Báo cáo Y học: Modulation of inositol 1,4,5-triphosphate concentration by prolyl endopeptidase inhibition ppt

Modulation of inositol 1,4,5-triphosphate concentration
by prolyl endopeptidase inhibition
Ingo Schulz
1
, Bernd Gerhartz
1
, Antje Neubauer
1
, Andreas Holloschi
2
, Ulrich Heiser
1
, Mathias Hafner
2
and Hans-Ulrich Demuth
1
1
Probiodrug AG, Halle, Germany;
2
Department of Molecular Biology and Cell Culture Technology, Mannheim University of
Applied Sciences, Germany
Prolyl endopeptidase (PEP) is a proline-specific oligopepti-
dase with a reported effect on learning and memory in dif-
ferent rat model systems. Using the astroglioma cell line
U343, PEP expression was reduced by an antisense
technique. Measuring different second-messenger concen-
trations revealed an inverse correlation between inositol
1,4,5-triphosphate [Ins(1,4,5)P
3
] concentration and PEP
expression in the generated antisense cell lines. However, no

effect on cAMP generation was observed. In addition,
complete suppression of PEP activity by the specific inhi-
bitor, Fmoc-Ala-Pyrr-CN (5 l
M
) induced in U343 and other
cell lines an enhanced, but delayed, increase in Ins(1,4,5)P
3
concentration. This indicates that the proteolytic activity of
PEP is responsible for the observed effect. Furthermore, the
reduced PEP activity was found to amplify Substance
P-mediated stimulation of Ins(1,4,5)P
3
. The effect of reduced
PEP activity on second-messenger concentration indicates a
novel intracellular function of this peptidase, which may
have an impact on the reported cognitive enhancements due
to PEP inhibition.
Keywords: antisense; inositol 1,4,5-triphosphate; prolyl
endopeptidase; protease; second messenger; Substance P.
Prolyl endopeptidase (PEP; also called prolyl oligopepti-
dase) is a serine peptidase characterized by oligopeptidase
activity. It is grouped in family S9A in clan SC [1]. Enzymes
belonging to clan SC are distinct from trypsin-type and
subtilisin-type serine peptidases in their structure and the
order of the catalytic triad residues in the primary sequence
[2,3]. The recently reported three-dimensional structure of
PEP revealed a two-domain organization [4]. The catalytic
domain displays an a/b hydrolase fold in which the catalytic
triad (Ser554, His680, Asp641) is covered by a so-called
b-propeller domain. The propeller domain probably con-

trols the access of potential substrates to the active site of the
enzyme and excludes peptides containing more than 30
amino acids.
Although the enzymatic and structural properties of
PEP are well known, its biological function is far from
being fully understood [5,6]. Highly conserved in mam-
mals, it is ubiquitously distributed, with high concentra-
tions in the brain [7]. Recently, the enzyme became of
pharmaceutical interest because of a reported cognitive
enhancement induced by treatment with specific PEP
inhibitors [8,9]. In rats displaying scopolamine-induced
amnesia, PEP inhibition caused acetylcholine release in
the frontal cortex and hippocampus [10]. Furthermore,
administration of a PEP inhibitor to rats with middle
cerebral artery occlusion prolonged passive avoidance
latency and reduced the prolonged escape latency in the
Morris water maze task [11]. The potential of PEP
inhibitors as antidementia drugs was further supported by
reports of neuroprotective effects. Inducing neurodege-
neration in cerebellar granule cells led to increased
neuronal survival and enhanced neurite outgrowth in the
presence of a PEP inhibitor [12]. Moreover, the level of
m
3
-muscarinic acetylcholine receptor mRNA was found
to be increased after PEP inhibition, resulting in stimu-
lated phosphoinositide turnover.
It has been hypothesized that these effects are due to
modulation of neuropeptide bioactivity by PEP [13]. In vitro,
PEP is able to rapidly degrade several neuropeptides,

including Substance P and arginine-vasopressin, by limited
proteolysis [14,15]. Such neuropeptides are known to
influence learning and memory [16,17]. Administration of
Substance P can induce long-term potentiation, a well
established parameter for learning and memory [18].
Binding of Substance P to neurokinin 1 receptor stimulates
a G-protein-mediated increase in Ins(1,4,5)P
3
concentration
and release of Ca
2+
from intracellular stores within the
endoplasmic reticulum [19,20]. It is well established, but
unconfirmed for Substance P, that Ca
2+
release from these
stores is implicated in the induction of long-term potenti-
ation and in learning and memory [21]. In postsynaptic cells,
long-term potentiation is prevented by the inhibition of
Ins(1,4,5)P
3
receptors, demonstrating the crucial role of
Ins(1,4,5)P
3
formation and Ca
2+
release in this learning and
memory model [22]. It should be noted, however, that PEP
is primarily located in the cytosol [23], whereas the
interaction between the neuropeptides and their receptors

