procedure to obtain lymphoblast extracts produced more

Clinical Chemistry 49, No. 5, 2003
procedure to obtain lymphoblast extracts produced more
concentrated enzyme extracts with higher specific activities.
By performing the assay in the presence of various
concentrations of ornithine, we confirmed that AGAT is
inhibited by ornithine, as reported previously (7 ). This
observation may be of clinical interest in conditions of
hyperornithinemia. Decreased brain creatine has been
observed in patients affected with gyrate atrophy of the
choroid and retina (McKusick 258870). It has been suggested that in this condition the high concentrations of
ornithine caused by the inherited deficiency of ornithine␦-aminotransferase inhibit AGAT and, thus, creatine biosynthesis (8 ). In addition, in hyperornithinemia-hyperammonemia-homocitrullinuria syndrome, creatine excretion
has been shown to be low (9 )
In our in vitro assay, the amount of ornithine formed
posed no problem. Inhibition became significant at an
ornithine concentration of ⬃0.1 mmol/L (93% remaining
activity), whereas during incubations over 20 h, at most 3
nmoles of ornithine were formed (0.015 mmol/L).
It is known that creatine regulates AGAT at the pretranslational level. We investigated whether creatine also
inhibits AGAT activity directly. We observed no inhibition of AGAT by creatine at concentrations up to 5
mmol/L.
In conclusion, we present a sensitive and specific enzymatic assay for AGAT that enables enzymatic diagnosis of
AGAT deficiency. It may help in the diagnosis of more
patients with AGAT deficiency. Early recognition and
treatment may effectively prevent neurologic damage (1 ).
We thank Prof. Dr. Giovanni Cioni for supplying the
patient’s lymphoblast line. We gratefully acknowledge
Wjera Wickenhagen for technical assistance.
References
1. Item CB, Stöckler-Ipsiroglu S, Stromberger C, Mühl A, Alessandri MG, Bianchi
MC, et al. Arginine:glycine amidinotransferase deficiency: the third inborn
error of creatine metabolism in humans. Am J Hum Genet 2001;69:1127–33.
2. Bianchi MC, Tosetti M, Fornai F, Alessandri MG, Cipriani P, De Vito G, et al.
Reversible brain creatine deficiency in two sisters with normal blood creatine
level. Ann Neurol 2000;47:511–3.
3. van Pilsum JF, Taylor D, Zakis B, McCormick P. Simplified assay for
transamidinase activities of rat kidney homogenates. Anal Biochem 1970;
35:277– 86.
4. Ratner S, Rochovansky O. Biosynthesis of guanidinoacetic acid. I. Purification
and properties of transamidinase. Arch Biochem Biophys 1956;63:277–95.
5. Struys EA, Jansen EEW, ten Brink HJ, Verhoeven NM, van der Knaap MS,
Jakobs C. An accurate stable isotope dilution gas chromatographic-mass
spectrometric approach to the diagnosis of guanidinoacetate methyltransferase deficiency. J Pharm Biomed Anal 1998;18:659 – 65.
6. Ratner S, Rochovansky O. Biosynthesis of guanidinoacetic acid. II. Mechanism of amidine group transfer. Arch Biochem Biophys 1956;63:296 –315.
7. Sipilä I. Inhibition of arginine-glycine amidinotransferase by ornithine. Biochim
Biophys Acta 1980;613:79 – 84.
8. Näntö-Salonen K, Komu M, Lundbom N, Heinänen K, Alanen A, Sipilä I, et al.
Reduced brain creatine in gyrate atrophy of the choroid and retina with
hyperornithinemia. Neurology 1999;53:303–7.
9. Dionisi Vici C, Bachmann C, Gambarara M, Colombo JP, Sabetta G. Hyperornithinemia-hyperammonemia-homocitrullinuria syndrome: low creatine excretion and effect of citrulline, arginine, or ornithine supplement. Pediatr Res
1987;22:364 –7.
