.5 mg/l 2,4-D, 3.0 g/l phytagel and was supplemented for osmotic pre-culture before bombard-

ment with 72.9 g/l mannitol. Media for regeneration were prepared as callus induction media but without 2,4-D. Selective regeneration medium was supplemented with 100 mg/l paromomycin sulphate. pH of all media was adjusted to 5.8 prior autoclaving at 121 C and 1.5 bar for 20 min. Immature embryos were aseptically excised (Popelka and Altpeter, 2001) and placed scutellum side up on callus induction medium and cultured in the dark at 25 C. Immature embryos were pre-cultured for four to five days, osmotically pre-treated, bombarded, and transferred on fresh callus induction medium 12–16 h after bombardment. Three weeks later, calli were transferred into culture containers and were cultured at a 16 h photoperiod, 60 lEm)2 s)1 illumination at 25 C for shoot elongation and selection. The regeneration medium without growth regulators was supplemented with 100 mg/l paromomycin to suppress regeneration of non-transgenic plants. Regenerated, rooted plantlets were transferred to soil and were propagated under the same conditions as wildtype rye plants. Vectors and biolistic parameters Vector pJFnptII (Altpeter and Xu, 2000), containing the selectable marker-gene nptII, under control of the maize ubiquitin promoter with first intron and the 35-S terminator, inserted in the pPZP111 vector, was co-introduced by biolistic gene transfer into rye tissue cultures with p1Dx5 and p1Dy10 encoding the wheat HMW-GS 1Dx5 and 1Dy10, respectively under control of their endogenous promoters and flanked by approximately 1.5 kb of native flanking sequences. DNA coating of micro-particles was carried out using equal amounts of 0.6 and 1.0 lm microparticles from BIO-RAD and 3 lg DNA of the coprecipitated plasmids per precipitation reaction according to Sanford et al. (1991). Pre-cultured immature explants were bombarded with 35 lg micro-particles/bombardment using the BIOLISTIC Particle Delivery System (PDS)-1000/He, and 1100 psi rupture disks from Bio-Rad. Southern blot analysis Genomic DNA from transgenic plants was extracted according to Dellaporta et al. (1983), restriction digested with BglII, electrophoresed on

790 0.8% (w/v) agarose gel with 10–15 lg DNA per lane and blotted on Hybond-N membrane (Amersham) with alkaline transfer buffer (0.4 M NaOH, 0.6 M NaCl). Probes for hybridization were prepared by PCR using primers that initiate in the coding sequence of the transgene (D’Ovidio and Anderson, 1994) and p1Dx5 or p1Dy10 as a template respectively. Amplification product was electrophoresed on 1.0% (w/v) agarose gel, the fragment extracted using the QIAEXII gel extraction kit and [32P]-dCTP labeled using the random primer labeling kit from GIBCO-BRL. Hybridization was performed with Rothi-HybriQuick solution (Roth, Karlsruhe, Germany) for 24 h at 65 C in a rotating hybridization oven. The membrane was washed at 65 C with 4 · SSC for 30 min followed by 15–30 min washing with 2 · SSC, 0.1% (w/v) SDS and 1 · SSC, 0.1% (w/v) SDS. Hybridization signals were visualized by phosphor imaging system (Molecular Dynamics). Immunodetection of transgene expression For NPTII ELISA (5¢Prime fi 3¢Prime, Inc.; Boulder, CA) 40 lg of crude protein extracts (0.25 mM Tris-HCl, pH 7.8 and 0.1 mM PMSF) per micro well was used, following the manufacturers instructions. SDS-PAGE analysis of seed proteins Individual mature dry seeds were ground with a mortar and pestle. 200 ll pre-extraction buffer (2% SDS, 0.001% Pyronin Y, 10% glycerol, 0.063 M Tris-HCL ph 6.8) was added in an Eppendorf tube to eight mg of flour from individual seeds, was vortexed for 40 min at room temperature. The preextracts were centrifuged (10 min, 14,000 rpm) and the supernatant discarded. The pellet was resuspended in 200 ll extraction buffer (2% SDS, 5% b-mercaptoethanol, 0.001% Pyronin Y, 10% glycerol, 0.063 M Tris-HCL pH 6.8), vortexed for 40 min at room temperature and boiled for 5 min to denature the protein. The proteins were separated on SDS-PAGE: 3–5 ll of each sample was loaded in Ready gel, 10% Tris-HCl (Biorad). Two gels were run simultaneously in Tris-Glycine SDS buffer in a Miniprotean cell (Biorad) at a constant voltage of 40 mAmp. for 45 min. The gels were first fixed for 1 h in a solution of 1/20/79 (vol/vol/ vol) of 85% phosphoric acid, methanol, and water and then stained in Rotiblue (Roth, Karlsruhe,

