| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Article |
Departments of
1
Medical Microbiology and
2
Molecular Biology, University of Odense, Winsløwparken 19, DK-5000 Odense C, Denmark.
a Author for correspondence. Fax 45 65 915267; e-mail b.teisner{at}imbmed.ou.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 chains in trimeric and
monomeric forms; (c) thermal transition is an ongoing in
vivo process; and (d) this is important in the choice of
assay technology. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PINP was first isolated from amniotic fluid under the name fetal
antigen 2 (6), and amino acid sequencing identified the
high-molecular weight form as a homomer of the 1 chains of PINP
(7). Antibodies against the 1 chain of PINP have formed
the basis for immunoassays for quantification of PINP, i.e.,
electroimmunoassay and ELISA
(3)(6)(8), as well as RIA techniques
(9)(10).
These antibodies recognize two molecular forms of PINP, which appear as
distinct peaks after size chromatography (3)(6).
These molecular forms of PINP are readily analyzed by
electroimmunoassay and sandwich ELISA techniques
(3)(6)(9)(11), using
antibodies specific for the 1 chain of PINP. Moreover, the two
molecular forms of PINP interact with the reaction of immunological
identity, and both forms migrate at an apparent molecular mass of 27
kDa when analyzed in sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) (3)(11). However,
when the same antibodies were used in RIA with 125I-PINP as
tracer, only the high-molecular weight form of PINP could be detected
(9)(11)(12). A commercial RIA for
quantification of PINP became available recently, and this assay is
also based on 1-chain-specific antibodies. Like the RIA described
previously by Price and co-workers (9)(12), the
commercial RIA fails to detect the low-molecular weight form of PINP
(10)(11).
The anti-PINP antibodies used for ELISA development (3), the antibodies applied in the RIA described by Price et al. (9), and the antibodies from the commercial RIA detect both molecular forms of PINP (10)(11). Thus, the difference in the ability of these assays to detect the low-molecular weight form of PINP is most likely caused by different functional affinities of the antibodies for the high- and low-molecular weight forms, which is revealed by the diminished reactivity in the competitive (antibody-limiting) RIA system but not in the ELISA (excess reagent) (11).
The trimeric structure of PINP is achieved via the noncovalent forces within the collagenous part of the chains. The helical form of the collagens is a consequence of many reinforcing bonds, each of which is relatively weak, and it is well known that the thermal stability of soluble collagens is low: e.g., soluble calf skin collagen has a melting temperature of 39 °C (13). The collagen region of PINP is short compared with skin collagen; hence, the thermal stability of PINP should be lower.
The aim of the present study was to analyze the thermal stability and structures of the two molecular forms of PINP and to relate those observations to the choice of assay technology.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
immunoassays
Electroimmunoassay.
Rocket immunoelectrophoresis was performed
in 1.5 mm thick, 1% agarose (Litex HSA; Litex Industries) in 0.02
mol/L Tris-barbital buffer, pH 8.6, containing monospecific rabbit
anti-PINP prepared as described previously (6).
ELISA.
The sandwich ELISA technique using immunospecifically
purified rabbit anti-PINP as the capture antibody and the same antibody
conjugated with biotin as indicator has been described in detail
previously (3).
RIA.
The procollagen, intact 125I-PINP RIA kit was
purchased from Orion Diagnostica and used in accordance with the
manufacturer's recommendations. This RIA is identical to that
described by Melkko et al. (10) in 1996.
pinp purification
PINP was purified from amniotic fluid on a CNBr-activated
Sepharose (Pharmacia) matrix to which immunospecifically purified
rabbit anti-PINP antibodies were coupled. This matrix (3.5 mL) was
packed in a XK16 column (Pharmacia). The column was connected to a Fast
Performance Liquid Chromatography system (Pharmacia) and equilibrated
with phosphate-buffered saline, pH 7.3, containing extra 1 mol/L NaCl.
The capacity of the column was 80 mL of amniotic fluid (25 mg PINP/L).
