The PREPL A protein, a new member of the prolyl oligopeptidase family, lacking catalytic activity

Article abstract of DOI:10.1007/s00018-005-5262-5

The PREPL (previously called KIAA0436) gene encodes a putative serine peptidase from the prolyl oligopeptidase family. A chromosomal deletion involving the PREPL gene leads to a severe syndrome with multiple symptoms. Homology with oligopeptidase B suggested that the enzyme cleaves after an arginine or lysine residue. Several PREPL splice variants have been identified, and a 638-residue variant (PREPL A) was expressed in Escherichia coli and purified. Its secondary structure was similar to that of oligopeptidase B, but differential-scanning calorimetry indicated a higher conformational stability. Dimerization may account for the enhanced stability. Unexpectedly, the PREPL A protein did not cleave peptide substrates containing a P1 basic residue, but did slowly hydrolyse an activated ester substrate, and reacted with diisopropyl fluorophosphate. These results indicated that the catalytic serine is a reactive residue. However, the negligible hydrolytic activity suggests that the function of PREPL A is different from that of the other members of the prolyl oligopeptidase family. Birkhaeuser Verlag, 2005.

Full text of DOI:10.1007/s00018-005-5262-5

CMLS, Cell. Mol. Life Sci. 62 (2005) 2376–2381
1420-682X/05/202376-6
DOI 10.1007/s00018-005-5262-5
© Birkhäuser Verlag, Basel, 2005
CMLS Cellular and Molecular Life Sciences
Research Article
The PREPL A protein, a new member of the prolyl
oligopeptidase family, lacking catalytic activity
Z. Szeltnera, I. Alshafeeb, T. Juhásza, R. Parvarib and L. Polgára*
a Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, 1518 Budapest 112, P.O.
Box 7 (Hungary), Fax: +36 1 466 5465, e-mail: polgar@enzim.hu.
b Department of Developmental Molecular Genetics, Faculty of Health Sciences, Ben Gurion University
of the Negev, Beer Sheva 84105 (Israel)
Received 16 June 2005; received after revision 19 July 2005; accepted 9 August 2005
Online First 7 September 2005
Abstract. The PREPL (previously called KIAA0436) scanning calorimetry indicated a higher conformational
gene encodes a putative serine peptidase from the prolyl stability. Dimerization may account for the enhanced sta-
oligopeptidase family. A chromosomal deletion involving bility. Unexpectedly, the PREPL A protein did not cleave
the PREPL gene leads to a severe syndrome with multiple peptide substrates containing a P1 basic residue, but did
symptoms. Homology with oligopeptidase B suggested slowly hydrolyse an activated ester substrate, and reacted
that the enzyme cleaves after an arginine or lysine resi- with diisopropyl fluorophosphate. These results indicated
due. Several PREPL splice variants have been identified, that the catalytic serine is a reactive residue. However, the
and a 638-residue variant (PREPL A) was expressed in negligible hydrolytic activity suggests that the function of
Escherichia coli and purified. Its secondary structure PREPL A is different from that of the other members of
was similar to that of oligopeptidase B, but differential- the prolyl oligopeptidase family.
Key words. KIAA0436; oligopeptidase B; protein expression and isolation; hydrolytic activity; dimerization; dena-
turation.
A unique syndrome has been reported characterized by predicted to produce four different proteins (PREPLs
cystinuria, neonatal seizures, hypotonia, somatic and de- A, C, D and E) [2]. (PREPL A is named PREPL A_B in
velopmental delay, facial dysmorphism and lactic acidae- Parvari et al. [2]). Here we report on one protein product
mia. The molecular basis of this disorder was a deletion of the splice variants, named PREPL A, represented by a
of 179,311 bp on chromosome 2p21 (previously 2p16), full-length cDNA prepared by the oligo-capping method.
which included the type I cystinuria gene SLC3A1, The gene was cloned by the NEDO human cDNA se-
the protein phosphatase 2C gene, an unidentified gene quencing project (Acc. No. at Entrez NCBI: AK025649)
KIAA0436 (presently approved gene symbol PREPL, and supported by six ESTs (BG704689, BG717492,
prolyl endopeptidase-like) and several additional tran- AL602657, AL603269, AL710407 and BQ231866).
