Purification and characterization of a trypsin inhibitor from mouse seminal vesicle secretion

(1)

Purification

and Characterization

of a Trypsin Inhibitor

from Mouse Seminal Vesicle Secretion

Min-Long

Lai,*

Shun-Wen

Chen,?

and Yee-Hsiung

Chen*,T,’

*Institute

of

Biochemical

Science, College

of

Science, National

Taiwan

University;

and tlnstitute

of

Biological

Chemistry,

Academia

Sinica, Taipei, 10764 Taiwan,

Republic

of

China

Received February 14, 1991, and in revised form May 24, 1991

A Kazal-type

trypsin

inhibitor

in mouse seminal

vesicle

secretion

was purified

to homogeneity

via a series of pu-

rification

steps including

ammonium

sulfate

fraction-

ation,

affinity

chromatography

on a trypsin

Affi-Gel

10 column,

and HPLC

on a reverse

phase C, column.

It was

shown

to be a weak basic protein

with

an isoelectric

point

of 8.7 and to contain

no carbohydrate.

The protein

had

a specific

activity

of 184 U/mg

protein

in the inhibitory

effect on the trypsin

digestion

of N-benzoyl-Pro-Phe-

Arg-p-nitroanilide.

Analysis

of the kinetic

data for the

trypsin

digestion

of N-benzoyl-Phe-Val-Arg

‘I-amido-

4-methylcoumarin

revealed

that the protein

was a com-

petitive

inhibitor

with

an inhibitory

constant

(Ki) of 0.15

nM.

The molecular

mass of the protein

was determined

to be 7 kDa by both gel chromatography

and electropho-

resis. Results

of direct

amino

acid determinations

indi-

cated that this protein

corresponded

to the reading

frame

of MP12

cDNA

identified

from mouse prostate.

We found

that cleavage

only

at the reactive

site of this protein

(Arg’s-Ile20)

resulted

in its denaturation.

o ISSI Academic Press, Inc.

The seminal vesicles present in most adult male mam-

mals secrete a group of proteins that constitute a major

portion of seminal plasma (1, 2). The biological signifi-

cance of these proteins in mammalian reproductive phys-

iology is currently obscure. It was found that extirpation

of seminal vesicles from a mouse greatly reduced fertility

(3,4). Recently, seven and five major proteins have been

identified from mouse and rat seminal vesicle secretion

(SVS).’ (5, 6). The genetic expression of some of these

i To whom correspondence should be addressed. Fax: (02) 363.5038. ‘Abbreviations used: MSVS, mouse seminal vesicle secretion; TI, trypsin inhibitor; PSTI, pancreatic secretory trypsin inhibitor; PI, protease inhibitor; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; IEF, isoelectric focusing; PCR, polymerase chain reaction.

0003.9861/91 $3.00

Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

proteins has been shown to be dependent on the presence

of testosterone (6-8).

Protease inhibitors (PIs) have a physiological function

other than the inhibition of protease activity

(9). Exis-

tence of PIs in sex accessory tissues of mammals is well

known (10-12). They are believed to be important for

protection of genital tract epithelium against proteolytic

damage (13) and/or have a regulatory role in the fertil-

ization process (14, 15). Hence, the study of PIs in the

genital tract is an important subject of reproductive bi-

ology.

A cDNA (MP12) which codes for a Kazal-type PI has

been identified in the mouse ventral prostate by molecular

cloning (16). Androgen-dependent expression of this PI

mRNA as probed with MP12 cDNA has been observed

in male sex accessory tissues including ventral prostate,

seminal vesicle, and coagulating gland, but it is expressed

constitutively

in the pancreas (16). Meanwhile, a PI with

a molecular weight of 12 kDa (P12) has been identified

in the mouse ventral prostate, and it was suggested that

MP12 cDNA corresponds to the P12 protein (17). In the

present study, we purified and characterized one TI from

MSVS. This TI is a basic polypeptide of 57 amino acid

residues, which is similar to rat PSTI-II but different from

rat PSTI-I (9) on the basis of molecular size. It is not a

glycoprotein in nature, which is different from the P12

protein which is believed to be a glycoprotein (17). We

confirmed further that the primary structure of this TI

is identical to that deduced from MP12 cDNA.