The bound PINP was eluted with 5 g/L citric acid, pH 2.8, and the
PINP-containing fractions were identified by rocket
immunoelectrophoresis.
analysis of thermal stability
Samples of purified PINP, second trimester amniotic fluid, cord
serum, and human serum were exposed to 37 °C for up to 72 h in
a water-jacketed incubator (Forma Scientific). The control samples were
kept at 4 °C for the same period. After incubation, the samples were
subjected to size chromatography, and the distribution of PINP was
analyzed by rocket immunoelectrophoresis, ELISA, and RIA.
size chromatography
Aliquots (200 µL) of second trimester amniotic fluid, human
serum, cord serum, or purified PINP were applied to a Superose 12 HR
10/30 column connected to a Fast Performance Liquid Chromatography
system. The chromatography was performed with phosphate-buffered
saline, pH 7.3, at a flow rate of 0.5 mL/min, and 0.5-mL fractions were
collected.
sds-page
SDS-PAGE analysis of fractions from the size chromatography was
performed under reducing conditions on 4–20% gradient gels with
discontinuous buffers. Protein bands were detected by Coomassie
staining.
mass spectrometry
Matrix-assisted laser desorption/ionization mass spectrometric
analysis of samples (5–20 pmol) was performed on a PerSeptive
Biosystems Voyager STR instrument (PerSeptive Biosystems). Protein
samples were desalted using Poros R1 column material. The desalted
protein was eluted in 800 mL/L acetonitrile containing 2.0 g/L
trifluoroacetic acid, and a small fraction was mixed with an equal
volume of sinapinic acid (20 g/L in 700 mL/L acetonitrile).
Mass spectra were acquired in the linear mode, using an acceleration voltage of 20 kV and a delay time of 300 ns. Single-shot (n = 128) spectra were added, and the final spectrum was calibrated externally.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
shows the absorbance profiles at 280 nm and the results of
analytical rocket immunoelectrophoresis of PINP on the fractions. The
apparent difference in the absorbance profile and the profile of the
rocket heights is because after peaks were detected by the
ultraviolet detector, they traveled through more tubing to reach the
fraction collector; the volume difference between the detector and the
fraction collector was 0.5 mL, or the volume of one fraction. From the
results shown in Fig. 1
, fractions 22 and 27 were chosen to represent
the high- and low-molecular weight forms of PINP, respectively.
Fraction 22 was subjected to rechromatography before and after
incubation at 37 °C for 72 h, and as seen in Fig. 2
, incubation at 37 °C caused a pronounced change from the
high- to the low-molecular weight form of PINP. This change was
revealed by both the change in the absorbance profile at 280 nm and the
result of rocket immunoelectrophoretic analysis of the PINP. Size
chromatography of the high-molecular weight form of PINP after 0, 18,
and 24 h at 37 °C revealed a half-life of ~10 h.
|
|
In contrast, incubation of fraction 27 (the low-molecular weight form)
at 37 °C for 72 h did not produce marked changes in either the
absorbance profile at 280 nm or the distribution of PINP as analyzed by
rocket immunoelectrophoresis (Fig. 3
).
|
assay technology and thermal instability
The concentration of PINP in amniotic fluid, cord serum, and
nondiseased human serum was measured by ELISA and RIA before and after
incubation for 24, 48, and 72 h at 37 °C (Table 1
). Whereas only minor fluctuations were observed in the ELISA
result, the apparent PINP concentration as measured by RIA decreased to
6.4% (amniotic fluid), 11% (cord serum), and 6.6% (nondiseased human
serum) of the values obtained for control samples over a 72-h
incubation period at 37 °C. The other obvious difference between the
two assays is reflected in that the ratio
[PINP]ELISA:[PINP]RIA differed
markedly in different biological fluids: the ratio was 2.0 for amniotic
fluid, 2.9 for cord serum, and 1.1 for nondiseased human serum. To find
an explanation for these differences, we analyzed the size distribution
of PINP in amniotic fluid, cord serum, and nondiseased human serum and
compared the ELISA and RIA results for fractions (size chromatography)
before and after treatment at 37 °C for 72 h.
|
The size distribution of PINP as analyzed by ELISA in amniotic fluid,
cord serum, and nondiseased human serum without 37 °C treatment is
shown in Fig. 4
. As seen from Fig. 4
, the ratios between the high- and
low-molecular weight forms differed between the biological fluids, and
calculation revealed the ratios to be 1.64 for amniotic fluid, 0.67 for
cord serum, and 3.1 for nondiseased human serum.
|
The RIA measures mainly the high-molecular weight form of PINP (9)(10)(11); therefore, the differences in size distribution between the biological fluids seem to explain the differences in the concentration ratio measured using the ELISA and the RIA in amniotic fluid, cord serum, and nondiseased human serum. Moreover, the calibrator supplied by the commercial RIA contains only the high-molecular weight form, whereas the calibrator used in the ELISA contains both forms of PINP (11).
