Numerous excellent reviews on the characterization and protein chemistry of barley endosperm protein and hordein in particular have been published (Brandt et al., 1981); Ingverson, 1983; Miflin et al., 1984; Shewry & Miflin, 1982), however, no complete review of the molecular genetics of these proteins is in press. This paper hopes to concisely detail the published work on the molecular genetics of the endosperm proteins, hordein, protein Z and alpha- and beta- hordothionin. While alpha-amylase is also an important endosperm protein it has been reviewed elsewhere (Melzer & Kleinhofs, 1987).

Hordein

Hordein is a low lysine, high proline and glutamine, alcohol-soluble protein family found in barley endosperm. it is the major nitrogenous fraction of barley endosperm composing 35% to 55% of the total nitrogen in the mature grain (Kirkman et al., 1982). Hordein is synthesized on the rough endoplasmic reticulum (Matthews & Miflin, 1980) during later stages of grain fill and deposited within vacuoles in protein bodies (Miflin et al. 1981). Sodium dodecyl sulfate polyacrylamide gel electrophoresis resolves the hordein into four major peptide groups (Table 1), (Aragoncillo et al., 1981; Brandt et al., 1981; Holder & Ingverson, 1978; Salcedo et al., 1980; Shewry and Miflin, 1982).

Several groups have isolated cDNA clones for B hordeins (Brandt, 1979; Forde et al., 1981; 1985b; Hopp et al., 1983). Partial nucleotide sequence analysis of a single B1 hordein-like cDNA (pcl6) was carried out. This sequence yielded an

open reading frame of 111 nucleotides followed by three termination codons. The three C-terminal amino acids predicted by this sequence are identical to those determined by carboxypeptidase digestion of B1 hordein polypeptides (Miflin et al., 1981; Schmitt, 1979). There was also strong homology between internal predicted amino acid sequence and amino acid sequences determined from B1 polypeptides (Schmitt & Svendsen, 1980). Thus cDNA clone pc16 and its related clones were determined to be B1 hordein cDNA clones.

Hopp and co-workers (1983) isolated and Rasmussen and co-workers (1983) sequenced four additional hordein cDNA clones (pe hor2-1 through pc hor 2-4). clone pc hor2-4 is 7209 nucleotides long and codes for 181 C-terminal amino acids. Sixty one of 74 B1 hordein amino acid residues which can be compared (Schmitt & Svendsen, 1980) are identical. cDNA clone pc hor2-3 (257 bp long) encodes 54 C-terminal amino acid residues which show 9 amino acid substitutions and 21 nucleotide changes compared to pc hor2-4. Thus pc hor2-3 appears to code for a second B1 hordein. Clones pe hor2-1 and 2-2 code for the same B1 hordein as pc hor2-3, but are different sizes and contain different amounts of 3' non-coding region.

Both pc hor2-1 and 2-2 clones contain the putative polyadenylation signals AATAAA and a poly(A) tail. In clone pc hor2-2 the polyadenylation signal is 15 nucleotides upstream of the poly(A) tail Clone pc hor2-1 has an additional 63 which extend from the start of the ply(A) tail in clone pc hor2-2. There are two additional polyadenyulation signals in clone hor2-1 at 13 and 24 nucleotides upstream of the poly(A) tail. The nucleotide sequence of these two clones is identical. where they overlap, thus they may originate from transcripts from the same gene. Clone pc hor2-2 would originate from a transcript using the first polyadenylation signal, while pc hor2-1 would originate from a transcript using the third polyadenylation signal. it cannot be ruled out, however, that these two clones are derived from transcripts of two separate genes which recently arose through duplication of a single ancestral B1 hordein gene.