EXPERIMENTAL

Materials. Affi-Gel 10 was purchased from Bio-Rad Laboratories (Richmond, CA). N-Benzoyl-Pro-Phe-Arg-p-nitroanilide HCl, N-benzoyl-Phe-Val-Arg 7-amido-4-methylcoumarin, bovine pancreatic tryp sin, and soybean trypsin inhibitor were obtained from Sigma Chemical Co., (St. Louis, MO). Markers for IEF and thin layer ampholine PAGE plate were procured from Sartorius Gmbh (FRG). Sephadex G-50 was obtained from Pharmacia (Uppsala, Sweden). AMV reverse transcriptase and T7 DNA polymerase sequencing system were purchased from Pro- mega (Madison, WI). T4 DNA ligase was obtained from Bethesda Re-

(2)

TABLE I

Purification

of TI from MSVS”

Purification step Protein (mg/mouse) Activity (U/mouse) Specific activity (U/mg of protein) Purification fold Yield % MSVS solution 14.0 17.7 1.25 1 100

Ammonium sulfate 36% soluble fraction 1.67 11.0 6.58 5 62

Trypsin affinity chromatography 0.04 2.4 59.6 48 14

HPLC of Fig. 1A 0.01 1.8 176 141 10

HPLC of Fig. 1B 0.05 9.2 184 147 52

e Data were the average of 20 mice, 9 weeks old. See text for details.

search Laboratories (Gaithersburg, MD). Restriction enzymes such as BumHI, EcoRI, and SmuI were purchased from Boehringer Mannheim Gmbh (FRG). Taq polymerase was obtained from Perkin-Elmer Cetus Co., (Norwalk, CT). Geneclean kit was purchased from Bio 101 Inc. (La Jolla, CA). All chemicals were reagent grade from commercial sources.

Extraction of TI from MSVS. The seminal vesicles of mature male mice (ICR) killed by cervical dislocation were carefully dissected to free them from the adjacent coagulating gland, and the secretion collected from 100 mice was expressed directly into 100 ml of ice cold 5% acetic acid. After stirring at 4°C for 30 min, the solution became clear. The solution was fractionated in 36% saturation of ammonium sulfate, and the mixture was adjusted to pH 2.0 with 6 N HCI. The precipitating material was removed by centrifugation at 8OOOg for 20 min. The solution was passed through a glass filter to remove lipid and dialyzed against 0.5% acetic acid and lyophilized.

SDS-PAGE and ZEF. Proteins were resolved by SDS-PAGE on a 15% gel slab (15 X 12 X 0.075 cm) according to the method of Schagger et al. (18). IEF on a thin layer ampholine PAGE plate (pH range 3-10) was performed by a LKB multiphor unit. The cathodic fluid consisted of 0.44% Arg-0.06% Lys, which was adjusted to pH 10 by the addition of ethylenediamine. The anodic fluid contained 0.33% Asp-0.37% Glu at pH 3.0. Focusing was conducted for 4 h at a constant power of 2.0 W with initial current of 10 mA and a voltage maximum of 1.7 kV.

Immobilization of trypsin. Coupling of protein to Affi-Gel 10 was according to the supplier’s specification. At 4”C, 5 mg of trypsin in 1.0 ml of 5 mM phosphate-O.4 M NaCl at pH 7.9 was mixed overnight with an equal volume of Affi-Gel 10, which was washed with the same buffer beforehand. Ethanolamine was then added to the gel mixture to a final concentration of 1.0 M to stop the coupling reaction. It was determined that 1 ml of the trypsin affinity gel could absorb 150 fig of soybean trypsin inhibitor.

Amino acid analysis and sequencing. Protein was hydrolyzed in the vapor phase of 7.0 M HCl containing 10% trifluoroacetic acid and 0.1% phenol at 158°C for 30 min, according to the method of Tsugita et al. (19). Amino acid composition was determined by the conventional method of Spackman et al. (20) on a Beckman System 6300 high per- formance analyzer. The amount of tryptophan was determined by a UV absorption method. The amino acid sequence was determined by automated Edman degradation with a gas-phase microsequenator (477A protein sequencer with on line 120A analyzer, Applied Biosystems, Foster City, CA).