The ELISA and RIA results for fractions from size chromatography of
nondiseased human serum with and without treatment at 37 °C are
shown in Fig. 5
. Before treatment, the RIA detected eluted PINP antigen
specific for the high-molecular weight form of PINP, whereas the ELISA
detected both forms. After incubation at 37 °C for 72 h, the
RIA failed to detect PINP in the fractions (detection threshold, 5
µg/L), whereas the results of the ELISA revealed a dramatic loss of
the high-molecular weight form and an increase in the low-molecular
weight antigen. Similar results were obtained with amniotic fluid and
cord serum as antigen sources.
|
analysis by sds-page and mass spectrometry
Fraction 22 (Fig. 1
) was incubated at 37 °C for 72 h
(control at 4 °C), and analytical size chromatography revealed a
shift in distribution identical to that shown in Fig. 2
. The
high-molecular weight form of PINP with or without 37 °C treatment
for 72 h was analyzed by SDS-PAGE, and the result is shown in Fig. 6
. Both the treated and untreated samples migrate in single bands
corresponding to a molecular mass of 27 kDa, and this result indicates
that the 1 chain remains intact in spite of the dramatic shift in
the size chromatography profiles observed in samples after treatment at
37 °C.
|
Mass spectrometric analysis of fractions 22 and 27 (Fig. 7
) was performed to verify the presence of intact PINP 1
chains in both fractions. Mixing of the protein with matrix solution,
necessary for mass spectrometry, causes dissociation of noncovalent
multimers into monomers. The resulting mass spectrum of fraction 22 is
shown in Fig. 7A
. The PINP monomeric form of the 1 chain was
identified by this analysis as several peaks appearing with a mass
difference of 16 Da. The difference between the calculated molecular
mass of the 1 chain (14 213.62 Da) and the measured value
corresponds to the presence of six to nine hydroxyprolines in the
collagenous C-terminal part of the intact 1 chain. Exactly the same
peak pattern was seen by mass analysis of fraction 27 (Fig. 7B
),
indicating intact 1 chain in both fractions.
|
Peptide mapping of the proteins present in the two fractions, using
endoproteinase Lys-C digestions (data not shown), further confirmed the
presence of intact 1 chain in both fractions.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1
chain, appears in two distinct molecular forms when subjected to size
chromatography (3)(6)(11). These
forms interact with the reaction of immunological identity, and after
SDS-PAGE both forms migrate in single bands corresponding to an
apparent molecular mass of 27 kDa (11). These observations
suggest that the two molecular forms of PINP represent different
(monomeric or multimeric) structures of the intact 1 chains.
The ratio between the two molecular forms varies in biological fluids,
as demonstrated here in amniotic fluid, cord serum, and pooled
nondiseased human serum. We have also observed major differences in the
ratio of the two molecular forms in sera obtained from apparently
healthy individuals, patients treated with continuous peritoneal
dialysis, and the dialysis fluid. The amount of PINP recovered in the
24-h dialysis fluid exceeded the circulating amount by >100%,
suggesting local synthesis of PINP in the peritoneal cavity induced by
an inflammatory reaction to the applied treatment (11). The
hypothesis tested in the present study is that the PINP is released
from the procollagen molecule as a trimer represented by the
high-molecular weight form and that thermal transition at 37 °C
produces monomeric 1 chains.
The incubation of the high-molecular weight form of PINP purified from amniotic fluid at 37 °C for 72 h followed by size chromatography produced a pronounced shift from the high- to the low-molecular weight form. The half-life of the high-molecular weight form at 37 °C was calculated to be ~10 h. In contrast, the low-molecular weight form remained unchanged after incubation at 37 °C for 72 h. Could the thermal instability of the high-molecular weight form be caused by the use of pH 2.8 citric acid buffer in the purification procedure and the lack of protecting proteins in the purified state?
