Báo cáo khoa học: Respective roles of the catalytic domains and C-terminal tail peptides in the oligomerization and secretory trafficking of human acetylcholinesterase and butyrylcholinesterase potx

Respective roles of the catalytic domains and C-terminal
tail peptides in the oligomerization and secretory
trafficking of human acetylcholinesterase and
butyrylcholinesterase
Dong Liang
1,2
, Jean-Philippe Blouet
1
, Fernanda Borrega
1
, Suzanne Bon
1
and Jean Massoulie
´
1
1 Laboratoire de Neurobiologie, CNRS UMR 8544, Ecole Normale Supe
´
rieure, Paris, France
2 Key Laboratory of Brain Functional Genomics, MOE&STCSM, Shanghai Institute of Brain Functional Genomics, East China Normal
University, China
In vertebrates, butyrylcholinesterase (BChE
T
) and the
T splice variant of acetylcholinesterase (AChE
T
)
consist of a catalytic domain of approximately 500
residues, followed by C-terminal tail (t) peptides [1,2].
These peptides of 41 and 40 residues, respectively, con-
tain seven strictly conserved aromatic residues, includ-
ing three evenly spaced tryptophans, and a cysteine

located at position )4 from the C-terminus. The
t peptide plays an important role in the biosynthesis of
cholinesterases, particularly their folding and export.
For example, it has been shown that it induces the
misfolding of a significant fraction of newly synthe-
sized AChE polypeptides, and that this effect depends
on hydrophobicity since it was maintained when the
aromatic residues were replaced by leucines. The t pep-
tide also reduces export, as indicated by the fact that
Keywords
acetylcholinesterase; butyrylcholinesterase;
cysteines; oligomers; secretion
Correspondence
J. Massoulie
´
, Laboratoire de Neurobiologie,
CNRS UMR 8544, Ecole Normale
Supe
´
rieure, Paris, France
Fax: +33 1 44 32 38 87
Tel: +33 1 44 32 38 91
E-mail:
(Received 13 August 2008, revised
25 September 2008, accepted
24 October 2008)
doi:10.1111/j.1742-4658.2008.06756.x
Butyrylcholinesterase (BChE) and the T splice variant of acetylcholinester-
ase that is predominant in mammalian brain and muscles (AChE
T

) possess
a characteristic C-terminal tail (t) peptide. This t peptide allows their
assembly into tetramers associated with the anchoring proteins ColQ and
PRiMA. Although the t peptides of all vertebrate cholinesterases are
remarkably similar and, in particular, contain seven strictly conserved
aromatic residues, these enzymes differ in some of their oligomerization
properties. To explore these differences, we studied human AChE (Aa) and
BChE (Bb), and chimeras in which the t peptides (a and b) were exchanged
(Ab and Ba). We found that secretion was increased by deletion of the
t peptides, and that it was more efficient with a than with b. The patterns
of oligomers were similar for Aa and Ab, as well as for Ba and Bb, indicat-
ing a predominant influence of the catalytic domains. However, addition of
a cysteine within the aromatic-rich segment of the t peptides modified the
oligomeric patterns: with a cysteine at position 19, the proportion of tetra-
mers was markedly increased for Aa(S19C) and Ba(S19C), and to a lesser
extent for Bb(N19C); the Ab(N19C) mutant produced all oligomeric forms,
from monomers to hexamers. These results indicate that both the catalytic
domains and the C-terminal t peptides contribute to the capacity of cho-
linesterases to form and secrete various oligomers. Sequence comparisons
show that the differences between the t peptides of AChE and BChE are
remarkably conserved among all vertebrates, suggesting that they reflect
distinct functional adaptations.
Abbreviations
AchE
T
, T splice variant of acetylcholinesterase; BChE, butyrylcholinesterase; DEPQ, 7-[(diethoxyphosphoryl)oxy]-1-methylquinolinium iodide;
Nbs
2
, 5,5¢-dithiobis(2-nitrobenzoic acid); PRAD, proline-rich attachment domains; t, tail.
94 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works

the ratio of secreted to cellular AChE was strongly
increased when it was deleted; this effect was sup-
pressed by mutation of the aromatic residues to
leucines [3–6].
However, the major function of the t peptides is that
they allow the assembly of tetramers of AChE
T
[7] and
of BChE
T
[8] and their association with the structural
proteins ColQ and PRiMA [9,10]. These heteromeric
structures are based on a tight association between
four t peptides, also named tryptophan (W) amphi-
philic tetramerization domains, and the poly-proline
motifs, or proline-rich attachment domains (PRADs)
of ColQ and PRiMA [11–13]. In addition, the BChE
tetramers that circulate in the blood plasma have
recently been shown to incorporate a similar proline-
rich peptide derived from the protein lamellipodin [14].
Crystallographic analyses of a complex of synthetic t
and PRAD peptides showed that four a-helical t pep-
tides form a coiled-coil around the PRAD, which is
arranged in a poly-proline II helix [15].
The assembly of cholinesterase homo-tetramers or
PRAD-associated tetramers is entirely conditioned by
the presence of a t peptide because truncated AChE
subunits lacking the t peptide only produce secreted
monomers [16]. This peptide constitutes an autono-
mous interaction module, necessary and sufficient for

tetramerization and association with PRAD-containing
proteins, because addition of a t peptide at the C-ter-
minus of green fluorescent protein or alkaline phos-
phatase allowed the formation of tetramers associated
with an N-terminal fragment of ColQ [17]. However,
the catalytic domains are also involved in quaternary
interactions that certainly participate in the formation
and stability of these oligomers. In particular, the tet-
ramers are formed of two pairs of subunits, in which
a
7,8
and a
10
helices from each subunit form a four
helix bundle, with a hydrophobic contact zone [16,18].
The respective contributions of the catalytic domains
and the t peptides in oligomers has not been evaluated.
The formation of AChE
T
tetramers associated with
PRAD-containing proteins is physiologically important
because it ensures their functional localization by ColQ
in the basal lamina at neuromuscular junctions [19], as
well as by PRiMA in cell membranes, particularly
in the brain [20]. Similarly, the formation of BChE
T
tetramers conditions the secretion of this enzyme and
its stability in the bloodstream.
Injection of AChE or BChE offers a very efficient
protection against poisoning by anti-cholinesterase

agents, such as organophosphorous pesticides, but
monomers and dimers are much more rapidly elimi-
nated than tetramers after injection in the circulation
[21–27]. Although the half life of recombinant
enzymes, even monomers, in the bloodstream can be
considerably increased by derivatization with polyeth-
ylene glycol [28–33], it may be interesting to produce
these enzymes as recombinant proteins in a stable tet-
rameric form, which also present a greater thermal sta-
bility than monomers or dimers [34].
Mutants lacking the cysteine located at )4 from the
C-terminus do not form stable dimers, but can form
tetramers, particularly in the presence of a PRAD-con-
taining protein. It is likely that transient dimerization
occurs as a first step in the assembly of tetramers,
either associated or not with a PRAD. We recently
found that addition of a second cysteine at an appro-
priate position in the t peptide of Torpedo AChE
greatly increased the formation and secretion of homo-
tetramers [4]. We therefore explored the possibility that
mutations in the t peptides of human AChE
T
and
BChE
T
might induce their assembly into tetramers.
Because these two enzymes differ in their capacity to
form oligomers, we investigated the respective roles of
the catalytic domain and of the t peptides. For this
purpose, we constructed chimeric proteins, in which