Sugar analysis. Protein was digested with 3.0 M trifluoroacetic acid in a siliconized tube at 100°C for 6 h. The digested sample was evaporated to dryness, redissolved in an adequate amount of water, and subjected to chromatography on a Dionex Bio LC System (Dionex Corp., Sun- nyvale, CA) attached to a pulse amperometric detector.

Assay of TI actiuity. N-Benzoyl-Pro-Phe-Arg-p-nitroanilide HCl was used as substrate and hydrolysis of the substrate was followed by the change in absorbance at 405 nm (21, 22). One enzyme unit corresponded to the hydrolysis of 1 pmol of substrate per minute by bovine

pancreatic trypsin. The activity of TI was based on its inhibitory effect on the digestion of substrate by trypsin. Trypsin and TI were mixed and preincubated for 5 min before the addition of substrate. One inhibitor unit corresponded to the reduction of one enzyme unit. The specific activity of TI was expressed in inhibitor units per milligram of protein.

Kinetic analysis. The trypsin digestion of N-benzoyl-Phe-Val-Arg 7-amido-4-methylcoumarin was followed by the change in fluorescence at 460 nm with excitation wavelength at 380 nm. The kinetic data for the inhibitory effect of TI on the trypsin activity were analyzed by Dix- on’s plot for determining the inhibitory constant (K,) of a tight binding enzyme inhibitor (23, 24):

K(l + [SIIK,) = K

PI

where K,,, may be regarded as a “pseudo-equilibrium constant” and K is the distance between two neighboring intersection points on the abscissa of the plot of the velocity of the enzyme reaction (ordinate) vs inhibitor concentration (abscissa). A plot of K against [S] will be a straight line of which the intersection with the vertical axis gives Ki and that with the base line gives K,,,.

RNA isolation and cDNA preparation. Total cellular RNA was isolated from adult MSV according to a previous method (25). Single- stranded cDNAs were prepared on the polyadenylated fraction of MSV RNA by standard procedure (25) using AMV reverse transcriptase.

Polymeruse chain reaction, cloning and analysis. Based on MP12 cDNA of mouse ventral prostate, we synthesized one oligonucleotide of

CCCAGATCTTCGACAATGAAGGT, which is in the 5’-noncoding region, and the other oligonucleotide of CCCACGTTGCCTTTCATTA-

CGG, which is complementary to the 3’.untranslated region (see Fig. 5). Those two oligonucleotides were employed as the primer pair for PCR, which amplified the single-stranded cDNAs of MSV with Taq polymerase for 30 cycles: 94°C 1.5 min; 4O“C, 2.0 min; 72”C, 3.0 min. The reaction mixture was subjected to electrophoresis on a 1.8% agarose gel. The amplified DNA, which was extracted from the gel with a Gene- clean kit, was ligated to pUC19 via a SmaI site and introduced into Escherichiu coli strain DH5a using a transformation technique (26). Positive clones containing the cDNA insert were confirmed by EcoRI- BumHI digestion. The cDNAs concerned were sequenced by the dideoxy T7 DNA polymerase technique using either oligonucleotide of the primer pair employed for PCR as the primer (27). Each base was determined at least three times in both orientations.

RESULTS

Purification of MSVS TI

Table I summarizes the steps for the purification.

MSVS was soluble in 5% acetic acid. The soluble MSVS

was fractionated initially by 36% saturation of ammonium

(3)

ab

0000 49.890

0 000

Elution

Time,Min

39.910

FIG. 1. Purification of MSVS TI by reverse-phase HPLC. Chroma- tography was carried out on a Waters C4 300 A column (3.9 X 300 mm) equilibrated with 0.1% TFA, and the effluent was monitored at 280 nm. (A) The TI fraction isolated from affinity chromatography on a trypsin Affi-Gel 10 column (see text for details) was applied, and the column was eluted with a linear gradient of O&60% acetonitrile at a flow rate of 1.0 ml/min for 50 min. (B) The soluble fraction of MSVS in 36% saturation of ammonium sulfate (see text for details) was subjected directly to chromatography, and the column was eluted with a linear gradient of 15-40% acetonitrile at a flow rate of 1 ml/min for 40 min. Peaks a, b, and c are denoted.

