Tài liệu Báo cáo khoa học: Purification and structural characterization of a D-amino acid-containing conopeptide, conomarphin, from Conus marmoreus docx

LINK DOWNLOAD MIỄN PHÍ TÀI LIỆU "Tài liệu Báo cáo khoa học: Purification and structural characterization of a D-amino acid-containing conopeptide, conomarphin, from Conus marmoreus docx": http://123doc.vn/document/1048361-tai-lieu-bao-cao-khoa-hoc-puri-cation-and-structural-characterization-of-a-d-amino-acid-containing-conopeptide-conomarphin-from-conus-marmoreus-docx.htm

Purification and structural characterization of a D-amino
acid-containing conopeptide, conomarphin, from
Conus marmoreus
Yuhong Han
1,2,
*, Feijuan Huang
3,
*, Hui Jiang
1,4
,LiLiu
1
, Qi Wang
1
, Yanfang Wang
1
, Xiaoxia Shao
1
,
Chengwu Chi
1,2
, Weihong Du
3,
* and Chunguang Wang
1
1 Institute of Protein Research, Tongji University, Shanghai, China
2 Institute of Biochemistry and Cell Biology, Shanghai Institute of Biology Sciences, Chinese Academy of Sciences, China
3 Department of Chemistry, Renmin University of China, Beijing, China
4 Research Institute of Pharmaceutical Chemistry, Beijing, China
Conus snails are a group of predatory mollusks living
in tropical oceans all over the world. They can pro-
duce highly diversified conotoxins for predation and
defense. Conotoxins are believed to number about
50 000, and could serve as a rich source of active
compounds for neuroscience research and nervous
system disease therapy [1]. Conotoxins are mainly
disulfide bond-rich peptides of 10–40 residues. A
small number of conotoxins have zero (Table 1) or
only one disulfide bond. All conotoxins are classified
into different families on the basis of the Cys frame-
work in the primary sequence and their different tar-
gets [1].
It is known that each conotoxin is encoded by an
individual mRNA. The original translation products of
conotoxin genes are in most cases composed of a sig-
nal peptide, a propeptide, and mature conotoxins at
the C-terminus. On the basis of the conserved signal
peptide sequences, conotoxins of different families are
grouped into several major superfamilies: A, M, O, I,
T, P, L, and S [2]. For example, most of the M-super-
family conotoxins have the ‘-CC-C-C-CC-’ pattern,
Keywords
conomarphin; Conus marmoreus;
D-Phe;
M-superfamily; NMR structure
Correspondence
C. Wang, Institute of Protein Research,
Tongji University, 1239 Siping Road,
Shanghai 200092, China
Fax: 86 21 65988403
Tel: 86 21 65984347
E-mail: chunguangwang@mail.tongji.edu.cn
W. Du, Department of Chemistry, Renmin
University of China, 59 Zhong Guan Cun
Street, Beijing 100872, China
Fax: +86 10 62516444
Tel: +86 10 62512660
E-mail: whdu@chem.ruc.edu.cn
*These authors contribute equally to this
paper
(Received 7 November 2007, revised 24
January 2008, accepted 22 February 2008)
doi:10.1111/j.1742-4658.2008.06352.x
Cone snails, a group of gastropod animals that inhabit tropical seas, are
capable of producing a mixture of peptide neurotoxins, namely conotoxins,
for defense and predation. Conotoxins are mainly disulfide-rich short pep-
tides that act on different ion channels, neurotransmitter receptors, or
transporters in the nervous system. They exhibit highly diverse composi-
tions, structures, and biological functions. In this work, a novel Cys-free
15-residue conopeptide from Conus marmoreus was purified and designated
as conomarphin. Conomarphin is unique because of its d-configuration
Phe at the third residue from the C-terminus, which was identified using
HPLC by comparing native conomarphin fragments and the corresponding
synthetic peptides cleaved by different proteases. Surprisingly, the cDNA-
encoded precursor of conomarphin was found to share the conserved signal
peptide with other M-superfamily conotoxins, clearly indicating that cono-
marphin should belong to the M-superfamily, although conomarphin
shares no homology with other six-Cys-containing M-superfamily conotox-
ins. Furthermore, NMR spectroscopy experiments established that cono-
marphin adopts a well-defined structure in solution, with a tight loop in
the middle of the peptide and a short 3
10
-helix at the C-terminus. By con-
trast, no loop in l-Phe13-conomarphin was found, which suggests that
d-Phe13 is essential for the structure of conomarphin. In conclusion, cono-
marphin may represent a new conotoxin family, whose biological activity
remains to be identified.
1976 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS
but some of them have been found to act on the
Na
+
-channel, K
+
-channel, and nicotinic acetylcholine
receptor [3–5]. Furthermore, different disulfide bond
linkages are formed with this same Cys framework
[3,6,7]. The mechanisms that lead to conservation of
signal peptides, particularly as compared to the
sequence hyperdivergence of mature conotoxins,
remain a subject for further study.
Another striking feature of conotoxins is the high
content of different post-translational modifications
[8]. A variety of post-translational modifications have
been found in conotoxins, such as C-terminal amida-
tion, hydroxylation of Pro, Val or Lys, c-carboxylation
of Glu, glycosylation of Ser or Thr, bromination of
Trp, sulfation of Tyr, cyclization of N-terminal Gln,
and epimerization of several different residues. It is
quite rare for such a variety of post-translational mod-
ifications to take place in a defined cluster of gene
products. Cone snails have probably developed a com-
plicated but delicate machinery to carry out these
modifications, which might be critical for the structure
and function of conotoxins.
Epimerization, namely converting an amino acid
residue in a peptide chain from the l-configuration
to the d-configuration, was first identified in dermo-
phin, an opiate-like peptide from the skin of the
American frog Phyllomedusa [9]. Later, this was
found in other toxins and peptides, such as the spi-
der toxin x-agatoxin [10], C-type natriuretic peptide
from the Australian platypus [11], fulicin from Afri-
can giant snails [12], and contryphan from cone
snails (Table 2). The first d-residue containing cono-
toxin, contryphan-R, was purified from Conus radia-
tus [13]. To date, a series of contryphans has been
identified [14–17]. A group of I-superfamily conotox-
ins has also been found to contain a d-residue [18].
