[CANCER RESEARCH 61, 957–962, February 1, 2001]
Identification of Precursor Forms of Free Prostate-specific Antigen in Serum of
Prostate Cancer Patients by Immunosorption and Mass Spectrometry
Jochen Peter,1 Carlo Unverzagt,2 Thomas N. Krogh, Ole Vorm, and Wolfgang Hoesel3
Institut für Organische Chemie und Biochemie, Technische Universität München, 85748 Garching, Germany [J. P., C. U.]; PROTANA A/S, DK-5230 Odense M, Denmark
[T. N. K., O. V.]; and Roche Diagnostics GmbH, 82372 Penzberg, Germany [W. H.]
ABSTRACT
The structural features of the free prostate-specific antigen (F-PSA)
present in human blood have not been clarified up to now, and it is,
therefore, not known why F-PSA is not complexed by the protease inhibitors that are present in human blood in large amounts. This lack of
information is mainly attributable to the low amount of F-PSA in serum,
which makes the isolation and structural characterization very difficult,
especially when only limited amounts of individual sera are available. It
has now been demonstrated that F-PSA occurs as a mixture of different
pro-PSA forms (zymogen forms) in the sera of prostate cancer patients,
and that, in some of these sera, a form with the regular NH2 terminus of
PSA is present as well. Among the five serum samples investigated, all
contained the (ⴚ7), (ⴚ5), and (ⴚ4) pro-PSA forms, whereas the (ⴚ1) and
(ⴚ2) forms were only present in three of them. These three samples also
contained the form with the regular NH2 terminus. The (ⴚ3) and (ⴚ6)
pro-PSA forms have not been detected thus far. The F-PSA has been
isolated by immunosorption from the individual sera using streptavidincoated magnetic beads. The pro-PSA forms were identified by matrixassisted laser desorption ionization time-of-flight mass spectrometry after
producing peptides by endoproteinase from Lysobacter enzymogenes digestion of the SDS-PAGE-separated F-PSA band. The structural identity
of the (ⴚ7)pro-PSA form was further proven by sequencing of that
particular peptide using electrospray ionization quadrupole time-of-flight
mass spectrometry.
INTRODUCTION
PSA41 is, because of its sensitivity and organ specificity, a very
valuable protein marker in human blood for population screening,
diagnosis, and monitoring of patients with PCa (1–5). There are,
however, some limitations on the use of PSA in PCa diagnosis.
Patients with BPH also show PSA concentrations up to about 15
ng/ml in serum. Therefore, PSA concentrations below 15 ng/ml cannot be used to distinguish between PCa and BPH (3). The determination of the ratio of not complexed F-PSA:T-PSA increases the
specificity in the range of 4 –15 ng/ml T-PSA to some extent, but there
is still much need for improvement (3, 6, 7). This could be accomplished by looking for structural differences between the PSA from
PCa and BPH patients. If differences were to be found, they could be
used to develop better diagnostic means for the differentiation between these two conditions. An interesting target for that purpose
poses the structural analysis of F-PSA from serum, although the low
Received 8/9/00; accepted 11/20/00.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
1
Present address: National Institute of Environmental Health Sciences (NIH/NIEHS),
Building 101, Room F011, Research Triangle Park, NC 27709.
2
Present address: Lehrstuhl für Bioorganische Chemie, Universität Bayreuth, Gebäude
NW 1, 95440 Bayreuth, Germany.
3
To whom requests for reprints should be addressed, at Roche Diagnostics GmbH,
Nonnenwaldstrasse 2, 82372 Penzberg, Germany. Phone: 49-8856-603274; Fax: 49-8856603341; E-mail: [email protected].
4
The abbreviations used are: PSA, prostate-specific antigen; F-PSA, free PSA; T-PSA,
total PSA; PCa, prostate cancer; BPH, benign prostatic hyperplasia; MALDI, matrixassisted laser desorption ionization; TOF, time-of-flight; MS, mass spectrometry; Q-TOF
MS, quadrupole TOF MS; endo Lys-C, endoproteinase from Lysobacter enzymogenes;
ACT, antichymotrypsin.
concentration of this analyte in human serum (8) turns this approach
into a challenging task.