sulfate precipitation at pH 2.0. Around 62% of total TI

activity remained in the solution. Further purification by

affinity chromatography on a trypsin Affi-Gel 10 column

preequilibrated with 0.4

M

NaCl-0.1

M

Tris at pH 7.9

gave a TI fraction which was eluted from the column with

0.1

M

glycine at pH 3.0 (not shown). Only 14% of the

total TI activity was recovered from the column. Subjec-

tion of the TI fraction to HPLC on a reverse phase C,

300A column gave six major peaks (Fig. 1A). TI activity

appeared only in peak a. Direct resolution of the proteins

in the ammonium sulfate solution described above by

HPLC greatly improved the yield of TI, which appeared

at peak c shown in Fig. 1B. Around 52% of the total TI

activity was recovered with this procedure. However, re-

peated HPLC was necessary to obtain pure TI. It should

be mentioned that the purity of TI was increased 147-

fold after HPLC. The specific activity of TI thus purified

was 184 U/mg of protein. Estimation based on the activity

indicated that TI comprised 0.7% of the soluble MSVS

proteins.

Preliminary Characterization of MS VS TI

MSVS was resolved by SDS-PAGE

into seven major

protein bands (Fig. 2A, lane 2). In accordance with the

nomenclature of Chen et al. (6), these bands were labeled

I-VII

in decreasing order of size.

The absorbance, E :&, , of peak a protein (Fig. 1A) was

determined to be rather low at 4.2. Sugar analysis revealed

no carbohydrate present in the protein. A single band of

7 kDa with the isoelectric point at 8.7 was detected on

analysis of the protein by both SDS-PAGE (Fig. 2A, lane

3) and IEF in pH 3.0-10.0 (Fig. 2B, lane 2), indicating

that the protein had been purified to homogeneity and

was distinct from the main proteins (cf. lanes 2 and 3 of

Fig. 2A). The molecular weight of the protein was deter-

mined from the partition coefficient, K,, , in the gel chro-

matography (Fig. 3) agreed with that estimated from

SDS-PAGE.

The K,, determined from the position trac-

ing for TI activity in the gel chromatography of the soluble

MSVS showed no difference from that of peak a protein,

suggesting no association between peak a protein and

other protein(s) in MSVS.

A

1

2

3

FIG. 2. (A) Protein components of soluble MSVS. A suitable amount of protein was subjected to SDS-PAGE on a 15% slab gel (15 X 12 X 0.075 cm). (1) Protein standard: phosphorylase b (94 kDa), bovine serum albumin (67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20 kDa), n-lactalbumin (14 kDa), snake venom cardiotoxin (7 kDa); (2) MSVS (50 pg); (3) peak a protein of

Fig. 1 (10 pg). (B) Isoelectric focusing patterns: (1) protein markers; (2) peak a protein of Fig. 1 (5.0 pg). IEF was performed in thin layer ampholine PAGE plate (pH 3-10). Protein patterns were visualized after staining the gels with Coomassie blue. The isoelectric point of peak a protein appeared at 8.7.

(4)

2 1

\

Lysozyme 14300,0.295 Y

05

IO3 104 105

Molecular Weight

FIG. 3. Molecular weight determination of MSVS TI by gel filtration. Peak a protein of Fig. 1, soluble MSVS, and marker proteins were applied individually to a Sephadex G-50 column (1.2 X 100 cm). The column was eluted with Tris 0.1 M at pH 7.8. Trypsin inhibitor activity was measured. The partition coefficient of each protein was calculated from its elution volume (VJ, void volume ( VJ, and total bed volume (V,) of the gel in accordance with K., = (V, - V,)/( V, - VJ. The molecular weight and Kav of each marker protein are shown in the figure. The K., of peak a protein as denoted with an arrow in the figure was estimated to be 0.43, which corresponded to a molecular weight of 7 kDa. The same molecular weight was determined from the trypsin inhibitor activity tracing on the chromatogram of soluble MSVS.