Recently, two other families of d -residue-containing
conopeptides, conophan and conomap, were identified
biochemically [19,20]. In comparison with other post-
translational modifications, such as C-terminal amida-
tion, Pro hydroxylation, or Glu c-carboxylation,
which exist in many conotoxin families, residue epi-
merization from the l-configuration to the d-configu-
ration is relatively rare [8].
In this work, we purified a novel 15 amino acid pep-
tide from the venom of Conus marmoreus. This peptide
was found to be particularly unique because it contains
no Cys residues, a hydroxylated Pro at position 10,
and a d-Phe at position 13. Its cDNA sequence indi-
cates that it belongs to the M-superfamily, albeit it
shares no sequence homology with other M-superfam-
ily conotoxins. Its solution structure, including the
Table 1. The identified cysteine-free conopeptide families. c, c-carboxylated glutamate; *, amidated C-terminus;

, glycosylation; O, hydroxy-
proline;
V, D-Val; F, D-Phe; NMDA, N -methyl-D-aspartate.
Family Example Sequence Target Reference
Conantokin Conantokin-G GEccLQcNQcLIRcKSN* NMDA receptor McIntosh et al. [51]
Contulakin Contulakin-G ZSEEGGSNAT

KKPYIL Neurotensin receptor Craig et al. [52]
Conorfamide Conorfamide GPMGWVPVFYRF* FR amide receptor Maillo et al. [53]
Conomap Conomap-Vt A
FVKGSAQRVAHGY* Unknown Dutertre et al. [19]
Conophan Conophan gld-V AOANS
VWS Unknown Pisarewicz et al. [20]
Conomarphin Conomarphin DWEYHAHPKONS
FWT Unknown This work
Table 2. D-Amino acid-containing peptides from different organisms. D-Amino acid residues are underlined. *, amidated C-terminus; O,
hydroxyproline; c, c-carboxylated glutamate.
Organism Example Sequence Position Reference
Cone snail Conomarphin DWEYHAHPKONS
FWT )3 This work
r11a GOSFCKADEKOCEYHADCCNCCLSGICAOSTNWILPGCSTSSF
FKI )3 Buczek et al. [35]
Contryphan-R GCO
WEPWC* )5 Jimenez et al. [13]
Glacontryphan-M NcScCP
WHPWC* )5 Hansson et al. [16]
Conophan gld-V AOANS
VWS )3 Pisarewicz et al. [20]
Conomap-Vn A
FVKGSAQRVAHGY* +2 Dutertre et al. [19]
Snail Fulicin F
NEFV* +2 Ohta et al. [12]
Frog Dermorphin Y
AFGYPS* +2 Montecucchi et al. [9]
Spider x-Agatoxin EDNCIAEDYGKCTWGGTKCCRGRPCRCSMIGTNCECTPRLIMEGL
SFA )3 Kuwada et al. [10]
Australian
platypus
C-type natriuretic
peptide
L
LHDHPNPRKYKPANKKGLSKGCFGLKLDRIGSTSGLGC +2 Torres et al. [11]
Defensin-like peptide I
MFFEMQACWSHSGVCRDKSERNCKPMAWTYCENRNQKCCEY +2 Torres et al. [39]
Y. Han et al. A
D-amino acid-containing conomarphin
FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS 1977
effect of the d-Phe on structure, was also studied. The
unusual structure of conomarphin further demon-
strates the high diversity of conotoxins.
Results
Purification and primary sequence of
conomarphin
As previously reported, the crude venom of
C. marmoreus was separated into two peaks on a gel
filtration column (Fig. 1A); the second one contained
mainly peptides [6]. Separation using an RP-HPLC
C-18 column provided at least 19 major peaks as well
as many minor peaks from the peptide fraction
(Fig. 1B). Every major peak was collected, repurified,
and sequenced. The peak that eluted at 53 min corre-
sponded to a novel 15 amino acid peptide with the
sequence DWEYHAHPKONSFWT, where O is
hydroxylated Pro. The M
r
of the peptide, 1931.1,
matched perfectly with the calculated one, 1931.05,
indicating that this was the complete sequence of this
peptide. This sequence shares no homology with that
of any other known conotoxin. This peptide was
named conomarphin, indicating that it is a Cys-free
conotoxin and originated from C. marmoreus.
D-Phe13 in conomarphin
In order to obtain more material for structural studies,
conomarphin was chemically synthesized using stan-
dard Fmoc–l-amino acids. However, to our great sur-
prise, the synthesized peptide did not have the same
retention time on a C-18 HPLC column as the natural
one (supplementary Fig. S1), although the peptides
have identical sequences and relative molecular masses.
The only explanation for this is that natural conomar-
phin must have one or more d-amino acids.
The strategy of protease digestion was employed to
locate the d-amino acid(s) in conomarphin. As Lys9 is
the only basic residue in this peptide, trypsin was first
used to cleave both natural conomarphin and the syn-
thetic peptide DWEYHAHPKONSFWT. The diges-
tion of natural conomarphin gave the expected results:
two fragments and the intact peptide with relative
molecular masses identical to the corresponding calcu-
lated ones (supplementary Fig. S2A). However, for the
synthetic all-l-amino acid conomarphin, trypsin
cleaved at two sites, Lys9-Hyp10 and Asn11-Ser12
(Fig. 2B). The second cleavage site was unexpected, as
trypsin usually only cleaves after a basic residue. Com-
parison between the two digestion products demon-
strated that the difference between natural and
synthetic conomarphin came from the C-terminal
fragment ONSFWT, which had an identical relative
molecular mass but a different retention time (P2 in
supplementary Fig. S2A and P3 in supplementary
Fig. S2B).
To narrow the range of the possible position of the
d-amino acid, chymotrypsin was used to digest natu-
ral conomarphin and the synthetic peptide DWEY-
HAHPKONSFWT. The cleaved fragments were
analyzed on a C-18 HPLC column and were assigned
on the basis of their relative molecular masses. The
shorter C-terminal fragment SFWT of natural cono-
marphin and the synthetic peptide exhibited different
Fig. 1. Purification of conomarphin from the venom of
C. marmoreus. (A) The crude venom was separated into two main
peaks on a Sephadex G-25 column (100 · 2.6 cm). (B) The second
peak was further separated on an HPLC C-18 column
(9.4 · 250 mm) with an elution gradient of 0–10 min 100%
Buffer A, 10–20 min 0–27% Buffer B, 20–25 min 27% Buffer B,
25–58 min 27–42.5% Buffer B, and 58–63 min 42.5–100%
Buffer B. Buffer A is 0.1% trifluoroacetic acid and Buffer B is 0.1%
trifluoroacetic acid in 70% (v ⁄ v) acetonitrile. The flow rate was
2mLÆmin
)1
. The peak labeled with an asterisk is conomarphin.