we associated the catalytic domain of each enzyme
with the t peptide of the other. For convenience, the
large catalytic domains are designated by capital letters
(A and B), whereas the small t peptides are designated
by lower case letters (a and b), so that the wild-type
AChE and BChE are Aa and Bb, and the chimeras are
Ab and Ba. Comparisons of wild-type enzymes and
chimeras, as well as of various mutants, show that
both domains contribute critically to the oligomeriza-
tion and to the efficiency of secretion.
Results
Exchange of t peptides between human AChE
and BChE
The T variants of human AChE and human BChE are
composed of a catalytic domain of approximately 500
residues, followed by small C-terminal t peptides of 40
and 41 residues, respectively. In the present study, the
catalytic domains are indicated by capital letters ( A and
B) and the C-terminal peptides by lower case letters (a
and b), so that the wild-type enzymes are abbreviated as
Aa and Bb. The C-terminal t peptides of human AChE
T
(a) and BChE
T
(b) are highly homologous, with 24 iden-
tical residues (60%), including the seven aromatic resi-
dues and the cysteine located at )4 from the C-terminus,
being strictly conserved among all vertebrate cholines-
terases (Fig. 1A). However, they present significant dif-
ferences, particularly between the residues immediately

following the catalytic domain. Some of the differences
D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 95
between the peptides a and b might be important for the
processing and the activity of AChE and BChE, notably
those involving charged residues, the presence in peptide
b of an additional tryptophan (W8) and the presence of
six instead of five residues between the aromatic-rich
region and the cysteine. Both peptides are predicted to
form amphiphilic a helices, in which the aromatic resi-
dues are clustered in a sector of approximately 140°
(Fig. 1B).
To analyze the oligomerization properties due to the
t peptides of human AChE and BChE, we constructed
chimeras Ab and Ba in which we exchanged these pep-
tides; we also deleted the C-terminal peptides, produc-
ing the truncated enzymes A and B. The different
constructs were expressed in transiently transfected
COS cells, and we analyzed the cellular and secreted
cholinesterase activities (Fig. 2), as well as the oligo-
meric patterns, revealed by sedimentation profiles in
sucrose gradients (Fig. 3).
The C-terminal t peptides do not influence the
catalytic activity of AChE and BChE
We examined the possible influence of the C-terminal
peptides on the AChE and BChE activities by
comparing the catalytic rates per active site. The active
sites were titrated with the irreversible inhibitor
7-[(diethoxyphosphoryl)oxy]-1-methylquinolinium iodide
(DEPQ) (see Experimental procedures). The slopes of

residual activity, plotted as a function of the amount
of DEPQ, were identical for A, Aa and Ab, with
acetylthiocholine as substrate, as well as for B, Ba
and Bb, using either acetylthiocholine or butyrylthio-
choline as substrates. Because of excess substrate inhi-
bition, AChE presented a maximal rate for
approximately 2 mm acetylthiocholine. The rates of
hydrolysis of acetylthiocholine and butyrylthiocholine
(at 6 mm) by BChE were approximately 14% and
24% of the rate of hydrolysis of acetylthiocholine
(at 2 mm) by AChE.
Influence of the C-terminal t peptides on activity,
secretion and oligomerization
As expected, the truncated mutants A and B, without
t peptides, produced only monomers, sedimenting
around 4S (not shown). The levels of cellular activity
were lower for these mutants than for the wild-types but
secretion was increased (Fig. 2A,B), in agreement with
our previous conclusions that t peptides induce a partial
misfolding of the polypeptides, as well as an intracellular
degradation of a fraction of active subunits [3].
Cells expressing wild-type human AChE (Aa)
secreted approximately 15% of their content per hour
and produced mostly monomers and dimers, with a
small proportion of tetramers (less than 10% of the
A
B
Fig. 1. Structures of AChE and BChE t peptides. (A) Sequences of the C-terminal t peptides of human AChE and BChE. These peptides
(a and b) are encoded by 3¢ exons from the cholinesterase genes; in the present study, we numbered from their first residue. The seven
aromatic residues, which are conserved in all vertebrate cholinesterases, are shown in blue; acidic residues are shown in red and basic resi-

dues in green; the cysteines are indicated by orange arrowheads and the residues that have been mutated to cysteines in the present study
are underlined. Residues that differ between peptides a and b and were mutated in b are indicated by vertical lines above the sequence
(those which were mutated as a group are joined by an horizontal line). (B) En face view of the a helices formed by the N-terminal regions
of peptides a and b . Colours are as in (A), except that cysteines are shown in orange and residues that were mutated to cysteines are indi-
cated by orange circles.
Oligomerization and secretion of AChE and BChE D. Liang et al.
96 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
secreted activity). For human BChE (Bb), the rate of
secretion was only approximately 5% of the cellular
content per hour. This enzyme formed a higher pro-
portion of oligomers, mostly dimers in the cells, and
tetramers represented approximately 30% of the
secreted enzyme, together with comparable proportions
of dimers and monomers (Fig. 3A).
The fact that Bb forms a higher proportion of olig-
omers, but is less efficiently secreted than Aa, is quite
paradoxical because secretion generally increases with
the degree of oligomerization. Clearly, the assembly
of tetramers is not restricted by the fact that each
BChE subunit possesses nine N-linked glycans [35],
whereas AChE has only four. This was confirmed by
050100150200250
0 50 100 150 200 250
-
A-
Aa
Aa S19C
Aa S38C
-
Ab

Ab SSVGL
Ab N19C
Ab N19C N18S
Ab N19C MD22VH
Ab N19C N18S MD22VH
-
B-
Ba
Ba S19C
-
Bb
Bb SSVGL
Bb A12C
Bb H15C
Bb N19C
Bb D23C
Bb N26C
Bb S37C
Bb N19C N18S
Bb N19C MD22VH
Bb N19C N18S MD22VH
-
Cellular activity
A
B
Secreted activity
0
1
2
3

4
5
6
A-
Aa
Aa S19C
Aa S38C
Ab
Ab SSVGL
Ab N19C
Ab N19C N18S
Ab N19C MD22VH
Ab N19C N18S MD22VH
B-
Ba
Ba S19C
Bb
Bb SSVGL
Bb A12C
Bb H15C
Bb N19C
Bb D23C
Bb N26C
Bb S37C
Bb N19C N18S
Bb N19C MD22VH
Bb N19C N18S MD22VH
Ratio of secreted to cellular
activity (% of the wild type)
Fig. 2. Cellular and secreted activities produced by human AChE, BChE and mutants used in the present study. (A) Cellular and secreted

activities. A and B represent AChE and BChE from which the t peptides were deleted; Aa and Bb represent the wild-type enzymes with
their t peptides, Ab and Ba represent chimeras in which the t peptides were exchanged; mutations in the t peptides are indicated. For each
mutant, the cellular and secreted activities are shown by bars to the left and the right. AChE and BChE activities were determined by the
Ellman assay with acetylthiocholine and butyrylthiocholine as substrates, respectively: AChE activities are indicated as grey bars and BChE
activities as hatched bars. AChE and BChE activities were normalized to the wild-type enzymes (Aa and Bb, respectively). (B) Ratio
of secreted to cellular activity. Note that the double mutation M22V ⁄ D23H is abbreviated as MD22VH.
D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 97
the fact that mutants lacking some of the N-glycosyl-
ation sites, which were provided by O. Lockridge [36],
did not produce a higher proportion of tetramers
(not shown).
For the chimeras Ab and Ba, the rates of secretion
were approximately 5% and 15% of the cellular con-
tent per hour, respectively, and therefore appeared to
be mainly determined by the t peptides. By contrast,
Fig. 3B shows that the sedimentation profiles were
very similar for Aa and Ab, and for Ba and Bb,
except that the BChE species sedimented faster than
their AChE counterparts, in agreement with the
higher mass of BChE subunits [37]. This indicates a
predominant influence of the catalytic domain on
oligomerization.
Role of the C-terminal cysteine
Mutation of the cysteine located at )4 from the C-ter-
minus to a serine in the a or b peptides suppressed the
formation of Aa or Bb dimers, but not the production
of a small proportion of tetramers (not shown). These
mutations increased the ratio of secreted to cellular
activity in both cases (Fig. 2). However, in the case of