805
Identification and Properties of Glycated Monoclonal
IgA That Affect the Fructosamine Assay, Kiyotaka Fujita,1* Linda K. Curtiss,1 Ikunosuke Sakurabayashi,2 Fumiko
Kameko,3 Nobuo Okumura,3 Fumiko Terasawa,3 Minoru Tozuka,4 and Tsutomu Katsuyama4 (1 Department of Immunology, The Scripps Research Institute, 10550 North Torrey
Pines Rd., La Jolla, CA 92037; 2 Department of Clinical
Laboratories, Omiya Medical Center, Jichi Medical
School, Amanuma, Saitama 330-8503, Japan; 3 Department of Biomedical Laboratory Sciences, School of Health
Sciences, Shinshu University, Asahi, Matsumoto, Nagano
390-8621, Japan; 4 Department of Laboratory Medicine,
Shinshu University School of Medicine, Asahi, Matsumoto, Nagano 390-8621, Japan; * author for correspondence: fax 858-784-9144, e-mail [email protected])
Serum fructosamine is used to monitor short-term (1–3
weeks) average glycemia (1, 2 ), but controversy remains
whether it is influenced by the serum concentrations of
albumin (3, 4 ), IgA (5, 6 ), or monoclonal IgA (7, 8 ). We
recently demonstrated glycation of monoclonal IgA and
the presence of IgA-albumin complexes (9 ), but the relationship between the glycation of monoclonal IgA and the
IgA-albumin complexes was not clear.
In this study, we measured immunoglobulin-albumin
complexes, serum fructosamine, and the glycation of
immunoglobulin in patients with monoclonal proteinemia and in patients with polyclonal hyper-IgA.
We obtained serum samples from 40 patients with
M-proteinemia with no history of diabetes and whose
plasma glucose concentrations were ⬍1.1 g/L. The Mprotein was IgG in 17 of the patients [mean (SD), 29.2
(21.0) g/L], IgA in 13 [18.5 (16.4) g/L], and IgM in 10 [16.8
(11.3) g/L]. We also analyzed sera from 15 nondiabetic
patients with hepatitis who had polyclonal hyper-IgA [5.1
(1.1) g/L] as controls.
Serum fructosamine was measured at 37 °C in an automated analyzer (JCA-RX 10 Clinalyzer; Japan Electron
Optics Laboratory) with reagents from Roche Diagnostics
Corporation.
Serum protein electrophoresis was performed on agarose gels. Immunoelectrophoresis was based on the
method of Scheidegger (10 ). Fructosamine was detected
on agarose gels by incubating the gel for 30 min at 37 °C
with a 2⫻ concentration of a fructosamine reagent.
IgA was isolated from serum with use of a jacalinagarose (Funakoshi Corporation) column (11 ) and 0.8
mol/L galactose to elute the IgA. The IgA fraction was
dialyzed against 0.2 mol/L Tris-HCl buffer (pH 8.0),
concentrated by ultrafiltration, loaded on a DEAE-Sephacel (Amersham Pharmacia Biotech) ion-exchange column,
and eluted with a linear gradient of 0 – 0.5 mol/L NaCl in
the 0.2 mol/L Tris-HCl buffer.
Sera were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), with or without 2-mercaptoethanol (2-ME), and electrophoretically
transferred to polyvinylidene difluoride (PVDF) membranes (12 ). Immunostaining was performed with rabbit
polyclonal antisera (Dako) as first antibodies and peroxi-
806
Technical Briefs
dase-conjugated goat anti-rabbit IgG (Dako) as second
antibody.
To detect fructosamine on PVDF membranes, we
blocked the membranes for 60 min at room temperature in
buffer containing 30 g/L bovine serum albumin. After
washing, the membranes were incubated for 30 min at
37 °C with a 2⫻ concentration of the fructosamine reagent.
Differences between the groups were verified by the
Student t-test. Correlations were studied by means of the
Pearson correlation coefficient.
Serum fructosamine concentrations were higher in patients with IgA-type [3.47 (0.93) mmol/L; P ⬍0.05] than
with IgG-type [2.55 (0.48) mmol/L] or IgM-type Mproteinemia [2.51 (0.42) mmol/L]. Concentrations of fructosamine and of serum IgA (but not IgG or IgM) in
patients with M-proteinemia were significantly correlated
(␥ ⫽ 0.698; P ⬍0.05). By contrast, in the 15 patients with
hepatitis and polyclonal hyper-IgA, mean (SD) fructosamine was 2.62 (0.23) mmol/L and all values were
within our reference interval (1.94 –3.16 mmol/L); in
addition, fructosamine was not significantly correlated
with the IgA concentrations (␥ ⫽ 0.279).
The sera from the patients with IgG- or IgM-type
M-proteinemia had fructosamine only at the position of
albumin, but 11 of 13 sera with IgA-type M-proteinemia
stained for glycoprotein at the position of the M-protein
band as well as the albumin band (Fig. 1 in the Data
Supplement that accompanies the online version of this
Technical Brief at http://www.clinchem.org/content/
Vol49/issue5/). Sera from the patients with polyclonal
hyper-IgA did not stain for glycoproteins at the position
of IgA but did stain at the position of the albumin band.