The reason for the presence of F-PSA in human blood is not yet
clear, but this F-PSA should be a proteolytically inactive form, otherwise it would be complexed by protease inhibitors (e.g., ␣1-ACT
and ␣2-macroglobulin), which are present in the blood in large
amounts (9 –11). Two different explanations have been proposed thus
far for this finding: (a) the F-PSA could exist as an inactive form
resulting from a nick in the PSA sequence, probably at the lysine at
position 145; and (b) alternatively pro-PSA forms (zymogen forms)
could be present that still carry all or parts of the signal sequence (12,
13). Experimental evidence for the presence of a nicked form was
presented by Noldus et al. (14), who isolated partially purified F-PSA
from a 230-ml pool of 59 PCa sera (T-PSA, ⬎2000 ng/ml). Gas phase
sequencing provided evidence for the presence of a form with the
regular NH2 terminus and a nicked form. There was no indication for
the presence of pro-PSA forms, but the F-PSA yield of their preparation was only about 25%, and it still contained impurities. Thus, the
authors might have missed some minor PSA forms that could also
have been present. On the other hand, Mikolajczyk et al. (15) found
evidence for the presence of PSA forms larger than the F-PSA from
seminal fluid when they separated the PSA from 75 ml of pooled PCa
sera (containing 50 –100 ng/ml T-PSA) by a combination of immunosorption and hydrophobic interaction liquid chromatography. The
HIC high-performance liquid chromatography showed that about 25%
of the F-PSA eluted at a similar position as the (⫺4) pro-PSA form
from cultured prostate cells, but there was no further characterization.
The authors, using Western blotting techniques after SDS-PAGE
separation, did not find clear evidence for the presence of nicked
F-PSA.
Different pro-PSA forms have been produced and clearly identified
(mostly by NH2-terminal sequencing) in several cell culture systems
after the cloning of the prepro-PSA gene and isolation of the expressed PSA forms. In a Syrian hamster tumor cell line, pro-PSA
forms with seven and five amino acids in the precursor part were
identified by Kumar et al. (16). Lövgren et al. (17) identified the same
forms after cloning and expression of prepro-PSA in BHK cells as
well as in a baculovirus system. In the latter system, a form with three
amino acids corresponding to the precursor part was identified as well.
Additionally a (⫺7)pro-PSA form was expressed and characterized in
Escherichia coli cells (18).
Here we describe the unequivocal identification of different proPSA forms in the serum from an individual PCa patient. The F-PSA
was isolated by immunosorption, and the intact molecule and its
peptides were analyzed by MALDI-TOF MS. Moreover, the relative
quantities of the pro-PSA forms and the F-PSA form with the regular
NH2 terminus were investigated in four other PCa serum samples.
MATERIALS AND METHODS
Serum A (50 ml) from a patient suffering from PCa (T-PSA, 1890 ng/ml;
F-PSA, 180 ng/ml) and reference serum were obtained from Bioclinical
Partners Inc, Franklin, MA. The other four PCa sera (S1, S5, S6 and S9, 2 ml
each) were from the sera collection of Roche Diagnostics GmbH (Penzberg,
Germany). These sera contained between 6900 and 8500 ng/ml of T-PSA and
957
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
PRECURSOR FORMS OF F-PSA IN PCa SERUM
between 640 and 960 ng/ml of F-PSA. PSA from seminal fluid was a product
of Scripps Laboratories (San Diego, CA). The monoclonal antibodies used
were from the protein chemistry department of Roche Laboratory Diagnostics.
Streptavidin-coated magnetic beads of 2.5-␮m diameter were the same as those
used in the Roche Laboratory Diagnostics automatic immunanalyzer ELECSYS.
All of the other products used were either from Roche Molecular Biochemicals or from Merck GmbH (Darmstadt, Germany) if not indicated otherwise.
Among these products were PBS, 50 mM potassium phosphate, and 150 mM
sodium chloride (pH 7.4).
Isolation of F-PSA by Immunosorption
The immunosorption methods used have recently been described in detail
(19). The following protocol was used here: a suspension (2.5 ml) of streptavidin-coated magnetic beads (10.7 mg/ml) was placed in a 10-ml tube and
washed with PBS. After the addition of 4 ml of biotinylated anti-F-PSA-M30IgG (monoclonal antibody from mouse, used in the Roche Diagnostics ELECSYS test for F-PSA; 25 ␮g/ml in PBS containing 1% BSA and 0.1% Tween
20), the suspension was incubated for 30 min. The beads were collected,
washed three times with PBS containing 20 mM N-octylglucoside and incubated with 8 ml of the PCa serum A (containing 180 ng/ml F-PSA) for 1 h to
bind the F-PSA to the beads. The beads were collected and washed as
described before, followed by a very short washing step with 200 ␮l of water
to remove most of the detergent. The beads were then incubated under shaking
with 500 ␮l of 1 M propionic acid for 1 h. After magnetic separation of the
beads, the supernatant was removed, lyophilized in a vacuum concentrator, and
stored at ⫺20°C if further analysis was not performed immediately after
lyophilization. In case of the other four PCa sera (F-PSA ⬎ 600 ng/ml) a
Fig. 2. MALDI-TOF MS analysis of the peptide pattern obtained after endo Lys-C in
gel digestion of F-PSA from the serum of a PCa patient isolated by immunosorption (A)
and of F-PSA from seminal fluid (B). The numbers above the respective PSA peptide
peaks, the molecular mass of the peptides; the numbers in brackets, the positions of the
amino acids in the intact PSA molecule; Mox, oxidized methionine.