Bb, the cellular activity was decreased and secretion
was increased, suggesting that the presence of this cys-
teine retains the enzyme intracellularly. In case of Aa,
the cellular activity of AChE was also decreased by
approximately 50% but secretion was not modified,
suggesting that degradation was increased by suppres-
sion of the cysteine.
Thus, it appears that the effect of a C-terminal
cysteine on the trafficking of cholinesterase in the
secretory pathway largely depends on the nature of the
preceding catalytic domain.
Oligomerization might be affected by the distance
between the aromatic core and the C-terminal cysteine,
which forms inter-catenary disulfide bonds. There are
five residues between Y31 and this cysteine in peptide a,
and six in peptide b, because of an additional residue,
T32. To evaluate the possible influence of this residue
on oligomerization, we deleted T32 in Bb and we
mutated peptides a (CSDL to SCDL) and b (SCVGL to
CSVGL), to modify the number of residues between the
cysteine and the aromatic core. We found that these
mutations had no effect on either the levels of cellular
and secreted activities, or on the distribution of oligo-
meric forms (not shown). Similarly, these mutations did
not modify the secretion or the oligomerization of
mutants possessing a cysteine at position 19 (see below).
Thus, the addition or subtraction of one residue in the
interval between the aromatic residues and the cysteine
had no influence, suggesting that this peptidic segment
represents a flexible spacer, in agreement with previous

studies on Torpedo AChE [38].
Role of cysteines in oligomerization – effects of
introducing an additional cysteine
In a previous study, we found that mutating residue
19 in the t peptide of Torpedo AChE considerably
increased the production and secretion of tetramers
Cell extract
Medium
G
1
G
2
G
4
G
1
G
2
G
4
G
1
G
2
G
4
G
1
G
2

G
4
G
3
G
6
G
1
G
2
G
4
G
1
G
2
G
4
G
1
G
2
G
4
G
1
G
2
G
4

Aa
Bb
Ba
Ab
Aa
AChE activity (arbitrary units)BChE activity (arbitrary units)
Sedimentation coefficients
55101015 15
Wild-type t peptides
AB
With added cysteine
19C
Bb
Ba
Ab
19C
19C
19C
Fig. 3. Sedimentation profiles indicating the proportions of oligo-
meric forms produced by AChE, BChE, chimeras and mutants. (A)
Left panels: Aa, Ab, Ba, Bb. (B) Right panels: mutants containing a
cysteine at position 19 (S19C in peptide a, N19C in peptide b). The
profiles corresponding to cell extracts are shown with filled sym-
bols (d, AChE;
, BChE) and a continuous line, and those corre-
sponding to the medium with empty symbols (s, AChE; h, BChE)
and a dashed line. The peaks corresponding to tetramers, dimers
and monomers are indicated as G
4
,G

2
and G
1
, respectively. Note
that the molecular forms of BChE and its mutants sediment slightly
faster than the corresponding AChE molecular forms.
Oligomerization and secretion of AChE and BChE D. Liang et al.
98 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
[4]. We therefore introduced similar mutations in a
and b, and analyzed the resulting activities and
molecular forms produced by expressing the four
cholinesterase combinations in COS cells. This
mutation did not modify the level of secretion for
Aa
19C
, increased it for Ba
19C
, and decreased it for
Ab
19C
and Bb
19C
(Fig. 2). The fact that the cellular
activity was unchanged or decreased, whereas sec-
retion was decreased, indicates that the N19C
mutation in peptide b induced an intracellular degra-
dation of Ab
19C
and Bb
19C

. As observed in the pre-
ceding section, the ratio of secreted to cellular
activity again appeared to depend essentially on the
t peptides: it was much higher for Aa
19C
and Ba
19C
than for Ab
19C
and Bb
19C
. The 19C mutations
enhanced the difference between the two peptides
because the secreted ⁄ cellular ratio was increased with
peptide a
19C
compared to a and decreased with pep-
tide b
19C
compared to b.
By contrast to the oligomeric patterns obtained
without a cysteine at position 19, we observed a
much stronger similarity between enzymes possessing
the same C-terminal peptide (a
19C
or b
19C
) than
between those possessing the same catalytic domain
(Fig. 3B). Thus, mutation S19C in a

19C
strongly
increased the proportion of tetramers, which became
the predominant secreted species for both Aa
19C
and
Ba
19C
: these mutants produced very similar patterns
of molecular forms. The effect of mutation N19C in
peptide b
19C
had little effect on the distribution of
secreted molecular forms of Bb
19C
. In the case of
Ab
19C
, the effect was more complex: the cells con-
tained mostly 4S monomers but secreted a variety of
oligomers, mostly monomers, dimers, trimers, tetra-
mers and hexamers (see below). The fact that the
oligomeric forms were very low or undetectable in
the cells suggests that they were secreted very rapidly
after their assembly. The results were identical when
the 19C mutations were combined with mutations
that modified the distance between the C-terminal
cysteine and the aromatic residues, as indicated above
(not shown).
Taken together, these results show that the t pep-

tides possessing a cysteine at position 19 had a stron-
ger effect on the secretability of cholinesterases than
wild-type t peptides, and exerted a dominant influence
on oligomerization.
Effects of introducing cysteines at different
positions in BChE
Our previous study of Torpedo AChE showed that the
pattern of oligomerization depended critically on
the position at which a cysteine was introduced in the
t peptide [4]. Because the presence of a cysteine
induced tetramerization at position 19 of peptide a,
but not at position 19 of peptide b, we explored the
effects of cysteines at other positions in BChE. We
mutated residues that, similar to N19, are located
within the aromatic-rich segment of peptide b, but are
oriented in the opposite sector of the a helix, produc-
ing mutants A12C, H15C, D23C and N26C (Fig. 1B).
We also added a second cysteine near the C-terminus
(S37C), changing the C-terminal peptide from SCVGL
to CCVGL.
These mutations had little effect on the cellular or
secreted activities compared to wild-type BChE, except
that the secreted ⁄ cellular ratio presented a minimum
with a cysteine at position 19, and was notably
increased in the mutant possessing two C-terminal
cysteines (S37C). As shown in Fig. 4, the sedimenta-
tion profiles of cellular enzyme varied mostly in the
proportions of monomers and dimers, whereas tetra-
mers remained low. The ratio of dimers to monomers
was markedly increased with cysteines in the N-termi-

nal region of peptide b: b
12C
and even more for b
15C
.
We previously reported a similar observation in the
case of mutants of Torpedo AChE [4]. The proportion
of tetramers was higher in the medium, and was maxi-
mal with mutation N19C. Therefore, position 19
appears to be the most favorable for tetramerization,
as previously observed in the case of Torpedo AChE.
Is the difference between oligomerization and
secretion caused by individual residues that differ
between peptides a and b?
Because a and b peptides only differ at a few positions,
we introduced point mutations to reduce these differ-
ences. We made these mutations in Ab
19C
because the
level of activity, secretion and molecular forms of this
mutant were strikingly different from those of Aa
19C
(Fig. 1B). We thus mutated the first three residues of
peptide b (GNI) as a group; W8 and E9 together and
separately; G13, N18, M22 and D23 together and sepa-
rately; and N29 and D30 together, replacing these
residues by the corresponding ones in a. We also
mutated KES to QDR, and VG to DL at the C-termi-
nus. We observed no marked effect of any of these
mutations on the cellular or secreted activities, except

that mutation W8R increased both cellular activity and
secretion, in agreement with the notion that aromatic
residues induce degradation of AChE through an endo-
plasmic reticulum associated degradation process [3].
In all these mutants, the cellular extracts contained
only a trace of tetramers, as observed for Ab and
D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 99
Ab
19C
(Fig. 5). The sedimentation profiles of the
secreted enzyme were similar to those obtained with
Ab
19C
, except that the proportion of tetramers (G
4
)
was somewhat increased with N18S. The M22V muta-
tion mostly increased the 13.5S species, and the D23H
mutation did not increase G
4
by itself, but their combi-
nation, M22V ⁄ D23H, induced a significant increase in
the proportion of secreted tetramers.
Hoping to obtain a higher yield of secreted tetra-
mers, we then combined the N18S and M22D ⁄ D23H
mutations in Ab
19C
. The combination of mutations
N18S, N19C, M22D and D23H, abbreviated as S, pro-

duced the highest proportion of secreted G
4
tetramers
and the highest secreted ⁄ cellular activity ratio. Because
these mutations appear to favor the production of
tetramers with the b peptide, we introduced them,
separately and together, in Bb
19C
. However, the
resulting Bb
S
mutant did not produce a higher
proportion of tetramers than Bb
19C
(Fig. 5).
Stokes radius and mass of oligomers
We wished to further characterize the oligomers of Ab
19C
and other mutants, some of which sedimented faster than
tetramers, at 12.3S and 13.5S. Because cholinesterase
oligomers may be associated with elongated proteins
such as collagen ColQ, their mass cannot be simply
deduced from their sedimentation coefficient, but rather
from a combination of their Stokes radius and sedimen-
tation coefficient [39]. We therefore used gel filtration
chromatography to determine the Stokes radius of oligo-
mers secreted by the mutant Ab
N19C-N29D-D30H
, which
was chosen because it produces the complete variety of