A
D-amino acid-containing conomarphin Y. Han et al.
1978 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS
retention times, which indicated that one or more of
the four C-terminal amino acids of natural conomar-
phin should be in the d-configuration (supplementary
Fig. S3).
As chymotrypsin cleaved the Asn11-Ser12 peptide
bond of natural conomarphin, it was deduced that
Ser12 of the SFWT fragment was not in the d-configu-
ration. The C-terminal Trp14-Thr15 peptide bond
could be cleaved with a large amount of chymotrypsin
(data not shown), which suggested that Thr15 was an
l-amino acid. Thus, the possible positions for the
d-amino acid residue in conomarphin were determined
to be Phe13, Trp14, or both.
To check these three possibilities, four peptides
were synthesized, all l-configuration SFWT, S
FWT
with d-Phe, SF
WT with d-Trp, and SFWT with
d-Phe and d-Trp. Under the same HPLC elution
conditions, the retention times of SFWT and S
FWT
were clearly not the same as the retention time of
the C-terminal tetrapeptide fragment from natural
conomarphin. However, S
FWT and SFWT did elute
at the same time as the natural fragment (supplemen-
tary Fig. S4), which suggested that there is only one
d-amino acid residue in natural conomarphin, d-Phe13
or d-Trp14.
Finally, two full-length conomarphin sequences
with either d-Phe13 or d-Trp14 were chemically syn-
thesized and compared to natural conomarphin. The
coelution results unambiguously demonstrated that
conomarphin contains d-Phe13, as the synthetic pep-
tide with d-Phe13 but not the one with d-Trp14 coe-
luted with natural conomarphin (supplementary
Fig. S5).
cDNA structure of conomarphin
The cDNA encoding conomarphin was obtained by
chance when gene cloning of the M-superfamily cono-
toxins was carried out from C. marmoreus in our labo-
ratory. Besides the clones for other conventional
M-superfamily conotoxins with six Cys residues, one
clone encoded a precursor comprising the exact cono-
marphin sequence at the C-terminus (Fig. 2). This was
entirely unexpected, as conomarphin shares no
sequence homology with other M-superfamily cono-
toxins.
Nevertheless, the cDNA structure of conomarphin
was similar to those of other conotoxin cDNAs. The
cDNA-encoded precursor of conomarphin consisted of
a conserved M-superfamily signal peptide of 25 resi-
dues, a proregion of 29 residues, the conomarphin
mature peptide, and the two additional residues, Leu
and Val, at the C-terminus, which are cleaved during
maturation.
Fig. 2. cDNA and the deduced precursor
sequence of conomarphin. The signal pep-
tide is shadowed and the mature peptide is
underlined. The polyA signal AATAAA in the
3¢-UTR is also underlined. The cDNA of con-
omarphin has been deposited in the Gen-
bank database with the accession number
EU048276.
Y. Han et al. A
D-amino acid-containing conomarphin
FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS 1979
Sequence-specific resonance assignments
Complete proton resonances for both conomarphin
and l-Phe13-conomarphin were assigned by well-
established methods [21], which were pioneered by
Wuthrich and successfully applied to various animal
conotoxins [22,23]. The spin systems were identified
on the basis of both DQF-COSY and TOCSY spec-
tra. For conomarphin, 14 of the 15 spin systems were
found in the ‘fingerprint’ region of a 120 ms TOCSY
spectrum (supplementary Table S1), which were veri-
fied in a relevant DQF-COSY spectrum. For l-Phe13-
conomarphin, sequence-specific resonance assignments
(supplementary Table S2) were performed using the
same strategy. The results of sequential d
aN(i,i +1)
connectivities found in the C
a
H-NH fingerprint region
of the NOESY spectra for conomarphin and l-Phe13-
conomarphin are represented in supplementary
Fig. S6A,B, respectively. The NOESY data acquired
at 300 K and pH 3 for conomarphin and l-Phe13-
conomarphin showed a large number of NOEs, which
suggested that the structures of the two peptides were
sufficiently constrained for distance–geometry calcula-
tions.
Structural calculation, refinement, and evaluation
The NMR experimental data were converted into dis-
tance and angle constraints as usual, providing enough
constraints for the structure calculation of conomar-
phin. The three-dimensional structure of conomarphin
was determined from NMR data using the same strat-
egy previously used for structural studies of conotoxins
and their analogs [24–27].
Most NOESY crosspeaks were assigned and inte-
grated, with concomitant cycles of structure calcula-
tions for evaluation of distance and angle constraint
violations as well as assignments of additional peaks
based on the preliminary structure. For conomarphin,
the process study led to 172 NOE-based distance
restraints, of which 105 were derived from intraresidue
NOEs, 50 from sequential backbone NOEs, 14 from
medium-range NOEs, and three from long-range
NOEs (Table 3). Eight dihedral angle constraints were
used from J coupling constants. For l-Phe13-conomar-
phin, 160 NOE-based distance restraints (44 sequential
NOEs, 104 intraresidue NOEs, and 12 medium NOEs)
and six dihedral angle constraints were used to build
up the structure (Table 3).
At this stage of the structure elucidation process, the
cyana program was used to provide hydrogen bond
information. Hydrogen–deuterium exchange-out exper-
iments indicated the hydrogen bonds that might exist
between the slow-exchange amide protons and their
nearby oxygen or nitrogen atoms. Thus, two hydrogen
bonds related to four distance constraints were
Trp14(HN)–Asn11(CO) and Thr15(HN)–Ser12(CO)
for conomarphin. The l-Phe13-conomarphin had the
same hydrogen bonds as conomarphin. With the addi-
tional hydrogen bond distance constraints, another
round of minimization was performed as previously
described [22].
