Page 1
Isolation, characterization, and cDNA sequencing of
a-1-antiproteinase-like protein from
rainbow trout seminal plasma
Monika Maka,1, Pawel Makb, Mariusz Olczakc, Agata Szalewiczc, Jan Glogowskia,Adam Dubinb, Wieslaw Watorekc, Andrzej Ciereszkoa,*
a Institute of Animal Reproduction and Food Research, Polish Academy of Sciences, Tuwima 10, 10-747 Olsztyn, PolandbFaculty of Biotechnology, Jagiellonian University, Gronostajowa 7, 30-387 Krakow, Poland
cWroclaw University, Institute of Biochemistry and Molecular Biology, Tamka 2, 50-137 Wroclaw, Poland
Received 22 October 2003; accepted 3 February 2004
Abstract
Seminal plasma of teleost fish contains serine proteinase inhibitors related to those present in blood. These inhibitors can be bound to Q-
Sepharose and sequentially eluted with a NaCl gradient. In the present study, using a two-step procedure, we purified (73-fold to
homogeneity) and characterized the inhibitor eluted as the second fraction of antitrypsin activity (inhibitor II) from Q-Sepharose. The
molecular weight of this inhibitor was estimated to be 56 kDa with an isoelectric point of 5.4. It effectively inhibited trypsin and
chymotrypsin but was less effective against elastase. It formed SDS-stable complexes with cod and bovine trypsin. Inhibitor II appeared to be
a glycoprotein. Carbohydrate content was determined to be 16%. N-terminal Edman sequencing allowed identification of the first 30 N-
terminal amino acids HDGDHAGHTEDHHHHLHHIAGEAHPQHSHG and 25 amino acids within the reactive loop
IMPMSLPDTIMLNRPFLLFILEDST. The N-terminal sequence did not match any known sequence, however, the sequence within the
reactive loop was significantly similar to carp and mammalian a1-antiproteinases. Both sequences were used to construct primers and obtain
a cDNA sequence from liver. The mRNA coding the protein is 1675 nt in length including a single open reading frame of 1281 nt that
encodes 426 amino acid residues. Analysis of this sequence indicated the presence of putative conserved serpin domains and confirmed the
similarity to carp a1-antiproteinase and mammalian a1-antiproteinase. Our results indicate that inhibitor II belongs to the serpin superfamily
and is similar to a1-antiproteinase.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Fish; Semen; Serpin; Proteinase inhibitor
1. Introduction
Serine proteinase inhibitors (serpins) are a dominant
class of proteinase inhibitors in blood plasma [1]. The
serpin superfamily has evolved over 500 million years with
representatives found in viruses, plants, protozoa, insects,
and higher vertebrates [2]. They are involved in maintaining
homeostasis through multiple regulatory functions, such as
blood coagulation, the complement cascade, fibrynolysis,
protein folding, cell migration, and differentiation, modula-
tion of inflammatory responses, prohormone conversion and
intracellular proteolysis [1]. Serpins account for more than
10% of plasma proteins. The main serine proteinase inhib-
itor in blood is a1-proteinase inhibitor (a1-PI), also called
a1-antiproteinase or a1-antitrypsin. This inhibitor has a
broad specificity towards serine proteinases, but neutrophil
elastase seems to be its physiological target [3].
In contrast to blood, seminal plasma of teleost fish is
characterized by a very low (1–3 mg/ml) protein concen-
tration. We have found previously [4] that fish seminal
plasma exhibits a distinct antitrypsin activity. Further elec-
trophoretic studies revealed that there are a number of
species-specific antitrypsin bands in seminal plasma with
different affinity towards cod and bovine trypsin [5]. Our
0304-4165/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbagen.2004.02.001
* Corresponding author. Tel.: +48-89-535-7426; fax: +48-89-535-
7421.
E-mail address: [email protected] (A. Ciereszko).1 Present address: Department of Immunology, Collegium Medicum,
Jagiellonian University, Czysta 18, 31–121 Krakow, Poland.
www.bba-direct.com
Biochimica et Biophysica Acta 1671 (2004) 93–105
Page 2
results indicated that, opposite to mammals and birds, most
serine proteinase inhibitors of fish seminal plasma are of
high molecular weight and are related to blood plasma
inhibitors. This suggests that they may belong to the serpin
family. The presence of serpins in fish seminal plasma has
been indicated by Huang et al. [6]. These authors found that
the proteinase inhibitor of the serpin family (p62), isolated
from carp perimeningeal fluid, is also present in numerous
tissues, including milt. This protein has been cloned and
recognized as the fish equivalent of a1-antitrypsin [7].
However, according to our knowledge, isolation and puri-
fication of serine proteinase inhibitors from fish seminal
plasma has never been attempted.
We have found recently that the antitrypsin activity of
rainbow trout seminal plasma can be sequentially resolved
into three fractions (I–III) after ion-exchange chromatogra-
phy on Q-Sepharose [8]. Antiproteinase of fraction II
(inhibitor II) can be detected in seminal and blood plasma
using cod and bovine trypsin and chymotrypsin. In the
current study, we report the purification and physicochem-
ical and kinetic characterization of this inhibitor. The N-
terminal and reactive loop amino acid sequences have also
been obtained. Using information derived from these
sequences a cDNA from rainbow trout liver was obtained.
Analysis of characteristics of purified protein and the cDNA
sequence demonstrate that properties of this inhibitor are
similar to a1-proteinase from carp perimeningeal fluid.
2. Materials and methods
2.1. Source of milt
Milt was obtained by stripping rainbow trout (Oncorhyn-
nchus mykiss Walbaum, 1792) maintained at the Rutki
Salmonid Research Laboratory, Institute of Inland Fisheries,
Olsztyn, Poland. Seminal plasma was obtained by centrifu-
gation (10,000� g, 10 min, 4 jC) and stored at � 20 jC.
2.2. Determination of protein
Protein concentration was determined by the method of
Bradford [9] using Sigma reagents and bovine serum
albumin as a standard.
2.3. Polyacrylamide gel electrophoresis (PAGE)
Slab gel native (PAGE) and sodium dodecyl sulfate
(SDS-PAGE) electrophoresis were conducted according to
the method of Laemmli [10] using the Mighty Small II
electrophoresis system (Hoefer-Amersham Biosciences, Pis-
cataway, NJ, USA) as was described in the manufacturer’s
instructions. Antitrypsin activity after PAGE was identified
in gels using bovine and cod trypsin or chymotrypsin,
according to Uriel and Berges [11]. Proteins in gels were
stained with Coomassie Brilliant Blue R-250. Molecular
weights of proteins were estimated with the use of the
Kodak 1D program (Eastman Kodak Company, New Ha-
ven, USA).
2.4. Measurement of antitrypsin activity
Antitrypsin activity was measured by the inhibition of
cod trypsin amidase according to the method of Geiger and
Fritz [12] with modifications previously described [5].
2.5. Purification of inhibitor from seminal plasma
Seminal plasma was dialyzed for 24 h against 20 mM
Tris–HCl buffer pH 7.6, (buffer A) at 4 jC. Ion-exchange
chromatography was carried out as previously described
[8]. Briefly, the dialysate was centrifuged and applied to a
column X/K 16/10 of HiLOad Q-Sepharose connected to a
FPLC system (Amersham Biosciences, Uppsala, Sweden).
