| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
kan Stenman3
1
Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, Canada M5G 1X5.
2
Department of Laboratory Medicine and Pathobiology,
University of Toronto, Toronto, Ontario, Canada M5G 1X5.
3
Department of Clinical Chemistry, Helsinki University
Central Hospital, Helsinki, Finland.
a Author for correspondence. Fax 416-586-8628; e-mail ediamandis{at}mtsinai.on.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We developed and characterized 11 monoclonal antibodies with high affinities for PSA (Kavalues from 1.1 x 108 to 1.8 x 1010L/mol), only 3 of which cross-react with human glandular kallikrein (hK2). Using these antibodies and PSA antibodies developed by others, in conjunction with time-resolved fluorometry, we developed ultrasensitive sandwich immunoassays specific for the free form of PSA.
Results: The analytical detection limit of these immunoassays is
0.001 µg/L. To our knowledge, this is the most sensitive free PSA
assay reported to date. The free PSA immunoassays exhibit <1%
cross-reactivity with PSA-
1-antichymotrypsin, show no
cross-reactivity with hK2, and correlate well with established free PSA
kits. The 11 antibodies developed by our group, in
conjunction with 4 commercially available antibodies, were used to
generate a putative epitope map of the PSA molecule.
Conclusion: The highly sensitive free PSA immunoassays may be used for measuring PSA subfractions in female serum, an application currently impossible with other reported free PSA immunoassays. © 1999 American Association for Clinical Chemistry
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PSA usually is released into the blood at low concentrations (<4
µg/L) in healthy males (6). PSA forms stable complexes
with two major extracellular serine protease
inhibitors,1-antichymotrypsin (ACT) and
2-macroglobulin (7). PSA
complexed to ACT is the predominant immunoreactive form of PSA in
serum, whereas <30% of the total PSA occurs in the noncomplexed
"free" form (8)(9). The formation of complexes
between PSA and 2-macroglobulin masks all PSA epitopes,
leaving no epitopes accessible for antibody recognition
(10); however, binding of ACT to the PSA molecule allows PSA
antigenicity to be retained because several PSA epitopes remain
unmasked (9). Thus, PSA-ACT and free PSA are two molecular
forms that may be monitored using immunological methods such as
immunoassay.
Because of the correlation of PSA with tumor volume and tissue specificity (11), its use as a tumor marker has flourished over the past decade. PSA immunoassays are widely used to detect early-stage prostate cancer, to evaluate disease progression, and to assess therapeutic response (12). Furthermore, PSA concentrations may be used to identify postsurgical residual disease or tumor recurrence (13). In addition to the total serum PSA concentration (free PSA plus PSA-ACT), the ratio of free to total PSA has become an important variable for distinguishing between males with benign and malignant prostatic disease. The percentage of free serum PSA is lower in males with prostatic carcinoma than in those with benign prostatic hyperplasia or with no apparent prostate pathology (10)(14).
Despite the extensive use of PSA as a clinical marker of prostate cancer, much is yet to be learned about the antigenic determinants on the PSA molecule (15). An understanding of the epitope distribution may yield insight into the antigenic characteristics of PSA and its interaction with anti-PSA antibodies. This knowledge may facilitate the development of antibodies that would augment the usefulness of diagnostic PSA immunoassays, e.g., antibodies that are highly specific for PSA or the various serological forms of PSA. A current concern with PSA antibodies is cross-reactivity with other members of the kallikrein family, in particular, human glandular kallikrein (hK2), which has 80% amino acid sequence identity with PSA (3)(16)(17).
The purpose of the present investigation was to produce and characterize new anti-PSA antibodies, and to examine the epitope configuration of the PSA molecule. We report here the development of 11 anti-PSA monoclonal antibodies and their characterization with respect to affinity, cross-reactivity with hK2, epitope specificity, and practicability for assay development. This characterization enabled the development of highly sensitive immunofluorometric assays that differentiate between the serological forms of PSA through the selective detection of the free form of PSA.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Screening for antibodies specific for PSA.