Direct analysis

of amino acid composition

(Table II)

indicated

that peak a protein

contained

no tryptophan

and only a small amount of tyrosine,

which accounts

for

the low absorbance

of this protein at 280 nm. The protein

was found to have a net excess of basic residues

which

could give rise to the slightly basic isoelectric

point. The

amino acid composition

data generated directly from the

analysis

of protein

were generally compatible

with the

putative mature protein derived from MP12 cDNA

(16).

It was of interest

to note that peak b protein,

which

showed no trypsin

inhibitory

activity,

shared with peak

a protein the same amino acid composition

(not shown).

Having measured the inhibitory

activity of peak a pro-

tein on the trypsin

digestion of N-benzoyl-Phe-Val-Arg

7-amido-4-methylcoumarin

in 2-100

pM

(one case is

shown in Fig. 4A), we found that the K in Eq. [l] varied

with the substrate concentration.

This behavior indicated

that peak a protein inhibited

the trypsin

catalysis

com-

petitively

according to the explanation

of Dixon (23). The

plot based on Eq. [l] as shown in Fig. 4B gave Ki and K,,,

to be 0.15

nM

and 50

pM,

respectively.

Primary

Structure

of MSVS TI

More than one type of PI with different

molecular size

may exist in one organ of an animal. For instance,

two

types of PSTIs are present in porcine and rat pancreatic

juice (9, 28, 29). Rat PSTI-I

consists

of 61 and PSTI-II

of 56 amino acid residues (9). They are highly homologous

and show the same trypsin

inhibitory

activity

but have

different

physiological

functions.

These two inhibitors

originate

from different

genes, but their cDNAs

have

identical

5’-noncoding

regions

and nearly identical

3’-

noncoding regions (30, 31), suggesting that 5’-noncoding

regions as well as 3’-noncoding

regions of PSTI-related

cDNAs

of one organ may be highly conserved.

Having

this background

in mind, we amplified the cDNAs

of MSV

by PCR using oligonucleotides

corresponding

to 5’- and

3’-noncoding

region of MP12 as primer pair (see Exper-

imental section).

Fourteen

colonies containing

the am-

plified cDNAs

were identified

from the transformed

E. coli strain DH5a.

The 14 cDNA

inserts

contained

the

same number of nucleotides

(309 bp) and showed the same

nucleotide

sequences, which confirmed

completely

those

in MP 12 cDNA

(369 bp). Apparently,

a PI mRNA

cor-

responding

to MP12 cDNA did express in MSV. The nu-

cleotide

sequences

of the amplified

cDNA

and MP12

cDNA

and the deduced amino acid sequence are shown

in Fig. 5. Further

analyses of peak a, b, and c proteins

purified

from HPLC

(Fig. 1) by direct determination

of

a portion of the primary

structure

were undertaken.

We

found no evidence of modification

of amino acid residues

during sequence analysis

from the results

of sequence

analysis.

Results

of amino

acid determinations

were

aligned with the cDNA-derived

sequence (Fig. 5).

Automated

Edman degradation of either peak a or peak

c protein (both were biologically

active) for 30 cycles gave

reliable data. The same amino acid was identified for the

TABLE II

Amino Acid Composition

of

MSVS TI

Amino acid mol (W) mol/mol of protein” Deduced value*

Asx 8.9 5.12 Glx 7.6 4.37 Ser 4.3 2.46 Thr 4.6 2.62 GUY 10.4 5.98 Ala 9.3 5.33 Arg 6.7 3.82 Pro 6.9 3.95 Val 8.6 4.91 Met 0.3 0.18 Ile 5.2 2.99 Leu 3.7 2.11 Trp 0.3 0.16 Phe 2.2 1.28 CYS 9.2” 5.28’ LYS 7.1 4.09 His 1.7 1.01 Tyr 2.8 1.63 57.29 6 4 2 57 n Average of four determinations based on a molecular weight of 7000 Da.

* Value derived from the MP12 cDNA sequence. ’ Three aspartic acid and two asparagine. d Four glutamic acid.