The column was equilibrated with buffer A. All antitrypsin
activity was bound under these conditions. The bound
proteins were eluted with a linear NaCl gradient (0–0.5
M) in buffer A at a flow rate of 60 ml/h. Three fractions of
antiproteinase activity were obtained as described previ-
ously [8], eluted at a conductivity range of 1.2–1.4, 13.1–
14.4, and 14.8–17.6 mS. The second fraction was pooled
and lyophilized. The lyophilisate was diluted in 2 ml H2O
and applied to a column X/K HiLoad 16/60 of Superdex
200 equilibrated with 50 mM Tris–HCl buffer (pH 7.6).
Elution was conducted with the same buffer at a flow rate
of 60 ml/h. Fractions of the pure inhibitor were lyophi-
lized, then resuspended in H2O and dialyzed against 20
mM Tris–HCl buffer, pH 7.6. The preparation was stored
at � 20 jC.
2.6. Determination of molecular weight
Molecular weight of the native inhibitor was estimated
using gel filtration in an X/K HiLoad 16/60 Superdex 200
column (Amersham Biosciences) equilibrated with 0.05 M
Tris–HCl buffer (pH 7.6) containing 0.15 M NaCl. The
column was calibrated with low and high molecular weight
protein standards (Amersham Biosciences).
The molecular weight of denatured inhibitor was estima-
ted by 12.5% SDS-PAGE slab gels. A low molecular weight
kit 14–94 kDa supplied by Amersham Biosciences was
used for calibration of gels. Molecular weights of protei-
nases and proteinase inhibitors were estimated with the use
of the Kodak 1D program (Eastman Kodak Company).
2.7. Complex formation between the inhibitor with cod and
bovine trypsin
The inhibitor (4 Ag) was incubated in 20 mM Tris–HCl
buffer (pH 7.6) for 15 min with a molar excess of either cod
or bovine trypsin. The samples were then boiled in the
presence of SDS and analyzed using SDS-PAGE.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–10594
Page 3
2.8. Determination of isoelectric point
Chromatofocusing was performed using 0.5� 20 cm of
PBEk94 column (Amersham Biosciences) equilibrated
with 0.025 M imidazole-HCl buffer (pH 7.4). Elution was
performed using 8-fold diluted Polybuffer 74-HCl (pH 5.0).
Isoelectrofocusing of proteinase inhibitor was carried out in
ready IEF gels pH 5–8 using a Ready Gels Cell with
cathode and anode IEF buffers (Bio-Rad, Hercules, CA,
USA). The isoelectric Focusing Marker Kit (Sigma, St.
Louis, MO, USA) was used for calibration of gels.
2.9. Interaction with lectins
Interaction of the inhibitor with the following digoxige-
nin-labeled lectins were tested: Galanthus nivalis agglutinin
(GNA) with affinity for terminal D-mannosyl residues in the
high mannose or hybrid type oligosaccharides; Sambucus
nigra agglutinin (SNA) with affinity for sialic acid bound
via a(2–6) to D-galactose or N-acetyl-D-galactosamine;
Maackia amurensis agglutinin (MAA) with affinity for
sialic acid bound via a(2–3) to D-galactose; Datura stra-
monium agglutinin (DSA) with affinity for terminal D-
galactose h(1–4) N-acetyl-D-glucosamine in complex or
hybrid glycans and peanut agglutinin (PNA) indicating
galactose-h(1–4) N-acetylgalactosamine (the disaccharide
usually forming core the unit of O-glycans). The inhibitor
(1 Ag) was directly bound to nitrocellulose (NC) mem-
branes using the dot-blot method. The blots were then
blocked and incubated with lectins according to the proce-
dure described in the DIG Glycan Differentiation Kit
(Boehringer, Mannheim, Germany) with slight modifica-
tions concerning blocking procedure and lectin detection
(0.25% gelatin in tris-buffered saline (TBS) and horseradish
peroxidase (HRP)-labeled Fab fragments of sheep anti-
digoxygenin antibodies were used, respectively). Carboxy-
peptidase Y, transferrin, fetuin, and asialofetuin were used
as control proteins. The enzyme conjugated probes were
visualized by peroxidase reaction using 0.06% 4-chloro-1-
naphtol in 50 mM Tris–HCl (pH 7.6) containing 0.01%
H2O2. Reactivity with ConA was tested on blots by the
ConA-HRP method according to Olden and Yamada [13]
and Clegg [14]. Activity of HRP on blots was detected
using 4-chloro-1-naphtol (0.06%) in 50 mM Tris–HCl
buffer (pH 7.5) containing 0.01% H2O2.
2.10. Enzymatic deglycosylation
Enzymatic deglycosylation of denatured inhibitor was
performed with N-glycosidase F (PNGase F, EC 3.2.2.18)
from Flavobacterium meningosepticum (Oxford Glycosys-
tem, UK). Inhibitor II (6 Ag) dissolved in the 20 mM
sodium phosphate buffer (pH 7.5) containing 50 mM
EDTA and 0.5% SDS was denatured by boiling for 2
min. Then n-octylglucoside was added to a final concen-
tration of 2%, followed by PNGase F in the proportion 25
mU of endoglycosidase activity/mg of protein. In the
control sample PNGase F was omitted. The reaction was
carried out at 37 jC for 18 h. The final reaction products
were analyzed by SDS-PAGE. Time-dependent deglycosy-
lation of inhibitor II was carried out under the same
conditions. At indicated time intervals equal aliquots were
withdrawn and the reaction was stopped by heat denatur-
ation in the SDS-PAGE sample buffer. The intermediate
products were then analyzed by SDS-PAGE. The effective-
ness of deglycosylation was checked by the reactivity with
ConA. For this purpose, samples resolved in SDS-PAGE
were transferred onto NC membranes using 25 mM Tris-
192 mM glycine/20% (v/v) methanol buffer (pH 8.3)
according to Towbin et al. [15]. Electrotransfer was per-
formed for 1 h at 200 mA in a Fastblot B33 semidry
apparatus (Biometra, Goettingen, Germany). Glycoproteins
bound to nitrocellulose were detected using the ConA-HRP
method as described earlier. Molecular weights of native
and deglycosylated inhibitor II were estimated with the use
of the Kodak 1D program.
2.11. Measurement of association rate constant
Concentrations of active bovine trypsin, cod trypsin
(Sigma), and bovine chymotrypsin (Sigma), were deter-
mined by active-site titration with p-nitrophenyl-pV-guani-
dinebenzoate [16]. Porcine elastase (ICN Biomedicals, Inc.,
Costa Mesa, USA) was titrated against a standardized
solution of human a-1-proteinase inhibitor as a secondary
standard using MeO-Suc-Ala-Ala-Pro-Val-pNA (Sigma) for
measurement of residual enzyme activity. The concentration
of active trypsin inhibitor from rainbow trout seminal
plasma was determined with the use of a standardized
solution of bovine trypsin, using Bz-Arg-pNA (BAPNA)
as substrate. The second-order association rate constants of
the inhibitor with various serine proteinases were deter-
mined by the method of Bieth [17]. Briefly, equimolar
mixtures of enzyme and inhibitor were incubated for in-
creasing times at room temperature in 0.2 M Tris–HCl
buffer (pH 8.0). Residual activities of enzymes were mea-
sured with the following proteinase substrates: BAPNA
(trypsin), Suc-Ala-Ala-Pro-Phe-pNA (chymotrypsin), and
MeO-Suc-Ala-Ala-Pro-Val-pNA (elastase). These substrates
were purchased from KabiVitrum, Stockholm, Sweden. The
association rate constant for each enzyme and inhibitor as
well as delay times of inhibition were calculated according
formulas of Bieth [17]. The final inhibitor II concentration
used in calculations was 116 nM and was equivalent to a
final inhibitor concentration in plasma calculated from the
purification table.