volume of 100 ␮l of the sera was used for immunosorption, and the amounts
of reagents were adjusted accordingly. For these sera, biotinylated IgG from
the monoclonal antibody M36, which recognizes free as well as complexed
PSA and is used in the Roche Diagnostics ELECSYS test for T-PSA was used
for the immunosorption, because we wanted to isolate the PSA/ACT complex
as well as the F-PSA by separation on SDS-PAGE (see Fig. 4).
SDS-PAGE and endo Lys-C Digestion
SDS-PAGE was performed under nonreducing conditions using the MiniPROTEAN II gel electrophoresis system and precast 12% ready gels from
Bio-Rad, using essentially the protocol as described by Laemmli (20). The gels
were silver stained and the F-PSA band was digested with endo Lys-C as
described by Shevchenko et al. (21).
MS
Fig. 1. Determination of the average molecular mass of F-PSA from PCa serum (B) and
seminal fluid (A) by MALDI-TOF MS.
Determination of the Molecular Mass of F-PSA. The samples were
analyzed in a Voyager Biospectrometry Workstation MALDI mass spectrometer equipped with delayed extraction, operating in the positive mode of
detection. The spectrometer contains a nitrogen laser operating at 337 nm. TOF
spectra were produced at 25 kV acceleration voltage by averaging 80 single
spectra. A matrix consisting of a solution of ferulic acid (4-hydroxy-3-methoxy
958
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
PRECURSOR FORMS OF F-PSA IN PCa SERUM
Table 1 Molecular mass of the peptides obtained after endo Lys-C digestion of F-PSA from PCa serum and seminal fluid compared with the theoretically expected values
Lys-C peptides from F-PSA,
isolated from PCa serum
([M ⫹ H]⫹ ions)
Lys-C peptides from F-PSA,
isolated from seminal
plasma ([M ⫹ H]⫹ ions)
Theoretical values of
Lys-C-derived
peptides ([M ⫹ H]⫹ ions)
Positions of the amino
acid residues of the
peptidesa
801.4
N.O.b
1320.6
1383.7
1546.8
1659.9
1827.9
1939.1
2588.2
3395.9
3524.6
4137.9
4231.8
801.4
1077.5
N.O.
1383.7
N.O.
N.O.
N.O.
1939.1
2588.3
3395.7
3524.7
4137.9
4231.8
801.47
1077.50
1320.49
1383.64
1546.81
1659.88
1827.97
1939.09
2588.31
3395.74
3524.66
4137.93
4232.18
222–227
1–9
(⫺2)–9
171–182
(⫺4)–9
(⫺5)–9
(⫺7)–9
222–237
146–167
84–113
114–145
183–221
47–83
The position of the peptides in the amino acid sequence of pro-PSA is indicated in the last column.
Negative refers to pro-PSA peptides.
b
N.O., not observed.
a
cinnamic acid; 10 mg/ml) in formic acid:water:acetonitrile (1:3:2, v/v/v) was
used for all determinations. PSA from semen was used as a reference solution
at a concentration of 2 pmol of protein/␮l of distilled water. The eluates from
the immunosorption procedures were dissolved in 10 ␮l of distilled water. An
aliquot (0.5 ␮l) of this protein solution was mixed with 1 ␮l of the matrix
solution on the target plate and allowed to dry at room temperature, prior to
insertion into the mass spectrometer. All of the spectra were calibrated externally using the singly charged ion of BSA ([M ⫹ H], 66,431 Da) and the
doubly charged ion of horse skeletal apomyoglobin ([M ⫹ 2H]2⫹, 16,953 Da)
as references.
Analysis of F-PSA Peptides. The endo Lys-C in-gel digests of the purified
F-PSA were analyzed by MALDI-MS on either a Bruker Reflex III or a
PerSeptive Biosystems STR mass spectrometer both equipped with delayed
extraction and reflectors. All of the spectra were acquired in positive ion mode
using a reflector mode for higher resolution. Samples were prepared on top of
a fast evaporating thin-layer preparation of 4-hydroxy-␣-cyano-cinnamic acid
(25 g/liter) and nitrocellulose (2 g/liter) in acetone. Sample (0.75 ␮l) was
mixed with 0.75 ␮l of 2% trifluoroacetic acid, allowed to dry, and washed with
10 ␮l of ultra-high-quality water. Two hundred single-shot spectra were
averaged and externally calibrated using Angiotensin I.