Ab oligomers (Fig. 6A). The major oligomers were iso-
lated from gradient fractions. By comparison with the
standard proteins b-galactosidase and alkaline phospha-
tase, we obtained Stokes radii values, as indicated in
Fig. 6B. We then determined the mass of these oligomers,
assuming that it is proportional to the product of the sed-
imentation coefficient and Stokes radius, as expected for
proteins of similar density. The values thus obtained indi-
cated a globular structure because the mass was in fact
proportional to S
3 ⁄ 2
. This relationship allowed us to
determined the mass of the minor components, sediment-
ing at 8.5S and 12.3S (Fig. 6A). Figure 6C shows that
Cell extract
Medium
G
1
G
2
G
4
G
1
G
2
G
4
G
1

G
2
G
4
G
1
G
2
G
4
G
1
G
2
G
4
G
1
G
2
G
4
Bb
A12C
BChE activity (arbitrary units)
Sedimentation coefficients
51015
Bb
H15C
Bb

N19C
Bb
D23C
Bb
N26C
Bb
S37C
Fig. 4. Sedimentation profiles of mutants of human BChE (Bb)
with cysteines at positions 12, 15, 19, 23 and 26. The profiles
obtained for Bb
19C
, also shown in Fig. 3, are repeated here for
comparison with the other mutants. The symbols are as in Fig. 3.
Tetramers, dimers and monomers are indicated as G
4
,G
2
and G
1
,
respectively. The mutations replacing various residues by cysteines
in the C-terminal peptide are indicated.
Oligomerization and secretion of AChE and BChE D. Liang et al.
100 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
the masses of the six observed oligomers represent multi-
ples of the smaller one, demonstrating that they represent
monomers (G
1
), dimers (G
2

), trimers (G
3
), tetramers
(G
4
), pentamers (G
5
) and hexamers (G
6
).
Trimers, pentamers and hexamers were only formed
with an additional cysteine (S19C). Thus, mutants of
Ab can associate into these different multimers,
illustrating the versatility of associations between
t peptides possessing a cysteine at position 19. As
noted above, most of these oligomers were observed in
the medium but not in cells. By contrast, BChE only
formed monomers, dimers and tetramers.
Discussion
The C-terminal t peptides do not influence
cholinesterase activity
The catalytic domain of cholinesterases is associated
with two major types of C-terminal peptides: the
h peptides contain a signal for the post-translational
addition of a glycolipid anchor and cysteines that
allow the formation of disulfide-linked dimers, and the
t peptides allow the formation of a variety of oligo-
mers. These peptides are not required for catalytic
activity because truncated enzymes, which are reduced
to their catalytic domains, are fully active. Previous

studies showed that oligomers of AChE
T
subunits pos-
sessed the same turnover rate per site, but this did not
exclude a possible influence of the nature of C-terminal
peptides. To examine this question, we titrated the
active sites of truncated, wild-type and chimeric cho-
linesterases with the irreversible inhibitor DEPQ, and
compared their activities with the substrates acetylthio-
choline and butyrylthiocholine. We found that the cat-
alytic rate per active site only depends on the catalytic
domain: it was identical for truncated enzymes (A or
B) and with enzymes possessing either a or b C-termi-
nal peptides, in agreement with previous studies [40]
showing that the variants AChE
T
, AChE
R
and a trun-
cated mutant possessed the same K
m
value and excess
substrate inhibition. These results also show that the
catalytic activity is not influenced by the oligomeric
state of the enzymes, and thus justifies quantitative
comparisons between the activities of the various
mutants investigated in the present study.
Effect of the C-terminal t peptides on folding
and secretion
The t peptides of cholinesterases form amphiphilic

a helices with a sector containing their seven conserved
aromatic residues. This organization is critical for the
association of cholinesterase tetramers with anchoring
proteins containing a PRAD, and most probably also
for the assembly of homomeric tetramers. However,
we have previously shown that the presence of aro-
matic residues in the t peptide reduces the production
and secretion of AChE, at two distinct checkpoints
[41]. First, it induces a partial misfolding of newly
synthesized polypeptides; this effect depends on the
G
1
G
2
G
4
G
3
G
6
G
1
G
2
G
4
G
6
cell extract
medium

G
1
G
2
G
4
G
6
G
1
G
2
G
4
G
1
G
2
G
4
G
3
G
6
G
1
G
2
G
4

G
3
G
6
G
1
G
2
G
4
G
3
Sedimentation coefficients
51015
Ab N18S
N19C
Ab N19C
M22V
Ab N19C
D23H
Ab N19C
M22V D23H
Ab N18S N19C
M22V D23H
G
1
G
2
G
6

G
1
G
2
G
6
Bb N18S
N19C
Bb N18S N19C
M22V D23H
Bb N19C
M22V D23H
AChE activity (arbitrary units)
BChE activity (arbitrary units)
51015
Ab N19C
Fig. 5. Effect of mutations suppressing differences between a and
b, on the distribution of oligomeric forms. (A) Left panels and top
right panel: Ab
19C
. (B) Lower right panels: Bb
19C
. Sedimentation
patterns are shown as in Fig. 3. The sedimentation profiles of the
Ab
19C
mutant (top right panel) are repeated for comparison with
those obtained with additional mutations, which suppressed some
of the differences with peptide a. Note that the effects of muta-
tions M22V and D23H are not additive.

D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 101
hydrophobic character of these residues because the
same effect was observed when they were replaced by
leucines [5]. Second, they target a fraction of active
AChE subunits towards degradation by endoplasmic
reticulum associated degradation rather than secretion;
this effect depends on the presence of aromatic resi-
dues, rather than on hydrophobicity. This quality con-
trol process may ensure that only correctly assembled
subunits are efficiently exported from the cells.
The present results confirm that the production,
secretion and oligomerization of human AChE and
BChE are strongly influenced by their t peptides. In
agreement with previous results, secretion was consid-
erably increased for both enzymes when the t peptides
were deleted. Using chimeras in which these peptides
were exchanged (Ab, Ba), we further showed that the
ratio between secreted and cellular activities, which
may be taken as an index of secretability, was essen-
tially determined by the t peptide. The rate of secretion
with the t peptide from AChE (a) was more than two-
fold higher than with the t peptide from BChE (b).
With modified t peptides possessing a cysteine near the
center of the aromatic cluster (S19C in a and N19C
in b), this difference became more than six-fold.
The respective roles of the catalytic domains
and t peptides in oligomerization
Although the truncated A and B mutants only pro-
duced monomers, the Aa, Ab, Ba and Bb enzymes all

formed oligomers, including tetramers. Because these
tetramers were obtained without co-expression with a
PRAD-containing protein, they most probably repre-
sent homotetramers, in which the four t peptides may
form a coiled coil complex with all aromatic residues
oriented inwards, but without a central PRAD. This
hypothesis is supported by the fact that, although the
presence of a PRAD only induces the assembly of tet-
ramers, expression of some mutants without a PRAD
produces tetramers together with other oligomers,
including molecular forms sedimenting as trimers, pen-
tamers and hexamers. The odd-numbered complexes
are not likely to represent heteromeric associations
containing other proteins because they only occur with
some Ab mutants with an added cysteine, and their
masses correspond exactly to those expected for multi-
ples of AChE subunits. Because the formation of these
unusual oligomers appears to depend strictly on the
presence of an additional cysteine, they are probably
stabilized by a network of inter-catenary disulfide
bonds, linking all subunits together.
The Ab
19C
chimera formed all oligomeric forms
from monomers to hexamers, illustrating the versatility
of oligomeric associations based on the t peptide, in
association with the catalytic domain. It should be
noted that hexamers have been observed in transfected
COS cells expressing wild-type rat AChE, and
appeared as a transient mode of association, which

could be dissociated into monomers, dimers and tetra-
mers (e.g. in the presence of Triton X-100) [7]. By
Ab N19C
N29D-D30H
secreted activity
G
1
G2
G3
G4
G5
G6
Sedimentation coefficients
51015
Arbitrary units
G6
0 0.2 0.4 0.6 0.8 1
Arbitrary scales
Elution coefficient (Ve-Vo)/(Vt-Vo)
-galactosidase Alkaline phosphatase
G4
G2
G1
01234567
0
100
200
300
400
500