Hybridoma
supernatants were screened using an ELISA to select antibodies
showing specificity to PSA. Goat anti-mouse immunoglobulin
(0.5µg in 100 µL of 50 mmol/L Tris buffer, pH 7.8) was immobilized
onto polystyrene microtitration wells (Dynatech Laboratories,
Inc.). This incubation was performed overnight at room
temperature. The wells were washed six times with wash solution (150
mmol/L NaCl, 50 mmol/L Tris, 1 mmol/L NaN3, 0.5 mL/L
Tween 20), and the following series of incubations was performed at
room temperature (volumes and amounts are per well): (a) 100
µL of hybridoma supernatant diluted 10-fold in a bovine serum albumin
(BSA) diluent (50 mmol/L Tris, pH 7.8, containing 60 g/L BSA) for 30
min; (b) 100 µL of biotinylated PSA (0.1 µg) in 60 g/L
BSA for 30 min; and (c) 100 µL of alkaline
phosphatase-labeled streptavidin (0.005 µg; Jackson
Immunoresearch) in 60 g/L BSA for 15 min. Each well was washed
six times between each incubation. Diflunisal phosphate (DFP; 10 mmol/L
stock in 10 mmol/L NaOH, prepared in house), diluted 10-fold in
substrate buffer (100 mmol/L Tris, pH 9.1, 150 mmol/L NaCl, 1 mmol/L
MgCl2, 7.5 mmol/L NaN3), was added to each
reaction and incubated for 10 min. Developing solution (1 mol/L Tris,
400 mmol/L NaOH, 3 mmol/L EDTA, 2 mmol/L Tb3+) was then
added to the DFP solution, and the resulting fluorescence was measured
on a Cyberfluor 615 Immunoanalyzer (Nordion International). These
time-resolved immunofluorometric procedures are discussed in more
detail elsewhere (19)(20).
Screening for antibodies specific for free PSA or PSA-ACT.
Purified PSA-ACT was a gift from Dr. Thomas Stamey. Hybridoma
supernatants containing PSA-specific antibodies were screened further
to distinguish those specific for the free form of PSA from those that
recognize both free PSA and PSA-ACT. Microtitration wells were coated
with goat anti-mouse immunoglobulin as described above and washed, and
the following series of incubations was performed at room temperature
(volumes and amounts are per well): (a) 100 µL of
hybridoma supernatant diluted 1:10 in 60 g/L BSA for 30 min;
(b) 100 µL of free PSA (1 µg) or PSA-ACT (3 µg) in 60
g/L BSA for 1 h; (c) 100 µL of biotinylated rabbit
anti-PSA polyclonal antibody (0.25 µg; Medix Biotech Inc.), which
recognizes both captured free PSA or PSA-ACT, in assay buffer (50
mmol/L Tris, pH 7.8, 100 mL/L goat serum, 60 g/L BSA, 50 mL/L mouse
serum, 10 g/L bovine 
-globulin, 5 mL/L Tween 20, 500 mmol/L
KCl) for 1 h; and (d) alkaline phosphatase-labeled
streptavidin as described previously. DFP and developing solution were
added, and the fluorescence was measured.
Ascites production.
Ascites fluid containing the selected
antibodies was obtained from BALB/c mice that had been injected
intraperitoneally with cloned hybridoma cells. Mice were initially
injected with pristane to prepare the peritoneal cavity for tumor
growth.
Antibody purification.
Antibodies were purified from hybridoma
culture supernatant and ascites fluid by affinity chromatography using
the Affi-Gel Protein A MAPS II system (Bio-Rad Laboratories) according
to the manufacturer's specifications.
Antibody isotyping.
The immunoglobulin subclass was determined
using isotyping reagents from Sigma Immunochemicals, according to the
manufacturer's instructions.
Determination of affinity constants.
The affinity of each
antibody was determined using the BIAcore 1000 system, which is based
on surface plasmon resonance (SPR; Biacore AB). This system
consists of a light source that provides incident light, a detection
unit that measures reflected light, and a gold-coated sensor chip. SPR
occurs when energy from the incident light reacts with free electrons
on the gold film, thereby decreasing the amount of reflected light. The
SPR signal is visualized as a dip in the intensity of the reflected
light. Briefly, this technique involved immobilizing the antibody onto
the surface of the sensor chip with a rabbit anti-mouse polyclonal
antibody. PSA was injected across the surface of the sensor chip.
