FEBS 17723 FEBS Letters 396 (1996) 285 288
Disulphide structure of a sunflower seed albumin: conserved and variant
disulphide bonds in the cereal prolamin superfamily
T.A. Egorov ~, T.I. Odintsova b, A.Kh. Musolyamov ~, R. Fido c, A.S. Tatham c, P.R. Shewry c,*
~Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, 32 Vavilova Str., Moscow 117984, Russian Federation
~' Vavilov Institute of General Genetics, Russian Academy of Sciences, 3 Gubkina Str., Moscow 117809, Russian Federation
~IA CR-Long Ashton Research Station, Department of Agricultural Sciences, University of Bristol, Long Ashton, Bristol BSI8 9AF, UK
Received 2 September 1996; revised version received 20 September 1996
Abstract Disulphide mapping of a methionine-rich 2S albumin
from sunflower seeds showed four intra-chain disulphide bonds
which are homologous with those in a related heterodimeric
albumin from lupin seeds (conglutin 5). Similar conserved
disulphide bonds are also present in c~-gliadin and T-gliadin
storage proteins of wheat, but a lower level of conservation is
present in a further related group of proteins, the cereal
inhibitors of c~-amylase and trypsin. These differences may
relate to the different functions of the proteins.
Key words." Seed; Cereal; Sunflower; Wheat; Albumin ;
Inhibitor; Gliadin; Disulfide bond
The cereal prolamin superfamily of seed proteins comprises
three major groups of proteins: the storage prolamins of bar-
ley, wheat and rye, the cereal inhibitors of m-amylase and
trypsin and 2S albumin storage proteins present in seeds of
a range of dicotyledonous plants [1 3]. Although these three
groups of proteins differ widely in their properties (including
Mr and amino acid sequences) they are characterized by the
presence of conserved cysteine residues. This conservation is
assumed to reflect a crucial role of these cysteine residues in
stabilizing the protein structure via disulphide bond forma-
tion, but this remains to be established as few components
have been disulphide mapped, and no detailed comparisons
have been reported.
2S albumin storage proteins have been characterized from a
diverse range of plants, including legumes (lupins, peas), cru-
cifers (radish, oilseed rape, Arabidopsis), cotton, sunflower,
Brazil nut and castor bean . These are typically heterodi-
meric proteins, with large and small subunits of Mr about
9000 and 4000, respectively, which arise from post-transla-
tional proteolysis of a precursor protein. One such component
has been disulphide mapped, the conglutin 5 of lupin. How-
ever, post-translational proteolysis does not occur in sun-
flower, and the albumin fraction consists of monomeric pro-
teins of Mr about 10000 18000 [5,6]. We have therefore
determined the disulphide structure of one such sunflower
albumin, a methionine-rich component SFA8 . The pattern
of disulphide bonds is identical to that in the heterodimeric
conglutin 5  and comparison with structures reported for
members of the prolamin and inhibitor groups shows the pres-
ence of conserved and variant disulphide bonds in this super-
family of seed proteins.
*Corresponding author: Fax: +44 (0)1275 394299.
0014-5793/96l$12.00 © 1996 Federation of European Biochemical Societies.
PHS0014-5793(96)01 1 17-9
2. Materials and methods
5 M cyanogen bromide in acetonitrile was obtained from Aldrich.
Guanidine hydrochloride and Tris were obtained from Sigma. Se-
quence grade trifluoroacetic acid (TFA) was from Applied Biosystems
and HPLC grade acetonitrile from Merck. Water was purified using a
Milli Q System (Millipore). All other reagents were analytical grade.
2.2. Protein purification
SFA8 was purified from sunflower Hybrid 246 as described by 
with an additional final separation by RP-HPLC on a SynChropak
RP-P C18 Semi-preparative column (10X 250 ram) with a linear gra-
dient (25% to 55%) of acetonitrile containing 0.07% (v/v) TFA in
0.05% (v/v) aqueous TFA.
2.3. Cyanogen bromide cleavage
0.5 mg (--40 nmol) of purified SFA8 was dissolved in 100 I.tl of 70%
(v/v) TFA and 25 I11 of 5 M CNBr in acetonitrile (about 200 molar
excess over methionine residues) added. The reaction mixture was
flushed with argon and incubated at 22°C for 18 h in the dark. After
drying the peptides were dissolved in 50 pl of 5 M guanidine hydro-
chloride in 0.1% (v/v) TFA and separated by RP-HPLC using an
Aquapore RP-300 C8 column (4.6 x 220 mm) with a linear gradient
(0 to 40%) of acetonitrile containing 0.08°/,, (v/v) TFA in 0.1% (v/v)
2.4. Identification of disulphide-bonded peptides
About 10% of each dried fraction was redissolved in 50 ~tl of 0.1 M
Tris-HC1 buffer, pH 8, containing 5 mM EDTA and 5 M guanidine
hydrochloride. Reduction of disulphide bonds and alkylation of sul-
phydryl groups was carried out by adding 1.4 ~tmol of dithiothreitol
dissolved in 1 ~tl of distilled water, standing for 30 min at 22°C, and
adding 1 ~tl (9.3 pznol) of 4-vinylpyridine. The reaction mixture was
then separated by RP-HPLC as described for the cyanogen bromide
digest, in order to identify fractions giving two peaks.