2.12. Amino acid sequence analysis
The N-terminal amino acid sequence analysis was per-
formed on a gas-phase sequencer (Model 491, Perkin
Elmer-Applied Biosystems, Foster City, CA, USA) at the
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 95
Page 4
BioCenter (Jagiellonian University, Krakow, Poland). The
phenylthiohydantoin derivatives were analyzed by on-line
gradient high performance liquid chromatography on a
microgradient delivery system model 140C equipped with
programmable absorbance detector model 785A (both from
Perkin Elmer-Applied Biosystems). Sequence comparisons
were performed using the database SWISS-PROT (http://
www.ncbi.nlm.nih.gov/blast).
2.13. Determination of N-terminal sequence of the inhibitor
Inhibitor II (1 nM) was desalted by HPLC using reversed
phase C-18 column (Waters, Milford, USA.) equilibrated
with 0.1% trifluoroacetic acid. Inhibitor II was eluted by a
gradient of 0.07% trifluoroacetic acid with 80% acetonitrile.
Eluted inhibitor was lyophilized, dissolved in water, and
sequenced.
2.14. Determination of sequence of peptide from reactive
center obtained after cleavage at P1–P1V
Inhibitor II was incubated for 15 min with bovine trypsin
at a molar ratio of 1.3:1.0 in order to obtain peptide from the
reactive center. After incubation, the residual trypsin activity
was inhibited by 4 mM of synthetic inhibitor PVBLOC
(Boehringer). Samples were boiled and subjected to tris-
tricine SDS-PAGE [18] and electroblotted onto PVDF mem-
brane using 10 mM 3-(cyclohexylamino)-1-propanesulfo-
noic acid-NaOH (pH 11.0) containing 10% methanol. The
membrane was stained in 0.1% CBB in 40% methanol, 1%
acetic acid and destained in 50% methanol. The band of a 3
kDa peptide was cut off and sequenced.
2.15. Determination of sequence of peptide from reactive
center obtained after cleavage at P4
Inhibitor II cleaved above P1–P1V was obtained after
incubation of the inhibitor with proteinase V8 from Staph-
ylococcus aureus at a mass ratio of 200:1. The reaction was
stopped with PVBLOC when residual inhibitory activity
was approximately 20% in order to avoid further proteolysis
of other than the exposed reactive loop sites of the inhibitor.
Further steps including electrophoresis, electroblotting, and
sequencing were performed as described above.
2.16. cDNA isolation from rainbow trout liver and sequence
analysis
Fresh liver tissue was immediately frozen in liquid
nitrogen and stored at � 80 jC. The tissue was homoge-
nized using a mortar and pestle followed by QIAShredder
microcolumns (Qiagen GmbH, Hilden, Germany). Total
RNA from 40 mg of homogenized tissue was purified using
the RNeasy Mini Kit (Qiagen) as recommended by the
manufacturer. The integrity of the total RNA was checked
with formaldehyde agarose electrophoresis.
To identify the cDNA of the inhibitor we designed a set of
degenerate PCR primers, based on the sequence from the N-
terminus of the protein and sequences of the peptides
generated by enzymatic digestion. A pair of forward 5V-
CA(CT)CA(CT)AT(ACT)GCIGGIGA(AG)GC-3V and re-
verse 5V-TT(ACGT)AGCATIATIGT(AG)TC(ACGT)GG-3V
primers gave a discrete PCR product of 1085 bp. The product
was cloned into pCR4-TOPO vector (Invitrogen Corpora-
tion, Carlsbad, CA, USA) and sequenced. The full-length
cDNA was further synthesized using the GeneRacer Kit
(Invitrogen) according to the manufacturer’s protocol. For
5VRACE and 3VRACE, a reverse 5V-GGTGTTGGCTTTGC-
TCAGGCTCCTG-3Vprimer, and a forward 5V-TTACCCA-
CAAGGCTGTGCTGAGCG-3V primer were designed,
respectively. Reverse transcription was carried out at 55 jC
for 1 h, using the Thermoscript reverse trancriptase (Invi-
trogen). All PCR reactions were performed in the presence of
Dynazyme Ext Polymerase (Finnzymes Oy, Espoo, Finland).
The same conditions of touchdown PCR were applied to
amplify the 5Vand 3VRACE products: initial incubation at 94
jC for 3 min was followed by five cycles of denaturation (94
jC, 30 s), annealing and extension (72 jC, 60 s), then five
cycles of denaturation (94 jC, 30 s), annealing and extension
(70 jC, 60 s), then 25 cycles of denaturation (94 jC, 30 s),
annealing (65 jC, 30 s) and extension (72 jC, 30 s), and
finally 10 min at 72 jC. The reaction volume was 50 Al. The
5Vand 3VPCR products were cloned into pCR4-TOPO vector
and sequenced. Motif scan in protein sequence was per-
formed using the Prosite database (http://us.expasy.org/
prosite/) according to Falquet et al. [19]. The putative
signal peptide was determined using SignalP V1.1 (http://
www.cbs.dtu.dk/services/SignalP) according to Nielson et
al. [20]. Molecular weight and isoelectric point of the
translated protein was calculated using ProtParm Tool
program (http://us.expasy.org/cgi-bin/protparam).
3. Results
3.1. Isolation of inhibitor
The two-step isolation procedure led to a 73-fold purified
inhibitor with a yield of about 20% (Table 1). During ion-
exchange chromatography the inhibitor was eluted before
two main plasma proteins and for this reason it was possible
Table 1
Isolation of inhibitor II from rainbow trout seminal plasma
Fraction Total
protein
(mg)
Total
activity (U)*
Specific
activity
(U mg� 1)
Fold Yield
(%)
Seminal plasma 14.1 2.728 0.19 1 100.0
Q-Sepharose 0.11 0.654 5.94 31.3 23.9
Superdex 200 0.04 0.558 13.95 73.4 20.4
*U—one unit was defined as inhibitor amount that inhibits activity of
one international enzymatic unit of cod trypsin.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–10596
Page 5
to obtain relatively pure preparations after this step (Fig. 1).
Final purification has been achieved using gel filtration. The
inhibitor migrated during PAGE and SDS-PAGE as a single
band. It could be visualized after PAGE with the zymogram
method using either cod trypsin, bovine trypsin, or bovine
chymotrypsin.
3.2. Physicochemical characteristics
The molecular weight of the inhibitor, estimated using gel
filtration and SDS-PAGE (in reducing and non-reducing conditions) was determined to be 56 kDa. The inhibitor
formed SDS-stable complexes with bovine and cod trypsin
(Fig. 2). Molecular weights of the inhibitor, bovine trypsin,
cod trypsin, inhibitor/bovine trypsin complex, and inhibitor/
cod trypsin complex were determined to be 56, 22.3, 21.6,
65, and 66 kDa, respectively. Two bands of lower molecular
weight were also seen which suggests proteolysis of the
Fig. 4. Deglycosylation of inhibitor II. Lane 1—Molecular weight
standards; lane 2—inhibitor II, control sample; lane 3—18 h incubation
of inhibitor II with PNGase F. (A) CBB staining of the gel, (B) ConA-HRP
staining of NC membrane.