takes place on the cell surface.
Correspondence to H U. Demuth, Probiodrug AG,
Weinbergweg 22, Biocenter, D-06120 Halle (Saale), Germany.
Fax: + 49 345 5559901, Tel.: + 49 345 5559900,
E-mail:
Abbreviations: Fmoc-Ala-Pyrr-CN, 9H-fluorenyl-9-ylmethyl
N-[2-(2-cyano-1-pyrrolidinyl)-1-methyl-2-oxoethyl]carbamate;
NHMec, 7-(4-methyl)coumarylamide; PEP, prolyl endopeptidase.
Enzyme: prolyl endopeptidase (EC 3.4.21.26).
(Received 16 July 2002, revised 20 September 2002,
accepted 7 October 2002)
Eur. J. Biochem. 269, 5813–5820 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03297.x
Here we show a novel effect of PEP inhibition that may
be related to long-term potentiation and learning and
memory. Using antisense cell lines with reduced PEP
expression as well as specific inhibitors, we were able to
show an inverse correlation between Ins(1,4,5)P
3
concen-
tration and PEP activity. The data presented strongly
suggest an indirect involvement of PEP in second-messenger
pathways with potential cross-talk to signal transduction
mediated by neuropeptides.
EXPERIMENTAL PROCEDURES
Construction of antisense vector
To obtain the coding sequence for the catalytic domain of
PEP, total RNA from 1 · 10
7
cells of the human glioma cell
line U343 was isolated with TRIzolÒ reagent (Gibco BRL).

Then 4 lg total RNA was converted into cDNA by
RT-PCR using hexanucleotide primers and Moloney
murine leukaemia virus (M-MLV) reverse transcriptase
(Promega). The resulting cDNA pool (4 lL) was then
amplified with the Expand
TM
PCR System (Roche) using a
pair of PEP-specific primers (5¢-CATATGCTGTCCTTC
CAGTACC-3¢;5¢-GATTCCGCTGTCAGGAGGAAG
CACG-3¢). The resulting PCR fragment contained the
entire ORF. By PCR, using two nested primers (5¢-CAT
ATGGGAATTGATGCTTCTGATTAC-3¢;5¢-GAATTC
TGGAATCCAGTCGACATTCAG-3¢), a 0.9-kb fragment
was generated containing the catalytic domain of the
enzyme (amino acids 442–731 of human PEP). This
fragment was cloned into pPCR-Script Cam (Stratagene).
The EcoRI restriction sites of the subcloned vector and of
the nested reverse primer were used to ligate the fragment
into the mammalian expression vector pIRESneo (Clon-
tech). The resulting transformants were analysed by PCR to
determine if the insert was present in antisense orientation,
and the correct nucleotide sequence was verified by DNA
sequencing (GATC Biotech AG).
Cell culture, transfection and stable cell lines
The human glioma cell lines U343 and LN405 were
maintained in Dulbecco’s modified Eagle’s medium con-
taining 10% fetal bovine serum (Gibco BRL) at 37 °Cina
5% CO
2
and 10% CO

2
atmosphere, respectively. The
neuroblastoma cell line SH-SY5Y was grown in Dulbecco’s
modified Eagle’s medium containing 5% fetal bovine serum
in a 10% CO
2
atmosphere. All media contained 60 lgÆmL
)1
gentamicin (Gibco BRL). The mammalian expression
vectors were transfected into U343 cells using Polyfectin
reagent (Biontex, Munich, Germany) according to the
manufacturer’s protocol. Stable transfectants were selected
in medium containing 400 lgÆmL
)1
G418 (Duchefa,
Haarlem, the Netherlands). Single clones were isolated with
cloning rings (Clontech).
Prolyl oligopeptidase assays
Cells (1 · 10
7
) were harvested by washing twice in NaCl/P
i
(Gibco BRL) and resuspended in 200 lL assay buffer
(50 m
M
Hepes, pH 7.5, 200 m
M
NaCl, 1 m
M
EDTA, 1 m

M
dithiothreitol). Cell lysis was achieved by three cycles of
thawing and freezing, and then the cells were removed from
the incubation flask with a cell scraper. The lysate obtained
was centrifuged at 18 000 g for 1 min, and the supernatant
transferred to a fresh tube. All steps were performed on
ice. The protein concentration in the supernatant was
determined by the method of Bradford [24]. PEP activity
was measured in the assay buffer using the fluorogenic
substrate Z-Gly-Pro-NHMec (10 l
M
) (Bachem, Heidelberg,
Germany) on a Kontron spectrofluorimeter SFM 25
(excitation wavelength 380 nm, emission wavelength
460 nm) equipped with a four-cell changer and controlled
by an IBM-compatible personal computer. The data
obtained were analysed with the software
FLUCOL
[25].
SDS/PAGE and immunoblotting
To generate a polyclonal antibody against human PEP,
rabbits were immunized with a peptide containing the
N-terminal PEP sequence of amino acids 10–25. Specific
antibodies (S449) were purified from rabbit serum using
an affinity chromatography column with the immobilized
peptide. Analytical electrophoresis in SDS/polyacrylamide
gels was performed as described by Laemmli with
separation gels containing 12% acrylamide [26]. The
separated cell extracts were transferred to a nitrocellulose
membrane (Schleicher & Schuell) following a standard

procedure [27]. PEP and actin were detected by the
polyclonal antibody S449 (1 : 400 dilution) and monoclo-
nal antibody anti-actin (1 : 2500 dilution, Sigma, A2066),
respectively, and visualized by chemiluminescence accord-
ing to the manufacturer’s protocol (SuperSignal
TM
, West
Pico; Pierce). Semiquantitative analysis of Western-blot
results was performed using densitometry software
(
GELSCAN
3D; BioSciTec, Marburg, Germany).
Assay of Ins(1,4,5)
P
3
Cells were grown in 25 cm
2
culture flasks to nearly 100%
confluence. Ins(1,4,5)P
3
concentration was determined by
an isotope dilution method (Amersham Phamacia Biotech)
using 0.5 · 10
6
cells per measurement. To inhibit intracel-
lular PEP, the cells were washed twice with NaCl/P
i
and
incubated for up to 24 h in Optimem 1 medium (Gibco
BRL) supplemented with 5 l