One cDNA clone (pc251) from the original cDNA library of Forde and co-workers (1981) has been identified by hybrid- release translation to code for C hordein (Miflin et al., 1983). A second cDNA clone (pcP387) was isolated from a new cDNA library by colony hybridization using pc251 as probe (Forde et al., 1985b). Both are less than 500 bp and represent some coding sequence as well as 3' untranslated sequence (Forde et al., 1985b). Clone pcP387 is 475 bp, of which 315 code for 105 amino acids at the C-terminus of the polypeptide. Clone pc251 is 310 bp with only 150 bp coding region. Both have been sequenced (Forde et al., 1985b) and are identical except for three nucleotides in the coding region and two nucleotides in the 3' untranslocated region. One of the nucleotide changes results in substitution of phenylalanine for leucine in clone pc251. There are two polyadenylation signals, one 6 nucleotides upstream and one 60 nucleotides upstream of the 3' end. Neither clone contains a poly(A) tail.

The most interesting feature of the two C hordein cDNA clones is that they are composed almost completely of a tandemly repeated sequence. The prototype repeat: CCACAACAACCAATTCCCCAGCAA is 24 nucleotides long and is found 13 times in clone pcP387. This repeat encodes the octapeptide Pro-Gln-Gln-Pro-Phe-Pro-Gln-Gln, obviously rich in proline and glutamine. Nucleotide substitutions and codon deletions and insertions are common and indicate degeneration of the repeats without altering the protein function. The amino acid sequence predicted from the pcP387 DNA sequence coincides very well with the amino acid composition derived from intact C hordein protein by Shewry and co-workers (1980). Clone pcP387 codes for only 20% of mature C hordein, thus the remaining 80% of the protein must contain similar repeat sequences. Direct sequencing of two small unordered chy-motryptic peptides of C hordein by Tatham and co-workers (1985) show the same octapeptide repeat.

Comparing nucleotide and amino acid sequences of C hordein with that of B1 and B3 hordein subfamilies revealed homology between the tandem repeats of C hordein and similar repeats found in the 5' region of both B hordein cDNAs. The B hordeins are divided into two domains based on the occurrence of a high proline content (33% to 40%) in the 5' end (domain 1) and proline-poor (8% to 11%) in domain 2. It is in domain I where the homology with the octapeptide of C hordein occurs. The consensus sequences for the repeats from the B and C hordein cDNAs are identical at the amino acid level and differ only at the third positions of codon 6 and 7 on the nucleotide level.

The authors have three possible explanations for the occurrence of the proline/glutamine-rich repeated octapeptide in both the C and B hordeins. One, part of the coding region of the C hordein gene may have inserted into the 5' end of a gene coding for an ancestral domain 2-like protein. Two, the C hordeins could have arisen by deletion of the domain 2 sequence from a B hordein gene, followed by duplication. Lastly, pre-existing reiterated sequences independently acquired the sequence necessary for transcription and translation. In one case it would occur by fusion to the 5' end of the domain 2-like gene.

Forde and co-workers (1983) identified three cDNA clones which code for part of a D hordein gene. All three clones (pcl33 pc150 pc135) were identified as D hordein clones by hybrid-release translation and SDS-PAGE of translation products. No further characterization of these clones was reported. They were used, however, to isolate two hexaploid wheat cDNA clones coding for high molecular weight gluten polypeptides by in situ colony hybridization (Forde et al., 1983). These wheat cDNAs were sequenced and shown to contain two interspersed tandem repeat sequences. It is likely that barley D hordein contains repetitive sequences as well. The only similarity between the hexaploid wheat gluten repeats and those of the B and C hordeins is the high content of glutamine.

Molecular analysis of storage protein mutants has been made possible by the isolation of cDNA clones for the B and C hordeins. The developmental regulation of hordein gene expression was studied by Rahman and co-workers (1984). They found hordein accumulation is controlled at the transcription level. Nitrogen (Giese & Hopp, 1984; Rahman et al., 1983) and sulfür nutrition (Shewry et al., 1983) also affect hordein expression at the transcription level. Kreis and co- workers (1983) analyzed the expression of hordein in mutant Risø 56. In vitro translations, RNA dot-blots, and Northern blots indicated that no B1 hordein mRNA is present in Risø 56. Southern blots of the genomic DNA of cv. Carlsberg II and Risø 56 revealed a deletion of at least 85 kb in mutant Risø 56 (Kries et al., 1983). Only two of eleven Hind III fragments which hybridized to pc179 (a B1 hordein cDNA) in Carlsberg II hybridized to Risø 56.