Fig. 3. Isoelectric focusing of inhibitor II. Isoelectrofocusing of proteinase
inhibitor was carried out using ready IEF gels (pH 5–8) in the Ready Gels
Cell and cathode and anode IEF buffers (Bio-Rad). Isoelectric Focusing
Marker Kit (Sigma) was used for calibration of gels. (1) Carbonic
anhydrase I, carbonic anhydrase II (pI 6.6, pI 5.9); (3) methyl red (pI 3.8);
(4) h-lactoglobulin (pI 5.1); (2) inhibitor II (pI 5.42).
Fig. 2. Formation of SDS-stable complexes between inhibitor II and cod
and bovine trypsin. The inhibitor (4 Ag) was incubated in 20 mM Tris–HCl
buffer (pH 7.6) for 15 min with a molar excess of either cod or bovine
trypsin. Samples were then boiled in the presence of SDS and analyzed
using SDS-PAGE. (1) Molecular weight standards; (2) bovine trypsin; (3)
cod trypsin; (4) inhibitor II; (5) complex inhibitor II/bovine trypsin; (6)
complex inhibitor II/cod trypsin.
Fig. 1. Electrophoretic profiles of fraction II after HiLoad Q-Sepharose. A
column was equilibrated with 20 mM Tris–HCl buffer and proteins were
eluted using a linear NaCl gradient (0.0–0.5 M). Fractions 109–115 were
collected and further purified using gel filtration.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 97
Page 6
complexes. During chromatofocusing the inhibitor was elut-
ed at pH 5.43. The isoelectric point estimated by electro-
focusing was determined to be 5.42 (Fig. 3).
3.3. Interaction with specific lectins
The glycoprotein character of inhibitor II was first shown
by a positive reaction with ConA. Using digoxigenin-
labeled lectins, an intense reaction of the inhibitor with
DSA and MAA, a weak reaction with SNA, and no reaction
with GNA and PNA were observed (data not shown). This
result indicates the presence of complex glycans, with
branches both sialylated and with non-sialytated. Based
on the intense reaction with MAA and a weak reaction
with SAA, the sialic acid appeared to be mainly a(2–3)
terminally linked to galactose and to a lesser extent linked
a(2–6) to galactose or N-acetylgalactosamine. Inhibitor II
does not contain high mannose and hybrid glycans as was
evidenced by the negative reaction with GNA. The negative
reaction with PNA indicates the lack of O-glycosidically
linked glycans.
3.4. Estimation of carbohydrate content based on deglyco-
sylation of the inhibitor
After deglycosylation with PNGase F the molecular
weight of inhibitor II decreased to about 47 kDa as shown
in SDS-PAGE (Fig. 4A). Deglycosylated inhibitor II did not
react with ConA which indicates that a complete deglyco-
sylation occurred (see Fig. 4B) Thus, inhibitor II contains
about 16% (w/w) carbohydrate.
3.5. Time-dependent deglycosylation
SDS-PAGE analysis of the inhibitor subjected to degly-
cosylation revealed the presence of one intermediate of
molecular weight 50.5 kDa appearing within the first hour
of PNGase F treatment (Fig. 5A). As shown by its reactivity
with ConA, this modified protein was still glycosylated
(Fig. 5B). A completely deglycosylated protein was ob-
served only after 24 h of deglycosylation, forming a single
protein band, non-reactive with ConA (Fig. 5). Thus, the
sugar moiety of the inhibitor contains at least two N-linked
oligosaccharide chains.
3.6. Association rate constant
The percentage of active enzymes and inhibitor were as
follows: cod trypsin 58%, bovine trypsin 65%, bovine
chymotrypsin 56%, human elastase 59%, inhibitor II 74%.
Titrated enzymes and inhibitor were used for association rate
constants determination. The obtained constants values were
used for delay times calculations, i.e. the time required for
complete inhibition of proteinase in vivo (see Table 2).
3.7. Amino acid sequencing
N-terminal sequencing identified the following 30 amino
acids HDGDHAGHTEDHHHHLHHIAGEAHPQHSHG.
After complex formation of inhibitor II with trypsin the
following 21 amino acids of released peptide (P1V–P21V)
were identified: SLPDTIMLNRPFLLFILEDST. Sequenc-
ing of the peptide obtained after cleavage with proteinase
V8 produced the following 10 amino acids IMPMSLPDTI.
It appeared that proteinase V8 cleaved inhibitor II at P4
(Fig. 6).
Fig. 5. Time-dependent deglycosylation. Lanes 1–7: samples of purified inhibitor II after 0, 15, 30 min, 2, 4, 6, and 24 h of PNGase F treatment. S—molecular
weight standards (see Fig. 4). (A) CBB staining of the gel, (B) ConA-HRP staining of NC membrane.
Table 2
Association rate constants and inhibition delay times for inhibitor II
Proteinase kass (M� 1 s� 1) t (s)
Cod trypsin 2� 106 20
Bovine trypsin 11�106 4
Bovine chymotrypsin 4� 106 10
Porcine elastase 1�105 430Fig. 6. Cleavage of inhibitor II with trypsin and proteinase V8.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–10598
Page 7
3.8. cDNA cloning
The nucleotide sequence and the deduced amino acid
sequence are shown in Fig. 7. A putative ATG codon is
located at nucleotide 33. The open reading frame is pre-
dicted to encode a protein of 426 amino acids with a
calculated molecular weight of 47,529 Da and pI 5.86. Both
peptide sequences determined by Edman sequencing
matched perfectly to the amino acid sequence deduced from
the analyzed cDNA. Hydrolysis of a putative signal peptide
should produce mature protein containing 407 amino acids
with a calculated molecular weight of 45,520.5 Da and pI
Fig. 7. Nucleotide sequence and deduced amino acid sequence of a1-antiproteinase-like protein. Sequences determined by amino acid sequence analysis are
underlined and attachment sites for N-linked sugar chains are double underlined. The sequence has been deposited in EMBL Nucleotide Sequence Database
with accession number AJ558113.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 99
Page 8
5.87. However, the N-terminal sequence obtained by amino
acid sequence analysis lacked two first amino acids in
comparison with a putative N-terminal sequence deduced
from cDNA. Motif search of the sequence revealed the
presence of the sequence IMLNRPFLLFI at P6V–16V that
has been identified as the ‘‘serpine signature’’ PROSITE
PDOC00256 (LIVMFU)-x-(LIVMFYAC)-(DNQ)-
(RKHQS)-(PST)-F-(LIVMFY)-(LIVMFYC)-x-(LIVM-
FAH). A histidine-rich region (PROSITE PDOC50099) has
been also localized at the N-terminal region. Two potential
N-glycosylation sites Asn267 and Asn279 (PROSITE
PDOC 00001) have been detected between 267–269
(NFT) and 279–282 (NSS). Five putative PKC phosphor-
ylation sites (PROSITE PDOC00005, pos. 235–237, 256–
258, 365–367, 410–412, 424–426) and CK2-phosphory-
lation sites (PROSITE PDOC00006, pos. 49–52, 314–317,
330–333, 383–386, 391–394) were identified. Potential N-
mirystyolation sites (PROSITE PDOC00008) have been
localized at pos. 3–8, 90–95, 147–152, 339–344, and
376 – 381 and a tyrosine sulfation site (PROSITE
PDOC00003) at pos. 255–269.