M
PEP inhibitor Fmoc-Ala-
Pyrr-CN. All measurements were carried out in quadrupli-
cate. The calculation of Ins(1,4,5)P
3
concentration and the
statistical analysis (t test) were performed using
PRISM
3.0
(Graph Pad Software).
Stimulation assay
Wild-type and PEP antisense U343 cell lines were cultured
in duplicate in 21-cm
2
culture dishes (Greiner, Frickenhau-
sen, Germany) until confluence. Before stimulation, the cells
were washed twice in NaCl/P
i
andpreincubatedfor10hin
Optimem 1 medium containing 1.6 lgÆmL
)1
leupeptin
(Sigma), 0.86 lgÆmL
)1
chymostatin (Sigma), and
40 lgÆmL
)1
bacitracin (Sigma) at 37 °Cand5%CO
2
.

Substance P (Bachem) was added to a final concentration of
1 l
M
, and the incubation was stopped at the indicated time
by rapidly aspirating the medium and adding 0.4 mL ice-
cold trichloric acid. Preparation of samples and measure-
ment of Ins(1,4,5)P
3
concentration were performed as
described above.
5814 I. Schulz et al.(Eur. J. Biochem. 269) Ó FEBS 2002
cAMP bioactivity assay
U343 cells were transfected with a reporter plasmid, pCRE-
EGFP containing a cassette of a minimal promoter and
three cAMP-responsive elements [28]. Media containing
400 lg G418 were used to select stable transfectants. Cells
were seeded at a density of 1 · 10
4
cells per well in a 96-well
plate (Greiner). After 24 h, the medium was replaced by
dilutions of forskolin or Fmoc-Ala-Pyrr-CN in serum-free
medium. The fluorescence was measured by a Bio Assay
fluorescence microplate reader (Perkin-Elmer, U
¨
berlingen,
Germany) at 485 nm excitation wavelength and 538 nm
emission wavelength. Data were calculated using the Prism
3.0 (Graph Pad).
RESULTS
Suppression of PEP expression in U343 cells

A cell line with sufficiently high concentrations of PEP was
required to investigate the cellular role of PEP. The
astroglioma cell line U343 showed the highest amount of
PEP (active and protein concentration) out of six cell lines
tested (U-138 MG, LN 2308, T 98p31, U343, SY5Y,
LN405).
Two different approaches were used to influence the
intracellular activity of PEP in U343 cells. Fmoc-Ala-Pyrr-
CN is a potent and specific inhibitor of PEP [29] with a K
i
of
70 p
M
against recombinant human PEP (data not shown).
This inhibitor is able to penetrate the cell membrane and
inhibit PEP intracellularly [30]. In U343 cells, total inhibi-
tion was achieved within 1 min by adding 5 l
M
Fmoc-
Ala-Pyrr-CN to the medium, and inhibition persisted for up
to 24 h without the addition of fresh inhibitor. A completely
different approach to reducing PEP activity was also used,
namely generation of antisense cell lines with reduced
expression of the target enzyme. U343 cells were transfected
with the antisense vector, and 120 clones were isolated using
cloning rings. From these clones, eight stable cell lines were
established, and all had reduced PEP activity (Table 1).
However, antisense cell lines 1, 13 and 110 lost their
antisense effect during the prolonged cultivation. Most of
the established cell lines displayed reduced PEP activity of

 50%. Cell line as11 showed the greatest reduction in PEP
activity, 30% compared with wild-type U343 cells. Western-
blot analysis confirmed the results obtained by activity
measurements (Fig. 1, Table 1). In all antisense cell lines,
the reduced proteolytic activity resulted from decreased
expression of PEP. The generated antisense cell lines did not
show a common change in phenotype, but individual
changes were observed. U343 as11 cell line showed
increased trypsin sensitivity, increased cell volume (three-
fold), and was no longer able to grow to 100% confluence.
Modulation of Ins(1,4,5)
P
3
concentration dependent
on PEP
To characterize the intracellular function of PEP,
Ins(1,4,5)P
3
concentration in the antisense cell lines was
measured. In U343 wild-type cells, it was  0.26 ±
0.02 pmol per 10
6
cells (n ¼ 4). It was increased in all
generated antisense cell lines (Fig. 2). The increase in
Ins(1,4,5)P
3
concentration correlated with reduced PEP
activity in the antisense cell lines tested (Fig. 2C, correlation
coefficient 0.997).
An alternative approach to suppressing PEP activity in

U343 cells was utilized. The cells were incubated for 3 h in
the presence of the specific inhibitor Fmoc-Ala-Pyrr-CN
(5 l
M
). In confirmation of the results obtained with the
antisense cell lines, basal Ins(1,4,5)P
3
concentration was
increased in cells treated with PEP inhibitor (Fig. 2).
However, the observed change in Ins(1,4,5)P
3
concentration
was only 0.16 pmol per 10
6
cells. This is much smaller than
thechangeinIns(1,4,5)P
3
concentration (0.66 pmol per 10
6
cells) observed in cell line as11, which still contained 30%
Table 1. Remaining activities and expression patterns of PEP in human glioma U343 antisense cell lines. Specific activity is expressed as mean ± SD.
All antisense cell lines show reduced remaining activity and expression intensity compared with wild-type cell line U343. Remaining acti-
vity ¼ percentage of the activity found in wild-type U343 cells; remaining expression ¼ densitometric analysis of Western blot, n ¼ 2.
Cell line
Specific activity
(mUÆmg
)1
)
Remaining
activity (%)

