|
|
||||||||
Articles |
a Author for correspondence. Fax 44 0171-600-1439; e-mail R.K.Iles{at}mds
| Abstract |
|---|
|
|
|---|
Methods: We measured immunoreactive LH, hCG, free
-subunit,
and free ß-subunit hCG (hCGß), as well as ß-core, using
the S504 RIA and Triton UGP enzyme immunoassay in 274 urine samples
from women with nonmalignant gynecological conditions. The molar
cross-reaction of each assay with purified ß-LH-core was determined.
Results: Cross-reaction with ß-LH-core was 100% in the LH and
the S504 ß-core assay, 5% in the Triton UGP assay, and <0.1% in
the hCG, free
-subunit, and free hCGß assays. Median urine
concentrations of all analytes showed an age-dependent increase. LH and
free
-subunit concentrations were ~103 pmol/mol
creatinine; hCG and S504 ß-core were ~102 pmol/mol
creatinine; free hCGß and Triton UGP ß-core were in the tens
of pmol/mol creatinine. The S504 ß-core concentrations were 10% of
those of LH. S504 ß-core was strongly correlated with LH, but not
with hCG or with free hCGß (LH, r2 = 0.45;
hCG, r2 = 0.26; free hCGß,
r2 = 0.03). The concentrations of
ß-core detected by the Triton UGP assay, which has a 5%
cross-reaction with ß-LH-core, were 2% of LH and 5% of the S504
ß-core concentrations. Triton UGP values correlated strongly with LH
concentrations, but less well with S504 ß-core, intact hCG, and free
hCGß (LH, r2 = 0.44; S504 ß-core,
r2 = 0.33; hCG,
r2 = 0.32; free hCGß,
r2 = 0.19).
Conclusions: Immunoreactive ß-core in women free of malignancies reflects cross-reaction with concentrations of the metabolite of LH, ß-LH-core, within the health-related reference interval.© 1999 American Association for Clinical Chemistry
| Introduction |
|---|
|
|
|---|
-subunit of the glycoprotein hormone (1)(2)(3)(4).
However, in vivo, this expression currently is detected by measurement
of urinary ß-core fragment, termed hCGßcf by the IFCC, but also
known as ß-core or urinary gonadotropin peptide (UGP). The
rationale for this approach was that many more cancer patients were
found to have increased urinary concentrations of hCGßcf than had
increased, or even detectable, concentrations of hCG or hCGß in their
serum. The physiological reasons for this have now been clarified. Free
hCGß has a much shorter half-life [metabolic clearance rate
(MCR), 19.0
mL · min-1 · m-2]
than intact hCG (MCR, 1.9
mL · min-1 · m-2)
Furthermore, aberrant glycosylation of the molecule could reduce its
circulating blood half-life to a matter of seconds (MCR, 495
mL · min-1 · m-2)
(5)(6)(7). hCGßcf/ß-core is a major immunoreactive degradation product of hCG and its free ß-subunit and is often the only form of hCG present in the urine of cancer patients (8)(9). The process by which hCGßcf/ß-core arises starts in the circulation with cleavage or "nicking" of amide bonds between residues 47 and 48 and, less frequently, between residues 42 and 43 of hCGß, either as part of the intact heterodimer or as a free subunit (10)(11). This nicking occurs at the center of a major hCG/luteinizing hormone (LH) receptor binding loop (12)(13). These nicked molecules are then taken up by the kidneys and further degraded to hCGßcf/ß-core, which is excreted into the urine (14). Although missing 53 amino acids of the COOH terminus, 5 amino acids of the NH2 terminus, and residues 4154, hCGßcf/ß-core retains many antigenic epitopes of its parent molecule because it retains a core, or central knot, of three sets of disulfide bond pairs, which are responsible for much of the three-dimensional shape of hCGß (8).