scripts represented by ESTs [1]. A comprehensive verifi- The initiator methionine of this isoform is preceded by a
cation of all ESTs and transcripts in the genomic interval stop codon, indicating that this protein is complete. The
of the PREPL gene identified several splice variants initiator methionine is in a strong Kozak consensus se-
quence (AACATGG, the ATG start codon in the context
RNNatgG, where R is a purine), obeying the rule that
* Corresponding author.
the translational start site in vertebrate mRNA nearly

CMLS, Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2377
always conforms to the consensus sequence [3]. This ione-S-transferase (GST)-tag that could be cleaved with
splice variant was also identified in oran-gutan, rat and thrombin. The gene was expressed in the Escherichia coli
mouse (97%, 92% and 91% identity, respectively; fig. 7B BL21(DE3)pLysS strain grown in six 500-ml LB cultures
in Parvari et al. [2]).
containing 100 µg/ml ampicillin and 25 µg/ml chloram-
PREPL A displays sequence homology with the serine phenicol at 37°C. When the OD600 reached a value of
peptidases of the prolyl oligopeptidase family (S9 of clan about 0.2, an additional 100 µg/ml ampicillin was added.
SC). Members of the family include dipeptidyl pepti- At an OD600 of about 0.6, a new portion of ampicillin and
dase IV, oligopeptidase B, acylaminoacyl peptidase and 0.1 mM isopropyl
b-D-thiogalactopyranoside (IPTG)
the prototype prolyl oligopeptidase, all of which are of were added. The cultures were incubated overnight at
physiological and pharmacological importance [4]. The 26°C. The cells were collected by centrifugation and son-
strongest homology was found with oligopeptidase B icated in 100 ml of ice-cold 20 mM phosphate buffer, pH
(EC 3.4.21.83). PREPL A is expressed in many tissues, 7.3, containing 0.1 M NaCl. The sonicate was centrifuged
including brain, testis, kidney, liver, colon, muscle and at 20,000 g, and Triton X-100 (1% final concentration)
lung (fig. 8 in Parvari et al. [2] and the GNF Expression was added to the supernatant to increase the solubility
Atlas track at the UCSC genome browser in Human: of the fusion construct. Purification of the GST-protein
http://genome.ucsc.edu/cgi-bin/hgTracks?position=chr2: was carried out according to the batch/column protocol
44457553-44500274&hgsid=42213428&knownGene=p of Pharmacia, so that 4 ml of glutathione Sepharose-4B
ack&hgFind.matches=AB007896).
affinity adsorbent was added to the clear sonicate, mixed
The three-dimensional structure determination of prolyl gently for 1 h and then poured into a column. After a
oligopeptidase provided the first structural information thorough washing with the phosphate buffer used for the
about these enzymes. It was shown that the carboxyl- sonication, the elution was repeated four to five times
terminal peptidase domain of prolyl oligopeptidase with 10 mM reduced glutathione in 50 mM Tris.HCl, pH
displayed an a/b hydrolase fold, and its catalytic triad 8.0. The eluted protein solutions were combined, con-
(Ser554, His680, Asp641) was covered by the central tun- centrated on an Amicon PM30 membrane up to 2 mg/ml
nel of an unusual b propeller [5]. This domain made the protein concentration, and the glutathione was washed
enzyme an oligopeptidase by excluding large structured out with 50 mM phosphate buffer, pH 7.6, containing 1
peptides from the active site. The peptidase domain is mM ethylenediaminetetraacetic acid (EDTA). The GST-
built up of residues 1–72 and 428–710, and the residues PREPL A protein concentration was determined at 280
between these two portions constitute the propeller do- nm, using Mr = 99640 and A280(0.1%) = 1.24. The yield of
main. The propeller domain is based on a sevenfold repeat the GST-tagged protein was 40–50 mg. To eliminate the
of four-stranded antiparallel b sheets, which are twisted GST-tag, 2-mercaptoethanol was added to the solution to
and radially arranged around their central tunnel.