Remaining
expression (%)
U343-wt 5.00 ± 0.14 100.0 100.0
U343-as2 2.64 ± 0.08 53.0 57.0
U343-as11
a
1.52 ± 0.04 30.0 20.0
U343-as40 2.70 ± 0.06 54.0 65.0
U343-as60 2.12 ± 0.04 42.0 33.0
U343-as70 2.18 ± 0.08 43.0 61.0
U343-as110 4.20 ± 0.10 84.0 71.0
a
Changes in phenotype.
Fig. 1. Western-blot analysis of PEP expression in established antisense
cell lines. The PEP activity remaining in each antisense cell line cor-
responds to the signal intensity in the Western-blot analysis. First,
1 · 10
7
cells from each cell line were extracted and analysed as des-
cribed in Experimental procedures. Then 20 lg total protein was
loaded per lane. Purified recombinant human PEP was used as a
positive control (75 ng). Western blots were probed with PEP-specific
antibody S449 (1 : 400) and anti-actin (1 : 2500) and detected by
chemiluminescence.
Ó FEBS 2002 Ins(1,4,5)P
3
increase by prolyl endopeptidase (Eur. J. Biochem. 269) 5815
PEP activity, and calls into question the correlation between
PEP activity and Ins(1,4,5)P
3

concentration. Therefore,
Ins(1,4,5)P
3
concentration was investigated over an exten-
ded period of total inhibition. As shown in Fig. 3,
Ins(1,4,5)P
3
concentration in U343 cells increased during
incubation. After 12 h, the total amount of Ins(1,4,5)P
3
(1.24 ± 0.24 pmol per 10
6
cells; n ¼ 4) was higher than the
concentration measured in cell line as11 (0.95 ± 0.05 pmol
per 10
6
cells; n ¼ 4), which has 30% remaining enzyme
activity. To confirm the observed effect, two other cell lines,
SY5Y and LN405, were incubated in the presence of Fmoc-
Ala-Pyrr-CN for 24 h (Fig. 3). Both cell lines showed an
increase in Ins(1,4,5)P
3
concentration with a similar
dependence on the incubation time to that in U343 cells.
However, the increase in Ins(1,4,5)P
3
concentration was
smaller. The PEP activity of SY5Y cells (1.39 ±
0.03 mUÆmg
)1

) and LN405 cells (1.65 ± 0.04 mUÆmg
)1
)
is  1.7-fold and 1.4-fold lower, respectively. This confirms
the observed activity dependence of PEP inhibition on
Ins(1,4,5)P
3
concentration.
Influence of PEP inhibition on the cAMP pathway
In addition to Ins(1,4,5)P
3
, the effect of PEP inhibition on
another second messenger, cAMP, was investigated. Using
a reporter plasmid (pCRE-EGFP) containing three cAMP-
responsive elements, the increase in cAMP concentration
was measured from the expression of enhanced green
fluorescent protein (EGFP) via activation of the cAMP-
responsive element. Incubation with Fmoc-Ala-Pyrr-CN
had no positive effect on the cAMP pathway, whereas
control experiments stimulating transfected U343 cells with
forskolin resulted in an increase in EGFP expression (not
shown).
PEP-dependent Ins(1,4,5)
P
3
accumulation after
stimulation with Substance P
To investigate whether the observed effect on Ins(1,4,5)P
3
concentration represents a novel interaction between the

biological activity of neuropeptides and PEP, Substance P
was chosen to stimulate U343 cells. Substance P, a neuro-
peptide known to be degraded by PEP in vitro [15,31], is
reported to influence learning and memory via a receptor-
mediated signalling cascade including the second messenger,
Ins(1,4,5)P
3
[32,33]. Using RT-PCR, the occurrence of
Substance P-specific neurokinin receptor 1 in U343 cells
was confirmed (data not shown). Acknowledging that
Fig. 3. Time course of Ins(1,4,5)P
3
concentration in different cell lines
treated with the PEP inhibitor Fmoc-Ala-Pyrr-CN. Whereas PEP
activity was completely inhibited after 1 min of a single treatment with
5 l
M
Fmoc-Ala-Pyrr-CN, the Ins(1,4,5)P
3
concentration required
12 h to reach maximum concentration. Results are presented as
mean ± SEM from experiments carried out in quadruplicate. (j)
U343; (d) SY5Y; (.)LN405.
Fig. 2. Analysis of Ins (1,4,5)P
3
concentration in various U343 cell lines.
(A) Reduced PEP activity induces increased Ins(1,4,5)P
3
concentration
in stable transfected cell lines. Human glioma cell line U343 was