Studies with mutant Risø 1508 (Hopp et al., 1983; Kreis et al., 1984; Thompson & Bartels, 1983) indicate a major reduction in the B hordein mRNAs, specifically the B1 hordein mRNA. B1 hordein mRNA was reduced to 5% and B3 mRNA 40% in Risø 1508 compared to the parental variety Bomi (Kreis et al., 1984). C hordein mRNA was not detected in Risø 1508 (Kreis et al., 1984). D hordein related mRNAs increased approximately two fold in Risø 1508 compared to Bomi. Hopp and co-workers (1983) and Thompson & Bartels (1983) found similar reductions in B1 hordein mRNA as well as a reduction in its template activity in vitro.

Southern blots of genomic Risø 1508 DNA probed with BI hordein cDNA detect no major deletions or insertions compared with Bomi genomic blots (Hopp et al., 1983; Kreis et al., 1984). This was expected since the 1508 mutant is presumed to be "regulatory" and is located on chromosome 7, not chromosome 5 where the Hor loci are found.

Southern blot hybridizations have been used to estimate the number of hordein genes per haploid genome. Shewry and co-workers (1985) estimated that about 20-30 C hordein genes exist per haploid barley genome (cv. Sundance). About 10 genes for B1 hordein were determined per haploid barley (cv. Bomi) genome (Brandt et al., 1985; Kreis et al., 1984).

Two genomic clones, both of B1 hordein, have been isolated (Brandt et al., 1985; Forde et al., 1985a). Forde and co-workers (1985a) used a library constructed from partial EcoRI digested barley DNA (cv. Sundance) cloned into Charon 32. Plaques were screened by plaque hybridization with a mixture of pB7 (a B3 Hordein cDNA clone) and pBll (a BI Hordein cDNA clone) (Forde et al., 1985b) used as probes. Three clones were plaque purified. A single clone, gamma HvBH3.4, had a small 2.9 kb EcoR I fragment which contained the entire B1 hordein gene. It was subeloned, sequenced, and found to contain an open reading frame of 878 nucleotides which is nearly identical to pB11 the B1 hordein cDNA. There are 4 mismatches (one a replacement substitution) plus an addition of 12 nucleotides at positions 774-785 in the genome clone pBHRI85 compared with the cDNA clone pB11.

A second B1 hordein genomic clone (hor2-4) was selected by Brandt and co-workers (1985) from the variety Carlsberg II. Sequencing revealed an open reading frame coding for 271 amino acids which was identical to the 164 carboxy-terminal amino acid sequence predicted by clone pe hor2-4 (Rasmussen et al., 1983). The nucleotide sequences were also identical except for the substitution of C for A at position 353 in the genomic DNA. The 17 N-terminal amino acids of the pc hor2-4 cDNA show no resemblance to those predicted by the genomic clone and are believed to be the result of reverse transcription error.

Neither genomic clone contains introns. Both clones contain what appear to be identical signal sequences of 19 amino acids. This is in agreement with in vitro translations which yield larger percursor polypeptides (Brandt & Ingerson, 1978; Matthews & Miflin, 1980) and with the mechanisms of hordein processing into the vacuole (Cameron-Mills et al., 1978; Matthews & Miflin, 1980). The genomic clones can be divided into two domains based on proline and glutamine content and the occurrence of an octapeptide repeat in domain 1 (Forde et al., 1985a).

S1 nuclease mapping allowed Forde et al. (1985a) to locate the mRNA cap site of pBHR184. The transcription start site is at position -51 relative to the ATG codon.

The coding regions of the two genomic B1 hordein clones show very close homology except for a 69 bp deletion in the amino terminal domain 1 region of clone gamma hor2-4. The 23 amino acid deletion is in the octapeptide repeat region and may indicate that deletions and insertions in this region account for much of the size polymorphism in B hordeins. There are 54 additional nucleotide changes resulting in 24 amino acid substitutions. Seven of the amino acid changes result in changes in charge and may account for IEF heterogeneity (Faulks et al., 1981). It is most likely that the two clones represent different genes contained within the B1 hordein locus.