2.5. N-terminal amino acid sequencing
N-terminal amino acid sequence analysis was carried out with a
Model 816 protein sequencer (Knauer, Berlin) equipped with a Model
120A PTH Analyser (Applied Biosystems) operated according to the
manufacturer's instructions. Peptide samples were applied onto Im-
mobilon membrane in 30% (v/v) acetonitrile containing 0.1% (v/v)
N-terminal amino acid sequencing of the SFA8 fraction
purified from sunflower Hybrid 246 showed that it was iden-
tical for 20 residues to the sequence reported by Kortt et al.
. The latter showed the presence of eight cysteine residues
and 16 methionines in a protein of 103 residues (Fig. 1A).
With the exception of two adjacent residues (Cys-51 and
Cys-52), all the cysteines are separated by at least one methio-
nine. The purified protein was therefore cleaved with cyano-
gen bromide and 11 fractions were purified by RP-HPLC
(Fig. 2). Re-separation of each fraction after reduction and
All rights reserved.
ZA. Egorov et al./FEBS Letters 396 (1996) 285~88
A PROTEIN SEQUENCE
B PEPTIDE SEQUENCES
87AHNL P I ECNLM 97
C DISULPEIDE MAP
ii 24 51 52 62 64 94 I01 1
Fig. 1. Disulphide bond mapping of the sunflower albumin SFA8. (A) The protein sequence reported by Kortt et al. . (B) The sequences of
disulphide-bonded peptides isolated as fractions 4, 7 and 11. (C) Disulphide map of the protein.
alkylation showed that three of them (nos. 4, 7, 11) consisted
of two or more peptides linked by disulphide bonds. These
fractions were therefore subjected to N-terminal sequence
analysis, both before and after reduction of disulphide bonds.
This showed that fraction 4 consisted of two peptides linked
by a disulphide bond between Cys-64 and Cys-101 and frac-
tion 7 of two peptides with a disulphide bond between Cys-11
and Cys-62. Fraction 11 consisted of three peptides with
Cys-24 and Cys-94 linked to the adjacent cysteine residues
Cys-51 and Cys-52, but it was impossible to identify the pre-
cise bonds involving these two residues (Fig. 1B). The disul-
phide structure of SFA8 was therefore established as shown in
The disulphide structure of one heterodimeric 2S albumin,
namely conglutin 8 from lupin which has nine cysteines, has
T.A. Egorov et aL/FEBS Letters 396 (1996) 285-288
/0 20 30 40
Fig. 2. Separation of products of cyanogen bromide cleavage of intact SFA8 (approx. 0.5 mg) by reversed-phase HPLC on an Aquapore RP-
300, C8 column (4.6 × 220 ram). Peptides were eluted with a linear gradient of acetonitrile in the presence to TFA (from 0% B to 40% B for 90
min) at a flow rate of 0.5 ml/min. Solvent A: 0.1% (v/v) aqueous TFA; solvent B: acetonitrile containing, 0.08% (v/v) TFA. Fractions 4, 7 and
11 contained disulphide-bonded peptides.
been reported . Despite a low level of overall sequence
identity to SFA8 (17% homology), eight of the cysteine resi-
dues in conglutin 8 are clearly homologous, based on position
and sequence context, to those in SFA8 (Fig. 3). These cys-
teine residues, called A-H in Fig. 3, also form the same 4
disulphide bonds, two of which are inter-chain in conglutin
8. In neither case was it possible to discriminate between the
bonds involving the adjacent cysteine residues (C and D). The
ninth cysteine residue in conglutin 8, Cys-45, is not present in
SFA8 and is unpaired. This is called Cys J in Fig. 3.
Six of the eight cysteine residues present in SFA8 and con-
glutin 8 (cysteines B, C, D, F, G, H) are also conserved in the
C-terminal domains of two types of monomeric wheat prola-
min, called c~- and [3-gliadins (Fig. 3). These six conserved
cysteine residues also form the same three disulphide bonds
in the ~x- and [3-gliadins as in the 2S albumins (B to C or D, C
or D to G, F to H), while the 7-gliadin also contains an
additional pair of cysteine residues (called K and L in Fig.
3) which form a fourth disulphide bond [9-11].