Germany) for 2–15 h. Protein bands were visualized by destaining in an aqueous solution of 25% methanol for approximately 5 min or until a clear background was obtained. Screening of segregating progeny Generative progenies were produced by self fertilization of the rye inbred lines and progenies homozygous for the HMW-GS were identified by SDS-PAGE analysis of the self fertilized T2 sub lines. In sub lines considered as homozygous, the presence of HMW-GS was detected in all analyzed 30 single seed extracts per sub line. Protein chemistry of rye flour Samples of T2 grain from transgenic and wildtype rye, grown under identical conditions were milled to obtain white flour type 630 with an ash content of 0.67%. The protein content (N · 5.7) of the flour samples was determined according to Dumas. For the extraction of Osborne fractions, non-defatted flour (100 mg) was extracted stepwise three times with 1.0 ml of 0.4 mol/l NaCl + 0.067 mol/l HKNa PO4 (pH 7.6) at room temperature (RT  20 C) (albumins/globulins), twice with 1.0 ml of 60% (v/v) ethanol at RT (un-reduced, ethanol-soluble prolamins), and twice with 1.0 ml of 50% (v/v) 1-propanol + 2 mol/l urea + 0.05 mol/l Tris-HCl (pH 7.5) + 1% (w/v) dithioerythritol (DTE) under nitrogen at 60 C (polymerized glutelins) (Wieser et al., 1998). Each suspension was centrifuged for 15 min at 6000 · g and RT. The corresponding supernatants were combined and diluted to 5.0 ml (salt extracts) and 2.0 ml (alcoholic extracts), respectively. A portion of the ethanol extract (200 ll) was dried in a stream of nitrogen and reduced with the glutelin extraction solvent (200 ll) for 20 min at 60 C. RP-HPLC of the filtered extracts was performed using an instrument with a solvent module 126, a System Gold software (Beckman, Munich, Germany) and a Nucleosil 300-5-C8 column (Machery-Nagel, Du¨ren, Germany). The applied conditions were described in detail previously (Gellrich et al., 2003). The injection volumes were 250 ll for albumins/globulins, 100 ll for un-reduced and reduced, ethanol-soluble prolamins and 120 ll for ethanol-insoluble, polymerized glutelins. Eluted proteins were detected at 210 nm

791 and quantified using the corresponding absorbance areas at 210 nm which have been shown to be highly correlated with the amount of protein, independent of protein type (Wieser et al., 1998). For GP-HPLC, the un-reduced, ethanol-soluble prolamin fraction was separated on a Superdex 200 HR column (Pharmacia Biotech, Freiburg, Germany), separation range Mr ¼ 10,000– 600,000), at RT using 0.05 mol/l of sodium phosphate buffer (pH 6.9) + 0.1% (w/v) SDS as elution solvent (Gellrich et al., 2003). The injection volume was 100 ll, the flow rate was 0.6 ml/ml and the detection wavelength was 210 nm.

Acknowledgements We thank E. Gru¨tzemann and K. Scha¨fer and U. Schu¨tzler for excellent technical assistance. Expression cassette of HMW-GS 1Dx5 and 1Dy10 was kindly provided by Ann Blechl and Olin Anderson (ARS, USDA, Albany, CA, USA).

References Altpeter, F., Vasil, V., Srivastava, V. and Vasil, I.K. 1996. Integration and expression of the high-molecular-weight glutenin subunit 1Ax1 gene into wheat. Nature Biotechnol. 14: 1155–1159. Altpeter, F. and Xu, J. 2000. Rapid production of transgenic turfgrass (Festuca rubra L.) plants. J. Plant Physiol. 157: 441– 448. Barro, F., Rooke, L., Bekes, F., Gras, P., Tatham, A.S., Fido, R., Lazzeri, P.A., Shewry, P.R. and Barcelo, P. 1997. Transformation of wheat with high-molecular-weight subunit genes results in improved functional properties. Nature Biotechnol. 15: 1295–1299. Blechl, A.E. and Anderson, O.D. 1996. Expression of a novel high molecular weight glutenin subunit gene in transgenic wheat. Nature Biotechnol. 14: 875–879. Blechl, A.E., Le, H.Q. and Anderson, O.D. 1998. Engineering changes in wheat flour by genetic transformation. J. Plant Physiol. 152: 703–707. Bushuk W. 2001. RYE: Production, Chemistry, and Technology, Second Edition AACC, St Paul, MN. Castillo, A.M., Vasil, V. and Vasil, I.K. 1994. Rapid production of fertile transgenic plants of rye (Secale cereale L.). Bio/ Technology 12: 1366–1371. Dellaporta, S.L., Wood, J. and Hicks, J.B. 1983. A plant DNA minipreparation: Version II. Plant Mol. Biol. Rep. 4: 19–21. De Bustos, A. and Jouve, N. 2003. Characterisation and analysis of new HMW-glutenin alleles encoded by the GluR1 locus of Secale cereale. Theor. Appl. Genet. 107: 74–83.