Increased urinary hCGßcf/ß-core was proposed originally as a
general marker of gynecological cancer (vaginal, cervical, ovarian, and
endometrial) (15)(16)(17). However, immunoreactive
"ß-core" is detectable in urine of women with no malignant
disease. Indeed, there is a general increase with age, especially after
menopause (18). Thus, cutoff values between normal and
abnormal must be adjusted according to menopausal status. Subsequent
studies demonstrated that much of this immunoreactive ß-core could be
attributed to cross-reaction with a metabolite of LH, which we termed
ß-LH-core and which was almost certainly produced by the same
degradation pathways that led to hCGßcf (19).
Nevertheless, using a highly specific assay, Alfthan et al.
(20) suggested that there was still an underlying increase
in hCGßcf with age. Recently, ß-LH-core was purified by Birken et
al. (21) at Columbia University, New York. We have assessed
the molar cross-reactivity of ß-LH-core in our in-house ß-core,
hCG, LH, free
-subunit, and free hCGß assays along with a recently
developed ß-core assay, Triton UGP, which includes a pretreatment LH
"scavenger" antibody step to remove cross-reacting LH metabolites.
We used these assays to measure immunoreactive ß-core in 254 women with nonmalignant conditions to determine whether the detected hCGßcf could be accounted for by cross-reactivity with ß-LH-core.
| Materials and Methods |
|---|
|
|
|---|
hormone assays
Intact hCG was measured with an in-house IRMA with a polyclonal
anti-
-subunit capture antibody conjugated to
l,l'-carbonyldiimidazole (CDI)-activated cellulose and a
125I-radiolabeled monoclonal antibody 1/07
(Quantum Bioscience) to epitopes on the hCGß C-terminal peptide. The
calibration curve ranged from 2.6 to 813 pmol/L.
LH was measured using an LH IRMA produced by North East Thames Immunoassay, St. Bartholomew's Hospital. This uses a polyclonal anti-LH antiserum for capture onto a CDI-activated cellulose solid phase and the same antibody, affinity purified and 125I-radiolabeled, for detection. The calibration curve ranged from 34 to 1333 pmol/L.
Free
-subunit was measured by use of an in-house RIA with an
antiserum (S781) raised against the free
-subunit and affinity
absorbed on a column of intact hCG conjugated to CNBr-activated
Sepharose. The tracer was 125I-radiolabeled
purified free
-subunit (NIH preparation CR123
-subunit). The
calibration curve ranged from 62.5 to 6250 pmol/L.
Free hCGß was measured using an in-house IRMA. A polyclonal antiserum
(S752) raised in sheep against free hCGß was conjugated to
CDI-activated cellulose. Free hCGß captured onto this solid phase was
detected by 125I-radiolabeled monoclonal antibody
1/07, which recognizes epitopes on the hCGß C-terminal peptide. A
calibration curve was constructed by plotting bound
-activity
values against increasing concentration of hCGß (NIH preparation
CR123). The cross-reactivity of this assay is detailed the
Results, the calibration curve ranged from 22 to 1110
pmol/L.
The in-house ß-core RIA used a polyclonal antibody to hCGßcf (S504) in a late addition competition assay using 125I-radiolabeled purified hCGßcf (18). This assay is known to cross-react with ß-LH-core (19). The calibration curve ranged from 9 to 500 pmol/L.
The commercial Triton ELISA uses a scavenger anti-LH antibody in the initial diluent to bind LH-like material. The assay used an immobilized monoclonal antibody against hCGßcf and a peroxidase-conjugated polyclonal antibody to hCGß (and thus epitopes retained by hCGßcf). The calibration curve ranged from 2 to 16 pmol/L.