The simulated annealing calculations were carried
out starting with 100 random structures, and the 20
final structures selected were in good agreement with
the NMR experimental constraints, for which the
NOE distance and torsion angle violations were smal-
ler than 0.2 A
˚
and 3°, respectively. The atomic rmsd
values about the mean coordinate positions of cono-
marphin were 0.54 ± 0.24 A
˚
for the backbone atoms
(N, CR, and C) and 1.25 ± 0.30 A
˚
for all heavy
atoms, and the values for l-Phe13-conomarphin were
0.44 ± 0.16 A
˚
and 1.22 ± 0.22 A
˚
, respectively.
Finally, the 20 best models with the lowest residual
target function and lowest rmsd values were further
Table 3. Structural statistics for the family of 20 structures of
conomarphin and
L-Phe13-conomarphin.
Structural statistics Conomarphin
L-Phe13-
conomarphin
Assigned NOE crosspeaks 172 160
Intraresidue 105 104
Sequential (|i – j| = 1) 50 44
Medium range 14 12
Long range 3 0
AMBER energies (kcalÆmol
)1
)
Bond 4.98 ± 0.15 4.87 ± 0.18
Angle 62.05 ± 1.11 64.28 ± 1.14
Dihedral 131.15 ± 1.88 124.25 ± 1.45
Van der Waals )80.48 ± 2.64 )71.66 ± 3.44
Electrostatic energy )1064.45 ± 59.17 )992.47 ± 66.95
Egb (generalized born energy) )470.50 ± 64.27 )543.59 ± 69.59
Constraints 3.07 ± 0.41 2.22 ± 0.16
Total )766.95 ± 7.20 )785.47 ± 6.77
rmsd to mean coordinates (A
˚
)
Mean global backbone rmsd 0.80 ± 0.36 0.82 ± 0.36
Mean global heavy rmsd 1.78 ± 0.53 1.95 ± 0.43
Mean global backbone rmsd
(2–15)
0.54 ± 0.23 0.63 ± 0.30
Mean global heavy
rmsd (2–15)
1.59 ± 0.46 1.88 ± 0.44
Ramachandran statistics from
PROCHECK-NMR
Most favored regions (%) 87.8 68.9
Additional allowed
regions (%)
12.2 31.1
Generously allowed
regions (%)
00
Disallowed regions (%) 0 0
A
D-amino acid-containing conomarphin Y. Han et al.
1980 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS
refined for simulated annealing and restrained energy
minimization [24,28], using the sander module of the
amber 9.0 package. The resulting conformers con-
tained no significant violations of any constraint with
lower energy, better Ramachandran plots were chosen
to represent the three-dimensional solution structure of
conomarphin, and the mean structure was generated
by molmol.
Structural characterization and comparison
The program procheck was used to analyze the family
of 20 structures (Table 3). Figure 3 shows an overlay
of the backbone atoms for the 20 structures of cono-
marphin and l-Phe13-conomarphin (Protein Data
Bank codes: 2YYF and 2JQC). The overall rmsd
reported for the final 20 structures was influenced by
the disorder of the N-terminal residue Asp1. When
Asp1 was eliminated and the molecule consisted of
only residues 2–15, the mean global backbone rmsd
dropped markedly. Unlike the C-terminal portion,
the N-terminal portion of the molecule was poorly
resolved.
The three-dimensional structure of conomarphin was
characterized by one compact loop of five residues
from Ala6 to Hyp10 with a loop center at residue 8,
and another secondary structure region at the peptide
C-terminus from residues Asn11 to Trp14 with a 3
10
-
helix. The helix was supported by O
i
-HN
i +3
hydro-
gen bonds for Asn11(CO)–Trp14(NH) and Ser12(CO)–
Thr15(NH), which were confirmed by the slow solvent
exchange kinetics of the Trp14 and Thr15 amide
protons. The two observed small
3
J
HN-Ha
coupling
constants for residues Asn11 and Ser12, and the
d
NN(i,i +2)
,d
aN(i,i +2)
[Phe13(C
a
H)–Thr15(NH)] and
d
aN(i,i +3)
NOEs in the region of residues 11–14
support the presence of a short 3
10
-helix.
Similar to conomarphin, l-Phe13-conomarphin con-
tained a short 3
10
-helix near its C-terminus, from
Asn11 to Trp14, supported by O
i
-HN
i +3
hydrogen
bonds for Asn11(CO)–Trp14(HN) and Ser12(CO)–
Thr15(HN), and confirmed by the slow solvent
exchange kinetics of the amide protons of Trp14
and Thr15. The observation of two small
3
J
HN-Ha
cou-
pling constants for residues Asn11 and Ser12 and the
d
NN(i,i +2)
, d
aN(i,i +2)
[Ser12(C
a
H)–Trp14(NH) and
Asn11(C
a
H)–Phe13(NH)] and d
aN(i,i +3)
[Asn11
(C
a
H)–Trp14(NH)] NOEs in the region of resi-
dues 11–14 are in agreement with the presence of a
short 3
10
-helix. A random coil rather than a compact
loop in the region of residues 1–10 existed in
l-Phe13-conomarphin.
Fig. 3. The overlay of the backbone atoms
for the 20 converged structures of conomar-
phin (A) and
L-Phe13-conomarphin (C),
respectively. The N-terminal Asp1 is in a
poorly resolved region of the molecule. The
backbone peptide folding of conomarphin
(B) and
L-Phe13-conomarphin (D) is also
shown. A short 3
10
-helix at the C-terminus
is shown in red.
Y. Han et al. A
D-amino acid-containing conomarphin
FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS 1981
Discussion
Cone snails have developed a collection of highly
diversified peptide toxins over 50 million years of
evolution, so even after research for more than
20 years, conotoxin classification needs to be updated.
In this work, we purified a novel Cys-free 15-residue
peptide containing a d-amino acid from C. marmoreus.
This peptide is not homologous to any other identi-
fied conotoxin, and thus represents a new conotoxin
family.
C. marmoreus is a molluscivorous cone snail, the
major part of whose venom consists of peptide toxins
(Fig. 1). These toxins are mainly Cys-rich peptides [6],
except for one Cys-free fraction having a novel
sequence of DWEYHAHPKONS
FWT with Pro10
hydroxylated and Phe13 in the d-conformation. Sev-
eral families of Cys-free conopeptides from different
cone snails have been reported (Table 1). However,
this novel peptide does not exhibit obvious homology
with the others, so it was designated conomarphin.