transfected with a vector (pIRES) containing the coding sequence of
the PEP catalytic domain (amino acids 442–731) in antisense direction.
Thecelllinetransfectedwiththevectornotharbouringaninsert
(pIRES) was used as a negative control. (B) Wild-type U343 cells
treated with the specific PEP inhibitor, Fmoc-Ala-Pyrr-CN (5 l
M
)
show an increased Ins(1,4,5)P
3
concentration; Data were obtained in
quadruplicate (mean ± SD) and analysed using the unpaired t test
(***P <0.001;**P <0.01;*P < 0.05; n.s., not significant). (C) The
increase in Ins(1,4,5)P
3
concentration correlates with the remaining
PEP activity in the established antisense cell lines. Correlation factor
was estimated by linear regression (***P < 0.0005).
5816 I. Schulz et al.(Eur. J. Biochem. 269) Ó FEBS 2002
Substance P is an excellent in vitro substrate for PEP, we
investigated potential degradation of Substance P during
the incubation in the serum-free Optimem 1 medium of
U343 cells by MALDI-TOF MS analysis. However, during
the incubation time of 10 min used, no PEP-specific
degradation was observed (data not shown). This is in
agreement with the fact that no PEP activity is measurable
in the medium (detection limit  0.1 lUÆmg
)1
).
Stimulation of the wild-type U343 cells for 5 s with 1 l
M

Substance P led to a rise in Ins(1,4,5)P
3
concentration
(Fig. 4). Intriguingly, U343 cells treated with Fmoc-Ala-
Pyrr-CN and U343 cell line as2 had a higher concentration
of Ins(1,4,5)P
3
after Substance P stimulation. Comparing
the total values after Substance P stimulation, the
Ins(1,4,5)P
3
concentration again correlated with the
impaired PEP activity (Fig. 4). The change in Ins(1,4,5)P
3
concentration during Substance P stimulation is illustrated
in Fig. 5. To compare the stimulation-dependent increase in
Ins(1,4,5)P
3
concentration, the amount of Ins(1,4,5)P
3
in
the nonstimulated state was subtracted as a baseline. All
three cell lines, U343 wild-type untreated or inhibitor
treated and as2 cells, showed a similar stimulation pattern
over the time measured. Maximum Ins(1,4,5)P
3
concentra-
tion always occurred after 5 s stimulation. The stimulation
produced a rapid increase in the second-messenger concen-
tration followed by a slow decline, not reaching baseline

levels until 40 s. Whereas U343 wild-type and as2 cells
showed no consistent difference in Ins(1,4,5)P
3
concentra-
tion (Fig. 5B), the inhibitor-treated wild-type cells showed
increased stimulation of Ins(1,4,5)P
3
by Substance P over
the whole incubation (Fig. 5A). Estimation of cAMP
stimulation with forskolin did not reveal any difference
between wild-type U343 cells and antisense cell lines or
Fmoc-Ala-Pyrr-CN-treated cells.
DISCUSSION
First described in 1970 as an oxytocin-inactivating enzyme
[34], PEP is well understood with respect to its enzymatic
and structural properties, but its physiological function
remains unclear. However, over the past few years, it has
become of pharmaceutical interest because of reports of
improved learning and memory after application of specific
PEP inhibitors [10,35–37].
Fig. 4. Ins(1,4,5)P
3
concentrations in various U343 cell lines stimulated
by Substance P. Ins(1,4,5)P
3
concentrations were measured in U343
wild-type cells with or without incubation in the presence of 5 l
M
Fmoc-Ala-Pyrr-CN for 12 h and in antisense cell line U343–as2. Each
cell line was stimulated with 1 l

M
Substance P for 5 s after which
Ins(1,4,5)P
3
was extracted and measured. Data (mean ± SD) were
obtained in quadruplicate and statistical analysis was performed using
the paired t test.
Fig. 5. Kinetic profile of Ins(1,4,5)P
3
stimulation by Substance P in
U343 cells. The kinetic profiles of Ins(1,4,5)P
3
stimulation by Sub-
stance P show a significant increase in inhibitor treated U343 cells
(A, s), antisense cell line 2 (B, m) and untreated control cells (A, B, d).
Cells were stimulated with 1 l
M
Substance P and harvested at different
time points to extract Ins(1,4,5)P
3
. U343 wild-type cells were treated
with 5 l
M
Fmoc-Ala-Pyrr-CN for 12 h ahead of the experiment. All
data points, presented as mean ± SD, are from experiments carried
out in quadruplicate.
Ó FEBS 2002 Ins(1,4,5)P
3
increase by prolyl endopeptidase (Eur. J. Biochem. 269) 5817
PEP inhibitors are in general very specific because of

the proline residue in the P
1
position (Berger and
Schlechter nomenclature [38]). However, we used two
different methods of inhibition. Antisense cell lines
expressing reduced PEP enable investigation of the
biological function of nonenzymatic properties of this
two-domain protein. In addition, this technique avoids
possible unspecific effects of the reactive group of the
inhibitor. Eight stable antisense cell lines were developed
with PEP expression reduced by various amounts. In all
cell lines a strong correlation was observed between
reduced PEP expression and remaining enzyme activity
(Table 1). Although differences in cultivation and mor-
phology of these cell lines could be observed, no common
change in the phenotype was present. The observed
changes seem to be related to the method used to generate
antisense cell lines, in which the antisense encoding DNA
has to be inserted into the genome in a random manner.
Phenotypic changes in U343 cells were not seen when cells
were cultivated in the presence of PEP inhibitors.
A relationship between the physiological function of
neuropeptides and PEP has been suggested [13,14]. Inacti-
vation of the biological activity of the neuropeptides via
limited proteolysis by PEP has been hypothesized. However,
this hypothesis does not explain how an intracellular enzyme
such as PEP can interfere with the extracellular interaction
between the neuropeptide and its receptor. During cultiva-
tion of U343 cells, we were unable to detect any extracellular
activity of PEP, all activity being found in the cytoplasmic