Sequencing of pro-PSA peptides was carried out using nanoelectrospray
ionization Q-TOF MS, carried out on a SCIEX Qstar. The sample were
desalted on a Poros R2,20 column packed in a GEloader tip (Eppendorf) and
eluted into a nanoelectrospray capillary (Protana) by 50% methanol, 5% formic
acid. Spectra were acquired in positive ion mode.
RESULTS
Isolation of F-PSA from PCa Serum. The F-PSA from 8 ml of
PCa serum A (containing 1890 ng/ml of T-PSA and 180 ng/ml of
F-PSA) was purified by immunosorption on streptavidin-coated magnetic beads according to published methods (19) using a biotinylated
antibody that was specific for F-PSA only (monoclonal antibody M30,
which is being used in the Roche Diagnostics ELECSYS test for
F-PSA). After the immunosorption, the amount of F-PSA in the
PCa-serum was measured with an immunoassay for F-PSA and
showed no detectable level, which indicated a quantitative isolation of
the F-PSA. After elution of the bound F-PSA with 1 M propionic acid,
the molecular mass of the F-PSA was determined by MALDI-TOF
MS,5 and we observed an ion with an average value of 29,180 Da for
the intact PSA molecule (Fig. 1B). This molecular mass is about 600
Da above the value for F-PSA from seminal fluid (28,580 Da). This
reference spectrum is shown in Fig. 1A.
Recently, we reported that PSA can be liberated from the PSA/ACT
5
The direct immunosorption method was used in this case because it gives higher yield
of a reasonably pure F-PSA compared with the indirect method using digoxigeninLys.HCL for elution, which provides higher purity but lower yield (see Ref. 19).
complex by a nucleophilic cleavage reaction (22). Thus, for PCa
serum A, the PSA liberated from the complex was found to have a
molecular mass of 28,560 Da. This mass is similar to that of the
F-PSA from seminal fluid. The molecular mass difference of about
600 Da indicates that the F-PSA of this PCa serum is somewhat larger
than PSA isolated from seminal fluid or from the PSA/ACT complex.
This finding led us to speculate that pro-PSA forms were present in
the PCa serum.
Analysis of F-PSA Peptides. F-PSA (180 ng) isolated from PCa
serum A separated by SDS-PAGE. The silver-stained band of F-PSA
was cut out, alkylated under reducing conditions, and digested by
endo Lys-C as described under Methods. Fig. 2 shows the peptide
patterns obtained by MALDI-TOF MS from this band (A) and a PSA
band from seminal fluid (B) which had been treated similarly. Table
1 contains a list of molecular masses for all of the PSA peptides observed
in these two preparations and the expected theoretical values.
It is obvious that the peptides containing the NH2-terminal part of
PSA are different in these two samples. The expected regular NH2terminal peptide of PSA (amino acids 1- 9) was only present in the
PSA from seminal fluid and could not be found in the F-PSA from this
PCa serum. From the latter sample, several peptides could be assigned
to forms carrying an elongation at the NH2 terminus. The data
obtained are in accordance with the theoretical values expected from
the pro-PSA forms with 7, 5, 3, and 2 additional amino acids at the
regular NH2 terminus (see Table 1). All of the other peptides originating from PSA in the two samples were identical. Moreover, the
peptide patterns that were obtained from PSA released from the
PSA/ACT complex (22) and PSA from seminal fluid were identical to
each other and showed only the NH2-terminal peptide amino acids
1–9. In these cases, pro-PSA forms were absent.
The sequence coverage obtained by endo Lys-C digestion was
calculated to be 83%. One minor peptide (amino acids 168 –170; Mr
346.2) and the peptide carrying the predicted N-glycosylation site
were missing (amino acids 10 – 46). We assume that the N-glycopeptide was not observed because of signal suppression. Because all of
the mass spectrometric experiments were carried out in reflector mode
and the glycans were not derivatized, significant metastable losses of
the glycans may have occurred, and these altered fragments are
normally not detected in reflector mode MALDI-MS.
Sequencing of the Pro-PSA Peptide (Amino Acids ⴚ7 to ⴙ9) by
Electrospray Ionization Q-TOF MS. To unequivocally identify the
pro-PSA forms the longest pro-PSA peptide (amino acids ⫺7 to ⫹9;
Mr 1827.97) was sequenced by Q-TOF MS. The resulting spectrum is
displayed in Fig. 3, and the observed masses were assigned to the
resulting sequences as shown in Table 2.