It is very surprising that, on the basis of the con-
served signal peptide sequence, conomarphin belongs
to the M-superfamily, a major conotoxin superfamily.
All of the conventional M-superfamily conotoxins have
three disulfide bonds, although their disulfide linkages
and targets are significantly different from each other
[3–7]. Now, with conomarphin, the M-superfamily
may become the most diversified conotoxin super-
family. It is interesting that conotoxin precursor signal
peptides are rather conserved, whereas mature peptides
are very diversified. Gene structure exploration of this
conotoxin superfamily would give some hints, as has
been done for the A-superfamily of conotoxins [29].
Obviously, conomarphin maturation involves several
different post-translational modifications. Apart from
cleavage of the signal peptide and the propeptide,
which happens in the maturation process for every
conotoxin [8], the removal of the two additional resi-
dues Leu and Val at the C-terminus is rather unique
to conomarphin. To our knowledge, the cleavage of
the C-terminal Leu and Val has not been reported pre-
viously. The enzyme responsible for this cleavage and
the recognition sequence for post-translational modifi-
cations, namely hydroxylation of Pro10 and epimeriza-
tion of l-Phe13 to d-Phe13, remain to be clarified.
Hydroxyproline has been found in many conotoxins
with or without disulfide bonds [7,20,30]. It has also
been found that Pro hydroxylation happens with high
specificity; that is, only one Pro residue is hydroxylated
in many conotoxins. However, the physiological role
of specific Pro hydroxylation in conotoxins is still
elusive.
The epimerized d-Phe13 is another striking feature
of conomarphin. The l-amino acid to d-amino acid
epimerization in a polypeptide chain is quite rare and
not well understood, although some d-amino acids
have been known for a long time to act as neurotrans-
mitters. d-Amino acid-containing peptides or toxins
have been found in mollusks [12,18–20], arthropods
[10], amphibians [9] and even mammals [11]. They are
produced by a ribosome protein translation pathway
based on their mRNAs [17,18,31], so epimerization
must occur on an incorporated l-amino acid of a pep-
tide chain. So far, epimerization enzymes have been
found in frog skin [32], spider venom [33], and the
venom of the Australian platypus [34], but they are
completely different with respect to sequence and
mechanism.
Although the detailed mechanism of epimerization is
unclear, epimerization in short peptides has been
found to have a ‘position rule’; namely, epimerization
occurs only at three positions: position 2 at the N-ter-
minus (+2), and positions 3 and 5 at the C-terminus
()3 and )5) (Table 2). It is noteworthy that epimeriza-
tion at each of these three positions has been found in
cone snails, such as the +2 position in conomap [19],
the )3 position in conomarphin and r11a [35], and the
)5 position in contryphan [16]. However, epimeriza-
tion happens mainly at the +2 position in other
organisms. Probably, cone snails have developed a
more advanced system to achieve this difficult modifi-
cation at different positions. This is not surprising,
because of the well-known high content of post-trans-
lational modifications in conotoxins [8]. It is notewor-
thy that from the single species C. marmoreus, two
epimerization positions have been found, the )3 posi-
tion for conomarphin and the )5 position for glacon-
tryphan-M [16]. It is not known whether they are
modified by the same enzyme system but with different
recognition sequences. It is also worth pointing out
that the l-amino acid to d-amino acid epimerization
seems to be complete for conopeptides, whereas both
isoforms coexist in defensin-like peptides and natri-
uretic peptides from the Australian platypus [11].
With the help of such a developed post-translational
modification system, conotoxins exhibit amazing struc-
tural diversity. In this work, we found that conomar-
phin, despite being a short peptide of 15 residues, is
well structured in solution (Fig. 3 and supplementary
Fig. S7). The d-Phe13 of conomarphin has a signifi-
cant effect on the structure of the peptide; a tight loop
around Pro8 and a short 3
10
-helix at the C-terminus
were identified. However, for l-Phe13-conomarphin,
there was no loop in the middle and the peptide chain
seemed to be rather straight, whereas the C-terminal
A D-amino acid-containing conomarphin Y. Han et al.
1982 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS
helix was two residues longer. Superimposition of the
C-terminal helices of conomarphin and l-Phe13-cono-
marphin showed that the relatively large side chain of
l-Phe13 might cause spatial hindrance to Hyp10 and
Lys9 forming a loop (Fig. 4), so that the rest of the
peptide chain of l-Phe13-conomarphin extends roughly
along the axis orientation of the C-terminal helix.
The structures of several d-amino acid-containing
peptides have been determined, including glacontry-
phan-M [36], contraphan-R [37], contrypan-Vn [38],
DLP-2 and DLP-4 [39], and the C-terminal peptide of
x-agatoxin [10]. The NMR structure of excitatory
r11a was reported very recently [40]. There are one or
more disulfide bonds in all of these peptides, which
supports their rigid structures. Consequently, the ter-
minal d-amino acid has only a minimal effect on the
overall structure. However, in Cys-free conomarphin,
d-Phe13 has a critical role in maintaining the peptide
conformation. Converting d-Phe13 to l-Phe13 dra-
matically changes the structure (Fig. 4). It is plausible
that this d-Phe13 and the well-maintained structure of
conomarphin could be very important for its function,
the exploration of which will certainly be of great
interest.
In summary, a new conotoxin family, conomarphin,
was identified and structurally studied in this work.
Furthermore, the critical influence of a d-amino acid
on the conformation of a peptide was demonstrated. It
is noteworthy that this conotoxin family exists in all
three major feeding types of cone snails. Apart from
conomarphin purified from the mollusk-hunting
C. marmoreus, a similar sequence was identified on the
cDNA level from the worm-hunting Conus litteratus
[41], and a homologous peptide was purified from the
fish-hunting Conus achatinus (H. Jiang & C. X. Fan,
unpublished data). The widespread occurrence of con-
omarphins in fish, mollusks and worm-hunting cone
snails suggests that this family of peptides may have a
specific function.