IgA-␬-type M-protein also reacted with albumin-specific antiserum (Fig. 1). The same abnormal precipitin arcs
were observed in 11 of 13 sera from patients with IgAtype M-proteinemia that were glycosylated at the position
of M-protein band. Moreover, the mobility of the abnormal precipitin arcs corresponded to the position of the
glycated M-protein band. The patients’ albumin had a
macromolecular mass ⬎300 kDa, as determined by immunostaining using anti-albumin serum after SDS-PAGE
without 2-ME.
The abnormal precipitin arcs for albumin disappeared
when antisera specific for human albumin or IgA (␣-chain
specific) were incubated for 37 °C for 30 min with the
purified IgA-type M-protein (at a ratio of 6:1 by volume)
and supernates were concentrated by ultrafiltration. The
immunoelectrophoretic patterns of the sera from patients
with IgG- or IgM-type M-proteinemia or with polyclonal
hyper-IgA showed no abnormal precipitin arcs with albumin.
After SDS-PAGE with 2-ME, macroalbumin was not
detected by immunostaining with anti-albumin serum,
and IgA-type M-proteins stained for fructosamine at ⬃55
kDa, which corresponded to the ␣-chain, and also at 25
kDa, which corresponded to the light chain, in 11 of 13
cases. Sera from patients with IgG- or IgM-type Mproteinemia or with polyclonal hyper-IgA showed fruc-
tosamine glycosylation only at 65 kDa, which corresponded to albumin (Fig. 2 in the Data Supplement that
accompanies the online version of this Technical Brief at
http://www.clinchem.org/content/vol49/issue5/).
To investigate the possibility that monoclonal IgA was
glycated in the presence of glucose, we incubated a
patient’s serum that had no fructosamine glycosylation at
the M-protein band with d-glucose at 10 or 20 g/L.
Incubation of the monoclonal IgA serum with 10 or 20
g/L glucose at 37 °C for 60 min produced no fructosamine
at the position of the monoclonal IgA band on electrophoresis, but increased the fructosamine at the position of
albumin band (Fig. 3 in the Data Supplement).
Glycohemoglobin is an accepted measure of long-term
average glycemia. Because nonenzymatic glycation also
affects other proteins (13 ), glycated albumin or total
glycated protein (fructosamine assay) can be used as a
measure of short-term, average glycemia (1, 2 ). Wahid et
al. (14 ) stated that serum fructosamine is useful as a
maker of 5-year risk of developing diabetes mellitus in
patients exhibiting stress hyperglycemia; however, the
dependence of fructosamine concentrations on the concentrations of albumin and total protein has been reported
(15–19 ). Montagna et al. (7 ) and Nakamura et al. (8 )
reported that nondiabetic patients with IgA-type M-proteinemia have significantly increased serum fructosamine.
Fig. 1. Electrophoretic (A) and immunoelectrophoretic (B) patterns of a
serum from a patient with IgA-␬-type M-proteinemia.
Shown in B are immunoelectrophoretic patterns obtained with antisera to human
whole serum (AHWS), ␣-chain (A-IgA), ␬ light chain (A-␬), ␭ light chain (A-␭),
immunoglobulin (A-Igs), and albumin (A-Alb). NS, normal serum; PS, serum from
patient with IgA-␬-type M-proteinemia. The arrow indicates an abnormal precipitin
arc of albumin. Anode is to the right.
Clinical Chemistry 49, No. 5, 2003
We also found that the fructosamine concentrations in
patients with IgA-type M-proteinemia were significantly
higher than the fructosamine concentrations in patients
with IgG- or IgM-type M-proteinemia. By staining for
fructosamine after electrophoresis, we demonstrated that
this is the result of an increase in glycated monoclonal
IgA, i.e., 84.6% of the patients with IgA-type M-proteinemia had increased fructosamine at the monoclonal IgA
bands, and fructosamine was detected in both the heavy
and light chains.
The two patients with IgA-type M-proteinemia who did
not have fructosamine at the M-protein band had lower
IgA values than did the 11 patients with IgA-type Mproteinemia. This suggests that glycation of IgA molecules may be related to the high IgA concentrations [mean
(SD), 18.5 (16.4) g/L]. The sera from the patients with
polyclonal hyper-IgA had fructosamine only in the albumin band. Thus, glycation of IgA molecules appears to be
unique to monoclonal IgA.