fraction. Another possible relationship may be an intracel-
lular involvement of PEP in the receptor-mediated signalling
cascade of neuropeptides. The first hint of this unexpected
function came from a PEP knock-out mutant in the slime
mold Dictyostelium [39]. While trying to generate a Li
+
-
resistant mutant of Dictyostelium, the authors found that the
PEP knock-out mutant prevented typical effects of Li
+
by
increasing the Ins(1,4,5)P
3
concentration. Ins(1,4,5)P
3
, as a
central molecule in the signalling cascade of neuropeptides,
offers an intriguing starting point to investigate such an
unexpected relationship. Neuropeptides such as Substance P
are able to activate phospholipase C via their specific
receptors and do so by inducing the second messengers
Ins(1,4,5)P
3
and 1,2-diacylglycerol [19,20]. It is known that
Ins(1,4,5)P
3
binds to its receptor located in the membrane of
the endoplasmic reticulum and induces release of Ca
2+
,

which is believed to play a crucial role in learning and
memory [22].
Interestingly, in the mammalian cell lines U343, SY5Y,
and LN405, Ins(1,4,5)P
3
concentration increased according
to reduced expression of PEP and was dependent on the
proteolytic activity being suppressed by the inhibitor (Figs 2
and 3). The effects of the antisense approach and inhibitor
treatment on Ins(1,4,5)P
3
stimulation differ with respect to
concentration, probably because of the longer period of
reduced PEP expression in the antisense approach. However,
the increased Ins(1,4,5)P
3
observed in the antisense cell lines
leaves open the question of which domain of PEP is
responsible for this effect. The results obtained with the
specific inhibitor indicate an involvement of the catalytic
domain within the enzyme. The inhibitor used, Fmoc-Ala-
Pyrr-CN, interacts with the enzyme in a substrate-like
manner and restricts changes to the active site of the enzyme
[29,40]. This strongly suggests that the impaired proteolytic
activity of PEP is responsible for the elevated Ins(1,4,5)P
3
concentration. No effect of PEP inhibition on the alternative
signal-transduction pathway of neuropeptides such as
arginine-vasopressin with cAMP as second messenger was
observed.

The astroglioma cell line U343 expresses neurokinin 1
receptor, the specific receptor for the neuropeptide Sub-
stance P, and displays typical Ins(1,4,5)P
3
kinetics after
Substance P stimulation (Fig. 5) [41]. Both U343 antisense
cell lines and cells incubated with the PEP inhibitor showed
an amplified Ins(1,4,5)P
3
signal after Substance P stimula-
tion (Fig. 4), but the kinetic profile of the stimulation was left
unchanged (Fig. 5). This amplification supports the hypo-
thesis that PEP somehow influences the signalling cascade of
neuropeptides such as Substance P. However, the amplifi-
cation of the Ins(1,4,5)P
3
signal appeared to be partially due
to the increased baseline level of the second messenger and
partially due to enhanced efficacy of Substance P. This raises
the question of whether PEP influences the neuropeptide
signalling cascade at or before phospholipase C-catalysed
Ins(1,4,5)P
3
formation or is independent of this pathway.
Such an alternative pathway includes the dephosphorylation
of InsP
5
to Ins(1,4,5)P
3
by multiple inositol polyphosphatase

[42]. This enzyme was reported to have increased activity in
the PEP knock-out mutants of Dictyostelium [39]. Neither
Ins(1,4,5)P
3
, its precursor, nor enzymes such as phosphol-
ipase C or multiple inositol polyphosphatase are substrates
of PEP, therefore, the observed effect must be indirect. The
extremely delayed response of Ins(1,4,5)P
3
concentration to
total inhibition of PEP supports this suggestion (Fig. 3). In
addition, it is intriguing that the enzymatic activity of PEP
can be suppressed by a phosphorylated residue adjacent to
the P
1
proline residue [43].
In conclusion, the results presented strongly indicate a
novel type of interaction between the signal-transduction
cascades of neuropeptides such as Substance P and the
serine peptidase PEP, in addition to the well reported in vitro
direct inactivation. Because of its intracellular localization,
the effect of PEP on the signalling cascade offers a new way
in which PEP inhibitors may enhance learning and memory.
After submitting this manuscript, Williams and co-
workers [44] have published an article where they establish
a link between the mood-stabilizing drugs lithium, car-
bamazepine and valproic acid, and inositol depletion.
Inhibitors of prolyl endopeptidase reverse the effects of all
three drugs on sensory neuron growth cone area and
collapse, suggesting an influence on Ins(1,4,5)P