4. Discussion
According to our knowledge, this is the first report
concerning the isolation and characterization of a serine
proteinase inhibitor from fish seminal plasma. Isolation of
the inhibitor from rainbow trout seminal plasma was
achieved by sequential application of Q-Sepharose ion-
exchange chromatography and Superdex 200 gel filtration.
This procedure is much simpler and faster in comparison
with protocols employed for purification of proteinase inhib-
itors from fish blood or perimeningeal fluid [6,21]. Seminal
plasma of rainbow trout, like most teleost fish, is character-
ized by a very low protein concentration (1–3 mg/ml). The
inhibitor is one of the main proteins of seminal plasma (it can
be detected directly by CBB-staining in seminal plasma). Q-
Sepharose chromatography was very efficient and produced
the inhibitor preparations with a high purity.
The molecular weight of the isolated inhibitor was deter-
mined to be 56 kDa. This value was obtained using both gel
filtration and SDS-PAGE in reducing and non-reducing
conditions. Therefore, it may be concluded that the inhibitor
is composed of a single polypeptide chain. Rainbow trout
inhibitor has a similar weight to the a1-proteinase inhibitor
of carp serum (55 kDa, [22]) and a lower weight than carp
perimeningeal fluid serine proteinase inhibitor CP9 (62 kDa;
[6]). The molecular weight calculated from nucleotide se-
quencing (45.5 kDa) is in close agreement with the molec-
ular weight obtained after deglycosylation (47 kDa).
The isoelectric point of the inhibitor from rainbow trout
seminal plasma (pI = 5.4) is similar to values described for
other fish serpins, for example pI of salmon antithrombin
is within the range of 4.7–6.0 [23] but lower values of pI
(4.2–4.9) have been reported for mammalian a1-proteinase
inhibitor [3,24–27]. It should be noted that many mam-
malian a1-proteinases show microheterogeneity and multi-
ple isoforms can be visualized by isoelectric focusing
[3,27]. Our data indicate that inhibitor II is present as a
single band after isoelectric focusing (Fig. 4). In this
respect, inhibitor II may resemble horse a1-proteinase
[3]. The isoelectric point calculated from the amino acid
sequence (5.87) was higher than the pI of a native protein.
It is likely that components of a carbohydrate moiety, such
as sialic acid residues, contribute to a lower pI of the
native protein.
Inhibitor II appeared to be a glycoprotein and its carbo-
hydrate content falls within the range characteristic for
serpins (10–20%), including a1-antiproteinase [3]. Treat-
ment with PNGase F caused a complete deglycosylation of
inhibitor II. This indicates the presence of N-linked carbo-
hydrate side chains in the molecule of inhibitor II. Such
presence is a characteristic feature of most a1-antiprotei-
nases and carbohydrate side chains are N-glycosidically
linked to the Asn side chains [3]. N-linked but not O-linked
carbohydrate side chains were also found in serpin p62 of
carp perimeningeal fluid [6]. Binding of specific lectins to
the carbohydrate moieties indicates that N-linked carbohy-
drates of inhibitor II are sialylated, complex oligosacchar-
ides. The sialic acid appeared to be terminally linked mainly
through a(2–3) and to a lesser extent through a(2–6) to
galactose based on the intense reaction with M. amurensis
agglutinin and a weak reaction with S. nigra agglutinins. A
Fig. 7 (continued ).
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105100
Page 9
positive reaction with D. stramonium agglutinin suggests
the presence of terminal Galh(1–4)GlcNAc. This may
suggest that a partial loss of terminal sialic acid occurred
leaving some branches terminated by Galh(1–4)GlcNAc, as
was suggested for salmon antithrombin [23]. Failure of
inhibitor II to react with G. nivalis agglutinin suggests that
it does not contain N-linked, high mannose glycan. Overall,
our data indicated that inhibitor II is a glycoprotein and its
carbohydrate components may have a similar structure to
a1-antiproteinase. Analysis of cDNA revealed a presence of
two potential glycosylation sites (Fig. 7). This corresponds
well to the results of time-dependent deglycosylation of
inhibitor II that showed two oligosaccharide chains present
in native protein. Two glycosylation sites of rainbow trout
inhibitor are located at the same place as in a1-antiprotei-
nase of carp (Fig. 8).
The presence of multiple potential phosphorylation and
mirystyolation sites and the tyrosine sulfation site suggests
that rainbow trout a1-antiproteinase could be modified by
these post-translational modifications. Such modifications
have been described for some serpins, for example tyrosine
sulfation for heparin cofactor II [28] and tyrosine phosphor-
ylation for maspin [29]. It has been postulated that phos-
phorylation may be important modulator of biological
functions of uterine serpins [30]. On the other hand, no
phosphorylation has been found in some serpins in spite of
the presence of multiple phosphorylation sites [31]. Further
studies are necessary to explain if the above mentioned sites
of potential post-translational modifications are indeed uti-
lized in rainbow trout a1-antiproteinase.
Purified inhibitor formed SDS-stable complexes with
bovine and cod trypsin, a unique feature of serpins. Molec-
ular weights of complexes appeared to be lower than the
sum of the individual components. A similar phenomenon
was also observed for a1-proteinase inhibitor from carp
blood plasma [21] and rabbit plasma a1-antiproteinase [32].
Fig. 8. Homology alignment of rainbow trout (first row) and carp (second row) a1-AP. The same amino acids are marked in bold. Stars denote amino acids
strictly conserved in >70% of serpin sequences [2]. N-glycosylation sites are double underlined.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 101
Page 10
It is likely that the lower molecular weight of the com-
plexes were produced due to a partial degradation of the
inhibitor or trypsin. This phenomenon may be partially
attributed to the loss of the 4 kDa C-terminal peptide from
the reactive loop of the inhibitor. Such a peptide was
observed during complex formation between of the inhib-
itor with trypsin and was used for amino acid sequencing.
Since molecular weights of the complexes were lower by
12–13 kDa, this suggests that further proteolytic degrada-
tion of the complexes occurred. It is known that some
serpin-proteinase complexes show increased susceptibility
to cleavage by residual free enzyme [33]. Presumably the
incubation time employed in our experiment (15 min) was
too long and shorter periods are necessary to obtain non-
degraded complexes. Further studies aimed to test complex
formation in relation to time of incubation should clarify
this issue.
Cod trypsin, bovine trypsin and chymotrypsin could be
used for visualization of the inhibitor purified in the present
work. Data of kinetic analysis (Table 2) confirmed that the
inhibitor efficiently inhibits all of these three serine protei-
nases, giving a delay time (t) toward these enzymes in the
range of seconds to tens of seconds. Delay time is a kinetic
parameter that is five times greater than half-life time and
represents a period of time needed for almost complete
inhibition of enzyme activity. Obtained values testify that
the inhibition of trypsin and chymotrypsin is therefore fast
and effective. Obtained results agree with the data of
Aranishi [21] who found that a1-antiproteinase of carp
blood plasma inhibited bovine trypsin and chymotrypsin.
On the other hand, the inhibitor of rainbow trout seminal
plasma was less effective against elastase in comparison to
trypsin and chymotrypsin. Lower affinity towards elastase
may be a characteristic feature of this inhibitor, opposite to
mammalian a1-antiproteinases that have a high affinity
towards elastase and its biological role in humans is defined
by control of leukocyte elastase [34].