959
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
PRECURSOR FORMS OF F-PSA IN PCa SERUM
PCa serum A (in each case calculated as percentage of the total
amount of F-PSA present in that sample).
It was found that the (⫺4), (⫺5) and (⫺7) pro-PSA forms were
present in all of the four sera investigated. In contrast to PCa serum A
and serum S9, the regular NH2 terminus as well as the (⫺1) pro-PSA
form were detected in the other three sera. The values listed in Table
3 should be regarded with caution, considering that they were obtained in different experiments and that quantification of MALDITOF MS is generally problematic. However, it can be assumed that
the F-PSA form with a regular NH2 terminus represents a major part
of the F-PSA of the sera S1 and S 5. The (⫺3) and (⫺6) pro-PSA
forms could not be detected in any of the samples investigated thus far.
DISCUSSION
The structural properties of F-PSA present in human blood have not
yet been unequivocally determined. It seems obvious that F-PSA
should exist in a form that does not display protease activity, because
otherwise it would be complexed by the protease inhibitors present in
Fig. 3. Sequencing of the (⫺7 to ⫹9) pro-PSA peptide by Q-TOF MS analysis
([M⫹H]⫹, 1827.97 Da; see also Fig. 2A). Ions marked with numbers, the charge status of
these ions; ions marked with letters, the nomenclature for fragment ions as described by
Roepstorff and Fohlmann (25).
Table 2 m/z values of the signals observed in the Q-TOF MS sequencing analysis of
the (⫺7) pro-PSA peptide and the amino acid sequences corresponding to these values
Ion signal
(m/z)
Ion
(charge)
Measured
mass
Calculated
mass
141.109
169.102
276.152
436.190
283.174
565.241
751.312
808.353
865.357
964.421
1077.547
617.331
660.832
717.371
773.912
830.454
586.321
878.980
914.501
610.003
a2
b2
y2
y3
y4 (2⫹)
y4
y5
y6
y7
y8
y9
y10 (2⫹)
y11 (2⫹)
y12 (2⫹)
y13 (2⫹)
y14 (2⫹)
y15 (3⫹)
y15 (2⫹)
y16 (2⫹)
Parent ion
141.11
169.10
276.16
436.19
565.35
565.24
751.31
808.35
865.36
964.42
1077.55
1233.66
1320.66
1433.74
1546.82
1659.91
1756.96
1756.96
1828.00
1828.01
141.10
169.10
276.16
436.19
565.24
565.23
751.31
808.33
865.35
964.42
1077.50
1233.61
1320.64
1433.73
1546.81
1659.90
1756.95
1756.95
1827.99
1828.00
a
Amino acid sequencea
P
AP
EK
CEK
ECEK
ECEK
WECEK
GWECEK
GGWECEK
VGGWECEK
XVGGWECEK
RXVGGWECEK
SRXVGGWECEK
XSRXVGGWECEK
XXSRXVGGWECEK
XXXSRXVGGWECEK
PXXXSRXVGGWECEK
PXXXSRXVGGWECEK
APXXXSRXVGGWECEK
X, leucine/isoleucine.
It can be seen that each of the observed peaks is in perfect agreement with the Y-ion sequence resulting from the seven amino acids of
the precursor part and the nine amino acids of the regular NH2
terminus of PSA.
Analysis of Additional PCa Sera for the Presence of Pro-PSA
Forms. Four other sera (S1, S5, S6, and S9) of PCa patients with a
high content of T-PSA (⬎6000 ng/ml) were further examined for the
presence of pro-PSA peptides. The analysis followed the same protocol as described above for the analysis of F-PSA peptides, and Fig.
4 shows a representative silver-stained SDS-PAGE gel of the PSA
fractions isolated from these four sera. The biotinylated monoclonal
antibody M36, which recognizes T-PSA, and the immunosorption
method with 1 M propionic acid elution, as described above, were used
in these studies. The bands representing the PSA/ACT complex and
F-PSA could clearly be identified on the gel, although some other
contaminating protein bands were still present. The F-PSA bands were
cut out and treated as described above, digested with endo Lys-C and
the resulting peptides analyzed by MALDI-TOF MS. The result of the
analysis is shown in Table 3, where the relative signal heights of the
NH2-terminal PSA peptides obtained from the F-PSA of these four
sera are compared with the peptide signals obtained from F-PSA of
Fig. 4. Silver-stained SDS-PAGE gel of the PSA fractions isolated from four PCa sera
by immunosorption and elution with 1 M propionic acid. Lane 1, rabbit muscle phosphorylase b, 97.4 kDa; BSA, 66.2 kDa; hen egg white ovalbumin, 45 kDa; bovine carbonic
anhydrase, 31 kDa; soybean trypsin inhibitor, 21.5 kDa, hen egg white lysozyme, 14.4
kDa. Lane 2, PSA purified from seminal fluid; arrow, the position of PSA. Lanes 3– 6,
PSA purified from PCa sera S1, S5, S6, S9; arrows, the position of F-PSA and of the
PSA-ACT complex. That the position of F-PSA from seminal fluid compared with that
from the sera is slightly different resulted very likely from uneven running conditions for
the lanes.