(5)

Inhibitor,

nM

B

I I

-50 0 50 100

Substrate,

uM

FIG. 4. Effect of MSVS TI on trypsin kinetics. (A) Trypsin at 1.0

nM and peak a protein (Fig. 1) were incubated at 25’C for 5 min before adding the substrate of N-benzoyl-Phe-Val-Arg 7-amido-4-methylcoumarin to a final concentration of 20 FM. The relative velocity was expressed by using the velocity measured in the absence of inhibitor as 100%. (B) The plot of distance

K

vs substrate concentration according to Eq. [l].

two proteins at each cycle, showing that they are the same

protein molecule. The amino acid determinations con-

firmed both Ala as the NH,-terminal

residue and the

reading frame of MP12 cDNA. The NH,-terminal

se-

quence analysis of peak b protein for 30 cycles gave re-

liable data also. Two predominant amino acids could be

detected at each of the first 19 cycles, and a single amino

acid could be detected at each of the last eleven cycles

(Table III). The actual yields of the two sequences in an

individual cycle were such that both components appeared

to be present in nearly equal amounts. One of the two

sequences could readily be interpreted as representing the

NH,-terminal sequence of peak a protein while the other

was found to be a new sequence, which confirmed residues

20-49 of the reading frame of MP12 cDNA. These data

along with the amino acid composition data suggested

that peak b protein was produced from the digestion by

the immobilized trypsin at the peptide bond of Argl’-IlezO

of peak a protein. In summary, 49 out of 57 amino acid

residues in the putative mature protein predicted from

MP12 cDNA were confirmed by direct protein analysis.

DISCUSSION

According to the alignment of homologous protease in-

hibitors (16), MSVS TI shows a high degree of sequence

similarity with the vertebrate Kazal-type PIs. Our results

suggest that one type of PI similar to rat PSTI-II is likely

to be translationally produced in MSV. We have not found

yet another type of PI similar to rat PSTI-I in MSV. The

vertebrate Kazal-type PIs may possess similar, although

not identical, main chain conformation (16). They give

“temporary

inhibition”

to the related proteinases by a

common mechanism (15, 32). In each inhibitor molecule

there exists on the surface one peptide bond, the reactive

site, which combines with the enzyme in a substrate-like

manner and serves as a substrate for the enzyme. In the

absence of tertiary

structure for MSVS TI, which was

shown to be a competitive inhibitor, we superimposed the

primary structure of this protein onto the three-dimen-

sional structure of porcine PST1 in the complex formed

by PST1 and trypsinogen (33) and made some educated

guesses. Residues from His” to Asp22 of MSVS TI may

be the primary origin which is involved in the major polar

interaction with trypsin, and the reactive site peptide bond

is apparently at Arg’g-Ile20. The basic sequence of Arg43-

Lys44-Arg45, which plays an important role in the regu-

lation of temporary inhibition according to the study of

TABLE III

The NH,-Terminal Sequence

Analysis

of Peak b Protein of Fig. 1A

Step Amino acids NH2 terminus of MSVS TI” New sequence 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Ala, Ile LYS, Tyr Val, Asp Thr, Pro Gly, Val Lys - Glu, Gly Ala, Thr Ser, Asp - Gly His, Ile Asp, Thr Ala, Tyr Val, Ala Ala, Asn Gly, Glu Pro, Val Arg, Leu Phe Glu Asn Arg LYS Arg Ile Glu Pro Val Ala LYS Val Thr GUY LYS Glu Ala Ser His Asp Ala Val Ala GUY Pro Arg Ile Tyr ASP Pro Val GUY Thr Asp GUY Ile Ile ‘b Asp Pro Val Glr Thr Asp GUY Ile Thr Tw Ala Asn Glu Val Leu Phe Glu Asn k LYS Arg Ile Glu Pro Val ’ The NH,-terminal sequence analysis of peak a protein of Fig. 1A.

(6)

A

-40 -30 -20 -10 +l 10 20 30 40 50 GCACCCTGTATAGTTCTTCTGGCTTTTGCACCCAGATCTTCGAC~~~GGTGGCTG'rCATCTTTCTTCTCAGTGCTTTGGCCCTGCTGAGTTTAGCAG 60 70 80 90 100 110 120 130 140 150 GTAACACTTTTTCAGCTAAGGTGACTGGACTGGAAAAGAGGCTAGTTGCCA'rGATGCAGTGGCGGGATGTCCCAGAATTT~rGATC~r~r~rGTGGGACTGACGG 160 170 180 190 200 210 220 230 240 250 AATTACTTATGCCAATGAATGTGTTCTGTGCTTTGAAAACAGGAAACGC~rAGAGCCTGTCCTCATTCGAAAAGGTGGGCCT'rGC~AAGTCAAGATTT 260 270 280 290 300 310 320 GAACTCTGTTATGGCTACCGTAATGAAAGGCAACGTGGGTGGGTTCGTTGAATAAATCGCTTCATGAACACTT