Although no statistically significant homology between
N-terminal sequence of inhibitor II obtained by amino acid
sequencing and any known protein was found, an arbitrary
alignment suggested that there is some similarity to N-
terminal sequence of a homolog of serine proteinase inhib-
itor (CP9) isolated from common carp perimeningeal fluid
[7], as is shown in Fig. 8. This is in agreement with a rule
that amino-terminal regions of serpins are most variable [3].
It is known, however, that the sequence within the reactive
site region of serine proteinase inhibitors is highly con-
served [35]. The amino acid sequence within the reactive
center of inhibitor II showed a high degree of homology
with vertebrate a1-antiproteinases, including: 76% identity
with both a1-antitrypsin homolog precursor and serine
proteinase inhibitor CP9 from common carp perimeningeal
fluid [7], 70% identity with rabbit a1-antitrypsin F precursor
[36], 60% identity with a1-antiproteinase of the African
clawed frog (Xenopus leavis, [37]), 54% identity with
guinea pig a1-antiproteinase S precursor [35], and 52%
with ground squirrel a-antitrypsin-like protein [38] and
horse a1-antitrypsin (accession no., AAC83412).
The reactive site region of the a1-antiproteinase family
can be classified into orthodox and unorthodox types [35].
The orthodox type has the P3–P3V consensus sequence of
Xaa-Pro-Met-Ser-Xaa-Pro, where Xaa is Val, Leu, Ileu or
Met. These amino acids are regarded to be very similar in
molecular size and shape as well as in hydrophobicity, and
frequently interchange during evolution of many protein
families. Met at the P1 site is believed to be crucial in the
physiological activity of the orthodox a1-antiproteinases.
The P3–P3V sequence of inhibitor II is Met-Pro-Met-Ser-
Leu-Pro (Fig. 8). This agrees with the consensus of the
orthodox type. Due to the presence of Met at P1 mammalian
a1-antiproteinases are inactivated by oxidation [3] and a
similar phenomenon is expected to be true for fish a1-APs.
The sequences of the peptides obtained from the enzy-
matic digestion aligned perfectly to the deduced amino acid
sequence drawn from the mRNA isolated from the liver as
was anticipated based on our previous results indicating the
same electrophoretical and chromatographical behavior of
inhibitor II from blood and seminal plasma. The highest
homology alignments were found with antitrypsins from
carp (61% and 63%) and African frog (48%). Moreover,
numerous significant alignments were detected with 39–
43% identities with mammalian a1-antiproteinases. This
strongly suggests that rainbow trout serpin described in this
study belongs, similarly to carp antitrypsin, to clade ‘‘a’’ of
serpins (antitrypsin-like) as was classified by Irving et al. [2].
Putative conserved serpin domains have been identified
by BLAST program (Fig. 9). Moreover, most (44 of 51, Fig.
8) of strictly conserved amino acids in >70% of the serpin
sequences are found in both carp and rainbow trout serpins.
These amino acids are located in the sites important for
serpin conformation and their significance has been reviewed
by Irving et al. [2]. These regions include the hinge (P15–P9,
Fig. 9), located within the ‘‘serpin signature’’, that provides
mobility essential for the conformation change of the reactive
center loop (RCL); the breach and the shutter located on the
top and near the center of the A h-sheet, respectively; and the
gate, including strands s3C and s4C. The breach and shutter
facilitate sheet opening and accept the conserved hinge of the
RCL as it inserts through the gate. Preservation of critical
amino acids in the serpins likely contributes to the conser-
vation of their structure, despite the evolutionary distances
among serpins and their different functions.
A statistically significant histidine-rich motif has been
found between positions 22 and 62 using the program of
Falquet et al. [19]. Interestingly, histidine-rich motifs have
been identified in serpin pNiXa isolated from Xenopus
laevis. It was shown that this motif is involved in binding
of nickel ions [39,40]. Serpin pNiXa [41] is present in
oocytes and embryos and is engaged in the induction of
maturation of oocytes and may regulate proteolytic process-
es during embryonic development [40]. This serpin shows
38% homology to rainbow trout a1-antiproteinase described
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105102
Page 11
in this study and is also produced by the liver. A histidine-
rich domain is also present in a1-antiproteinase of X. laevis
([37], showing 48% homology with trout a1-antiproteinase).
The histidine-rich region can be also seen in the N-terminal
part of carp a1-antiproteinase. Altogether these results
suggest that the presence of a histidine-rich motif may be
a characteristic feature of serpins of lower vertebrates.
Similar to the majority of the mammalian a1-antiprotei-
nases, a mature rainbow trout a1-antiproteinase-like protein
contains only one Cys residue (the other two Cys residues
are located within the putative N-signal sequence). For this
reason, no disulfide bonds are anticipated in the protein
molecule. However, unlike mammals where the Cys residue
is located at position 272, rainbow trout protein has it
(similarly to carp) localized at the N-terminal region (pos.
56 of the deduced cDNA sequence).
Use of the SignalP program [20] allowed for prediction of
the cleavage site for the signal peptide between Ala19 and
Ala20. If this prediction is correct, it implies that additional
cleavage of Ala–Pro dipeptide from protein occurs during
protein maturation, because the N-terminal sequence
obtained after amino acid sequencing of purified protein
starts from His22. Such cleavage might be done by amino-
peptidase or dipeptydyl peptidases. Interestingly, such
enzymes have been identified in human seminal plasma,
including alanyl aminopeptidases and dipeptidyl peptidase
IV that cleaves off N-terminal dipeptides from peptides
containing proline or alanine [42,43]. Aminopeptidase ac-
tivity is also present in fish seminal plasma [44], however,
this activity has not yet been characterized. On the other
hand, although the calculated value of the combined cleav-
age site score [20] is highest for Ala20 (0.869), it is also high
for His22 (0.638). Therefore, it is also possible that the signal
peptide is cleaved between Pro21 and His22 and in such case
the predicted N-terminal sequence will match the sequence
obtained by amino acid sequencing. It seems that carp,
rainbow trout and X. laevis a-APs significantly differ in
the region of the signal peptide. Huang et al. [7] suggested
that a putative junction for the signal peptide and mature
protein is located between Ala16 and Asp17. No signal
peptide has been identified by Yoshida et al. [37] for a1-
antiproteinases of X. laevis.
The physiological role of a1-antiproteinase-like protein
in rainbow trout milt is unknown at present. This inhib-
itor does not appear to be specific for milt because it is
also present in blood and liver, and probably in other
tissues, as was reported for the serpin p62 of carp
perimeningeal fluid [6]. The inhibitor may participate in
protection of spermatozoa from proteolytic attack. Protec-
tion of spermatozoa may be important for successful
prolonged storage of spermatozoa in the spermatic duct
which is characteristic for rainbow trout. Potential target
proteinases for the inhibitor may originate from seminal
plasma, damaged spermatozoa, or blood cells. The pres-
ence of serine proteinases in seminal plasma of rainbow
trout has already been indicated in our laboratory, using
SDS-gelatin-PAGE [45]. Our studies on rainbow trout
seminal plasma also revealed a presence of potential
non-target proteinases for a1-antiproteinase, such as met-
alloproteinases [46,47]. Both blood cells, present in
rainbow trout milt [48], and damaged spermatozoa [49],
are a potential source of serine proteinases. The protective
role of rainbow trout a1-antiproteinase-like protein against
spermatozoa may also include its potential antiapoptotic
and bacteriostatic actions [50,51]. It has to be pointed out
that in addition to protease inhibitory functions, serpins
may influence cell behavior via both direct and indirect
ways. For example, a1-PI may play a role in influencing
tissue repair by directly stimulating fibroblast proliferation
and extracellular matrix production via classical mitogen-
activated signaling pathways [52]. Spermatogenesis con-
sists of many proliferation steps and it is possible that
this process may be controlled by proteinase inhibitors.