Antigen-antibody binding caused a change in the refractive index at the
surface of the sensor chip, thus altering the SPR signal. The BIAcore
system is discussed in more detail elsewhere (21).
Cross-reactivity of individual antibodies with hK2.
The
cross-reactivity of each PSA antibody with hK2 was assessed using
Western blotting. All necessary equipment for Western blot analysis was
obtained from Novex. Recombinant hK2 [a gift from Dr. Robert Wolfert
at Hybritech Inc., San Diego, CA; described in Ref.
(22)] was subjected to sodium dodecyl
sulfate-polyacrylamide gel electrophoresis under nonreducing
conditions on 4–12% Tris-glycine polyacrylamide gels. Separated
proteins were transferred electrophoretically to nitrocellulose
membranes. After the membranes were blocked overnight at 4 °C with
50 g/L nonfat dry milk in 20 mmol/L Tris-HCl, 137 mmol/L NaCl, 1 mL/L
Tween 20, pH 7.6, the membranes were cut into strips, and each strip
was incubated with one PSA monoclonal antibody, followed by horseradish
peroxidase-conjugated sheep anti-mouse secondary antibody (Amersham
Corp.). Probing with an antibody specific for hK2 served as a positive
control. PSA was electrophoresed in a separate well and was probed with
a PSA-specific antibody. Biotinylated molecular weight markers were
visualized by simultaneous reaction with a streptavidin-horseradish
peroxidase and the secondary antibody. The blots were incubated for 1
min with enhanced chemiluminescence reagents (ECL; Amersham Corp.), as
specified by the manufacturer, and exposed to x-ray film for detection
of immunoreactive protein bands.
antibody pairing
Pairing experiments were performed using 11 antibodies generated
in this study, in conjunction with 4 commercially available antibodies,
coded 8301 and 8311 (Diagnostic Systems Laboratories) and F1 and F2
(BiosPacific). A pairing study was performed to determine whether the
15 antibodies could be paired with a set of reference anti-PSA
antibodies included in the International Society for Oncodevelopmental
Biology and Medicine (ISOBM) workshop (see below). Two
additional pairing studies were performed to determine whether the 15
test antibodies could be paired with each other, either in the presence
of immobilized PSA or in sandwich configurations. These experiments led
to the generation of a PSA antigenic map and the development of
immunoassays specific for free PSA. The inability to pair, i.e., to
bind simultaneously to PSA, would indicate identical (no fluorescence
detected) or overlapping (low fluorescence detected) epitopes.
Pairing with ISOBM workshop antibodies.
Pairing studies were
performed to compare the epitope specificity of the 11 antibodies
generated here and 4 commercial antibodies with that of 12 reference
anti-PSA monoclonal antibodies included in the ISOBM workshop as
described elsewhere (23). The reference antibodies belong to
six major groups, and each group has a number of subgroups. One
microgram of each of the 15 antibodies was incubated with 0.005 µg of
purified PSA for 1 h. Fifty nanograms of 1 of the 14
reference antibodies, which had been labeled with europium
(Eu3+), was then applied for 1 h. This mixture was
added to microtitration wells that had been coated with a PSA
polyclonal antibody (Dako Diagnostics Canada Inc.) and incubated for 30
min. After the wells were washed, enhancement solution was added, and
the fluorescence was measured. If the test antibody shared epitope
specificity with the ISOBM workshop reference antibody, the labeled
reference antibody would be unable to bind and no signal would be
generated.
Antibody pairing by competitive binding to immobilized PSA.
Additional investigation into epitope recognition by the 11 in-house
antibodies and the 4 commercial antibodies was facilitated by
competitive binding to immobilized PSA. PSA antigen was immobilized on
microtitration wells, and the ability of any one antibody to block the
binding of another was determined. The ability to block binding
would indicate identical or overlapping epitope specificity.