cross-reactivity studies
Hormones and subunits were NIH preparations (CR123) of intact hCG,
free hCGß, and
-subunit of hCG; WHO International Reference
Preparations of LH (LH-68/40), follicle-stimulating hormone (FSH;
FSH-83/575), and thyrotropin (TSH; TSH-80/558; National Institute for
Biological Calibrations and Control, Potters Bar, Herts, UK). Purified
hCGßcf was donated by Drs. R. Wehmann and D. Blithe (NIH, Bethesda,
MD). Recently purified ß-LH-core was kindly donated by Dr. S. Birken
(Presbyterian Medical Center, Irving Center for Clinical Research,
Columbia University, New York, NY). Cross-reactivity was
determined on a molar basis using molecular weights calculated from the
established primary structures: intact hCG,
Mr 36 700; free hCGß,
Mr 22 200; free
-subunit,
Mr 14 500
(22)(23); hCGßcf and ß-LH-core,
Mr 10 000
(21)(24); LH,
Mr 29 000; FSH,
Mr 33 000; and TSH,
Mr 30 000 (19).
Cross-reactivities were calculated from the molar equivalents that produce 50% B/Bo displacement for the RIA and at concentrations in the IRMA and enzyme immunoassay at which calibrator and cross-reactant curves paralleled.
statistical analysis
Matched assay results for the samples were examined for
correlations using the Spearman rank test (adjusted for ties).
Calculations were performed using Astute (statistics add-in for
Microsoft Excel; DDU software, University of Leeds, UK).
| Results |
|---|
|
|
|---|
|
|
Age-dependent increases were seen (Fig. 2
) in the median values of apparent immunoreactive LH, hCG, free
hCGß, free
-subunit, S504 RIA ß-core, and Triton UGP ß-core.
|
LH and free
-subunit concentrations were
~103 pmol/mol creatinine, hCG and S504 ß-core
concentrations were ~102 pmol/mol creatinine,
and free hCGß and Triton UGP ß-core concentrations were ~10
pmol/mol creatinine (Table 2
).
|
A correlation analysis between ß-core concentrations and all other
hCG, LH, and free subunits analytes was conducted (Table 3
and Fig. 3
). The S504 ß-core concentrations were 10% of those of
LH, and there was a stronger correlation of S504 ß-core
concentrations with LH than with hCG or free hCGß within individual
sample (LH, r2 = 0.45; hCG,
r2 = 0.26; free hCGß,
r2 = 0.03; P = 0.60).
The concentrations of ß-core detected by the Triton UGP assay were
2% of the LH and 5% of the S504 ß-core concentrations. Triton UGP
ß-core values similarly correlated more strongly with LH and
S504 ß-core concentrations than with intact hCG and free hCGß (LH,
r2 = 0.44; S504 ß-core,
r2 = 0.33; hCG,
r2 = 0.32; free hCGß,
r2 = 0.19).
|
|
| Discussion |
|---|
|
|
|---|
Purified ß-LH-core did not cross-react in the free hCGß, free
-subunit, and intact hCG assays, but showed 100% cross-reaction in
the LH and S-504 ß-core assays. However, ß-LH-core showed 5%
cross-reaction in the Triton UGP ß-core, despite pretreatment of
samples with the LH scavenger antibody (Fig. 1
).
On a molar basis, immunoreactive LH was the most abundant form of
gonadotropic molecule in the urine of pre- and postmenopausal women.
Both LH and free
-subunit concentrations were in the
~103 pmol/mol creatinine range; S504 RIA
ß-core was in the ~102 pmol/mol creatinine
range; free hCGß and intact hCG were in the ~10 pmol/mol creatinine
range; and Triton UGP ß-core was in the ~1 pmol/mol creatinine
range. Thus, cross-reactivity with LH and LH-related ß-LH-core
is highly likely to influence the immunochemical measurement of other
related gonadotropic peptides. For example, an hCG assay with 0.1%
cross-reactivity with LH would measure apparent hCG concentrations in
the urine of women >60 years of age as 2.7 pmol/mol creatinine, simply
because of cross-reactivity.