Experimental procedures
Materials
Specimens of C. marmoreus were collected from Sanya near
the South China Sea. Sephadex G-25 was purchased from
Amersham Biosciences (Uppsala, Sweden), a ZORBAX
300SB-C18 semipreparative column was from Agilent Tech-
nologies (Santa Clara, CA, USA), and trifluoroacetic acid
and acetonitrile used for HPLC were from Merck (Darms-
tadt, Germany). Trypsin and tosyl phenylalanyl chloro-
methyl ketone-treated chymotrypsin were from Sigma
(St Louis, MO, USA). The 3¢-RACE kit and TRIzol
reagent were purchased from Invitrogen (Carlsbad, CA,
USA), and Taq DNA polymerase and the pGEM-T Easy
vector system were from Promega (Madison, WI, USA).
Fig. 4. Stereo view of the superimposed
structures of conomarphin (red) and
L-Phe13-conomarphin (gray). The side chain
of
L-Phe13 might cause spatial hindrance to
the side chain of Lys9 and Hyp10 in forming
the tight loop as in conomarphin. The side
chain of Lys9, Hyp10 and
D-Phe13 of cono-
marphin is shown in stick mode, and the
side chain of
L-Phe13 in L-Phe13-conomar-
phin is shown as a gray stick.
Y. Han et al. A
D-amino acid-containing conomarphin
FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS 1983
Purification
The purification procedure and conditions were exactly the
same as previously described [6]. Briefly, the crude venom
was first separated on a Sephadex G-25 column and the
second peak was then applied to a ZORBAX 300SB-C18
semipreparative column (9.4 · 250 mm) connected to an
HPLC instrument. The peptides were eluted with a gradient
of acetonitrile.
Peptide synthesis
Peptides were synthesized by solid-phase methods on an
ABI 433A peptide synthesizer using standard Fmoc chemis-
try and side-chain protection.
N-terminal sequencing and MS
N-terminal amino acid sequence analysis was performed by
automated Edman degradation on an ABI model 491A
Procise Protein Sequencing System (Applied Biosystems,
Foster City, CA, USA). A 20 pmol sample was loaded onto
a glass fiber filter previously conditioned with BioBrene
Plus (Applied Biosystems).
All purified and synthetic peptides were analyzed in the
scan type of Enhanced MS by Qtrap (Applied Biosystems).
The mass spectrometer, equipped with a TurboIonSpray
Source, was operated in positive ionization mode.
Protease digestion
The native and synthesized peptides were dissolved in
50 mm Tris ⁄ HCl (pH 7.8) and 20 mm CaCl
2
buffer to a
concentration of 1 l gÆlL
)1
. Trypsin was added to a ratio of
1 : 20. The digestion was carried out at 25 °C for 18 h, and
then quenched with 50% trifluoroacetic acid before HPLC
analysis.
The chymotrypsin digestion was performed in 50 mm
Tris ⁄ HCl (pH 7.8) and 20 mm CaCl
2
buffer with the same
peptide concentration and enzyme ratio. The reactions were
kept at 25 °C for 18 h, and analyzed by HPLC after being
quenched with 50% trifluoroacetic acid.
cDNA cloning
The cDNA of conomarphin was obtained unexpectedly
when the cDNA cloning of M-superfamily conotoxins was
performed from cDNAs reverse transcribed from total
RNAs of the venom duct of C. marmoreus, as previously
described [6]. The 5¢-primer corresponded to the highly
conserved M-superfamily signal peptide sequence
(5¢-ATGTTGAAAATGGGAGT(G ⁄ A)GTG-3¢), and the
3¢-primer was the abridged universal amplification primer
devoid of the poly(dT) tail from the 3¢-RACE kit. The
resulting PCR products were inserted directly into the
pGEM-T Easy vector for sequencing.
NMR experiments
Samples of conomarphin and l-Phe13-conomarphin for
NMR studies were prepared in either 99.99% D
2
O (Cam-
bridge Isotopes Lab) or 9 : 1 (v ⁄ v) H
2
O ⁄ D
2
O with 0.01%
trifluoroacetic acid, at pH 3 (uncorrected for the isotope
effect), with a final sample concentration of approximately
2mm. For experiments in D
2
O, the peptide was lyophilized
and redissolved in 99.99% D
2
O.
NMR spectra were collected on a Bruker-DRX 600 MHz
spectrometer using standard pulse sequences and phase
cycling at 300 K. Proton DQF-COSY [42], NOESY [43]
and TOCSY spectra [44] of samples in 99.99% D
2
O and
90 : 10 H
2
O ⁄ D
2
O, respectively, were acquired with the
transmitter set at 4.70 p.p.m. and a spectral window of
6000 Hz, as described previously [22].
Spectra were processed with topspin or xwinnmr soft-
ware. Phase-shifted sine-squared window functions were
applied before Fourier transformation. To identify the slow
exchange of backbone amide protons, the hydrogen–deute-
rium exchange experiments were carried out by dissolving
the lyophilized sample in D
2
O and recording a series of one-
dimensional spectra every 5 min for 1 h, and subsequently
every hour for 10 h. Chemical shifts were referenced to the
methyl resonance of 4,4-dimethyl-4-silapentane-1-sulfonic
acid as an internal standard. Complete sets of two-dimen-
sional spectra for both samples of conomarphin and
l-Phe13-conomarphin were recorded at 300 K and pH 3.
Restraint set generation
An initial survey of distance constraints was performed on
a series of NOESY spectra acquired at mixing times of 100,
200 and 350 ms. Buildup curves were produced that dem-
onstrated a leveling of the intensity of the NOE at mixing
times greater than 200 ms. Peak picking, spin system identi-
fication and volume integration of the NOESY crosspeaks
were performed with the interactive program sparky
(v. 1.113). Non-stereospecifically assigned atoms were trea-
ted as pseudo-atoms and given correction distances. A set
of distance restraints was generated from these data and
used as input for cyana (v. 2.1).
Six u dihedral angles were determined on the basis of the
3
J
NHa
coupling constants derived by analysis of a high-res-
olution one-dimensional proton spectrum of the conotoxin
conomarphin. For peaks that did not show a splitting
pattern, the
3
J
NHa
value was derived from a measure of
the line width at half the height of the signal. For
3
J
NHa
values < 5.5 Hz, the u angle was constrained in the
range )65±25°, and for
3
J
NHa
values > 8.0 Hz, it was
constrained in the range )120 ± 40° [21,45]. Backbone
A D-amino acid-containing conomarphin Y. Han et al.