Polystyrene microtitration plates were coated with free PSA (0.5 µg
in 100 µL of coating buffer per well) and incubated at room
temperature overnight. After the wells were washed, they were incubated
for 1 h with one of the 15 antibodies diluted in 60 g/L BSA (1
µg of antibody in 100 µL of BSA solution). The biotinylated form of
one of the antibodies (0.25 µg in 50 µL of 60 g/L BSA) was then
applied to each well and incubated for an additional 30 min. After the
wells were washed, alkaline phosphatase-labeled streptavidin was
applied as described above. DFP and developing solution were added, and
the fluorescence was measured.
Antibody pairing in sandwich immunoassays.
Sandwich
immunoassays were performed with all possible combinations of the 11
in-house and 4 commercial antibodies to determine which pairing
combinations enabled specific recognition of free PSA. In addition,
this analysis confirmed the findings from the previous experiment in
terms of epitope sharing or overlap. Microtitration wells were coated
with each antibody (0.2 µg in 100 µL) as described above. The wells
were washed after each of the following incubations at room
temperature: (a) 100 µL of 0.5 and 5 µg/L free PSA
diluted in a 60 g/L BSA diluent or 100 µL of 1.5 and 15 µg/L
PSA-ACT for 1 h; (b) 100 µL of biotinylated antibody
(0.05 µg in assay buffer) for 1 h; and (c) alkaline
phosphatase-labeled streptavidin as described above. DFP and developing
solution were added as described above.
free psa immunoassays
The above pairing experiments allowed us to identify two pairs of
antibodies that were suitable for the selective detection of free PSA:
antibody 9 for coating and antibody 2 for detection (designated 9/2);
and antibody F2 for coating and antibody F1 for detection (designated
F2/F1).
Cross-reactivity of free PSA immunoassays with PSA-ACT.
Positive cross-reactivity was determined by incubation of various
concentrations of PSA-ACT with the free PSA sandwich configurations.
Microtitration wells were coated with 0.2 µg of antibodies 9 or F2 as
described above. The wells were washed after each of the following
series of incubations at room temperature: (a) 100 µL of
PSA-ACT diluted in 60 g/L BSA to the following concentrations: 0, 0.15,
0.2, 2, and 10 µg/L; (b) 100 µL of biotinylated antibody
(0.25 µg in assay buffer) for 1 h; and (c) alkaline
phosphatase-labeled streptavidin as described above. DFP and developing
solution were added as described above.
Negative cross-reactivity was determined by incubating free PSA and PSA-ACT simultaneously. Microtitration wells were coated with 100 µL of each antibody (0.2 µg) as described. The wells were washed as described after each of the following series of incubations at room temperature: (a) PSA-ACT (diluted in 60 g/L BSA at concentrations of 0, 0.03, 0.15, 0.5, 2, and 10 µg/L) in the presence of 0, 0.01, or 0.05 µg/L free PSA (giving a total volume of 100 µL) for 1 h; (b) 100 µL of biotinylated antibody (0.25 µg in assay buffer) for 1 h; (c) alkaline phosphatase-labeled streptavidin as described above. DFP and developing solution were added as described above.
Cross-reactivity was calculated by dividing the concentration of PSA-ACT detected by the free PSA immunoassay by the actual PSA-ACT concentration.
Cross-reactivity of free PSA immunoassays with hK2.
The free
PSA assays developed were tested for cross-reactivity with hK2.
Microtitration wells were coated with 0.2 µg of antibodies 9 or F2 as
described above. The wells were washed after each of the following
series of incubations at room temperature: (a) 100 µL of
hK2 diluted in 60 g/L BSA at the following concentrations: 0, 0.01,
0.1, 1, 10, and 100 µg/L; (b) 100 µL of biotinylated
antibody (0.25 µg in assay buffer) for 1 h; and (c)
alkaline-phosphatase-labeled streptavidin as described above. DFP and
developing solution were added as described above.
Clinical samples.
To determine the accuracy of the free PSA
measurements obtained using the immunoassays generated in our
laboratory, we compared values with those obtained with the Immulite
Free PSA Kit (Diagnostic Products Corp.).