The use of pure preparations of a1-antiproteinase-like
protein may be beneficial in further experiments aimed
at the examination of its role in the reproduction of
rainbow trout.
Acknowledgements
This study was supported in part by State Committee for
Scientific Research, Project 5 P06D 027 19 and funds
appropriated to Faculty of Biotechnology, Jagiellonian
Fig. 9. Homology alignment of reactive site sequences of a1-AP of fish, amphibia and mammals. (1) Rainbow trout, (2) carp homolog precursor, (3) carp CP9,
(4) X. leavis, (5) rabbit, and (6) horse. Identical amino acids are marked in bold. Sh = sheet; St = strand, after Ref. [3].
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 103
Page 12
University and Institute of Biochemistry and Molecular
Biology, Wroclaw University. Thanks are due to Michael H.
Penn for editing the manuscript.
References
[1] J. Potempa, E. Korzus, J. Travis, The serpin superfamily of protein
inhibitors: structure, functions, and regulation, J. Biol. Chem. 269
(1994) 15957–15960.
[2] J.A. Irving, R.N. Pike, A.M. Lesk, J.C. Whisstock, Phylogeny of the
serpin superfamily: implications of patterns of amino acid conserva-
tion for structure and function, Genome Res. 10 (2000) 1845–1864.
[3] S.D. Patterson, Mammalian a1-antitrypsins: comparative biochemis-
try and genetics of the major plasma serpin, Comp. Biochem. Physiol.
100B (1991) 439–454.
[4] K. Dabrowski, A. Ciereszko, Proteinase inihibitor(s) in seminal plas-
ma of teleost fish, J. Fish Biol. 45 (1994) 801–809.
[5] A. Ciereszko, B. Piros, K. Dabrowski, D. Kucharczyk,M.J. Luczynski,
S. Dobosz, J. Glogowski, Serine proteinase inhibitors of seminal
plasma of teleost fish: distribution of activity, electrophoretic pro-
files and relation to proteinase inhibitors of blood, J. Fish Biol. 53
(1998) 1292–1305.
[6] C.-J. Huang, C.-C. Chen, H.-.J. Chen, F.-L. Huang, G.D. Chang, A
protease inhibitor of the serpin family is a major protein in carp
perimeningeal fluid: I. Protein purification and characterization, J.
Neurochem. 64 (1995) 1715–1720.
[7] C.-J. Huang, M.-S. Lee, F.-L. Huang, G.-D. Chang, A protease in-
hibitor of the serpin family is a major protein in carp perimeningeal
fluid: II. cDNA cloning, sequence analysis, and Escherichia coli ex-
pression, J. Neurochem. 64 (1995) 1721–1727.
[8] A. Ciereszko, M. Kwasnik, K. Dabrowski, B. Piros, J. Glogowski,
Chromatographic separation of trypsin-inhibitory activity of rainbow
trout blood and seminal plasma, Fish Shellfish Immunol. 10 (2000)
91–94.
[9] M.M. Bradford, A rapid and sensitive method for quantitation of
microgram quantities of protein utilizing the principle of protein
dye-binding, Anal. Biochem. 72 (1976) 248–254.
[10] U.K. Laemmli, Cleavage of structural proteins during the assembly of
the head bacteriophage T4, Nature 227 (1970) 680–685.
[11] J. Uriel, J. Berges, Characterization of natural inhibitors of trypsin and
chymotrypsin by electrophoresis in acrylamide-agarose gels, Nature
218 (1968) 578–580.
[12] R. Geiger, H. Fritz, Trypsin, in: H.U. Bergmeyer (Ed.), Methods
of Enzymatic Analysis, vol. 5. Verlag Chemie, Weinheim, 1983,
pp. 121–124.
[13] K. Olden, K.M. Yamada, Direct detection of antigens in sodium
dodecyl sulfate-polyacrylamide gels, Anal. Biochem. 78 (1977)
483–490.
[14] J.C. Clegg, Glycoprotein detection in nitrocellulose transfers of elec-
trophoretically separated protein mixtures using concanavalin A and
peroxidase: application to arenavirus and flavivirus proteins, Anal.
Biochem. 127 (1982) 389–994.
[15] H. Towbin, T. Staehelin, J. Gordon, Electrophoretic transfer of proteins
from polyacrylamide gels to nitrocellulose sheets: procedure and some
applications, Proc. Natl. Acad. Sci. U. S. A. 76 (1979) 4350–4354.
[16] T. Chase Jr., E.S. Show, Titration of trypsin, plasmin and thrombin
with p-nitrophenyl guanidobenzoate, Methods Enzymol. 19 (1970)
20–27.
[17] J. Bieth, Pathophysiological interpretation of kinetic constants of pro-
tease inhibitors, Bull. Eur. Physiopathol. Respir. 16 (1980) 183–195.
[18] H. Schagger, G. von Jagow, Tricine-sodium dodecyl sulfate-poly-
acrylamide gel electrophoresis for the separation of proteins in the
range from 1 to 100 kDa, Anal. Biochem. 166 (1987) 368–379.
[19] L. Falquet, M. Pagni, P. Bucher, N. Hulo, C.J. Sigrist, K. Hofmann,
A. Bairoch, The PROSITE database, its status in 2002, Nucleic Acids
Res. 30 (2002) 235–238.
[20] H. Nielson, J. Englbrecht, S. Brunak, G. von Heijne, Identification of
prokaryotic and eukaryotic signal peptides and prediction of their
cleavage sites, Protein Eng. 10 (1997) 1–6.
[21] F. Aranishi, Purification and characterization of serum serpin from
carp (Cyprinus carpio), Mar. Biotechnol. 1 (1999) 81–88.
[22] F. Aranishi, Purification and characterization of a1-proteinase inhib-
itor from carp (Cyprinus carpio) serum, Mar. Biotechnol. 1 (1999)
33–43.
[23] Ø. Andersen, R. Flengsrud, K. Norberg, R. Salte, Salmon antithrom-
bin has only three carbohydrate side chains, and shows functional
similarities to human h-antithrombin, Eur. J. Biochem. 267 (2000)
1651–1657.
[24] A. Koj, M.W. Hatton, K.L. Wong, E. Regoeczi, Isolation and partial
characterization of rabbit plasma alpha1-antitrypsin, Biochem. J. 169
(1978) 589–596.
[25] Y. Suzuki, K. Yoshida, T. Ichimiya, T. Yamamoto, H. Sinohara, Tryp-
sin inhibitors in guinea pig plasma: isolation and characterization of
contrapsin and two isoforms of a-1-antiproteinase and acute phase
response of four major trypsin inhibitors, J. Biochem. 107 (1990)
173–179.
[26] K. McGilligan, D.W. Thomas, Evaluation of assays for detecting
alpha-1-protease inhibitor during purification from rat serum, Anal.
Biochem. 193 (1991) 260–265.
[27] E. Mattes, H.P. Matthiessen, P.L. Tureck, H.P. Schwarz, Preparation
and properties of an alpha-1-protease inhibitor concentrate with high
specific activity, Vox Sang. 81 (2001) 29–36.