4.3 S
6.5 S
10.5 S
13.5 S
8.5 S
12.3 S
Masses of oligomers (kDa)
Numbers of subunits
G6
G5
G4
G3
G2
G1
AB C
Fig. 6. Determination of the Stokes radius and mass of AChE
B
oligomers. (A) Oligomers were isolated from sucrose gradients of medium
from cells expressing the Ab
19C-29D-30H
mutant. The profile of cellular activity was identical to that shown in Fig. 3B for the Ab
19C
mutant.
(B) Elution of oligomers in gel filtration chromatography. The elution parameters were defined as K
e
=(V
e
– V
o
) ⁄ (V

t
– V
o
), where V
o
corre-
sponds to the exclusion volume (blue dextran) and V
t
is the total volume (potassium ferricyanide). The Stokes radii were determined from
the linear relationship between the Stokes radius and the square root of [)log (K
e
)] using the standards b-galactosidase (6.9 nm, 16S,
464 kDa) and alkaline phosphatase (3.3 nm, 6.1S; 87 kDa). (C) The masses of the different oligomers were determined by their proportional-
ity to the product of the Stokes radius with the sedimentation coefficient. The masses of the minor 8.5S and 12.3S species were deter-
mined from the linear relationship with S
3 ⁄ 2
, observed for the other oligomers. The masses are found to be proportional to discrete degrees
of oligomerization, from 1 to 6, showing that the oligomers correspond to monomers (G
1
), dimers (G
2
), trimers (G
3
), tetramers (G
4
), penta-
mers (G
5
) and hexamers (G
6

).
Oligomerization and secretion of AChE and BChE D. Liang et al.
102 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
contrast, Bb
19C
only formed the classical monomers,
dimers and tetramers, possibly because of steric
constraints due to the catalytic domain or to its associ-
ated N-glycans.
Although the nature and proportions of oligomers
depended on the presence of the t peptides and their
cysteines, the catalytic domains also influenced the
oligomerization patterns. The cellular and secreted
oligomers formed by Aa and Ab were very similar, as
well as those formed by Ba and Bb, suggesting a pre-
dominant influence of the catalytic domains on oligo-
merization. This may be due in part to the difference
in N-glycosylation of AChE and BChE, which carry
four and nine N-glycan chains, respectively [35]; we
therefore compared the oligomeric patterns of
wild-type BChE and mutants in which part of the
N-glycosylation sites were mutated [36], but observed
no difference (not shown). The relative influence of the
C-terminal t peptide appeared to be strongly increased
when a cysteine was added at position 19 because the
patterns obtained for Aa and Ba were almost the same,
except for a shift in the sedimentation coefficients,
which are higher for BChE than for AChE.
By contrast to Aa and Ab, oligomers of Ba and Bb
represented a significant proportion of cellular activity,

indicating that AChE oligomers were secreted more
rapidly after assembly than BChE oligomers. It is
remarkable that the Ab
19C
oligomers were observed in
the medium but not in cell extracts. This could be
related to the presence of the peptide b
19C
which
reduces secretion but may be masked in the oligomers.
Thus, both the catalytic domain and the C-terminal
t peptides contribute to the control of oligomerization
and secretion, in a complex interplay.
Origin and significance of the difference between
the t peptides of AChE and BChE
The a and b peptides present a considerable sequence
similarity, with 60% identical residues, including the
seven aromatic residues and the cysteine, which play a
key role in the interaction properties of the t peptides.
In addition, both peptides are predicted to possess the
same tendency to form amphiphilic a helices. It was
therefore unexpected to observe a strong difference in
their influence on the oligomerization and secretion of
AChE and BChE. We tried to assign this difference
to some of the residues that distinguish the a and b
peptides. Because oligomerization also depends on the
catalytic domains, as indicated by the difference
between the molecular forms produced by Ab
19C
and

Bb
19C
, the linkage between the two domains might well
play a crucial role in the quaternary associations of
the cholinesterase subunits. The first three residues of
peptides a and b are indeed different, but their replace-
ment in Ab
19C
(GNI to DTL) had little effect on either
secretion or oligomerization. It is also noteworthy that
the effects of the combined mutations M22D ⁄ V23H
could not be simply accounted for by the effects of the
separate mutations M22V and D23V. This suggests
that the secretory trafficking of molecules containing
peptides a and b depends on global properties of the
peptides rather than on individual residues.
AChE and BChE are expressed differentially during
embryogenesis [42–44]. They appear to play distinct
roles, which may be based on their catalytic activity,
but also on protein–protein interactions [45], because
their catalytic domain is homologous to adhesion pro-
teins such as neuroligin [46,47]. For example, AChE
may be involved in neurite extension during brain
development [40,48,49]. Both catalytic and noncatalytic
functions clearly require appropriate oligomeric orga-
nization and localization and therefore depend on the
C-terminal t peptides, which may be directly involved
in distinct interactions with partner proteins.
The two cholinesterases present a complex relation-
ship with the development of Alzheimer’s disease, which

may be partly related to their C-terminal t peptides.
Both AChE and BChE are associated with senile pla-
ques in Alzheimer’s disease [50] but they appear to play
antagonistic roles: AChE promotes amyloid aggregation
and increases the neurotoxicity of the Ab peptide
in vitro, suggesting that it may participate in the patho-
genesis of the disease [51,52]; this appears to depend on
interactions of Ab peptides with the peripheral site of
AChE and not on its C-terminal t peptide, which has no
effect on Ab aggregation [53]. By contrast, the C-termi-
nal t peptide of BChE (peptide b) was found to reduce
Ab aggregation, possibly because of the presence of its
additional tryptophan (W8) located opposite to the aro-
matic cluster of the amphiphilic helix (Fig. 1B), so that
BChE might have a protective effect against Alzheimer’s
disease [53,54]. In this respect, it is worth recalling that,
although the human AChE t peptide (here termed
peptide a) is organized as an a helix, its AEFHRWS-
SYMVHWK fragment, which resembles a portion of
the amyloid Ab peptide (AEFRHDSGYEVHHQK),
was found to organize into b sheets and to form fibrils;
by contrast, the homologous fragment from BChE
(AGFHRWNNYMMDWK) did not possess this
property [55–57].
AChE and BChE probably arose from a gene dupli-
cation in the lineage of vertebrates and it is remarkable
that sequence differences between their t peptides are
strongly conserved, suggesting that they correspond to
distinct molecular interactions and the oligomerization
D. Liang et al. Oligomerization and secretion of AChE and BChE

FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 103
of these sister enzymes. The t peptides of higher mam-
mals, including rat, mouse, rabbit, horse, bovine, dog,
cat and primates, are identical and share 22 common
residues with the t peptide of chicken AChE (however,
there are more differences with marsupials) (Fig. 7).
The t peptides of BChE show more variation between
mammalian species: there are a few differences between
man and Maccaca mulatta, and the K variant of
human BChE, which occurs with high frequency in
European and American populations, consists of the
replacement of A6 by a threonine [58], but it does not
affect the assembly of tetramers [59]. Quite surpris-
ingly, the human t peptide shares 34 common residues
with that of chicken and only 13 of these residues are
common to both AChEs and BChEs (ten if one
considers marsupials), including the seven conserved
aromatic residues and the C-terminal cysteine. Thus,
the differences between the t peptides of AChE and
BChE are conserved in vertebrates, suggesting that
they are functionally significant. It is possible that,
although the physiological localization of AChE in
cholinergic tissues depends on its association with its
anchoring proteins ColQ and PRiMA, the major func-
tion of BChE rather depends on other interactions,
particularly on the formation of soluble tetramers
circulating in the bloodstream.
Production of recombinant AChE or BChE
tetramers
We have shown that mutation S19C in the C-terminal

peptide of human AChE (Aa
19C
) allows the production
of recombinant secreted AChE homotetramers. This
may be useful to obtain a stable form of AChE that
could serve in the decontamination of toxic anti-ChE
organophosphates, or could be injected as a pre-treat-
ment against possible intoxication by such compounds.
The chimeric protein Ba
19C
also formed homotetra-
mers. However, introduction of a cysteine at the
homologous position in Bb (Bb
19C
) only produced
approximately 30% of tetramers, and this proportion
was not significantly increased by several mutations
that abolished differences between peptides a and b.
This was unexpected because soluble BChE homotetra-
mers were considered to represent the major species of
this enzyme in the bloodstream. This suggests that the
physiological assembly of BChE tetramers depends on
the presence of a PRAD-containing protein or peptide,
in agreement with the recent discovery that human
plasma BChE contains a proline-rich peptide from
lamellipodin [14].
In conclusion, the catalytic domains of cholinesterases
and their C-terminal t peptides constitute modules of
quaternary interactions that are only partially indepen-
dent. The conservation of these peptides during the

evolution of vertebrates probably reflects their subtly
distinct functions, associated with the respective roles
of AChE and BChE in synaptic and nonsynaptic
contexts.
Experimental procedures
Mutations and constructs
The coding sequences of human AChE
T
(T variant, Aa)
and BChE
T
(Bb), inserted in the pGS vector, were gener-
ously provided by O. Lockridge. The residues of the t pep-
tides are numbered from the first residue following the
catalytic domain. To exchange the C-terminal t peptides,
BsiWI restriction sites were introduced at the junction
between regions encoding the catalytic domains and the
t peptides. Fragments encoding the t peptides were cut
between these sites and a downstream SacII site in the
vector, purified in agarose gels, and religated with the
appropriate complementary fragment. The nucleotides sepa-
rating the coding sequences of the catalytic domains and
C-terminal peptides were then removed by site-directed
mutagenesis with the method of Kunkel et al. [60]. Other
mutations were performed with this method.
Cell cultures and transfection
Transient transfection of COS cells was performed by the
DEAE-dextran method as described previously [61], usually
AChE T Chicken
GPPEDAEREWRLEFHRWSSYMGRWRTQFEHYSRQQPCATL

Man
DTLDEAERQWKAEFHRWSSYMVHWKNQFDHYSKQDRCSDL
Consensus chicken/man
AER-W EFHRWSSYM W QF-HYS-Q C L
BChE T Chicken
GNIDEAEREWKAGFHRWNNYMMDWKNQFNDYTSKKERCAGSN
Man
GNIDEAEWEWKAGFHRWNNYMMDWKNQFNDYASKKESCVGL
Consensus chicken/man
GNIDEA EWKAGFHRWNNYMMDWKNQFNDY-SKKE-C
Consensus AChE/BChE
AE W F-RW YM W QF Y C
Fig. 7. Sequence conservation between a and b in vertebrates. The conservation between a peptides and between b peptides is illustrated,
using human and chicken sequences as an example. The common consensus for both peptides includes only five residues, in addition to
the universally conserved aromatic residues and C-terminal cysteine.
Oligomerization and secretion of AChE and BChE D. Liang et al.
104 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
with 2 lg DNA per 60 mm dish. Transfected COS cells
were incubated at 37 °C, in a medium containing 10%
NUserum (Inotech, Dottikon, Switzerland), which had been
pre-treated with 10
)6
m soman (Centre d’Etudes du Bou-
chet, Vert-le-Petit, France) to inactivate serum cholinester-
ases. The cells and the culture media were collected
3–4 days after transfection. Intracellular and membrane-
bound enzyme was extracted for 15 min at 20 °C in a TMg
buffer (Tris–HCl 50 mm, pH 7,4; 1% Triton X-100, 10 mm
MgCl
2

) followed by centrifugation for 15 min at 10 000 g
at 4 °C. The culture medium containing the secreted
enzyme was also centrifuged at 10 000 g for 10 min to
remove cell debris before analysis.
Enzyme assays
Enzyme activity was quantified by the colorimetric
method of Ellman et al. [62]. The reaction medium con-
tained 0.5 mm acetylthiocholine or butyrylthiocholine,
0.5 mm 5,5¢-dithiobis(2-nitrobenzoic acid) (Nbs
2
) and
0.05 m Na
+
phosphate (pH 7). The reaction was moni-
tored at 414 nm in a Multiskan RC automatic plate
reader (Labsystems, Helsinki, Finland); the optical density
was recorded at 20 s intervals over a period of 10 min.
Alkaline phosphatase and b-galactosidase from Escherichia
coli were assayed with the chromogenic substrates
p-nitrophenyl phosphate and o-nitrophenyl galactoside,
respectively. For titration of active sites, the culture
media were adjusted to pH 7 with 1 m Tris–HCl (pH
9.5) and 50 lL samples were treated with 100 lLofa
series of dilutions of DEPQ in 10 mm Na
+
phosphate
buffer (pH 7.4) overnight at room temperature; the
remaining AChE activity was then determined with the
Ellman colorimetric method, by addition of 150 lLof
50 mm Na

+
phosphate (pH 7.4) buffer containing 1 mm
Nbs
2
and substrates. The final concentrations were 2 mm
acetylthiocholine for AChE and 6 mm acetylthiocholine
or butyrylthiocholine for BChE.
Sedimentation analyses and gel filtration
chromatography
Centrifugation in 5–20% sucrose gradients (50 mm
Tris–HCl, pH 7.5, 20 mm MgCl
2
, in the presence of 1%
Brij-96 was performed in a Beckman SW41 rotor (Beckman
Coulter, Fullerton, CA, USA) at 36 000 r.p.m. for 17 h
30 min at 4 °C. Approximately 40 fractions were collected
and cholinesterase activity was measured with the Ellman
colorimetric assay [62] using acetylthiocholine as a substrate
for AChE or BChE, and butyrylthiocholine for BChE. The
gradients contained E. coli b-galactosidase (16S) and
alkaline phosphatase (6.1S) as internal sedimentation
standards.
Gel filtration in Biogel A-1.5m (Bio-Rad Laboratories
Inc., Hercules, CA, USA) was performed to determine the
Stokes radii of complexes. The flow rate of the column
(length 95 cm, diameter 1.5 cm) was approximately
10 mLÆh
)1
, and fractions were collected every 10 min. The
void volume was determined with blue dextran and the

total volume with potassium ferricyanide; the standards
used were b-galactosidase (6.9 nm, 464 kDa) and alkaline
phosphatase (3.3 nm, 87 kDa).
Prediction of secondary structure
The secondary structure of mutated t peptides was pre-
dicted according to Rost [63], using the predictprotein
website ().
Acknowledgements
We thank Dr Oksana Lockridge for the generous gift of
vectors expressing wild-type and mutated human AChE
and BChE, and Dr Yaacov Ashani for DEPQ. This
work was supported by the US Defense Threat Reduc-
tion Agency under contract W91ZLK-06-C-0020, by the
Centre National de la Recherche Scientifique, and by
the Association Franc¸ aise contre les Myopathies.
References
1 Massoulie
´
J (2002) The origin of the molecular diversity
and functional anchoring of cholinesterases. NeuroSig-
nals 11, 130–143.
2 Massoulie
´
J & Bon S (2006) The C-terminal T peptide
of cholinesterases: structure, interactions, and influence
on protein folding and secretion. J Mol Neurosci 30,
233–236.
3 Belbeoc’h S, Massoulie
´
J & Bon S (2003) The C-termi-