Fifty male serum samples were collected from the Department of Pathology and Laboratory Medicine at Mount Sinai Hospital in Toronto, Canada. Polystyrene microtitration wells were coated with 0.5 µg of antibodies 9 or F2 as described above. The wells were washed as described above after being coated and after each of the following series of incubations: (a) 20 µL of serum + 50 µL of assay buffer for 2 h; (b) 100 µL of biotinylated antibody (0.25 µg in assay buffer) for 30 min; and (c) alkaline phosphatase-labeled streptavidin as described above. DFP and developing solution were applied as described above. The concentrations of the free PSA preparations used for calibration were 0, 0.05, 0.2, 1, 5, and 20 µg/L. All calibrators and samples were analyzed in duplicate. Linear regression and Pearson correlation analyses were performed to compare PSA measurements by different assays.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antibody isotyping.
The absorbance at 450 nm was highest when
goat anti-mouse IgG1 was used as the capture antibody. Therefore, all
of the PSA monoclonal antibodies produced are of the IgG1 subclass.
Affinity constants.
The affinity constant
(Ka) and Kd
(Ka-1) for each antibody is listed
in Table 1
. Ka values ranged from 1.1 x
108 to 2.5 x 1010 L/mol. The antibody
with the highest Ka was 8301. The
Ka for antibody 8 was not measurable using
first-order kinetics.
|
Cross-reactivity of antibodies with hK2.
Antibodies 4, 5, 11,
8311, and F2 recognized hK2 electrophoresed under denaturing conditions
(Fig. 1
). PSA loaded in an adjacent well was probed with a commercial
PSA antibody (8301) as a positive control.
|
antibody pairing and epitope map
Pairing studies were performed with the 11 in-house antibodies, in
conjunction with the 4 commercial antibodies, to generate a PSA epitope
map and to develop a sandwich-type immunoassay for free PSA. The
fluorescence generated with each antibody pair is shown in Fig. 2
. Two antibodies with identical or overlapping epitopes were
unable to bind to the same PSA molecule simultaneously, thus generating
a very low fluorescent signal, as shown by the filled circles.
Antibodies specific for antigenic sites that do not overlap are able to
"sandwich" the PSA molecule and generate a high fluorescent signal,
represented by the open circles. Intermediate levels of fluorescence
are shown as gray circles.
|
The ability of each of the 15 antibodies to inhibit binding of the 12
reference ISOBM workshop antibodies was also determined. This analysis
enabled the antibodies to be sub-classed into various groups, as
described by Paus et al. (23) (Fig. 3
). Group 1 represents antibodies binding to free PSA only,
whereas groups 2–6 represents antibodies binding to free PSA and
PSA-ACT. For group assignment, antibodies should inhibit 80% of the
PSA-binding to, and should be unable to pair with, one specific ISOBM
workshop antibody. Group 7 contains antibodies that did not inhibit the
binding of any of the ISOBM workshop antibodies.
|
When the 15 test antibodies were paired with each other in all possible
combinations in the presence of immobilized PSA or in a sandwich
configuration, the above results were confirmed and the positions of
the epitopes relative to each other were determined. This analysis
allowed us to generate a putative epitope map of PSA, shown in Fig. 3
.
Antibodies from group 5 (antibodies 1, 3, 7, and 8) were unable to pair
with 8311 (group 6); therefore, these antibodies recognize a cluster of
two closely spaced domains. Antibody 8301 (group 2) could not be paired
with antibodies in groups 1 (antibodies 9 and F1), 3 (antibodies 4, 5,
11, and F2), or 7 (antibodies 6 and 10). Furthermore, antibodies in
group 7 could not be paired with those in group 1. Therefore, a second
antigenic domain appears to contain epitopes recognized by groups 1, 2,
3, and 7. Finally, the site recognized by antibody 2 appears to be
unique because this antibody could be paired with any of the in-house
or commercial antibodies.
free psa immunoassays
The pairing studies allowed us to select two combinations of
antibodies (9/2 and F2/F1) for use in free PSA immunoassays. The
detection limit of each PSA assay was determined by the analysis of 20
replicates of the zero PSA calibrator. The detection limit of the
assay, defined as the PSA concentration that corresponded to the
fluorescence of the zero calibrator plus 2 SD, was determined to be
0.001 µg/L.
Cross-reactivity of free PSA immunoassays with PSA-ACT.