a final concentration of 2 mM, and the fusion protein was
Oligopeptidase B is a cytosolic enzyme found in Gram- digested with thrombin. After complete digestion control-
negative bacteria and trypanosomes [6–8]. Enzyme activ- led by SDS polyacrylamide gel electrophoresis, the un-
ity has been linked to the virulence of bacterial and eu- tagged protein was separated from thrombin and GST by
karyotic pathogens [9], making it a potential therapeutic FPLC chromatography on a MonoQ column (Pharmacia).
target. Oligopeptidase B hydrolyses peptides exclusively The chromatography was performed in 20 mM phosphate
at the carboxyl side of arginine or lysine residues, but buffer, pH 7.2, containing 1 mM 2-mercaptoethanol. The
cleavage is faster after two adjacent basic residues. Here GST-tag eluted in the flow-through fraction. The PREPL
we concentrate on PREPLA and demonstrate that though A protein was eluted with a linear gradient of NaCl (0–1.0
the catalytic triad and the S1 subsites of PREPL A and M) and concentrated using an Amicon PM30 membrane.
oligopeptidase B are similar, PREPL A dispenses with The pure PREPL A tends to precipitate above 2 mg/ml
peptidase activity using a broad range of substrates con- concentration. The concentration of the monomer was
taining P1 basic residues.
calculated from the absorbance at 280 nm using an Mr
value of 73,352 and A280(0.1%) of 1.13.
Materials and methods
Preparation of oligopeptidases. The overexpression
and purification of oligopeptidase B from E. coli [10]
Expression and purification of PREPLA. The PREPLA and porcine prolyl oligopeptidase [11] were described
cDNA was cloned and provided by the Kazusa DNA Re- earlier.
search Institute (accession at Entrez NCBI: AB007896).
The clone was inserted from the first 5’ methionine to Activity measurements. Samples were prepared in 50
the last amino acid (phenylalanine, position 638) into the mM phosphate buffer, pH 7.0, containing 0.7 µM protein
BamHI (5’) and EcoRI (3’) sites of the pGEX2T vector and 20 µM 4-methylumbelliferyl-p-guanidinobenzoate
(Pharmacia), which produced the protein with a glutath- (MUGB), and the initial rates of the reactions were re-

2378
Z. Szeltner et al.
New oligopeptidase B-like protein
corded with a Cary Eclipse fluorescence spectrometer us-
ing 323 and 446 nm excitation and emission wavelengths
(5-nm bandwidth each), respectively. Due to the slow re-
actions with PREPL A and prolyl oligopeptidase (about
2 h), the fluorescence data were registered every 5 min
to omit the photodecomposition of the fluorophore. The
reaction of oligopeptidase B was continuously measured
for a few minutes. The spontaneous hydrolysis of MUGB
was taken into account.
The reaction of PREPL A with diisopropyl-fluorophos-
phate (DIFP) was performed in 15 µl volume, using 40
µM protein and 940 µM DIFP and incubated for 0.5 and
2 h in 200 µl capped PCR tubes at 25 °C. The remaining
activity was determined for oligopeptidase B and PREPL
A with the MUGB substrate and for prolyl oligopepti-
dase with benzyloxycarbonyl (Z)-Gly-Pro-Nap (2-naph-
thylamide) at 340 and 410 nm excitation and emission
wavelengths, respectively.
Gel chromatography. A 100-µl of sample (10 µM) was
loaded on a Superose 12 column, which was equilibrated
with 50 mM phosphate buffer, pH 7.5, containing 150
mM NaCl, 1 mM EDTA, and 1 mM 1,4-dithioerythritol
(DTE).
Differential scanning calorimetry. Calorimetric meas-
urements were carried out with a Microcal VP-DSC dif-
ferential scanning calorimeter. Denaturation curves were
recorded between 25 and 80ºC, at a pressure of 2.5 atm
using a scan rate of 60ºC/h. The protein concentration
was 1.5–2.5 µM (0.1–0.2 mg/ml) in 50 mM phosphate
buffer, pH 8.0, containing 0.5 mM EDTA.