3
metabolism
by PEP which is demonstrated in the present investigation.
ACKNOWLEDGEMENTS
This study was supported by grants from the BMBF, project no. beo-
312302. The MALDI-TOF MS analysis of Dr Fred Rosche is gratefully
acknowledged. We are indebted to Dr S. Buckley and Dr S. Hinke for
critical reading of the manuscript and the stimulating discussion.
REFERENCES
1. Barrett, A.J., Rawlings, N.D. & Woessner, J.F. (1998) Handbook
of Proteolytic Enzymes. Academic Press, London.
2. Goossens, F.M.I., Vanhoof, G., Hendriks, D., Vriend, G. &
Scharpe, S. (1995) The purification, characterization and analysis
of primary and secondary-structure of prolyl oligopeptidase from
5818 I. Schulz et al.(Eur. J. Biochem. 269) Ó FEBS 2002
human lymphocytes. Evidence that the enzyme belongs to the
alpha/beta hydrolase fold family. Eur. J. Biochem. 233, 432–441.
3. Barrett, A.J. & Rawlings, N.D. (1992) Oligopeptidases, and the
emergence of the prolyl oligopeptidase family. Biol. Chem. Hoppe
Seyler 373, 353–360.
4. Fulop, V., Bocskei, Z. & Polgar, L. (1998) Prolyl oligopeptidase:
an unusual beta-propeller domain regulates proteolysis. Cell 94,
161–170.
5. Wetzel, W., Wagner, T., Vogel, D., Demuth, H.U. & Balschun, D.
(1997) Effects of the CLIP fragment ACTH 20–24 on the duration
of REM sleep episodes. Neuropeptides 31, 41–45.
6. Demuth, H.U., Neumann, U. & Barth, A. (1989) Reactions
between dipeptidyl peptidase IV and diacyl hydroxylamines:
mechanistic investigations. J. Enzyme Inhib. 2, 239–248.
7. Goossens, F.M.I., Vanhoof, G. & Scharpe, S. (1996) Distribution

of prolyl oligopeptidase in human peripheral tissues and body
fluids. Eur. J. Clin. Chem. Clin. Biochem. 34, 17–22.
8. Yoshimoto, T., Kado, K., Matsubara, F., Koriyama, N., Kaneto,
H. & Tsura, D.J. (1987) Specific inhibitors for prolyl
endopeptidase and their anti-amnesic effect. Pharmacobiodyn 10,
730–735.
9. De Nanteuil, G., Portevin, B. & Lepagnol, J. (1998) Prolyl
endopeptidase inhibitors: a new class of memory enhancing drugs.
Drugs Future 23, 167–179.
10. Toide, K., Iwamoto, Y., Fujiwara, T. & Abe, H. (1995)
JTP-4819: a novel prolyl endopeptidase inhibitor with
potential as a cognitive enhancer. J. Pharmacol. Exp. Ther. 274,
1370–1378.
11. Shinoda, M., Matsuo, A. & Toide, K. (1996) Pharmacological
studies of a novel prolyl endopeptidase inhibitor, JTP-4819, in
rats with middle cerebral artery occlusion. Eur. J. Pharmacol. 305,
31–38.
12. Katsube, N., Sunaga, K., Aishita, H., Chuang, D.M. &
Ishitani, R. (1999) ONO-1603, a potential antidementia
drug, delays age-induced apoptosis and suppresses overex-
pression of glyceraldehyde-3-phosphate dehydrogenase in cul-
tured central nervous system neurons. J. Pharmacol. Exp. Ther.
288, 6–13.
13. Shishido, Y., Furushiro, M., Tanabe, S., Shibata, S., Hashimoto,
S. & Yokokura, T. (1999) Effects of prolyl endopeptidase
inhibitors and neuropeptides on delayed neuronal death in rats.
Eur. J. Pharmacol. 372, 135–142.
14. Mentlein, R. (1988) Proline residues in the maturation and deg-
radation of peptide hormones and neuropeptides. FEBS Lett. 234,
251–256.

15. Wilk, S. (1983) Prolyl endopeptidase. Life Sci. 33, 2149–2157.
16. Bennett, G.W., Ballard, T.M., Watson, C.D. & Fone, K.C. (1997)
Effect of neuropeptides on cognitive function. Exp. Gerontol. 32,
451–469.
17. Huston, J.P. & Hasenohrl, R.U. (1995) The role of neuropeptides
in learning: focus on the neurokinin substance P. Behav. Brain Res.
66, 117–127.
18. Liu, X.G. & Sandkuhler, J. (1998) Activation of spinal N-methyl-
D
-aspartate or neurokinin receptors induces long-term
potentiation of spinal C-fibre-evoked potentials. Neuroscience 86,
1209–1216.
19. Abdel-Latif, A.A. (1989) Calcium-mobilizing receptors, poly-
phosphoinositides, generation of second messengers and contrac-
tion in the mammalian iris smooth muscle: historical perspectives
and current status. Life Sci. 45, 757–786.
20. Defea, K., Schmidlin, F., Dery, O., Grady, E.F., & Bunnett, N.W.
(2000) Mechanisms of initiation and termination of signalling by
neuropeptide receptors: a comparison with the proteinase-acti-
vated receptors. Biochem. Soc. Trans. 28, 419–426.
21. Voronin, L., Byzov, A., Kleschevnikov, A., Kozhemyakin, M.,
Kuhnt, U. & Volgushev, M. (1995) Neurophysiological analysis of
long-term potentiation in mammalian brain. Behav. Brain Res. 66,
45–52.
22. Komatsu, Y. (1996) GABAB receptors, monoamine receptors,
and postsynaptic inositol trisphosphate-induced Ca
2+
release are
involved in the induction of long-term potentiation at visual cor-
tical inhibitory synapses. J. Neurosci. 16, 6342–6352.