There is also significant sequence homology in the 3' downstream and 5' upstream regions. Three polyadenylation signals (AATAAA) occur in both clones in the region 50 to 150 bp downstream of the termination codon. A TATA-like box occurs at -79 bp in hor2-4 and at -78 in pBHR184. A CCAAT box-like sequence (CATCCAAACA) occurs at around -135 bp in both clones. Possibly even more interesting is the occurrence of a highly conserved 28 bp sequence at about -300 bp. The sequences are identical and show high homology with -300 sequences in genomic clones of zein (Langridge & Feix, 1983; Pedersen et al., 1982) and a -gliadin (Rafalski et al., 1984). Part of the -300 element has been found in the 5' regions of cloned legA, B and C genes of pea (Lycett et al., 1984, 1985). The -300 element may be critical in regulation of storage protein genes. Lycett and co-workers (1984) noted that this sequence shows striking homology with the SV40 enhancer core sequence (Weiher et al., 1983). This -300 element has not been detected in the 5' region of any other cloned cereal genes. Additional genomic clones of both B and C hordeins have been isolated by researchers at Carlsberg Laboratories, but are not yet fully characterized (Brandt, personal communication).

Protein Z

Protein Z is one of the major albumins of barley endosperm and is high in lysine, > 5% lysine residues. The purification and properties of protein Z are detailed in a paper by Hejgaard (1982). Rasmussen and co-workers (1984) cloned a cDNA coding for 18 amino acid residues of barley protein Z. The clone (pc paz1-1) was isolated from a-cDNA library by hybridizing with endosperm mRNA. Clone po pazl-1 was identified by comparison of its deduced amino acid sequence with the protein Z amino acid sequence obtained by automatic Edman degradation of the purified protein.

Using the cloned cDNA the authors were able to estimate the size of the protein Z mRNA at 1,800 b. This is sufficient to code for the 46,000 MW or 44,000 MW precursor peptides found in vitro translations (Rasmussen et al., 1984) plus leave 400 to 500 b for the 5' and 3' non-coding regions. The authors tentatively estimate five copies of the gene per haploid genome based on a Southern blot of genomic DNA digested with Eco RI and Hind III. Nielsen et al. (1983) utilized wheat barley addition lines to localize the protein Z locus (paz1) to chromosome 4. Immunological techniques were used rather than hybridization.

Six additional partial protein Z cDNA clones were isolated by Hejgaard and co-workers (1985) using pc pazl-1 (Rasmussen et al., 1984) as hybridization probe. Combining the sequence data of clones pc pazl-2 and pc pazl-3, which have a small overlapping region, yields of 695 bp sequence coding for approximately 40% of the carboxy-terminal portion of protein Z. One hundred forty base pairs of 3' non-coding sequence was cloned which contains two possible polyadenylation signals. Recently researchers at the Carlsberg Laboratories have isolated a genomic clone for protein Z (Brandt, personal communication). It is currently being characterized.

alpha- and beta-Hordothionin

Hordothionins are endosperm peptides which while orally innocuous are toxic to cultured mammalian cells, yeast, and phytopathogenic bacteria (reviewed in Garcia-Olmedo et al., 1982). They are small (5000 Da), high lysine and cystine polypeptides related to mistletoe viscotoxins and crambin of the cruciferae. Two types, alpha and beta, have been identified in barley.

Three cDNA's have been isolated which code for hordothionin (Hernandez-Lucas et al., 1986 and Ponz et al., 1986). The single alpha- (Ponz et al., 1986) and the two beta- hordothionin clones (Hernandez-Lucas et al., 1986) have been sequenced and compared. The two beta-hordothionin clones are identical except the clevage/poly(A) site of one is 4 nucleotides further downstream from the polyadenylation signal. There is 94% nucleotide homology between the coding sequences of the alpha- and beta-hordothionin clones. The deduced amino acid sequences differ in only 13 out of 127 positions. The a-and clones contain an 18 amino acid signal sequence which differs by only one amino acid at position 13.