This question was addressed by analysis of the thermal stability of native PINP in serum and amniotic fluid. Treatment of amniotic fluid and serum at 37 °C for 72 h produced a change of PINP antigen similar to that observed when purified PINP was used. These data indicate that PINP is unstable at body temperature and that a continuous shift in PINP size distribution takes place in body fluids. This finding underlines the importance of knowing the specificity of the assays used to assess collagen type I metabolism.
It was demonstrated previously that antibodies recognizing both
molecular forms of PINP fail to detect the low-molecular weight form of
PINP when applied in an RIA that uses 125I-labeled
PINP as the tracer (9)(10)(11). In contrast, the sandwich ELISA
technique has been demonstrated to detect the 1 chain independent of
molecular form (3)(11).
The present study also demonstrates that quantification of PINP in
amniotic fluid, cord serum, and nondiseased human serum, using the
ELISA technique, revealed no marked change in the results during
incubation of samples at 37 °C for up to 72 h. In contrast,
in the RIA, the apparent PINP concentration decreased up to 93% after
incubation at 37 °C. The decrease in the apparent concentration in
RIA during the first 24 h (Table 1
) is in good agreement with the
estimated half-life of the purified high-molecular weight form of PINP,
i.e., ~10 h. Moreover, the ratio between the results of PINP
quantification obtained using ELISA and RIA differed markedly between
these three biological fluids, with the ratio being 2.9 in cord serum
and 1.1 in nondiseased human serum. These differences seem to be
explained by the difference in size distribution of the PINP antigen
between the biological fluids and by the fact that RIA fails to detect
the low-molecular weight form of PINP. This was further substantiated
by the results obtained by ELISA and RIA on fractions (from size
chromatography) of nondiseased human serum, where 37 °C treatment
led to the disappearence of the signal in the RIA, whereas the ELISA
signal simply moved from the high- to the low-molecular weight
fractions.
It has been suggested that the low-molecular weight form of PINP is identical to the Col 1 fragment that has been produced by digestion of procollagen by bacterial collagenase (10). However, this artificial fragment is unrelated to the two molecular weight forms of PINP because Col 1 migrates in a band corresponding to 14 kDa in contrast to the 27-kDa band observed for both molecular forms of native PINP (10)(11). The RIA detects only the high-molecular weight form of PINP, which in accordance with our observations is eluted in a volume corresponding to a molecular mass of ~100 kDa from size chromatography and migrates in a band corresponding to 27 kDa in SDS-PAGE (10)(11). Thus we find no evidence for the hypothesis that the low-molecular weight form of PINP is related to the Col 1 fragment (11).
Analysis of the two molecular forms by mass spectrometry showed
identical spectra and a mass in accordance with the theoretical mass of
the intact 1 chain of PINP when the proline hydroxylations are taken
into account. These data clearly show that both molecular forms of PINP
contain intact 1 chains and further substantiate our hypothesis that
the high-molecular weight form represents the trimeric structure of
PINP and the low-molecular weight form represents the monomeric form of
intact 1 chain. The apparent difference in molecular mass estimated
by SDS-PAGE (27 kDa) and mass spectrometry (~14 kDa) is attributable
to aberrant migration in SDS-PAGE caused by the collagenous stretch in
PINP compared with the globular molecules used as molecular markers.
The observations described in the present study using thermal
transition, SDS-PAGE, and mass spectrometry strongly support our
hypothesis that the high-molecular weight form of PINP represents the
unstable trimeric structure and that the low-molecular weight form
represents the monomeric 1 chain released from the trimeric
structure at body temperature.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 chain of procollagen type I. APMIS 1992;100:1106-1114.
[Web of Science][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  P. Garnero, P. Vergnaud, and N. Hoyle Evaluation of a Fully Automated Serum Assay for Total N-Terminal Propeptide of Type I Collagen in Postmenopausal Osteoporosis Clin. Chem., January 1, 2008; 54(1): 188 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Johansen, K. Brasso, P. Iversen, B. Teisner, P. Garnero, P. A. Price, and I. J. Christensen Changes of Biochemical Markers of Bone Turnover and YKL-40 Following Hormonal Treatment for Metastatic Prostate Cancer Are Related to Survival Clin. Cancer Res., June 1, 2007; 13(11): 3244 - 3249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. KJAeR Role of Extracellular Matrix in Adaptation of Tendon and Skeletal Muscle to Mechanical Loading Physiol Rev, April 1, 2004; 84(2): 649 - 698. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