[28] C. Bohme, M. Nimtz, E. Grabenhorst, H.S. Conradt, A. Strathmann,
H. Ragg, Tyrosine sulfation and N-glycosylation of human heparin
cofactor II from plasma and recombinant Chinese hamster ovary cells
and their effects on heparin binding, Eur. J. Biochem. 269 (2002)
977–988.
[29] V.A. Odero-Marah, Z. Khalkhali-Ellis, G.B. Schneider, E.A. Seftor,
R.E.B. Seftor, J.G. Koland, M.J.C. Hendrix, Tyrosine phosphoryla-
tion of maspin in normal mammary epithelia and breast cancer cells,
Biochem. Biophys. Res. Commun. 295 (2002) 800–805.
[30] M.R. Peltier, L.C. Raley, D.A. Liberles, S.A. Benner, P.J. Hansen,
Evolutionary history of the uterine serpins, J. Exp. Zool. 288 (2000)
165–174.
[31] C.Y. Chen, U. Cronshagen, H.F. Kern, A novel pancreas-specific
serpin (ZG-46p) localizes to the soluble and membrane fraction of
the Golgi complex and the zymogen granules of acinar cells, Eur. J.
Cell Biol. 73 (1997) 205–214.
[32] A. Saito, H. Sinohara, Isolation and characterization of rabbit plasma
a-1-antiproteinase E, Biol. Chem. 379 (1998) 1367–1370.
[33] Y.S. Askew, S.C. Pak, C.J. Luke, D.J. Askew, S. Cataltepe, D.R.
Mills, H. Kato, J. Lehoczky, K. Dewar, B. Birren, G.A. Silverman,
SERPINB12 is a novel member of the human ov-serpin family that is
widely expressed and inhibits trypsin-like serine proteinases, J. Biol.
Chem. 276 (2001) 49320–49330.
[34] J. Travis, G.S. Salvesen, Human plasma proteinase inhibitors, Ann.
Rev. Biochem. 52 (1983) 655–709.
[35] Y. Suzuki, K. Yoshida, E. Honda, H. Sinohara, Molecular cloning and
sequence analysis of cDNAs coding for guinea pig alpha 1-antipro-
teinases S and F and contrapsin, J. Biol. Chem. 266 (1991) 928–932.
[36] A. Saito, H. Sinohara, Cloning and sequencing of cDNA coding for
rabbit alpha-1-antiproteinase F: amino acid sequence comparison of
alpha-1-antiproteinases of six mammals, J. Biochem. 109 (1991)
158–162.
[37] K. Yoshida, Y. Suzuki, H. Sinohara, Cloning and comparative se-
quence analysis of Xenopus laevis alpha1-antiproteinase, J. Biochem.
Mol. Biol. Biophys. 3 (1999) 59–63.
[38] N. Takamatsu, M. Kojima, M. Taniyama, K. Ohba, T. Uematsu, C.
Segawa, S. Tsutou, M. Watanabe, J. Kondo, N. Kondo, T. Shiba,
Expression of multiple alpha1-antitrypsin-like genes in hibernating
species of the squirrel family, Gene 204 (1997) 127–132.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105104
Page 13
[39] J. Haspel, F.W. Sunderman Jr., S.M. Hopfer, D.C. Henjum, P.W.
Brandt-Rauf, I.B. Weinstein, S. Nishimura, Z. Yamaizumi, M.R. Pin-
cus, A nickel-binding serpin, pNiXa, induces maturation of Xenopus
oocytes and shows synergism with oncogenic ras-p21 protein, Res.
Commun. Chem. Pathol. Pharmacol. 79 (1993) 131–140.
[40] F.W. Sunderman Jr., A.H. Varghese, O.S. Kroftova, S. Grbac-
Ivankovic, J. Kotyza, A.K. Datta, M. Davis, W. Bal, K.S. Kasparzak,
Characterization of pNiXa, a serpin of Xenopus laevis oocytes and
embryos, and its histidine-rich Ni(II)-binding domain, Mol. Reprod.
Dev. 44 (1996) 507–524.
[41] B.L. Beck, D.C. Henjum, K. Antonijczuk, O. Zahaia, G. Korza, J.
Ozols, S.M. Hopfer, A.M. Barber, F.W. Sunderman Jr., pNiXa, a
Ni(2+)-binding protein in Xenopus oocytes and embryos, shows iden-
tity to Ep45, an estrogen-regulated hepatic serpin, Res. Commun.
Chem. Pathol. Pharmacol. 77 (1992) 3–16.
[42] I. DeMeester, G. Vanhoof, A.M. Lambeir, S. Scharpe, Use of immo-
bilized adenosine deaminase (EC 3.5.4.4) for the rapid purification of
native human CD26 dipeptidyl peptidase IV (EC 3.4.14.5), J. Immu-
nol. Methods 189 (1996) 99–105.
[43] D. Fernandez, A. Valdivia, J. Irazusta, C. Ochoa, L. Casis, Peptidase
activities in human semen, Peptides 23 (2002) 461–468.
[44] F. Lahnsteiner, R.A. Patzner, T. Weismann, Testicular main ducts and
spermatic ducts in some cyprinid fishes: I. Morphology, fine structure
and histochemistry, J. Fish Biol. 44 (1994) 937–951.
[45] R. Kowalski, J. Glogowski, D. Kucharczyk, K. Goryczko, S. Dobosz,
A. Ciereszko, Proteolytic activity and electrophoretic profiles of
proteases from seminal plasma of teleosts, J. Fish Biol. 63 (2003)
1008–1019.
[46] R. Kowalski, M. Wojtczak, J. Glogowski, A. Ciereszko, Gelatinolytic
and anti-trypsin activities in seminal plasma of common carp: rela-
tionship to blood, skin mucus and spermatozoa, Aquat. Living
Resour. 16 (2003) 438–444.
[47] P.E. Desrochers, K. Mookhtiar, H.E. Van Wart, K.A. Hasty, S.J.
Weiss, Proteolytic inactivation of alpha 1-proteinase inhibitor and
alpha 1-antichymotrypsin by oxidatively activated human neutrophil
metalloproteinases, J. Biol. Chem. 267 (1992) 5005–5012.
[48] T. Wlasow, J. Glogowski, A. Ciereszko, Presence of blood cells in
rainbow trout milt, Pol. Arch. Fish. 7 (1999) 359–364.
[49] K. Inaba, M. Morisawa, Chymotrypsin-like protease activity associ-
ated with demembraned sperm of chum salmon, Biol. Cell 76 (1992)
329–333.
[50] Y. Miyamoto, T. Akaike, H. Maeda, S-nitrosylated human alpha(1)-
protease inhibitor, BBA Prot. Struct. Mol. Enzymol. 1477 (2000)
90–97.
[51] Y. Ikari, E. Mulvihill, S.M. Schwartz, Alpha 1-Proteinase inhibitor,
alpha 1-antichymotrypsin, and alpha 2-macroglobulin are the antia-
poptotic factors of vascular smooth muscle cells, J. Biol. Chem. 276
(2001) 11798–11803.
[52] K. Dabbagh, G.J. Laurent, A. Shock, P. Leoni, J. Papakrivopoulou,
R.C. Chambers, Alpha-1-antitrypsin stimulates fibroblast proliferation
and procollagen production and activates classical MAP kinase sig-
nalling pathways, J. Cell. Physiol. 186 (2001) 73–81.
M. Mak et al. / Biochimica et Biophysica Acta 1671 (2004) 93–105 105