It is interesting that the macroalbumin complexes that
were composed of IgA and albumin were detected in all
sera from patients with IgA-type M-protein containing
fructosamine at the M-protein band (11 of 13). The IgAalbumin complex was absent from the sera from patients
with IgG- or IgM-type M-proteinemia as well as from the
sera from patients with polyclonal hyper-IgA. Moreover,
monoclonal IgA that was not associated with albumin
was not glycated, although the glucose concentration in
the patient’s serum was increased. Thus, a major factor in
the glycation of IgA was the binding to albumin. Because
the half-life of albumin is longer (15–19 days) (20 ) than the
half-life of human IgA molecules (6 days) (21 ), the halflife of monoclonal IgA that binds albumin is longer, and
this could increase the glycation of the monoclonal IgA.
Why is the monoclonal IgA glycosylated and not the
polyclonal IgA? The mean IgA concentrations (18.5 g/L)
in patients with monoclonal IgA were clearly higher than
the IgA concentrations (5.1 g/L) in patients with polyclonal IgA. Among the patients with IgA concentrations
⬍10 g/L, 3 of the 5 patients with monoclonal IgA had
IgA-albumin complexes, but none of the 15 patients with
polyclonal IgA had IgA-albumin complexes (Fig. 4 in the
Data Supplement). Moreover, serum IgA is present in
several polymeric forms, ranging from monomers to
pentamers. In normal human serum, the monomeric form
usually predominates, but in monoclonal IgA serum, the
relative amount of each polymeric form varies. Polymerization of the monoclonal IgA might therefore play an
important role in its interaction with albumin. This idea is
supported by the fact that the IgA-albumin complexes
had a macromolecular mass ⬎300 kDa.
The formation of IgA-albumin complexes has been
noted previously (22–25 ). Tomasi and Hauptman (23 )
examined 49 IgA myeloma sera and found that 65% had
IgA-albumin complexes, suggesting that albumin was
bound predominantly to dimeric IgA.
Why is the abnormal precipitin arc for albumin faster
than that for IgA on immunoelectrophoresis? Serum protein electrophoresis usually separates the various compo-
807
nents of blood protein into bands or zones according to
their electrical charges. The isoelectric points of immunoglobulins are 6.4 –7.2, whereas that of albumin is 4.8 (26 ).
It can be considered that the abnormal precipitin arc for
albumin in the IgA-albumin complexes moves faster than
that for IgA because of the acidity of the albumin.
In conclusion, our findings throw doubt on the clinical
utility of fructosamine as a measure of hyperglycemic
status in patients with IgA-type M-proteinemia.
We thank Minoru Ogawa (Yokote Hospital), Yoshimasa
Aoki, and Dr. Mitsuaki Kameko (Nagano Municipal Hospital) for the kind gifts of patient sera.
References
1. Armbruster DA. Fructosamine: structure, analysis, and clinical usefulness
[Review]. Clin Chem 1987;33:2153– 63.
2. Lim YS, Staley MJ. Measurement of plasma fructosamine evaluated for
monitoring diabetes. Clin Chem 1985;31:731–3.
3. Baker JR, O’Connor JR, Metcalf PA, Lawson MR, Johnson RN. Clinical
usefulness of estimation of serum fructosamine concentration as a screening test for diabetes mellitus. Br Med J 1983;287:863–7.
4. MacDonald D, Pang CP, Cockram CS, Swaminathan R. Fructosamine measurements in serum and plasma. Clin Chim Acta 1987;168:247–52.
5. Rodriguez-Segade S, Lojo S, Camina MF, Paz JM, Del Rio R. Effects of
various serum proteins on quantification of fructosamine. Clin Chem 1989;
35:134 – 8.
6. Lloyd DR, Marples J. Fructosamine and IgA: no correlation in non-diabetics
[Letter]. Clin Chem 1989;35:1556 –7.
7. Montagna MP, Laghi F, Cremona G, Zuppi C, Barbaresi G, Castellana ML.
Influence of serum proteins on fructosamine concentration in multiple
myeloma. Clin Chim Acta 1991;204:123–30.
8. Nakamura F, Kaimori M, Takaya H, Fujita K, Suzuki N, Sakurabayashi I, et al.
Markedly elevated serum fructosamine in a non-diabetic patient with IgA-␬
type multiple myeloma. Rinsho Byori 1996;44:85–9.
9. Fujita k, Sakurabayashi I. Effects on M-protein on laboratory data. Rinsho
Byori 2001;49:682–5.