Table 3 Relative signal intensities of the precursor (NH2-terminal) peptides of F-PSA
obtained from the sera of five individual PCa patients, observed by
MALDI-TOF MS
The signal intensities are given as relative percentage of the total F-PSA present in that
particular serum sample (the sum of all NH2-terminal peptide signals obtained for each
serum represents 100% F-PSA).
PCa sera
Peptide calculated
mass (Da)
1 to 9
⫺1 to
⫺2 to
⫺4 to
⫺5 to
⫺7 to
(1077.50)
9 (1233.61)
9 (1320.49)
9 (1546.81)
9 (1659.88)
9 (1827.97)
S1
S5
S6
S9
Serum A
34
11
32
5
12
6
30
10
34
7
12
7
16
7
20
14
30
13
0
0
0
28
44
28
0
0
16
29
35
20
960
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
PRECURSOR FORMS OF F-PSA IN PCa SERUM
abundant amounts in the blood (mainly by ␣1-ACT and ␣2macroglobulin). Whether this inactivity is related to internal nicking of the PSA or the presence of pro-PSA forms has largely been
a matter of speculation (12, 13). Some evidence for the presence of
a nicked form came from data by Noldus et al. (14), but this was
not corroborated by Mikolajczik et al. (15), who did not find
substantial amounts of a nicked form. Instead their HIC highperformance liquid chromatography data provided evidence that
about 25% of the F-PSA seemed to be present as pro-PSA forms
when compared with the elution of a pro-PSA reference obtained
from cell cultures. A structural analysis of that peak was not
reported. The contradictory results may be related to the use of
pooled PCa sera. Moreover only low amounts of partially purified
F-PSA were available, which prevented a thorough structural analysis (14). In a recent paper by Hilz et al. (23), it was reported that
pro-PSA forms could not be detected in the F-PSA preparation that
was isolated by immunosorption from blood aspirated during transvesical prostatectomy of BPH patients. This was concluded from
the sequencing of the F-PSA by Edman degradation on polyvinylidene difluoride membranes. Furthermore, it was stated in that
paper that no pro-PSA forms could be detected in PCa and BPH
sera when probing with an antibody that had been raised against
the propeptide moiety of the zymogen.
The development of efficient immunopurification procedures using
streptavidin-coated magnetic beads (19) enabled us to isolate sufficient amounts of F-PSA from individual PCa sera for an analysis
using state-of-the-art MS. Thus, we could unequivocally demonstrate
the presence of pro-PSA forms in the serum of a PCa patient and
identify their structures. Moreover, by looking at other PCa sera, we
found that F-PSA with a regular NH2 terminus can also be present.
This could explain the results obtained by Noldus et al. (14) as well
as those of Mikolajczik et al. (15), because F-PSA, with a regular NH2
terminus, as well as pro-PSA forms can be present in PCa sera and
were, therefore, very likely present in the pooled PCa sera used by
those authors. In contrast to Mikolajczik et al. (15), the presence of
pro-PSA forms with seven and five amino acids in the pro-PSA part
could be detected in all of the PCa sera investigated by us. However,
the form with the regular NH2 terminus was only present in some of
the sera. Thus far, we have not investigated the lack of a complex
produced by protease inhibitors of this F-PSA form carrying the
regular NH2 terminus. It is tempting to speculate about the occurrence
of a nicked form, but this needs to be examined in additional experiments, using an approach different from the one described above
(e.g., using SDS-PAGE separation under reducing conditions in combination with suitable specific detection methods for the analysis of
F-PSA forms, which has not been done by us thus far).