Figure 1. Amino acid sequence alignment for PREPL A, and oli-
gopeptidase B from T. brucei and E. coli.
Circular dichroism. Measurements were carried out Results and discussion
in 20 mM phosphate buffer, pH 8.0, with a Jasco J-720
CD spectropolarimeter at 25°C, using thermostated Structural relationship between PREPL A and oli-
cuvettes of 0.1-cm path lengths. Protein concentration gopeptidase B. PREPL A exhibits higher homology to
was typically 0.1 mg/ml. Spectra were acquired be- oligopeptidase B (BAA01750, 25.5% identity with the
tween 200 and 250 nm at a scan speed of 20 nm/min, E. coli enzyme) than to porcine prolyl oligopeptidase
averaging three scans. For all CD spectra, the baseline (A37942, 19.1% identity) for which the three-dimen-
of the buffer was subtracted. Assignments to the second- sional structure is known. Figure 1 shows the alignment
ary structural elements for prolyl oligopeptidase were (ClustalW) of PREPL A (AB007896) with two ortho-
taken from PDB code 1h2w. The CD spectra analyses logues of oligopeptidase B, one from Trypanosoma bru-
were performed with the CDPro software package http: cei (AAC80459) and the other from E. coli (BAA01750).
//lamar.colostate.edu/~sreeram/CDPro).
The N terminus of prolyl oligopeptidase, as well as that
of oligopeptidase B as indicated by model building [12],
1-Chloro-2,4-dinitrobenzene assay. The GST-tagged contains a short two-stranded antiparallel b sheet and
protein was identified during purification by measuring two long a helices, which are connected to the C-termi-
the reaction of 1-chloro-2,4-dinitrobenzene (CDNB) nal catalytic domain [5]. These structural elements are
with reduced glutathione, which was proportional to the practically absent from the PREPL A variant.
GST concentration. The reaction was followed by the The catalytic triads of oligopeptidase B from T. brucei
increase in absorbance at 340 nm in 0.1 M potassium (Ser563, Asp 648, His683) and E. coli (Ser532, Asp617,
phosphate buffer, pH 6.5, containing 1 mM CDNB and 1 His652) are aligned with the corresponding amino acids
mM glutathione. Slow enzymatic rates were corrected for (Ser470, Asp556, His601 and Ser559, Asp645, His690) of
the spontaneous reaction.
the PREPLA (fig. 1). Tyr482 and Tyr451 of oligopeptidase

CMLS, Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2379
Table 1. Secondary-structure fractions for oligopeptidases.
Protein Method
Helix
0.224
b strand Turn
Unordered
0.261
POP
DSSP
(PDB)
0.292
0.224
CDSSTR 0.153
CONTILL 0.165
SELCON3 0.174
0.298
0.299
0.322
0.220
0.219
0.219
0.332
0.317
0.298
OPB
CDSSTR 0.171
CONTILL 0.161
SELCON3 0.239
0.280
0.256
0.265
0.209
0.218
0.228
0.343
0.365
0.286
PREPL ACDSSTR 0.190
CONTILL 0.206
0.279
0.293
0.295
0.216
0.216
0.220
0.313
0.285
0.291
SELCON3 0.203
Figure 2. Far-UV CD spectra of PREPL A (dash-dotted line), oli-
gopeptidase B from E. coli (thin line) and porcine prolyl oligopepti-
dase (thick line).
POP, prolyl oligopeptidase; OPB, oligopeptidase B. The data ob-
tained with the DSSP method refer to the crystallographic data of
prolyl oligopeptidase. The secondary structures are classified ac-
cording to Sreeramaa et al. [15]. The other three methods show the
far-UV CD spectra analyses with the CDPro software package. The
helix and the strand columns both include the sum of the regular and
distorted fractions.