The disulphide structures of three cereal cx-amylase/trypsin
inhibitors have been reported: the monomeric 0.28 (WAI-
0.28)  and tetrameric 0.53 (WAI-0.53)  (x-amylase in-
hibitors from wheat (Fig. 3), and a bifunctional c~-amylase/
trypsin inhibitor (RBI) from ragi (Indian finger millet) 
(not shown). Again it is possible to identify conserved cysteine
residues homologous to those present in the 2S albumins and
the gliadins, but the patterns of disulphide bond formation are
WAI-0.28 and RBI both have 10 cysteine residues, corre-
sponding to cysteines A to H and K in the gliadins and the 2S
albumins and a single unique cysteine residue called M. The
disulphide maps of these proteins were determined using
NMR spectroscopy (RBI) and mass spectroscopy (WAI-
0.28), showing identical patterns. Also, in both cases, it was
possible to distinguish between disulphide bonds involving the
adjacent cysteine residues (C and D), in contrast to the other
proteins. These studies show that 4 of the 5 disulphide bonds
(A to E, B to C, D to G and F to H) are conserved, with the
fifth bond between cysteine K (which is paired with L in the
7-gliadin) and M. The third inhibitor that has been disulphide-
mapped, WAI-0.53, shares 9 cysteine residues with WAI-0.28
and RBI, lacking cysteine residue E. The absence of cysteine E
is associated with the formation of a new disulphide bond
between cysteines A and H, leaving cysteine F unpaired.
These comparisons show a high degree of conservation of
disulphide bonds within and between the gliadins and 2S al-
bumins, involving cysteine residues B to C/D, C/D to G and F
to H. In addition, the y-gliadin and 2S albumins each contain
a single additional disulphide bond, between cysteines K to L
and A to E, respectively. However, only two disulphide bonds
involving the adjacent cysteine residues (C/D) are also con-
served in the inhibitors, with differences between RBI/WAI-
0.28 on the one hand and WAI-0.53 on the other. Thus, their
different disulphide bond patterns may relate to the mainte-
nance of active sites for their target enzymes.
 Kreis, M., Forde, B.G., Rahman, S., Miflin, B.J. and Shewry,
P.R. (1985) J. Mol. Biol. 183, 499-502.
 Kreis, M., Shewry, P.R., Forde, B.G., Forde, J. and Miflin, B.J.
(1985) in: Oxford Surveys of Plant Cell and Molecular Biology,
vol. 2 (Miflin, B.J. ed.) pp. 253 317, Oxford University Press,
 Kreis, M. and Shewry, P.R. (1989) Bio-Essays 10, 201 207.
 Shewry, P.R. (1995) Biol. Rev. 70, 375426.
 Kortt, A.A. and Caldwell, J.B. (1990) Phytochemistry 29, 2805-
 Anisimova, I.N., Fido, R.J., Tatham, A.S. and Shewry, P.R.
(1995) Euphytica 83, 15-23.
 Kortt, A.A., Caldwell, J.B., Lilley, G.G. and Higgins, T.J.V.
(1991) Eur. J. Biochem. 195, 329 334.
288 Download full-text
T.A. Egorov et al./FEBS Letters 396 (1996) 285 288
51 52 62 64
C D EF
G HI 1°3
8 20 37 1 17 18 29 31
A B || C D EF
45 65 73
J G H
184 191 192 204
t I i
150 151 163
",( i i
WAI - 0.28
WAI - 0.53
7 21 29 42 4354 56
I )k I
6 20 28 41 42
K F ]
Fig. 3. Comparison of the patterns of disulphide bonds formed by 2S albumins, cereal inhibitors and gliadins. Cysteine residues are labelled
A M as discussed in the text. The disulphide maps for conglutin 6, WAI-0.28, WAI-0.53, ct-gliadin and 7-gliadin are reported in [8-14].
 Lilley, G.G. and Inglis, A.S. (1966) FEBS Lett. 195, 235 241.
 Muller, S. and Wieser, H. (1995) J. Cer. Sci. 22, 21 27.
 Wieser, H, and Muller, S. (1996) in: Wheat Biochemistry (Scho-
field, J.D. ed.) Royal Society of Chemistry, London, in press.
 Egorov, T.A., Musolyamov, A.K., Barbashov, S.F., Zolotykh,
O., Andersen, J., Roepstorff, P. and Popineau, Y. (1994) in:
Wheat Kernel Proteins. Molecular and Functional Aspects, Pro-
ceedings of the International Meeting, Viterbo, Italy, pp. 4144.
 Poerio, E., Caporale, C., Carrano, L., Pucci, P. and Buonocore,
V. (1991) Eur. J. Biochem. 199, 595-600.
 Maeda, K., Wakabayashi, S. and Matsubara, H. (1993) J. Bio-
chem 94, 865-870.
 Strobl, S., Miihlhahn, P., Bernstein, R., Wiltscheck, R., Maskos,
K., Wunderlich, M., Huber, R., Glockshuber, R. and Holak,
T.A. (1995) Biochem. 34, 8281 8293.