The results presented above demonstrate that F-PSA in sera from
PCa patients occurs as a combination of different pro-PSA forms, and
that a form with the regular NH2 terminus can be present as well in
some sera. Among the individual sera, the presence of, and the relative
amounts of, the different PSA forms seem to vary. The origins of
these differences are not known. The five sera used in this study
contained very high amounts of PSA. In PCa and BPH sera, in which
there are low amounts of PSA (e.g., in the diagnostically interesting
range of 2–15 ng/ml), the relative amounts of the various F-PSA
forms could differ significantly from samples with a high PSA content
or might not be present at all. For this reason comparisons to the
results obtained by Mikolajczik et al. (15) and Hilz et al. (23), who
were using sera with a much lower PSA concentration, have to be
made with much caution because they may or may not be relevant.
However, to be able to analyze sera with low PSA content for the
presence of pro-PSA forms, we have prepared a monoclonal antibody
that binds to the three pro-PSA forms (⫺7,⫺6,⫺5), and used it to
develop an immunoassay that specifically recognizes these precursor
forms. A preliminary analysis of panels of PCa and BPH sera of high
and low PSA content showed that the pro-PSA forms could be
detected almost in all of the sera analyzed.6 Whether this test can be
used to better differentiate between BPH and PCa in the diagnostically
gray area of 2–15 ng/ml T-PSA compared with the state-of-the-art
methods (e.g., by using the ratio of F-PSA:T-PSA) will be evaluated.
In this context, a very recent report from Mikolajczik et al. (24),
dealing with the identification of pro-PSA forms in PCa and benign
transition zone prostate tissue, is also shedding light on the issue of
F-PSA isoforms. As reported earlier for sera (15), they identified the
(⫺4) and (⫺2) forms of pro-PSA in tissue extracts after immunopurification by NH2-terminal sequencing. In contrast to our report, they
again did not find pro-PSA forms with the longer precursor sequences
(e.g., ⫺7 or ⫺5), and the reason for this is not clear to us at this time.
Interestingly enough, the pro-PSA forms were present more in PCa
tissues than in BPH. Whether this reflects a similar situation in the
sera has to be elucidated in the future using a panel of specific
antibodies for the different pro-PSA forms.
REFERENCES
1. Stamey, T. A., Yang, N., Hay, A. R., McNeal, J. E., Freiha, F. S., and Redwine, E. A.
Prostatic-specific antigen as a serum marker for adenocarcinoma of the prostate.
N. Engl. J. Med., 317: 909 –916, 1987.
2. Chu, T. M. Prostate-specific antigen and early detection of prostate cancer. Tumor
Biol., 18: 123–134, 1997.
3. Sokoll, L. J., and Chan, D. W. Total, free and complexed PSA: analysis and clinical
utility. J. Clin. Ligand Assay, 21: 171–179, 1998.
4. Diamandis, E. P. Prostate-specific antigen—its usefulness in clinical medicine.
Trends Endocrinol. Metab., 9: 310 –316, 1998.
5. Polascik, T. J., and Oesterling, J. E. Prostate-specific antigen: a decade of discovery—
what we have learned and where we are going. J. Urol., 162: 293–306, 1999.
6. Becker, C., and Lilja, H. Individual prostate-specific antigen (PSA) forms as prostate
tumor markers. Clin. Chim. Acta, 257: 117–132, 1997.
7. Moore, R. A. Free PSA as a percentage of the total: where do we go from here? Clin.
Chem., 43: 1561–1562, 1997.
8. Wang, T. J., Slawin, K. M., Rittenhouse, H. G., Millar, L. S., and Mikolajczyk, S. D.
Benign prostatic hyperplasia-associated prostate-specific antigen (BPSA) shows
unique immunoreactivity with anti-PSA monoclonal antibodies. Eur. J. Biochem.,
267: 4040 – 4045, 2000.
9. Stenman, U. H., Leinonen, J., Alfthan, H., Ranniko, S., Tuhkanen, K., and Alfthan,
O. A complex between prostate-specific antigen and ␣1-antichymotrypsin is the major
form of prostate-specific antigen in serum of patients with prostate cancer: assay of
the complex improves clinical sensitivity for cancer. Cancer Res., 51: 222–226, 1991.
10. Lilja, H., Christensson, A., Dahlen, U., Matikainen, M. T., Nilsson, O., Pettersson, K.,
and Lövgren, T. Prostate-specific antigen in human serum occurs predominantly in
complex with ␣1-antichymotrypsin. Clin. Chem., 37: 1618 –1627, 1991.
11. Zhang, W. M., Finne, P., Leinonen, J., Vesalainen, S., Nordling, S., Ranniko, S., and
Stenman, U. H. Characterization and immunological determination of the complex
between prostate-specific antigen and ␣2-macroglobulin. Clin. Chem., 44: 2471–
2479, 1998.