B from T. brucei and E. coli, respectively (Tyr473 of pro-
lyl oligopeptidase), are part of the oxyanion-binding site
that stabilizes the transition state of the reaction. Tyr390
in PREPL A corresponds to this catalytically important
residue. However, there appear to be differences in the residues. We examined several substrates containing Arg
substrate specificity of the two enzymes. Oligopeptidases and Lys residues, including simple substrates like benzoyl
B of T. brucei and E. coli have a pair of acidic residues (Bz)-Arg-Nan (4-nitroanilide), Bz-Arg-Nap, Z-Arg-Arg-
at the S1 recognition site (Glu607, Glu609 and Glu576, Amc (7-amido-4-methylcoumarin) and intrinsically fluo-
Glu578, respectively), which is specific for basic amino rescent substrates that have 2-aminobenzoyl (Abz) and
acid residues (fig. 1) [13, 14]. In PREPL A, there is a pair 4-nitrophenylalanyl [Phe(NO2)] as fluorescent donor and
of glutamic acid residues (Glu514 and Glu516) at the acceptor/quencher, respectively. These substrates were
equivalent position, suggesting the enzyme is also specific Abz-Arg-Arg-Phe(NO2)-Ala-NH2,
for basic P1 amino acids. Oligopeptidase B hydrolyses Arg-Phe(NO2)-Ala-NH2, Abz-Thr-Arg-Arg-Phe(NO2)-
peptides exclusively at the carboxyl side of arginine or Ser-Leu-NH2, Abz-Thr-citrulline-Arg-Phe(NO2)-Ser-
lysine residues, but the cleavage is considerably faster Leu-NH2, Abz-Thr-Lys-Lys-Phe(NO2)-Ser-Leu-NH2 and
after two adjacent basic residues. Oligopeptidase B has an Abz-Gly-Phe-Arg-Pro-Phe(NO2)-Arg-Ala. However,
Abz-Ala-Ala-Arg-
additional pair of acidic residues (Glu490 and Glu492 in T. none of these peptides was hydrolysed during several
brucei, and Asp459 and Asp461 in E. coli) forming the S2 hours of incubation. Therefore, we examined MUGB, an
subsite for recognition of a P2 basic residue [13, 14]. The activated ester substrate, used for the-active site titration
equivalent acidic S2 residues are not present in PREPL A of trypsin, also specific for P1 Arg and Lys residues. A
(fig. 1), suggesting a substrate specificity that is similar to very low reaction rate was found with PREPL A, 540
but distinct from oligopeptidase B, such that it constitutes times lower than that observed with oligopeptidase B. An
a new member of the prolyl oligopeptidase family.
even lower rate was obtained with prolyl oligopeptidase,
The far-UV CD spectrum of proteins is characteristic of not specific for basic residues (table 2). We also exam-
their secondary structure. Figure 2 shows that the CD ined the reaction of Bz-Arg-ethyl ester by PREPL A, but
spectra of PREPL A, oligopeptidase B from E. coli and this simple alkyl ester was not hydrolysed either. There
porcine prolyl oligopeptidase are similar. The structural was also no cleavage in the substrates specific for prolyl
elements calculated from the spectra are compiled in ta- oligopeptidase and chymotrypsin. These results indicated
ble 1. The data confirm the conformational relationship that PREPLA had an active serine residue, but its catalyt-
among the three proteins.
ic machinery was damaged or its specificity was outside
the range of the substrates examined.
The catalytic activity of PREPLA. The similarity of the Interestingly, the fast reaction of oligopeptidase B with
S1 subsite of PREPL A to that of oligopeptidase B sug- MUGB was not stoichiometric, as in the case of trypsin,
gested that the enzyme exhibits specificity for P1 basic but the substrate was completely hydrolysed, providing

2380
Z. Szeltner et al.
New oligopeptidase B-like protein
Table 2. Activity and denaturation of oligopeptidases.
Protein
PREPL A
OPB
MUGB*
1
0.5 hour+
2 hours+
k (min–1)‡
0.43±0.04
0.42±0.03
45
10
0
35
0
540
POP
0.33
0
6.2±0.9,
0.25±0.03
* Relative activities from initial rate measurements.