The beta-hordothionin clones contain a single polyadenylation signal 59 nucleotides downstream of the termination codon. The alpha-hordothionin cDNA contains an additional polyadenylation signal 79 nucleotides further downstream. The two beta-hordothionin clones are unusual in that two different cleavage/poly(A) sites are under the influence of a single polyadenylation signal.

Ponz and co-workers (1986) used the alpha- cDNA clone as a probe to determine the hordothionin mRNA content of developing endosperms of Bomi and Risø 1508. Interestingly, the high-lysine mutant Risø 1508 had approximately half the hordothionin and hordothionin mRNA as its parental variety Bomi. Apparently the Risø 1508 mutation affects expression of hordothionin genes as well as beta hordein genes.

Conclusion

The molecular analysis of barley genes is just beginning. To date the endosperm proteins have received the greatest attention as evidenced by the material covered in this review.

However, the analysis of endosperm protein genes at the genomic level is just beginning, and the mechanisms by which these genes are regulated still must be investigated.

References

Aragoncillo, C., Sanchez-Monge, R. and Salcedo, G. (1981) Two groups of low molecular weight hydrophobic proteins from barley endosperm. J. Exp. Bot. 32, 1279-1286.

Brandt, A. (1979) Cloning of double stranded DNA coding for hordein polypeptides. Carlsberg Res. Commun. 44, 255-267.

Brandt, A. and Ingverson, J. (1978) Isolation and translation of hordein messenger RNA from wild type mutant in barley. Carlsberg Res. Commun. 43, 451-469.

Brandt, A., Ingverson, J., Cameron-Mills, V. Schmitt, J. M. and Rasmussen, S. K. (1981) Molecular aspects of storage protein synthesis in barley endosperms. In: Barley Genetics IV, pp. 614-622. Proc. fourth Int. Barley Genet. Symp., Edinburgh 1981. Edinburgh University Press.

Brandt, A., Montembault, A., Cameron-Mills, V. and Rasmussen, S. K. (1985) Primary structure of a B1 hordein gene from barley. Carlsberg Res. Commun. 50, 333-345.

Cameron-Mills, V., Ingversen, J. and Brandt, A. (1978) Transfer of in vitro synthesized barley endosperm proteins into the lumen of the endoplasmic reticulum. Carlsberg Res. Commun. 43, 91-102.

Faulks, A. J., Shewry, P. R. and Miflin, B. J. (1981) The polymorphism and structural homology of storage polypeptides (hordein) coded by the Hor 2 locus in barley (Hordeum vulgare L.). Biochem. Genet. 19, 841-858.

Forde, J., Forde, B. G., Fry, R. R., Kreis, M., Shewry, P. R. and Miflin, B. J. (1983) Identification of barley and wheat cDNA clones related to the high-Mr polypeptides of wheat gluten. FEBS Lett. 162, 360-366.

Forde, B. G., Heyworth, A., Pywell, J. and KreiB, M. (1985a) Nucleotide sequence of a BI hordein gene and the identification Of possible upstream regulatory elements in endosperm storage protein genes from barley, wheat and Maize. Nucl. Acids Res. 13, 7327-7339.

Forde, B. G., Kreis, M., Bahramian, M. B., Matthews, J. A., Miflin, B. J., Thompson, R. D., Bartels, D. and Flavell, R. B. (1981) Molecular cloning and analysis of cDNA sequences derived from PolyA+ RNA from barley endosperm: identification of B hordein related clones. Nucl. Acids Res. 9, 6689-6707.

Forde, B. G., Kreis, M., Williamson, M. S., Fry, R. P., Pywell, J., Shewry, P. R., Bunce, N. and Miflin, B. J. (1985b) Short tandem repeats shared by B- and C-hordein cDNAs suggest a common evolutionary origin for two groups of cereal storage protein genes. EMBO J. 4, 9-15.

Garcia-01medo, F., Carbonero, P. and Jones, B. L. (1982) Chromosomal locations of genes for wheat endosperm proteins. Adv. Cereal Sci. Technol. 5, 1-47.