12. Piironen, T., Petterson, K., Suonpaa, M., Stenman, U. H., Oesterling, J. E., Lövgren,
T., and Lilja, H. In vitro stability of free prostate-specific antigen (PSA) and
prostate.-specific antigen (PSA) complexed to ␣1-antichymotrypsin in blood samples.
Urology, 48 (Suppl. 6A): 81– 87, 1996.
13. Pettersson, K., Piironen, T., Seppälä, M., Luikkonen, l., Christensson, A., Matikainen,
M. T., Suonpaa, M., Lövgren, T., and Lilja, H. Free and complexed prostate-specific
antigen (PSA): in vitro stability, epitope map, and development of immunofluorometric assays for specific and sensitive detection of free PSA and PSA-␣1-antichymotrypsin complex. Clin. Chem., 41: 1480 –1488, 1995.
14. Noldus, J., Chen, Z., and Stamey, T. A. Isolation and characterization of free form
prostate specific antigen (f-PSA) in sera of men with prostate cancer. J. Urol., 158:
1606 –1609, 1997.
15. Mikolajczik, S. D., Grauer, L. S., Millar, L. S., Hill, T. M., Kumar, A., Rittenhouse,
H. G., Wolfert, R. L., and Saedi, M. S. A. A precursor form of PSA (pPSA) is a
component of the free PSA in prostate cancer serum. Urology, 50: 710 –714, 1997.
16. Kumar, A., Mikolajczik, S. D., Goel, A. S., Millar, L. S., and Saedi, M. S. Expression
of pro form of prostate-specific antigen by mammalian cells and its conversion to
mature, active form by human kallikrein 2. Cancer Res., 57: 3111–3114, 1997.
6
J. Peter, C. Hübner-Parajsz, P. Stieber, and W. Hoesel. Development and clinical
evaluation of an immunoassay for the measurement of pro-PSA forms in BPH and PCa
sera. Manuscript in preparation.
961
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
PRECURSOR FORMS OF F-PSA IN PCa SERUM
17. Lövgren, J., Rajakoski, K., Karp, M., Lundwall, A., and Lilja, H. Activation of the
zymogen form of prostate-specific antigen by human glandular kallikrein 2. Biochem.
Biophys. Res. Comm., 238: 549 –555, 1997.
18. Takayama, T. K., Fujikawa, K., and Davie, E. W. Characterization of the precursor
of prostate-specific antigen. J. Biol. Chem., 272: 21582–21588, 1997.
19. Peter, J., Unverzagt, C., Lenz, H., and Hoesel, W. Purification of prostate-specific
antigen from human serum by indirect immunosorption and elution with a hapten.
Anal. Biochem., 273: 98 –104, 1999.
20. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of the
bacteriophage T4. Nature (Lond.), 227: 680 – 685, 1970.
21. Shevchenko, A., Wilm, M., Vorm, O., and Mann, M. Mass spectrometric sequencing
of proteins of silver-stained polyacrylamide gels. Anal. Chem., 68: 850 – 858, 1996.
22. Peter, J., Unverzagt, C., and Hoesel, W. Analysis of free prostate-specific antigen
(PSA) after chemical release from the complex with ␣1-antichymotrypsin (PSAACT). Clin. Chem., 46: 474 – 482, 2000.
23. Hilz, H., Noldus, J., Hammerer, P., Buck, F., Luck, M., and Huland, H. Molecular
heterogeneity of free PSA in sera of patients with benign and malignant prostate
tumors. Eur. Urol., 36: 286 –292, 1999.
24. Mikolajczyk, S. D., Millar, L. S., Wang, T. J., Rittenhouse, H. G., Marks, L. S., Song,
W., Wheeler, T. M., and Slawin, K. M. A precursor form of prostate-specific antigen
is more highly elevated in prostate cancer compared with benign transition zone
prostate tissue. Cancer Res., 60: 756 –759, 2000.
25. Roepstorff, P., and Fohlman, J. Proposal for a common nomenclature for sequence
ions in mass spectra of peptides. Biomed. Mass Spectrom., 11: 601, 1984.
962
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.
Identification of Precursor Forms of Free Prostate-specific
Antigen in Serum of Prostate Cancer Patients by
Immunosorption and Mass Spectrometry
Jochen Peter, Carlo Unverzagt, Thomas N. Krogh, et al.
Cancer Res 2001;61:957-962.
Updated version
Cited articles
Citing articles
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/61/3/957
This article cites 25 articles, 9 of which you can access for free at:
http://cancerres.aacrjournals.org/content/61/3/957.full.html#ref-list-1
This article has been cited by 11 HighWire-hosted articles. Access the articles at:
/content/61/3/957.full.html#related-urls
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 2001 American Association for Cancer
Research.