nal T peptide of acetylcholinesterase enhances degrada-
tion of unassembled active subunits through the ERAD
pathway. EMBO J 22, 3536–3545.
4 Belbeoc’h S, Falasca C, Leroy J, Ayon A, Massoulie
´
J
& Bon S (2004) Elements of the C-terminal t peptide
of acetylcholinesterase that determine amphiphilicity,
homomeric and heteromeric associations, secretion and
degradation. Eur J Biochem 271, 1476–1487.
5 Falasca C, Perrier NA, Massoulie
´
J & Bon S (2005)
Determinants of the t peptide involved in folding,
degradation and secretion of acetylcholinesterase.
J Biol Chem 280, 878–886.
6 Massoulie
´
J, Bon S, Perrier N & Falasca C (2005) The
C-terminal peptides of acetylcholinesterase: cellular traf-
ficking, oligomerization and functional anchoring. Chem
Biol Interact 157-158, 3–14.
7 Bon S & Massoulie
´
J (1997) Quaternary associations
of acetylcholinesterase. I Oligomeric associations of
T subunits with and without the amino-terminal domain
of the collagen tail. J Biol Chem 272, 3007–3015.
D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 105

8 Blong RM, Bedows E & Lockridge O (1997) Tetramer-
ization domain of human butyrylcholinesterase is at the
C-terminus. Biochem J 327, 747–757.
9 Krejci E, Thomine S, Boschetti N, Legay C, Sketelj J &
Massoulie
´
J (1997) The mammalian gene of acetylcho-
linesterase-associated collagen. J Biol Chem 272, 22840–
22847.
10 Perrier AL, Massoulie
´
J & Krejci E (2002) PRiMA,
the membrane anchor of acetylcholinesterase in brain.
Neuron 33, 275–285.
11 Bon S, Coussen F & Massoulie
´
J (1997) Quaternary
associations of acetylcholinesterase; II. the polyproline
attachment domain of the collagen tail. J Biol Chem
272, 3016–3021.
12 Noureddine H, Schmitt C, Liu WQ, Garbay C, Massoulie
´
J & Bon S (2007) Assembly of acetylcholinesterase tetra-
mers by peptidic motifs from the proline-rich membrane
anchor, PRiMA: competition between degradation and
secretion pathways of heteromeric complexes. J Biol
Chem 282, 3487–3497.
13 Noureddine H, Carvalho S, Schmitt C, Massoulie
´
J&

Bon S (2008) Acetylcholinesterase associates differently
with its anchoring proteins ColQ and PRiMA. J Biol
Chem 283, 20722–20732.
14 Li H, Schopfer LM, Masson P & Lockridge O (2008)
Lamellipodin proline rich peptides associated with
native plasma butyrylcholinesterase tetramers. Biochem
J 411, 425–432.
15 Dvir H, Harel M, Bon S, Liu WQ, Vidal M, Garbay C,
Sussman JL, Massoulie
´
J & Silman I (2004) Structural
insight into tetramerization of the synaptic form of ace-
tylcholinesterase: interaction of a polyproline II helix
with four WWW motifs. EMBO J 23, 4394–4405.
16 Morel N, Leroy J, Ayon A, Massoulie
´
J & Bon S
(2001) Acetylcholinesterase H and T dimers are associ-
ated through the same contact; mutations at this inter-
face interfere with the C-terminal T peptide, inducing
degradation rather than secretion. J Biol Chem 276,
37379–37389.
17 Simon S, Krejci E & Massoulie
´
J (1998) A four-to-one
association between peptide motifs: four C-terminal
domains from cholinesterase assemble with one proline-
rich attachment domain (PRAD) in the secretory path-
way. EMBO J 17 , 6178–6187.
18 Sussman JL, Harel M, Frolow F, Oefner C, Goldman

A, Toker L & Silman I (1991) Atomic structure of ace-
tylcholinesterase from Torpedo californica: a prototypic
acetylcholine-binding protein. Science 253, 872–879.
19 Feng G, Krejci E, Molgo J, Cunningham JM, Massou-
lie
´
J & Sanes JR (1999) Genetic analysis of collagen Q:
roles in acetylcholinesterase and butyrylcholinesterase
assembly and in synaptic structure and function. J Cell
Biol 144, 1349–1360.
20 Perrier NA, Khe
´
rif S, Perrier AL, Krejci E, Dumas S,
Mallet J & Massoulie
´
J (2003) Expression of PRiMA,
the membrane anchor of acetylcholinesterase, in the
mouse brain. Eur J Neurosci
18, 1837–1847.
21 Raveh L, Ashani Y, Levy D, De La Hoz D, Wolfe AD
& Doctor BP (1989) Acetylcholinesterase prophylaxis
against oprganophosphate poisoning; quantirtative
correlation between protection and blood-enzyme level
in mice. Biochem Pharmacol 38, 529–534.
22 Broomfield CA, Maxwell DM, Solana RP, Castro CA,
Finger AV & Lenz DE (1991) Protection by butyrylcho-
linesterase against organophosphorus poisoning in non-
human primates. J Pharmacol Exp Ther 259, 633–638.
23 Maxwell DM, Castro CA, De La Hoz DM, Gentry
MK, Gold MB, Solana RP, Wolfe AD & Doctor BP

(1992) Protection of Rhesus monkeys against soman
and prevention of performance decrement by preteat-
ment with acetylcholinesterase. Toxicol Appl Pharmacol
115, 44–49.
24 Doctor BP, Blick DW, Caranto G, Castro CA, Gentry
MK, Maxwell DM, Murphy MR, Schutz M, Waibel K
& Wolfe AD (1993) Cholinesterases as scavengers for
organophosphorus compounds: protection of primate
performance against soman toxicity. Chem Biol Interact
87, 285–293.
25 Maxwell DM, Saxena A, Gordon RK & Doctor BP
(1999) Improvements in scavenger protection against
organophosphorus agents by modification of cholines-
terases. Chem Biol Interact 119-120, 419–428.
26 Chitlaru T, Kronman C, Velan B & Shafferman A
(2001) Effect of human acetylcholinesterase subunit
assembly on its circulatory residence. Biochem J 354,
613–625.
27 Doctor BP & Saxena A (2005) Bioscavengers for the
protection of humans against organophosphate toxicity.
Chem Biol Interact 157-158, 167–171.
28 Cohen O, Kronman C, Chitlaru T, Ordentlich A, Velan
B & Shafferman A (2001) Effect of chemical modifica-
tion of recombinant human acetylcholinesterase by
polyethylene glycol on its circulatory longevity. Biochem
J 357, 795–802.
29 Chilukuri N, Parikh K, Sun W, Naik R, Tipparaju P,
Doctor BP & Saxena A (2005) Polyethylene glycosyla-
tion prolongs the circulatory stability of recombinant
human butyrylcholinesterase. Chem Biol Interact 157-