Giese, H. and Hopp, H. E. (1984) Influence of nitrogen nutrition on the amount of hordein, protein Z and ~ - amylase mesenger RNA in developing endosperms of barley. Carlsberg Res. Commun. 49, 365-383.

Hejgaard, J. (1982) Purification and properties of protein Z - a major albumin of barley endosperm. Physiol. Plant 54, 174-182.

Hejgaard, J., Rasmusson, S. K., Brandt, A. and Svendsen, I. (1985) Sequence homology between barley endosperm protein Z and protease inhibitors of the -antitrypsin family. FEBS Lett. 180, 89-94.

Hernandez-Lucas, C., Royo, J., Paz-Ares, J., Ponz, F., Garcia- Olmedo, F. and Carbonero, P. (1986) Polyadenylation site heterogeneity in mRNA encoding the precursor of the barley toxin -hordothionin. FEBS Lett. 200, 103-106.

Holder, A. A. and Ingverson, J. (1978) Peptide mapping of the major components of in vitro synthesized barley hordein: Evidence for structural homology. Carlsberg Res. Commun. 43, 177-184.

Hopp, H. E., Rasmussen, S. K. and Brandt, A. (1983) Organizaiton and transcription of B1 hordein genes in high lysine mutants of barley. Carlsberg Res. Commun. 48, 20+- 216.

Ingversen, J. (1983) The molecular biology of storage protein synthesis in maize and barley endosperm. In: Seed Proteins. (eds. J. Daussant, J. Mosse, J. Vaughan), pp. 193-204. Academic Press, London.

Kirkman, M. A., Shewry, P. R. and Miflin, B. J. (1982) The effect of nitrogen nutrition on the lysine content and protein composition of the barley seed. J. Sci. Ed. Agric. 33, 115-127.

Kreis, M., Shewry, P. R., Forde, B. G., Rahman, S., Bahramian, M. B., and Miflin, B. J. (1984) Molecular analysis of the effects of the lys 3a gene on the expression of Hor loci in developing endosperms of barley (Hordeum vulgare L.). Biochem. Genet. 22, 231-255.

Kreis, M., Shewry, P. R., Forde, B. G., Rahman, S. and Miflin, B. J. (1983) Molecular analysis of a mutation conferring the high lysine phenotype on the grain of barley (Hordeum vulgare L.). Cell 34, 161-167.

Langridge, P. and Feix, G. (1983) A zein gene of maize is transcribed from 2 widely separated promoter regions. Cell 34, 1015-1022.

Lycett, G. W., Croy, R. D., Shirsat, A. H. and Boulter, D. (1984) The complete nucleotide sequence of a legumin gene from pea (Pisum. satiuum L.). Nucl. Acids Res. 12, 4493- 4506.

Lycett, G. W., Croy, R. D., Shirsat, A. H., Richards, D. M. and Boulter, D (1985) The 5'-flanking regions of three pea legumin genes: comparison of the DNA sequences. Nucl. Acids Res. 13, 6733-6734.

Matthews, B. A. and Miflin, B. J. (1980) In vitro synthesis of barley storage proteins. Planta 149, 262-268.

Melzer, J. M. and Kleinhofs, A. (1987) Molecular genetics of barley. In: Barley Genetics V. Proc. Fifth Int. Barley Genet. Sym., Kurashiki, 1986. (in press).

Miflin, B. J., Field, J. M. and Shewry, P. R. (1983) Cereal storage proteins and their effects on technological properties In: Seed Proteins (ed. J. Daussant, J. Mosse, J. Vaughan), pp. 255-319. Academic Press, London.

Miflin, B. J., Forde, B. G., Kreis, M., Rahman, S., Forde, J. and Shewry, P. R. (1984) Molecular biology of the grain storage proteins of the Triticeae. Phil. Trans. Royal Soc. (Lond.), B 304, 333-339.

Miflin, B. J., Shewry, P. R., Faulks, A. J., Forde, B. G., Kreis, M., Matthews, J., Bahramian, M. B., Thompson, R., Bartels, D. and Flavell, R. (1981) The analysis of the Hor -2 locus of barley. In: Barley Genetics IV, pp. 604- 611. Proc. Fourth Int. Barley Genet. Sym., Edinburgh, 1981. Edinburgh University Press.