1984 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS
dihedral constraints were not applied to
3
J
NHa
values
between 5.5 and 8.0 Hz.
The hydrogen bond acceptors for the slowly exchanged
amide protons were identified by analysis of the preliminary
calculated structures [46,47]. The hydrogen bond distance
restraints were added as target values of 1.8–2.2 A
˚
for
NH
i
–O
j
bonds and 2.8–3.2 A
˚
for N
i
–O
j
bonds, respectively.
Structural computation and refinement
The experimentally derived distance constraints, torsion
angle constraints and hydrogen bond constraints were
input for the molecular modeling protocol. One hundred
calculations with the program cyana were started with
random polypeptide conformations, and the 20 resulting
conformers with the lowest residual target function values
were analyzed.
The 20 structures with the lowest target functions were
submitted to molecular dynamics refinement with the
sander module of the amber 9 program as the starting
structure [47]. The molecular dynamics simulations were
performed using the parm03 force field and the GB ⁄ SA
implicit solvation system [48]. The visual analysis of cono-
marphin was done using molmol [49] software, and the
geometric qualities of the obtained structures were assessed
with procheck-nmr software [50].
Acknowledgements
This work was supported by the National Basic
Research Program of China (2004CB719900), the
National Natural Science Foundation of China
(20473013) and the Chinese Academy of Sciences for
Key Topics in Innovation Engineering (KSCX2-YW-
R-104). Chunguang Wang is supported by the Pro-
gram for young excellent talents in Tongji University
(2006KJ063) and Dawn Program of Shanghai Educa-
tional Development Foundation (06SG26).
References
1 Terlau H & Olivera BM (2004) Conus venoms: a rich
source of novel ion channel-targeted peptides. Physiol
Rev 84, 41–68.
2 Olivera BM (2006) Conus peptides: biodiversity-based
discovery and exogenomics. J Biol Chem 281,
31173–31177.
3 Shon KJ, Olivera BM, Watkins M, Jacobsen RB, Gray
WR, Floresca CZ, Cruz LJ, Hillyard DR, Brink A,
Terlau H et al. (1998) mu-Conotoxin PIIIA, a new pep-
tide for discriminating among tetrodotoxin-sensitive Na
channel subtypes. J Neurosci 18, 4473–4481.
4 Shon KJ, Grilley M, Jacobsen R, Cartier GE, Hopkins
C, Gray WR, Watkins M, Hillyard DR, Rivier J,
Torres J et al. (1997) A noncompetitive peptide inhibi-
tor of the nicotinic acetylcholine receptor from Conus
purpurascens venom. Biochemistry 36, 9581–9587.
5 Ferber M, Sporning A, Jeserich G, DeLaCruz R,
Watkins M, Olivera BM & Terlau H (2003) A novel
conus peptide ligand for K
+
channels. J Biol Chem
278, 2177–2183.
6 Han YH, Wang Q, Jiang H, Liu L, Xiao C, Yuan
DD, Shao XX, Dai QY, Cheng JS & Chi CW (2006)
Characterization of novel M-superfamily conotoxins
with new disulfide linkage. FEBS J 273, 4972–4982.
7 Corpuz GP, Jacobsen RB, Jimenez EC, Watkins M,
Walker C, Colledge C, Garrett JE, McDougal O, Li W,
Gray WR et al. (2005) Definition of the M-conotoxin
superfamily: characterization of novel peptides from
molluscivorous Conus venoms. Biochemistry 44,
8176–8186.
8 Buczek O, Bulaj G & Olivera BM (2005) Conotoxins
and the posttranslational modification of secreted gene
products. Cell Mol Life Sci 62, 3067–3079.
9 Montecucchi PC, de Castiglione R, Piani S, Gozzini L
& Erspamer V (1981) Amino acid composition and
sequence of dermorphin, a novel opiate-like peptide
from the skin of Phyllomedusa sauvagei. Int J Pept
Protein Res 17, 275–283.
10 Kuwada M, Teramoto T, Kumagaye KY, Nakajima K,
Watanabe T, Kawai T, Kawakami Y, Niidome T,
Sawada K, Nishizawa Y et al. (1994) Omega-agatoxin-
TK containing D-serine at position 46, but not syn-
thetic omega-[L-Ser46]agatoxin-TK, exerts blockade of
P-type calcium channels in cerebellar Purkinje neurons.
Mol Pharmacol 46, 587–593.
11 Torres AM, Menz I, Alewood PF, Bansal P, Lahnstein
J, Gallagher CH & Kuchel PW (2002) d-Amino acid
residue in the C-type natriuretic peptide from the
venom of the mammal, Ornithorhynchus anatinus, the
Australian platypus. FEBS Lett 524, 172–176.
12 Ohta N, Kubota I, Takao T, Shimonishi Y,
Yasuda-Kamatani Y, Minakata H, Nomoto K,
Muneoka Y & Kobayashi M (1991) Fulicin, a novel
neuropeptide containing a D-amino acid residue
isolated from the ganglia of Achatina fulica. Biochem
Biophys Res Commun 178, 486–493.
13 Jimenez EC, Olivera BM, Gray WR & Cruz LJ (1996)
Contryphan is a D-tryptophan-containing Conus
peptide. J Biol Chem 271, 28002–28005.
14 Jimenez EC, Watkins M, Juszczak LJ, Cruz LJ &
Olivera BM (2001) Contryphans from Conus textile
venom ducts. Toxicon 39, 803–808.
15 Massilia GR, Eliseo T, Grolleau F, Lapied B, Barbier
J, Bournaud R, Molgo J, Cicero DO, Paci M, Schinina
ME et al. (2003) Contryphan-Vn: a modulator of
Ca2
+
-dependent K
+
channels. Biochem Biophys Res
Commun 303, 238–246.
Y. Han et al. A D-amino acid-containing conomarphin
FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS 1985
16 Hansson K, Ma X, Eliasson L, Czerwiec E, Furie B,
Furie BC, Rorsman P & Stenflo J (2004) The first
gamma-carboxyglutamic acid-containing contryphan. A
selective L-type calcium ion channel blocker isolated
from the venom of Conus marmoreus. J Biol Chem 279,
32453–32463.