10. Scheidegger JJ. Une micro-méthode de I’immunoélectrophorèse. Int Arch
Allergy 1955;7:103–10.
11. Hortin GL, Trimpe BL. Lectin affinity chromatography of proteins bearing
o-linked oligosaccharides: application of jacalin-agarose. Anal Biochem
1990;188:271–7.
12. Towbin H, Staechelin T, Gordon J. Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 1979;76:4350 – 4.
13. Bunn HF. Nonenzymatic glycosylation of protein: relevance to diabetes. Am J
Med 1981;70:325–30.
14. Wahid ST, Sultan J, Handley G, Saeed BO, Weaver JU, Robinson ACJ. Serum
fructosamine as a marker of 5-year risk of developing diabetes mellitus in
patients exhibiting stress hyperglycaemia. Diabet Med 2002;19:543– 8.
15. Seng LY, Staley MJ. Plasma fructosamine is a measure of all glycated
proteins [Technical Brief]. Clin Chem 1986;32:560.
16. Howey JEA, Browning MC, Fraser CG. Assay of serum fructosamine that
minimizes standardization and matrix problems: use to assess components
of biological variation. Clin Chem 1987;33:269 –72.
17. Lemon M, Forrest ARW. Fructosamine activity of proteins in serum [Letter].
Clin Chem 1986;32:2101.
18. Zoppi F, Mosca A, Granata S, Montalbetti N. Glycated proteins in serum:
effect of their relative proportions on their alkaline reducing activity in the
fructosamine test. Clin Chem 1987;33:1895–7.
19. Singh J, Kulig KA. Effects of albumin and immunoglobulin A on fructosamine
assay. Clin Chem 1992;38:824 –30.
20. Kerr MA. The structure and function of human IgA. Biochem J 1990;271:
285–96.
21. Austin GE, Mullins RH, Morin LG. Non-enzymic glycation of individual plasma
proteins in normoglycemic and hyperglycemic patients. Clin Chem 1987;33:
2220 – 4.
22. Mannik M. Binding of albumin to ␥ A-myeloma proteins and Waldenström
macroglobulins by disulfide bonds. J Immunol 1967;99:899 –906.
23. Tomasi TB, Hauptman SP. The binding of ␣-1 antitrypsin to human IgA.
J Immunol 1974;112:2274 –7.
24. Hauptman SP, Sobczak G. Origin of immunoglobulin-albumin complexes.
Nature 1976;263:64 –7.
25. Eilat D, Fischel R, Zlotnick A. Albumin-immunoglobulin complexes in human
808
Technical Briefs
serum: classification and immunochemical analysis. Scand J Immunol
1981;14:77– 88.
26. Paraskevas F, Foerster J. Immunodiagnosis. In: Lee GR, Foerster J, eds.
Wintrobe’s clinical hematology. Baltimore: Williams & Wilkins, 1999:40pp.
Fully Automated Chemiluminometric Assay for Hyperglycosylated Human Chorionic Gonadotropin (Invasive
Trophoblast Antigen), Raj Pandian,1* Julie Lu,2 and Jolanta
Ossolinska-Plewnia2 (1 Quest Diagnostics, Nichols Institute,
33608, Ortega Highway, San Juan Capistrano, CA 92690;
2
Nichols Institute Diagnostics, San Clemente, CA 92673;
* author for correspondence: fax 949-728-4990, e-mail
[email protected])
Hyperglycosylated human chorionic gonadotropin (HhCG)
is a glycoprotein hormone secreted during embryonic
implantation and trophoblast invasion of the uterine wall
and is an early marker of pregnancy (1 ). Relative to hCG,
HhCG has a higher molecular mass (38.5– 40 kDa, depending on the amount of carbohydrate) and a higher
number of asparagine (N)-linked triantennary carbohydrates and serine (O)-linked tetrasaccharide core structures in the ␤-subunit (2 ). Although both are secreted
from the placenta and choriocarcinoma, HhCG is produced by mononucleated cytotrophoblasts, and hCG is
produced by syncytiotrophoblast cells (3– 6 ). Because the
cytotrophoblasts are primitive and invasive in nature,
HhCG is also called invasive trophoblast antigen (ITA)
(5 ).
Birken et al. (7 ) described a monoclonal antibody (B152)
specific for the ␤-subunit C-terminal peptide and the
O-linked oligosaccharide of HhCG. Although the epitope
for this antibody does not require sialic acid, the presence
of the O-linked tetrasaccharide core structure is essential
(1 ).