De Bustos, A., Rubio, P. and Jouve, N. 2001. Characterisation of two gene subunits on the 1R chromosome of rye as orthologs of each of the Glu-1 genes of hexaploid wheat. Theor. Appl. Genet. 103: 733–742. D’Ovidio, R. and Anderson, O.D. 1994. PCR analysis of genes encoding allelic variants of high-molecular-weight glutenin subunits at the Glu-D1 locus. Theor. Appl. Genet. 88: 175– 180. Gellrich, C., Schieberle, P. and Wieser, H. 2003. Biochemical characterization and quantification of storage protein (secalin) types in rye flour. Cereal Chem. 80: 102–109. He, G.Y., Rooke, L., Steele, S., Be´ke´s, F., Gras, P., Tatham, A.S., Fido, R., Barcelo, P., Shewry, P.R. and Lazzeri, P. 1999. Transformation of pasta wheat (Triticum turgidum L. var. durum L.) with HMW glutenin subunit genes and modification of dough functionality. Mole. Breeding 5: 377– 386. Kipp, B., Belitz, H.-D., Seilmeier, W. and Wieser, H. 1996. Comparative studies of high Mr subunits of rye and wheat. I. Isolation and biochemical characterisation and effects on gluten extensibility. J. Cereal Sci. 23: 227–234. Kipp, B. and Wieser H. 1999. Comparative Studies of High Mr Subunits of Rye and Wheat. II. Partial Amino Acid Sequences. J. Cereal Sci. 30: 303–313. Ko¨hler, P. and Wieser, H. 2000. Comparative studies of high Mr subunits of rye and wheat. III. Localisation of cysteine residues. J. Cereal Sci. 32: 189–197. Lukow, O.M., Payne, P.I. and Tkachuk, R. 1989. The HMW glutenin subunit composition of Canadian wheat cultivars and their association with bread-making quality. J. Sci. Food. Agric. 46: 451–460. Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Plant. Physiol. 15: 473–497. Payne, P.I., Nightingale, M.A., Krattiger, A.F. and Holt, L.M. 1987. The relation between HMW glutenin subunit composition and the bread-making quality of British-grown wheat varieties. J. Sci. Food Agric. 40: 51–65. Popelka, J.C. and Altpeter, F. 2001. Interactions between genotypes and culture media components for improved in vitro response of rye (Secale cereale L.) inbred lines. Plant Cell Rep. 20: 575–582. Popelka, J.C. and Altpeter, F. 2003a. Evaluation of rye (Secale cereale L.) inbred lines and their crosses for tissue culture response and stable genetic transformation of homozygous rye inbred line L22 by biolistic gene transfer. Theor. Appl. Genet. 107: 583–590. Popelka, J.C. and Altpeter, F. 2003b. Agrobacterium tumefaciens-mediated genetic transformation of rye (Secale cereale L.). Mol. Breeding 11: 203–211. Rooke, L., Barro, F., Tatham, A.S., Fido, R., Steele, S., Be´ke´s, F., Gras, P., Martin, A., Lazzeri, P.A., Shewry, P.R. and Barcelo, P. 2000. Altered functional properties of tritordeum by transformation with HMW glutenin subunit genes. Theor. Appl. Genet. 99: 851–858. Sanford, J.C., De Vit, M.J., Russel, J.A., Smith, F.D., Harpening, P.R., Roy, M.K. and Johnston, S.A. 1991. An Improved, helium-driven biolistic device. Technique 3: 3–16. Shewry, P.R., Halford, N.G. and Tatham, A.S. 1992. High molecular weight subunits of wheat glutenin. J. Cereal Sci. 15: 105–120.

792 Shewry, P.R., Miflin, B.J. and Kasarda, D.D. 1984. The structural and evolutionary relationships of the prolamin storage proteins of barley, rye and wheat. Phil. Trans. R. Soc. Lond. B 304: 297–308. Shewry, P.R., Popineau, Y., Lafiandra, D. and Belton, P. 2001. Wheat glutenin subunits and dough elasticity: findings of the EUROWHEAT project. Trends Food Sci. Tech. 11: 433–441. Vasil, I.K., Bean, S., Zhao, J., McClusskey, P., Lookhart, G., Zhao, H.-P., Altpeter, F. and Vasil, V. 2001. Evaluation of baking properties and gluten protein composition of field

grown transgenic wheat lines expressing high molecular weight glutenin gene 1Ax1. J. Plant Physiol. 158: 521–528. Wieser, H., Antes, S. and Seilmeier, W. 1998. Quantitative determination of gluten protein types in wheat flour by reversed-phase high-performance liquid chromatography. Cereal Chem. 75: 644–650. Wieser, H. and Zimmermann, G. 2000. Importance of amounts and proportions of high molecular weight subunits of glutenin for wheat quality. Eur. Food Res. Technol. 210: 324–330.