17 Jimenez EC, Craig AG, Watkins M, Hillyard DR,
Gray WR, Gulyas J, Rivier JE, Cruz LJ & Olivera
BM (1997) Bromocontryphan: post-translational
bromination of tryptophan. Biochemistry 36, 989–
994.
18 Buczek O, Yoshikami D, Watkins M, Bulaj G,
Jimenez EC & Olivera BM (2005) Characterization of
D-amino-acid-containing excitatory conotoxins and
redefinition of the I-conotoxin superfamily. FEBS J
272, 4178–4188.
19 Dutertre S, Lumsden NG, Alewood PF & Lewis RJ
(2006) Isolation and characterisation of conomap-Vt, a
D-amino acid containing excitatory peptide from the
venom of a vermivorous cone snail. FEBS Lett 580,
3860–3866.
20 Pisarewicz K, Mora D, Pflueger FC, Fields GB & Mari
F (2005) Polypeptide chains containing D-gamma-
hydroxyvaline. J Am Chem Soc 127, 6207–6215.
21 Pardi A, Billeter M & Wuthrich K (1984) Calibration
of the angular dependence of the amide proton-C alpha
proton coupling constants, 3JHN alpha, in a globular
protein. Use of 3JHN alpha for identification of helical
secondary structure. J Mol Biol 180, 741–751.
22 Du W, Han Y, Huang F, Li J, Chi C & Fang W (2007)
Solution structure of an M-1 conotoxin with a novel
disulfide linkage. FEBS J 274, 2596–2602.
23 McDougal OM & Poulter CD (2004) Three-dimensional
structure of the mini-M conotoxin mr3a. Biochemistry
43, 425–429.
24 Rogers JP, Luginbuhl P, Shen GS, McCabe RT,
Stevens RC & Wemmer DE (1999) NMR solution
structure of alpha-conotoxin ImI and comparison to
other conotoxins specific for neuronal nicotinic
acetylcholine receptors. Biochemistry 38, 3874–3882.
25 Al-Sabi A, Lennartz D, Ferber M, Gulyas J, Rivier JE,
Olivera BM, Carlomagno T & Terlau H (2004)
KappaM-conotoxin RIIIK, structural and functional
novelty in a K
+
channel antagonist. Biochemistry 43,
8625–8635.
26 Nielsen KJ, Adams DA, Alewood PF, Lewis RJ,
Thomas L, Schroeder T & Craik DJ (1999) Effects of
chirality at Tyr13 on the structure–activity relationships
of omega-conotoxins from Conus magus. Biochemistry
38, 6741–6751.
27 Hofmann MW, Poschner BC, Hauser S & Langosch D
(2007) pH-Activated fusogenic transmembrane LV-pep-
tides. Biochemistry 46, 4204–4209.
28 Zhang N, Chen X, Li M, Cao C, Wang Y, Wu G,
Hu G & Wu H (2004) Solution structure of BmKK4,
the first member of subfamily alpha-KTx 17 of scorpion
toxins. Biochemistry 43, 12469–12476.
29 Yuan DD, Han YH, Wang CG & Chi CW (2007) From
the identification of gene organization of alpha
conotoxins to the cloning of novel toxins. Toxicon 49,
1135–1149.
30 Fan CX, Chen XK, Zhang C, Wang LX, Duan KL,
He LL, Cao Y, Liu SY, Zhong MN, Ulens C et al. (2003)
A novel conotoxin from Conus betulinus, kappa-BtX,
unique in cysteine pattern and in function as a specific
BK channel modulator. J Biol Chem 278, 12624–12633.
31 Richter K, Egger R & Kreil G (1987) D-alanine in the
frog skin peptide dermorphin is derived from L-alanine
in the precursor. Science 238, 200–202.
32 Jilek A, Mollay C, Tippelt C, Grassi J, Mignogna G,
Mullegger J, Sander V, Fehrer C, Barra D & Kreil G
(2005) Biosynthesis of a D-amino acid in peptide link-
age by an enzyme from frog skin secretions. Proc Natl
Acad Sci USA 102, 4235–4239.
33 Shikata Y, Watanabe T, Teramoto T, Inoue A,
Kawakami Y, Nishizawa Y, Katayama K & Kuwada
M (1995) Isolation and characterization of a peptide
isomerase from funnel web spider venom. J Biol Chem
270, 16719–16723.
34 Torres AM, Tsampazi M, Tsampazi C, Kennett EC,
Belov K, Geraghty DP, Bansal PS, Alewood PF &
Kuchel PW (2006) Mammalian l-to-d-amino-acid-resi-
due isomerase from platypus venom. FEBS Lett 580,
1587–1591.
35 Buczek O, Yoshikami D, Bulaj G, Jimenez EC &
Olivera BM (2005) Post-translational amino acid
isomerization: a functionally important D-amino acid in
an excitatory peptide. J Biol Chem 280, 4247–4253.
36 Grant MA, Hansson K, Furie BC, Furie B, Stenflo J &
Rigby AC (2004) The metal-free and calcium-bound
structures of a gamma-carboxyglutamic acid-containing
contryphan from Conus marmoreus, glacontryphan-M.
J Biol Chem 279, 32464–32473.
37 Pallaghy PK, Melnikova AP, Jimenez EC, Olivera BM
& Norton RS (1999) Solution structure of contryphan-
R, a naturally occurring disulfide-bridged octapeptide
containing D-tryptophan: comparison with protein
loops. Biochemistry 38, 11553–11559.
38 Eliseo T, Cicero DO, Romeo C, Schinina ME, Massilia
GR, Polticelli F, Ascenzi P & Paci M (2004) Solution
structure of the cyclic peptide contryphan-Vn, a
Ca2
+
-dependent K
+
channel modulator. Biopolymers
74, 189–198.
39 Torres AM, Tsampazi C, Geraghty DP, Bansal PS,
Alewood PF & Kuchel PW (2005) D-amino acid residue
in a defensin-like peptide from platypus venom: effect
on structure and chromatographic properties. Biochem
J 391, 215–220.
40 Buczek O, Wei D, Babon JJ, Yang X, Fiedler B, Chen
P, Yoshikami D, Olivera BM, Bulaj G & Norton RS.
A D-amino acid-containing conomarphin Y. Han et al.
1986 FEBS Journal 275 (2008) 1976–1987 ª 2008 The Authors Journal compilation ª 2008 FEBS