B

-20 -10 +I 10 20 30 40 50

Deduced from ~KVAVIFLLSALALLSLAGNTFSAKVTGKEASCHDAVAGCPRIYDPVCGTDGITYANECVLCFENRKRIEPVLIRKGGPC MP12 cDNA

(Peak a protein) AKVTGKEAS-HDAVAG-PRIYDPV-GTDG

(New sequence deduced from Peak b protein) IYDPV-GTDGITYANE-VL-FENRKRIEPV

FIG. 5. (A) The nucleotide sequences of MP12 cDNA from Mills et al. (16). Bold characters indicate the primer pair used for PCR of this work and bars indicate the initial and stop codons of the open reading frame. (B) Comparison of amino acid sequences deduced from MP12 cDNA with those determined from Edman degradation of peak a and peak b proteins purified from HPLC shown in Fig. 1.

human PST1 (32), locates three-dimensionally

opposite

to the reactive site. The result of sequence analysis

for

peak b protein reveals that the immobilized trypsin

linked

to Affi-Gel 10 is able to digest only the bound inhibitor

at the reactive site but is inaccessible

to attack the basic

sequence. The peak b protein

is biologically

inactive,

which is contradictory

to previous

reports that one cleav-

age only at the reactive site of either porcine or bovine

PST1 did not result in their denaturation

(28, 34).

ently, post-translational

cleavage must be at the peptide

bond of Ser-Ala

in the signal peptide of MSVS TI and

at the peptide bond of Ala-Lys

in the signal peptide of

rat PSTI-II.

ACKNOWLEDGMENTS

Our results suggest that MSVS TI corresponds to MP12

cDNA. A question arises whether MSVS TI and P12 are

the same protein. The molecular size of a protein deduced

from MP12 cDNA should be smaller than that of P12

protein; note that the 80 amino-acid open reading frame

of the cDNAs should generate a primary translation

product of 9 kDa, which could decrease in size after re-

moval of the signal peptide to 7 kDa, which is the molec-

ular size of MSVS TI. Glycosylation of the 7-kDa core

protein to produce PI2 protein as suggested by Mills et

al. (16, 17) was not observed here. Perhaps glycosylation

of the P12 occurs only in the prostate, but not in the

seminal vesicle. Obviously, MSVS TI is different from

the P12 protein.

This work was partially supported by the National Science Council, Taiwan, Republic of China (Grant NSC-SO-0203-BOOl-05). Some of the work described in this paper forms part of a dissertation submitted by M.-L. Lai in partial fulfillment for the requirement of the degree of D.Sc. at National Taiwan University.

REFERENCES

1. Mann, T. (1964) The Biochemistry of Semen and of the Male Re- productive Tract, Wiley, New York.

2. Price, D., and Williams-Ashman, H. G. (1961) in Sex and Internal Secretion (Young, W. C., Ed.), 3rd ed., Vol 1, pp. 366-448, Williams & Wilkins, Baltimore.

3. Pang, S. F., Chow, P. H., and Wang, T. M. (1979) J. Reprod. Fertil. 56,129-132.

4. Peitz, B., and Olds-Clarke, P. (1986) Btil. Reprod. 35,608-617. 5. Ostrowski, M. C., Kistler, M. K., and Kistler, W. S. (1979) J. Bid.

Chem. 254,383-390.