Using IRMAs and ELISAs, investigators showed that (a)
HhCG rapidly increases in early pregnancy, attaining
substantially higher concentrations and decreasing earlier
than hCG (1, 8 ); (b) HhCG is increased in Down syndrome-affected pregnancies in both the first and second
trimesters (9 –11 ); and (c) the HhCG:hCG ratio appears to
be higher in those with invasive vs noninvasive trophoblastic disease (12 ).
The above HhCG assays were performed manually
using large sample volumes (200 ␮L) and long incubation
times (turnaround time, 1–2 days). We therefore developed an automated immunochemiluminometric assay
(ICMA) that uses two monoclonal antibodies: the HhCGspecific B152 antibody described above and a hCG ␤-subunit-specific antibody (B207). Both antibodies were purified from cell lines provided by Dr. O’Connor (Columbia
University, New York, NY). B152 was biotinylated with
long-chain NHS-biotin (13 ), and B207 was conjugated
with acridinium ester (14 ).
The Nichols Institute Diagnostics Advantage® instrument automatically pipetted 15 ␮L of sample into a
cuvette, followed by 25 ␮L of streptavidin-coated mag-
netic particles (4 g/L Dynal M-270), 70 ␮L of capture
antibody (6 mg/L B152), and 260 ␮L of buffer [0.1 mol/L
phosphate-buffered saline (PBS), pH 8.2, containing 50
g/L bovine serum albumin (BSA)]. During a 30-min
incubation at 37 °C, HhCG in the sample bound to
the B152 capture antibody, which in turn bound to the
magnetic particles. The magnetic particles were automatically washed three times to remove unbound
materials. Detection antibody [300 ␮L of 1 mg/L B207
in 0.5 mol/L PBS (pH 7.4) with 5 g/L protease-free
BSA, 60 mL/L normal mouse serum, and 1 g/L mouse
␥-globulin] was then added to the washed magnetic
particles. During this 10-min incubation at 37 °C, the
B207 antibody bound to a hCG-shared epitope on the
HhCG molecule, forming a sandwich complex. After
another three washes, the magnetic particle-containing
wells were transferred to the on-board luminometer.
Hydrogen peroxide (3.25 mL/L)- and sodium hydroxide
(0.25 mol/L)-containing solutions were automatically injected into the wells, initiating the chemiluminescence
reaction. The generated “flash” of light was quantified
and expressed as relative light units (RLU). The RLU are
directly proportional to the concentration of HhCG in the
sample. The relationship was linear up to 7500 ␮g/L,
at which concentration the curve plateaued (Fig. 1). No
hook effect was observed with concentrations as high
as 30 000 ␮g/L [HhCG from choriocarcinoma, prepared
in 0.5 mol/L PBS (pH 7.4) containing 1 g/L protease-free
BSA].
The calibration curve (Fig. 1, inset) was stored by the
instrument, so calibrators did not have to be included
every time the assay was performed. The limit of detection, based on the mean RLU for 20 replicates of the zero
calibrator plus 2 SD, was 0.1 ␮g/L. The limit of quantification, based on assays of serial dilutions of a sample
(HhCG ⫽ 2 ␮g/L) with the CV calculated from 10
observations at each dilution, was 0.2 ␮g/L (CV ⬍20%).
On the basis of results from three serum pools and one
urine pool, the intraassay CV for 20 replicates was ⬍3.5%
and the interassay CV was ⬍7.5%. Thus, the assay could
be performed in singlicate.
The assay had ⬍0.1% cross-reactivity with all glycoprotein hormones except hCG, which had a cross-reactivity ⬍3.5% (Table 1). The dose–response of each hCG
preparation (total, free ␤, nicked, and nicked free ␤-hCG)
was parallel to that of HhCG. Although hCG crossreactivity was minimal, it is not clear whether such
cross-reactivity was attributable to native hCG itself or to
contamination of hCG preparations with HhCG. The
latter is highly likely because of (a) the different crossreactivities reported for various hCG preparations [Table
1 and Refs. (5, 7 )]; (b) the similarities between dose–
response curves for HhCG and the cross-reactants; and (c)
the 0.9% cross-reactivity observed when recombinant
murine hCG was tested (Table 1). Because the reported
hCG cross-reactivity was small, HhCG could be measured
in the presence of hCG. The HhCG:hCG ratio may be
clinically useful (1, 15, 16 ).
To determine the suitability of various sample types,