23. Kimura, A., Yoshida, I., Takagi, N. & Takahashi, T. (1999)
Structure and localization of the mouse prolyl oligopeptidase gene.
J. Biol. Chem. 274, 24047–24053.
24. Bradford, M.M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Anal. Biochem. 72, 248–254.
25. Machleidt, W., Nagler, D.K., Assfalg-Machleidt, I., Stubbs, M.T.,
Fritz, H. & Auerswald, E.A. (1995) Temporary inhibition of
papain by hairpin loop mutants of chicken cystatin. Distorted
binding of the loops results in cleavage of the Gly(9)-Ala10 bond.
FEBS Lett. 361, 185–190.
26. Laemmli U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature (London) 227,
680–685.
27. Bjerrum, O.J. & Schafner-Nielson, C. (1986) Buffer systems and
transfer parameters for semidry electro-blotting with a horizontal
apparatus. In: Electrophoresis (Unn, M.J., ed.), pp. 315–319.
VCH, Weinheim.
28. Holloschi, A. & Hafner, M. (2002) A new green fluorescent
protein reporter cell line to measure the bioactivity of calcitonin.
Eur. J. Cell Biol. 79,(Suppl. 50), 95.
29. Li, J., Wilk, E. & Wilk, S. (1996) Inhibition of prolyl oligopepti-
dase by Fmoc-aminoacylpyrrolidine-2-nitriles. J. Neurochem. 66,
2105–2112.
30. Johnston, J.A., Jensen, M., Lannfelt, L., Walker, B. & Wil-
liams, C.H. (1999) Inhibition of prolylendopeptidase does
not affect gamma-secretase processing of amyloid precursor
protein in a human neuroblastoma cell line. Neurosci. Lett. 277,
33–36.
31. Welches, W.R., Brosnihan, K.B. & Ferrario, C.M. (1993) A

comparison of the properties and enzymatic activities of three
angiotensin processing enzymes: angiotensin converting enzyme,
prolyl endopeptidase and neutral endopeptidase 24.11. Life Sci.
52, 1461–1480.
32. Hasenohrl, R.U., Huston, J.P. & Schuurman, T. (1990) Neuro-
peptide substance P improves water maze performance in aged
rats. Psychopharmacology (Berl.) 101, 23–26.
33. Huston, J.P. & Hasenohrl, R.U. (1995) The role of neuropeptides
in learning: focus on the neurokinin substance P. Behav. Brain Res.
66, 117–127.
34. Walter, R., Shlank, H., Glass, J.D., Schwartz, I.L. & Kerenyi,
T.D. (1971) Leucylglycinamide released from oxytocin by human
uterine enzyme. Science 173, 827–829.
35. Shinoda, M., Miyazaki, A. & Toide, K. (1999) Effect of a novel
prolyl endopeptidase inhibitor, JTP-4819, on spatial memory and
on cholinergic and peptidergic neurons in rats with ibotenate-
induced lesions of the nucleus basalis magnocellularis. Behav.
Brain Res. 99, 17–25.
36. Shishido, Y., Furushiro, M., Tanabe, S., Taniguchi, A., Hashi-
moto, S., Yokokura, T., Shibata, S., Yamamoto, T. & Watanabe,
S. (1998) Effect of ZTTA, a prolyl endopeptidase inhibitor, on
memory impairment in a passive avoidance test of rats with basal
forebrain lesions. Pharmacol. Res. 15, 1907–1910.
37. Toide, K., Shinoda, M., Fujiwara, T. & Iwamoto, Y. (1997) Effect
of a novel prolyl endopeptidase inhibitor, JTP-4819, on spatial
memory and central cholinergic neurons in aged rats. Pharmacol.
Biochem. Behav. 56, 427–434.
38. Berger, A. & Schlechter, I. (1976) On the size of the active site in
proteases. I. Papain. Biochem. Biophys. Res. Commun. 27, 157–
162.

Ó FEBS 2002 Ins(1,4,5)P
3
increase by prolyl endopeptidase (Eur. J. Biochem. 269) 5819
39. Williams, R.S., Eames, M., Ryves, W.J., Viggars, J. & Harwood,
A.J. (1999) Loss of a prolyl oligopeptidase confers resistance to
lithium by elevation of inositol (1,4,5) trisphosphate. EMBO J. 18,
2734–2745.
40. Demuth, H.U., Schlenzig, D., Schierhorn, A., Grosche, G.,
Chapot-Chartier, M.P. & Gripon, J.C. (1993) Design of (omega-
N-(O-acyl) hydroxy amid) aminodicarboxylic acid pyrrolidides as
potent inhibitors of proline-specific peptidases. FEBS Lett. 320,
23–27.
41. Araki-Sasaki, K., Aizawa, S., Hiramoto, M., Nakamura, M.,
Iwase, O., Nakata, K., Sasaki, Y., Mano, T., Handa, H. & Tano,
Y. (2000) Substance P-induced cadherin expression and its signal
transduction in a cloned human corneal epithelial cell line. J. Cell
Physiol. 182, 189–195.
42. Craxton, A.,Caffrey,J.J., Burkhart,W.,Safrany,S.T. &Shears, S.B.
(1997) Molecular cloning and expression of a rat hepatic multiple
inositol polyphosphate phosphatase. Biochem. J. 328, 75–81.
43. Kaspari, A., Diefenthal, T., Grosche, G., Schierhorn, A. &
Demuth, H.U. (1996) Substrates containing phosphorylated
residues adjacent to proline decrease the cleavage by proline-spe-
cific peptidases. Biochim. Biophys. Acta 1293, 147–153.
44. Willams, R.S., Cheng, L., Mudge, A.W. & Harwood, A.J. (2002)
A common mechanism of action for three mood-stabilizing drugs.
Nature 417, 292–295.
5820 I. Schulz et al.(Eur. J. Biochem. 269) Ó FEBS 2002