158, 115–121.
30 Cohen O, Kronman C, Raveh L, Mazor O, Ordentlich
A & Shafferman A (2006) Comparison of polyethylene
glycol-conjugated recombinant human acetylcholinester-
ase and serum human butyrylcholinesterase as biosca-
vengers of organophosphate compounds. Mol
Pharmacol 70, 1121–1131.
31 Cohen O, Kronman C, Lazar A, Velan B & Shafferman
A (2007) Controlled concealment of exposed clearance
and immunogenic domains by site-specific polyethylene
glycol attachment to acetylcholinesterase hypolysine
mutants. J Biol Chem 282, 35491–35501.
Oligomerization and secretion of AChE and BChE D. Liang et al.
106 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works
32 Mazor O, Cohen O, Kronman C, Raveh L, Stein D,
Ordentlich A & Shafferman A (2008) Aging-resistant
organophosphate bioscavenger based on polyethylene
glycol-conjugated F338A human acetylcholinesterase.
Mol Pharmacol 74, 755–763.
33 Chilukuri N, Sun W, Naik RS, Parikh K, Tang L, Doc-
tor BP & Saxena A (2008) Effect of polyethylene glycol
modification on the circulatory stability and immuno-
genicity of recombinant human butyrylcholinesterase.
Chem Biol Interact 175, 255–260.
34 Massoulie
´
J, Bon S, Rieger F & Vigny M (1975) Molec-
ular forms of acetylcholinesterase. Croat Chem Acta 47,
163–179.
35 Lockridge O, Bartels CF, Vaughan TA, Wong CK,

Norton SE & Johnson LL (1987) Complete amino acid
sequence of human serum cholinesterase. J Biol Chem
262, 549–557.
36 Nachon F, Nicolet Y, Viguie
´
N, Masson P, Fontecilla-
Camps JC & Lockridge O (2002) Engineering of a
monomeric and low-glycosylated form of human
butyrylcholinesterase: expression, purification, charac-
terization and crystallization. Eur J Biochem 269,
630–637.
37 Vigny M, Gisiger V & Massoulie
´
J (1978) ‘Nonspecific’
cholinesterase and acetylcholinesterase in rat tissues:
molecular forms, structural and catalytic properties,
and significance of the two enzyme systems. Proc Natl
Acad Sci USA 75, 2588–2592.
38 Bon S, Dufourcq J, Leroy J, Cornut I & Massoulie
´
J
(2004) The C-terminal t peptide of acetylcholinester-
ase forms an alpha helix that supports homomeric
and heteromeric interactions. Eur J Biochem 271,
33–47.
39 Bon S, Vigny M & Massoulie
´
J (1979) Asymmetric and
globular forms of acetylcholinesterase in mammals and
birds. Proc Natl Acad Sci USA 76, 2546–2550.

40 Sternfeld M, Ming G, Song H, Sela K, Timberg R, Poo
M & Soreq H (1998) Acetylcholinesterase enhances neu-
rite growth and synapse development through alternative
contributions of its hydrolytic capacity, core protein, and
variable C termini. J Neurosci 18, 1240–1249.
41 Massoulie
´
J, Perrier N, Noureddine H, Liang D & Bon
S (2008) Old and new questions about cholinesterases.
Chem Biol Int 175, 30–44.
42 Layer PG, Alber R & Rathjen FG (1988) Sequential
activation of butyrylcholinesterase in rostral half som-
ites and acetylcholinesterase in motoneurones and
myotomes preceding growth of motor axons. Develop-
ment 102, 387–396.
43 Layer PG (1990) Cholinesterases preceding major tracts
in vertebrate neurogenesis. Bioessays 12, 415–420.
44 Layer PG, Weikert T & Alber R (1993) Cholinesterases
regulate neurite growth of chick nerve cells in vitro by
means of a non-enzymatic mechanism. Cell Tissue Res
273, 219–226.
45 Paraoanu LE & Layer PG (2005) Mouse AChE binds
in vivo to domain IV of laminin-1beta. Chem Biol Inter-
act 157 ⁄ 158, 411–413.
46 Krejci E, Duval N, Chatonnet A, Vincens P & Massou-
lie
´
J (1991) Cholinesterase-like domains in enzymes and
structural proteins: functional and evolutionary rela-
tionships and identification of a catalytically essential

aspartic acid. Proc Natl Acad Sci USA 88, 6647–6651.
47 Grifman M, Galyam N, Seidman S & Soreq H (1998)
Functional redundancy of acetylcholinesterase and neu-
roligin in mammalian neuritogenesis. Proc Natl Acad
Sci USA 95, 13935–13940.
48 Johnson G & Moore SW (2000) Cholinesterases
modulate cell adhesion in human neuroblastoma cells
in vitro. Int J Dev Neurosci 18, 781–790.
49 Sharma KV, Koenigsberger C, Brimijoin S & Bigbee
JW (2001) Direct evidence for an adhesive function in
the noncholinergic role of acetylcholinesterase in neurite
outgrowth. J Neurosci Res 63, 165–175.
50 Mesulam M, Carson K, Price B & Geula C (1992) Cho-
linesterases in the amyloid angiopathy of Alzheimer’s
disease. Ann Neurol 31, 565–569.
51 Inestrosa NC, Alvarez A, Perez CA, Moreno RD,
Vicente M, Linker C, Casanueva OI, Soto C & Garrido J
(1996) Acetylcholinesterase accelerates assembly of amy-
loid-beta-peptides into Alzheimer’s fibrils: possible role
of the peripheral site of the enzyme. Neuron 16, 881–891.
52 Inestrosa NC & Alarcon R (1998) Molecular interac-
tions of acetylcholinesterase with senile plaques. J Phys-
iol (Paris) 92, 341–344.
53 Diamant S, Podoly E, Friedler A, Ligumsky H, Livnah
O & Soreq H (2006) Butyrylcholinesterase attenuates
amyloid fibril formation in vitro. Proc Natl Acad Sci
USA 103, 8628–8633.
54 Podoly E, Bruck T, Diamant S, Melamed-Book N,
Weiss A, Huang Y, Livnah O, Langermann S, Wilgus
H & Soreq H (2008) Human recombinant butyrylcho-

linesterase purified from the milk of transgenic goats
interacts with beta-amyloid fibrils and suppresses their
formation in vitro. Neurodegener Dis 5, 232–236.
55 Greenfield SA & Vaux DJ (2002) Parkinson’s disease,
Alzheimer’s disease and motor neurone disease:
identifying a common mechanism. Neuroscience 113,
485–492.
56 Cottingham MG, Hollinshead MS & Vaux DJ (2002)
Amyloid fibril formation by a synthetic peptide from a
region of human acetylcholinesterase that is homolo-
gous to the Alzheimer’s amyloid-beta peptide. Biochem-
istry 41, 13539–13547.
57 Cottingham MG, Voskuil JLA & Vaux DJT (2003) The
intact human acetylcholinesterase C-terminal oligomeri-
zation domain is alpha-helical in situ and in isolation,
but a shorter fragment forms beta-sheet-rich amyloid
fibrils and protofibrillar oligomers. Biochemistry 42,
10863–10873.
D. Liang et al. Oligomerization and secretion of AChE and BChE
FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works 107
58 Bartels CF, Jensen FS, Lockridge O, van der Spek AF,
Rubinstein HM, Lubrano T & La Du BN (1992) DNA
mutation associated with the human butyrylcholinester-
ase K-variant and its linkage to the atypical variant
mutation and other polymorphic sites. Am J Hum Genet
50, 1086–1103.
59 Altamirano CV, Bartels CF & Lockridge O (2000) The
butyrylcholinesterase K-variant shows similar cellular
protein turnover and quaternary interaction to the wild-
type enzyme. J Neurochem 74, 869–877.

60 Kunkel TA, Roberts JD & Zakour RA (1987) Rapid
and efficient site-specific mutagenesis without pheno-
typic selection. Methods Enzymol 154, 367–382.
61 Morel N, Bon S, Greenblatt HM, Van Belle D, Wodak
SJ, Sussman JL, Massoulie
´
J & Silman I (1999) Effect
of mutations within the peripheral anionic site on the
stability of acetylcholinesterase. Mol Pharmacol 55,
982–992.
62 Ellman GL, Courtney KD, Andres V & Featherstone
RM (1961) A new and rapid colorimetric determination
of acetylcholinesterase activity. Biochem Pharmacol 7,
88–95.
63 Rost B (1996) PHD: predicting one-dimensional protein
structure by profile-based neural networks. Methods
Enzymol 266, 525–539.
Oligomerization and secretion of AChE and BChE D. Liang et al.
108 FEBS Journal 276 (2009) 94–108 Journal compilation ª 2008 FEBS. No claim to original French government works