+ Remaining activities (%) after incubation with DIFP for the given
time period.
Figure 3. DSC scans of PREPL A, E. coli oligopeptidase B (OPB)
and porcine prolyl oligopeptidase (POP).
‡ Denaturation rate constants in 4 M urea. The denaturation rate of
prolyl oligopeptidase conforms to a double exponential curve.
Km = 3.3 µM and kcat/Km = 37 mM–1 s–1, calculated from of PREPL A in the presence of 4 M urea followed first-
the Michaelis parameters. The rate constant measured order kinetics and produced a similar rate constant to
under first-order conditions (MUGB = 0.2 µM) and that observed with oligopeptidase B. However, prolyl
divided by the enzyme concentration gave a kcat/Km = oligopeptidase exhibited a double exponential curve,
46 mM–1 s–1. The reaction of PREPL A was not stoichio- having a fast and a slow phase (table 2). Such a differ-
metric either. Using the same enzyme concentration (0.7 ence was also found in the DSC measurements, where
µM), the initial rates were identical at 10, 20 and 40 µM prolyl oligopeptidase displayed two Tm values. (iv) We
MUGB concentrations, indicating that the PREPL A anticipated that PREPL A would be less stable than
was saturated at the substrate concentrations employed. prolyl oligopeptidase or oligopepidase B because of the
On the other hand, the increase in initial rates was pro- missing N terminus. However, this was not verified by
portional with the increase in the enzyme concentration the denaturation experiments. Apparently, two PREPL A
(0.15, 0.30, 0.69 and 1.25 µM), as expected.
molecules might associate at their open surface, provid-
DIFP is known to inhibit serine peptidases. Table 2 ing a more stable dimer structure. Indeed, size exclu-
shows that PREPL A reacts with DIFP but at a much sion chromatography on a Superose 12 column clearly
lower rate than oligopeptidase B and, in particular, prolyl demonstrated that PREPL A is a dimer with a molecular
oligopeptidase, which after 2-min incubation retains only mass of 145.5 kDa, as obtained from the calibration
2% activity. This again shows an enhanced nucleophilic curve (SIGMA Molecular Weight Market Kit, 12.4–200
reactivity of the catalytic serine of PREPL A and con- kDa). The molecular mass calculated for the monomer
firms that the hydrolysis of the activated ester substrate is is 73,352 Da, which was absolutely absent from the size
an enzymatic reaction.
exclusion chromatogram.
Conformational stability of PREPL A. The stability of Conclusion. While PREPL A belongs to the prolyl
PREPL A protein was determined with several independ- oligopeptidase family and obviously expresses in differ-
ent denaturation methods, using high temperature, acidic ent tissues, it does not exhibit an appreciable hydrolytic
and alkaline pH, and denaturation with urea. (i) The activity versus substrates containing a P1 basic residue.
pH-induced unfolding was determined fluorimetrically Amino acid sequence homology clearly indicates that
in a wide pH range. The internal fluorescence of PREPL PREPL A is more related to oligopeptidase B than to
A decreased at moderate acidic and alkaline pH values, prolyl oligopeptidase. Therefore, OPBL (oligopepti-
giving rise to a profile similar to that of oligopeptidase B dase B-like) would be a more appropriate acronym for
[10] and prolyl oligopeptidase [11]. Specifically, the con- the gene than the recently proposed PREPL (prolyl
formation of PREPL A did not change in the pH range oligopeptidase-like). Although the specificities of the
5.5–9.5. The fluorescence intensity only diminished S1 binding subsites for P1 basic residues are identical in
outside this range. (ii) The thermal stability of PREPL PREPL A and oligopeptidase B, the S2 sites are different,
A was determined with differential scanning calorimetry indicating that the two enzymes have distinct specifici-
(DSC). The protein exhibited a higher Tm value (51.8°C) ties. PREPL A also differs from oligopeptidase B, as it is
than oligopeptidase B (47.7 °C), and both proteins gave a dimer rather than a monomer. The insignificant activity
a single peak, whereas prolyl oligopeptidase displayed of the dimer suggests that its function might be different
a more complex temperature scanning profile with two from the known enzymes of the prolyl oligopeptidase
Tm values (44.6 and 52.8°C) (fig. 3). (iii) Denaturation family and its biological role remains to be established.