Positive cross-reactivity of PSA-ACT was <0.5% in the free PSA
immunoassays generated from in-house antibodies (9/2) and 
1% for the
immunoassay generated using the F2/F1 configuration. Negative
cross-reactivity was determined by the simultaneous incubation of
PSA-ACT and free PSA. Increasing concentrations of PSA-ACT did not
interfere with the signal generated by free PSA for both free PSA
assays, at ratios of PSA-ACT to free PSA up to 300.
Cross-reactivity of free PSA immunoassays with hK2.
The
cross-reactivity of both free PSA immunoassays (9/2 and F2/F1) with hK2
was determined. No signal was generated with hK2 at concentrations up
100 µg/L, indicating that this protein is not recognized by these
sandwich configurations.
Clinical samples.
To determine the accuracy of the PSA
measurements obtained using PSA immunoassays generated with in-house
antibodies and commercial antibodies, we compared male serum free PSA
values with those obtained with the Immulite Free PSA Kit.
Calibration curves were utilized to measure free PSA in male serum in
immunoassays using the 9/2 and F2/F1 configurations. The within-run and
between-day imprecision (CV) was <10% for four male sera containing
free PSA between 0.02 and 0.2 µg/L.
Linear regression and Pearson correlation analyses were performed to
compare free PSA measurements by different assays (Fig. 4
). The free PSA values obtained with the immunoassays using the
9/2 and F2/F1 combinations correlate fairly well with those obtained by
the Immulite Kit, with r values of 0.90 and 0.95,
respectively. The free PSA values obtained using the 9/2 or F2/F1
configuration, however, appeared to be consistently higher than those
obtained using the Immulite Kit because the slopes of the linear
regression plots were 0.6 and 0.7, respectively, and the
y-intercepts were 0.13 (P = 0.002) and 0.12
(P = 0.05), respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The potential of serum PSA measurements as a screening tool for prostate cancer remains controversial because of the lack of specificity of PSA for prostate cancer. Numerous nonmalignant conditions of the prostate, such as benign prostatic hyperplasia, are associated with increased PSA (26). The ratio of free to total PSA has become an important variable for distinguishing between males with benign and malignant prostatic pathology. Shortly after the discovery of different molecular forms of PSA in serum (8)(9)(10), it was demonstrated that PSA-ACT accounted for a higher fraction of serum PSA in patients with prostate cancer than in those with benign prostatic hyperplasia (10). Although the total PSA concentration alone is neither sensitive nor specific enough for the early diagnosis of prostate cancer, the ratio of free to total PSA may improve both sensitivity and specificity (10)(14). The development of highly sensitive immunofluorometric assays that differentiate between the different serological forms of PSA and those that selectively detect the free form of PSA would, therefore, have clinical relevance. Furthermore, ultrasensitive PSA analysis may have utility in breast cancer detection or diagnosis. It is now widely accepted that PSA is present in hormone-regulated female tissues such as the breast, and is detectable in female sera (27). Recent studies have shown that the molecular forms of PSA in serum may differ between women with and without breast cancer (28)(29), thus introducing an additional use for ultrasensitive immunoassays that discriminate between free PSA and PSA-ACT. Such studies could not be conducted with any commercially available free PSA assays because of insufficient sensitivity.
In the present study, 11 PSA monoclonal antibodies belonging to the IgG1 subclass were produced. The specificity of the antibodies for PSA was confirmed by immunohistochemical staining of healthy prostatic tissue. The affinity constants (Ka) of the antibodies produced ranged from 1.1 x 108 to 1.8 x 1010 L/mol. The affinity of antibody 8 could not be measured using first-order kinetics. It is likely that the binding of the first PSA molecule to the antibody inhibited the binding of a second molecule, leading to a two-step binding curve. The affinity could likely be predicted using a more complex model.
Pairing studies were performed to compare the epitope specificities of
the antibodies produced with that of each other, with commercially
available antibodies, and binding inhibition with ISOBM workshop
antibodies. Three different antigenic clusters on the PSA
molecule were distinguishable. One antigenic domain is formed by the
epitope for which only antibody 2 is specific. The antigenic
determinant of this antibody is unique, enabling it to be paired with
any other PSA antibody. This epitope is specific for free PSA and is
masked by the binding of PSA-ACT. The other two antigenic clusters are
composed of a number of epitopes, as shown in Fig. 3
. Antibodies F1 and
9, which belong to group 1 as described by Paus et al. (23),
preferentially bind free PSA. Two of the antibodies, 6 and 10, did not
inhibit the binding of any of the ISOBM antibodies. This was probably
because of their relatively low affinity, approximately
10-8 mol/L.