Nielsen, G., Johansen, H., Jensen, J. and Hejgaard, J. (1983) Localization on barley chromosome 4 of genes coding for a -amylase (Bmyl) and protein Z (Pazl). Barley Genet. Newal. 13, 55-57.

Pederson, K., Devereux, J., Wilson, D. R., Sheldon, E. and Larkins, B. A. (1982) Cloning and sequence analysis reveal structural variation among related zein genes in maize. Cell 29, 1015-1026.

Ponz, F., Paz-Ares, J., Hernandez-Lucas, C., Garcia-Olmedo, F. and Carbonero, P. (1986) Cloning and nucleotide sequence of a cDNA encoding the precursor of the barley toxin -hordothionin. Eur. J. Biochemc. 156, 131-135.

Rafalski, J. A., Scheets, K., Metzler, M., Peterson, D. M., Hedgcoth, C., and Soll, D. (1984) Developmentally regulated plant genes: The nucleotide sequence of a wheat gliadin genomic clone. EMBO J. 3, 1409-1415.

Rahman, S., Kreis, M., Forde, B. G., Shewry, P. R. and Miflin, B. J. (1984) Hordein-gene expression during development of the barley (Hordeum vulgare) endosperm. Biochem. J. 223, 315- 322.

Rahman, S., Shewry, P. R., Forde, B. G., Kreis, M. and Miflin, B. J. (1983) Nutritional control of storage- protein synthesis in developing grain of barley (Hordeum vulgare L.). Planta 159, 366-372.

Rasmussen, S. K., Hopp, H. E., and Brandt, A. (1983) Nucleotide sequences of cDNA clones for B1 hordein polypeptides. Carlsberg Res. Commun. 48, 187-199.

Rasmussen, S. K., Hopp, H. E., Brandt, A., Sevendsen, I. B. and Hejgaard, J. (1984) A cDNA clone for protein Z, a major barley endosperm albumin. Carlsberg Res. Commun. 49, 385-390.

Salcedo, G., Sanchez-Monge, R., Argamentaria, A. and Aragoncillo, C. (1980) The A-hordeins as a group of salt soluble hydrophobic proteins. Plant Sci. Lett. 19, 109-119.

Schmitt, J. M. (1979) Purification of hordein polypeptides by comumn chromatography using volatile solvents. Carlsberg Res. Commun. 44, 431-438.

Schmitt, J. M. and Svendsen, 1. (1980) Partial amino acid sequence from hordein polypeptide B1. Carlsberg Res. Commun. 45, 549-555.

Shewry, P. R., Bunce, N. A. C., Kreis, M. and Forde, B. G. (1985) Polymorphism at the Hor 1 locus of barley (Hordeum vulgare L.). Biochem. Genet. 23, 391-404.

Shewry, P. R., Franklin, J., Parmar, S., Smith, S. J. and Miflin, B. J. (1983) The effects of sulphur starvation of the amino acid and protein composition of cereal grain. J. Cereal Sci. 1, 21-31.

Shewry, P. R., Field, J. M., Kirkman, M. A., Faulks, A. J. and Miflin, B. J. (1980) The extraction, solubility and characterization of two groups of barley storage polypeptides. J. Exp. Bot. 31, 393-407.

Shewry, P. R. and Miflin, B. J. (1982) Genes for the storage proteins of barley. Qual. Plant. Plant Goods Hum. Nutr. 31, 251-267.

Tatham, A. S., Drake, A. F. and Shewry, P. R. (1985) A conformation study of a glutamine and proline-rich cereal seed protein, C hordein. Biochem. J. 226, 557-562.

Thompson, R. D. and Bartels, D. (1983) Hordein messenger RNA levels in wild type and mutant barley endosperm. Plant Sci. Lett. 29, 295-304.

Weiher, H., Konig, M. and Gruss, P. (1983) Multiple point mutations affecting the simian virus 40 enhancer. Science 219, 626-631.