CMLS, Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2381
Acknowledgements. Thanks are due to Ms. I. Szamosi and Ms. V.
Renner for excellent technical assistance. This work was supported
by the Wellcome Trust (grant No. 066099/01/Z) and the Hungarian
National Science Fund (grant No. T 046057), as well as a grant (No.
708/01) from the Israel Science Foundation to R. P.
8 Venalainen J. I., Juvonen R. O. and Mannisto P. T. (2004)
Evolutionary relationships of the prolyl oligopeptidase family
enzymes. Eur. J. Biochem. 271: 2705–2715
9 Caler E. V., Vaena de Avalos S., Haynes P. A., Andrews N. W.
and Burleigh B. A. (1998) Oligopeptidase B-dependent signal-
ing mediates host cell invasion by Trypanosoma cruzi. EMBO
J. 17: 4975–4986
10 Polgár L. and Felföldi F. (1998) Oligopeptidase B: cloning and
probing stability under nonequilibrium conditions. Proteins 30:
424–434
11 Szeltner Z., Renner V. and Polgár L. (2000) Substrate- and pH-
dependent contribution of oxyanion binding site to the catalysis
of prolyl oligopeptidase, a paradigm of the serine oligopepti-
dase family. Protein Sci. 9: 353–360
12 Gérczei T., Keserü G. M. and Náray-Szabó G. (2000) Construc-
tion of a 3D model of oligopeptidase B, a potential processing
enzyme in prokaryotes. J. Mol. Graph. Model. 18: 7–17
13 Juhász T., Szeltner Z., Renner V. and Polgár L. (2002) Role of
the oxyanion binding site and subsites S1 and S2 in the catalysis
of oligopeptidase B, a novel target for antimicrobial chemother-
apy. Biochemistry 41: 4096–4106
14 Morty R. E., Fülöp V. and Andrews N. W. (2002) Substrate
recognition properties of oligopeptidase B from Salmonella
enterica serovar Typhimurium. J. Bacteriol. 184: 3329–3337
15 Sreerama N., Venyaminov S. Y. and Woody R. W. (1999) Esti-
mation of the number of alpha-helical and beta-strand segments
in proteins using circular dichroism spectroscopy. Protein Sci.
1 Parvari R., Brodyansky I., Elpeleg O., Moses S., Landau D. and
Hershkovitz E. (2001) A recessive contiguous gene deletion of
chromosome 2p16 associated with cystinuria and a mitochon-
drial disease. Am. J. Hum. Genet. 69: 869–875
2 Parvari R., GonenY., Alshafee I., Buriakovsky S., Regev K. and
Hershkovitz E. (2005) The 2p21 deletion syndrome: characteri-
zation of the transcription content. Genomics 86: 195–211
3 Kozak M. (1996) Interpreting cDNA sequences: some insights
from studies on translation. Mamm. Genome 7: 563–574
4 Polgár L. (2002) The prolyl oligopeptidase family. Cell. Mol.
Life Sci. 59: 349–362
5 Fülöp V., Böcskei Z. and Polgár L. (1998) Prolyl oligopepti-
dase: an unusual beta-propeller domain regulates proteolysis.
Cell 94: 161–170
6 Kanatani A., Masuda T., Shimoda T., Misoka F., Lin X. S.,
Yoshimoto T. et al. (1991) Protease II from Escherichia coli:
sequencing and expression of the enzyme gene and charac-
terization of the expressed enzyme. J. Biochem. (Tokyo) 110:
315–320
7 Tsuru D. andYoshimoto T. (1994) Oligopeptidase B: protease II
from Escherichia coli. Methods Enzymol. 244: 201–215
8: 370–380