Pairing studies also led to the development of two novel ultrasensitive free PSA immunoassays that exhibit <1% cross-reactivity with PSA-ACT. Commercially available antibodies are used in one assay, whereas the other consists of antibodies generated in this study. Both immunoassays have a detection limit of 0.001 µg/L, which represents the lowest concentration of free PSA ever measured with an immunoassay. The functional sensitivity is usually two- to threefold higher, as described previously (19).
We here present evidence that among the antibodies produced, those of groups 3 and 6 all appear to react with an epitope that is also present on the hK2 molecule. Importantly, it is also clear that a number of PSA monoclonal antibodies are very specific for PSA, with no cross-reactivity from hK2. PSA immunoassays designed to use antibodies such as these as the coating antibody should be free of hK2 interference.
The free PSA immunoassays were compared with the corresponding assay used for routine clinical PSA analysis, the Immulite Free PSA Kit. The two free PSA immunoassays correlated well with the Immulite Free PSA assay, but with somewhat lower slopes and significantly different y-intercepts. This is likely because of underestimation of free PSA by the Immulite assays in comparison with the in-house immunoassays, which have higher analytical sensitivity (0.05 vs 0.001 µg/L). The cross-reactivity with hK2 is not responsible for the discordance because neither of the free PSA assays produced detect this kallikrein.
In summary, we have produced 11 PSA-specific monoclonal antibodies that were used, in conjunction with 4 commercial PSA antibodies, to characterize the antigenic domains of the PSA molecule and to develop highly sensitive immunoassays for free PSA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1-antichymotrypsin; hK2, human glandular kallikrein; BSA, bovine serum albumin; DFP, diflunisal phosphate; SPR, surface plasmon resonance; and ISOBM, International Society for Oncodevelopmental Biology and Medicine. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-seminoprotein. FEBS Lett 1987;225:168-172.
[Web of Science][Medline]
[Order article via Infotrieve]
1-antichymotrypsin. Urol Clin N Am 1993;20:681-686.
[Web of Science][Medline]
[Order article via Infotrieve]
1-antichymotrypsin. Clin Chem 1991;37:1618-1625.
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 1991;51:222-226.
1-antichymotrypsin as an indicator of prostate cancer. J Urol 1993;150:100-105.
[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. Leinonen, P. Wu, and U.-H. Stenman Epitope Mapping of Antibodies against Prostate-specific Antigen with Use of Peptide Libraries Clin. Chem., December 1, 2002; 48(12): 2208 - 2216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Grzeda, T. Le Bui, C. N. Warner, T. L. Pirucki, L. M. Dewey, M. Babich, and J. A. Maggiore Measurement of Prostate-specific Antigen by Use of a Novel Blood Collection and Analytical System Clin. Chem., August 1, 2002; 48(8): 1272 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Soukka, J. Paukkunen, H. Harma, S. Lonnberg, H. Lindroos, and T. Lovgren Supersensitive Time-resolved Immunofluorometric Assay of Free Prostate-specific Antigen with Nanoparticle Label Technology Clin. Chem., July 1, 2001; 47(7): 1269 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Rafferty, P. Rigsby, M. Rose, T. Stamey, and R. Gaines Das Reference Reagents for Prostate-specific Antigen (PSA): Establishment of the First International Standards for Free PSA and PSA (90:10) Clin. Chem., September 1, 2000; 46(9): 1310 - 1317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Black, M. Giai, R. Ponzone, P. Sismondi, H. Yu, and E. P. Diamandis Serum Total and Free Prostate-specific Antigen for Breast Cancer Diagnosis in Women Clin. Cancer Res., February 1, 2000; 6(2): 467 - 473. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), high fluorescence, >10 000 arbitrary units;
(
), moderate
fluorescence, 5000–10 000 arbitrary units; (
), low fluorescence,
<5000 arbitrary units).