The signal peptides of both MSVS TI and rat PSTI-II

have 23 amino acids, with sequences that are almost iden-

tical. While the NHz-terminal residue of rat PSTI-II

is

Lys, that of MSVS TI is Ala, which agrees with the sug-

gestion of Mills et al. (16) from analysis of MP12 cDNA

by a weight-matrix

method of von Heijne (35) for pre-

dicting the signal peptide sequence cleavage site. Appar-

6. Chen, Y. H., Pentecost, B. T., McLachlan, J. A., and Teng, C. T. (1987) Mol. Endocrinol. 1, 707-716.

7. Higgins, S. F., Burchell, J. M., and Mainwaring, I. P. (1976) Biochem. J. 158, 271-282.

8. Higgins, S. F., Colman, A., Fuller, F. M., and Jackson, P. J. (1981) Mol. Cell Endocrinol. 2 1, 255-262.

9. Uda, K. I., Ogawa, M., Shibata, T., Murata, A., Mori, T., Kikuchi, N., Yoshida, N. Tsunasawa, S., and Sakiyama, F. (1988) Bid. Chem. Hoppe-Seyler 369(Suppl), 55-61.

(7)

10. 11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

Fink, E., and Fritz, H. (1976) in Methods in Enzymology (Jakoby, W. B., and Wilchek, M., Eds.), Vol. 46, pp. 825-833, Academic Press, San Diego.

Fritz, H., Tschesche, H., and Fink, E. (1976) in Methods in Enzy- mology (Jakoby, W. B., and Wilchek, M., Eds.), Vol. 46, pp. 834- 847, Academic Press, San Diego.

Meloun, B., Cechovi, D., and Jonhkova, V. (1984) Hoppe-Seykr’s Z. Physiol. Chem. Rd. 364(Suppl.), 1665-1670.

Tschesche, H., Wittig, B., Decker, G., Miiller-Esterl, W., and Fritz, H. (1982) Eur. J. Biochem.

126, 99-104.

Huhtala, M. L. (1984) Hoppe-Seyler’s 2. Physiol. Chem. Bd. 365(Suppl.), 819-825.

Cechovi, D., and Jonakovi, V. (1981) in Methods in Enzymology (Lorand, L., Ed.), Vol. 80, pp. 7922803, Academic Press, San Diego. Mills, J. S., Needham, M., and Parker, M. G. (1987) EMBO J. 6,

3711-3717.

Mills, J. S., Needham, M., Thompson, T. C., and Parker, M. G. (1987) Mol. Cell. Endocrinol. 53,

111-118.

Schlgger, H., and von Jagow, G. (1987) Anal. Biochem.

166, 368-

379.

Tsugita, A., Uchida, T., Mewes, H. W., and Ataka, T. (1987) J. Biochem.

102, 1593-1597.

Spackman, D. H., Stein, W. H., and Moore, S. (1958) Anal. Chem.

30,

1190-1206.

Green, N. M., and Work, E. (1953) Biochem. J. 54, 347-352.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Lottenberg, R., Christensen, U., Jackson, C. M., and Coleman, P. L. (1981) in Methods in Enzymology (Lorand, L., Ed.), Vol. 80, pp. 341-361, Academic Press, San Diego.

Dixon, M. (1972) Biochem. J.

129,

197-202.

Greco, W. R., andHakala, M. T. (1979) J. Biol. Chem. 254,12,104- 12,109.

Maniatis, T., Fritsch, E. F., and Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Chung, C. T., Niemela, S. L., and Miller, R. H. (1989) Proc. Natl. Acad. Sci. USA 66, 2172-2175.

Tabor, S., and Richardson, C. C. (1987) Proc. Natl. Acad. Sci. USA

84,4767-4771.

Schneider, S. L., Stasiuk, L., and Laskowski, M., Sr. (1973) J. Biol. Chem. 248,

7207-7214.

Bartelt, D. C., and Greene, L. J. (1971) J. Biol. Chem. 246,

2218-

2229.

Horii, A., Tomita, N., Yokouchi, H., Doi, S., Uda, K., Ogawa, M., Mori, T., and Matsubara, K. (1989) Biochem. Biophys. Res. Commun. 162,X-159.

Hill, R. E., and Hastie, N. D. (1987) Nature 326, 96-99.

Kikuchi, N., Negata, K., Shin, M., Mitsushima, K., Teraoka, H., and Yoshida, N. (1989) J. Biochem.

106, 1059-1063.

Bolognesi, M., and Gatti, G. (1982) J. Mol. Biol.

162, 839-868.

Sealock, R. W., and Laskowski, M., Jr. (1973) Biochemistry

12,

3139-3146.