Given the data on specificity for the various assays used in this study, all the apparent ß-core measured by the S504 RIA in these samples could be accounted for by cross-reaction with ß-LH-core. This interpretation is supported by the fact that there was a strong correlation between the S504 RIA concentrations of ß-core and LH concentrations (r2 = 0.45), but not with hCG and hCGß concentrations (r2 = 0.26 and 0.03, respectively). Thus, ß-LH-core could represent 10% of the LH immunoreactivity of the samples measured using the LH IRMA (this IRMA cross-reacts 100% with ß-LH-core). The concentrations of ß-core detected by the Triton UGP assay were 5% of those detected by the S504 ß-core RIA and 2% of the LH concentrations. Given that the UGP assay has a 5% cross-reactivity with authentic ß-LH-core, this is consistent with all of the ß-core immunoreactivity found in the samples being attributable to cross-reaction with ß-LH-core.
Alfthan et al. (20) claimed a true increase in urinary hCGßcf with age. However, there were no data on the cross-reactivity of authentic ß-LH-core in their assay system. The median concentration of immunoreactive ß-core found in urine by Alfthan et al. was 70 pmol/mol creatinine for those <50 years of age and 230 pmol/mol creatinine for those >50 years of age. This is consistent with the concentrations (molar ratios) detected by the S504 ß-core RIA. Furthermore, their studies of male urinary hCGßcf showed no dramatic increase with age. Thus, we believe it is possible that the assay developed by Alfthan et al. also cross-reacts 100% with ß-LH-core.
Free intact hCG and free hCGß concentrations also rose with age.
However, hCG and free hCGß concentrations did not correlate
significantly with hCGßcf (see Table 3
) and reactivity of LH and
ß-LH-core in these assays was <0.1%. Intact hCG and free hCGß
showed a weak correlation (r2 = 0.37),
and hCGß concentrations were approximately 20% of those of intact
hCG. The origin of these true hCG species is unknown, although hCG and
free hCGß expression has been demonstrated in the nonmalignant
pituitary (28)(29) and in testicular
(30), prostatic (31), and urothelial epidermal
tissues (32).
In conclusion, most, if not all, ß-core immunoreactivity found in the urine of healthy pre- and postmenopausal women and patients with nonmalignant gynecological conditions can be attributed to cross-reaction of the assays with the urinary metabolite of LH, ß-LH-core. Any true hCGßcf, derived from extremely low-level expression of the hCGß gene cluster by pituitary gonadotrophs, is likely to be extremely low and much smaller than immunoreactivity from cross-reacting ß-LH-core metabolite. If assays can be developed that clearly distinguish between urinary ß-LH-core and hCGßcf, the issue of the clinical utility of hCGßcf as a tumor marker should be reexamined.
| Acknowledgments |
|---|
| Footnotes |
|---|
1 Nonstandard abbreviations: hCGß, free ß-subunit of human chorionic gonadotropin; hCGßcf, ß-core fragment of human chorionic gonadotropin; UGP, urinary gonadotropin peptide; MCR, metabolic clearance rate; LH, luteinizing hormone; ß-LH-core, ß-core fragment of luteinizing hormone; CDI, l,l'-carbonyldiimidazole; FSH, follicle-stimulating hormone; and TSH, thyrotropin. ![]()
| References |
|---|
|
|
|---|
-subunit and
-subunit dissociated from intact human chorionic gonadotropin, despite differences in sialic acid contents. Endocrinology 1987;121:1215-1220.
[Abstract]
-subunit. J Biol Chem 1973;248:6796-6809.
The following articles in journals at HighWire Press have cited this article:
![]() |
P. T.A. Borrelli, S. A. Butler, S. M. Docherty, E. M. Staite, A. L. Borrelli, and R. K. Iles Human Chorionic Gonadotropin Isoforms in the Diagnosis of Ectopic Pregnancy Clin. Chem., December 1, 2003; 49(12): 2045 - 2049. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. F. O'Connor, G. Kovalevskaya, S. Birken, and R. K. Iles Interference of Luteinizing Hormone {beta}-Core Fragment in Urinary Gonadotropin Assays Dr. Iles responds: Clin. Chem., December 1, 1999; 45(12): 2290 - 2292. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |