CLIN. CHEM. 41/10, 1495-1499 (1995)
#{149} Enzymes
and Protein
Markers
Generation and Characterization of Monoclonal Antibodies to Human Type-5 Tartrate-Resistant Acid Phosphatase: Development of a Specific Immunoassay of the lsoenzyme in Serum Paul Chamberlain,’ Juliet and Stephen D. Holmes”3
We have characterized
Compston,2
Timothy
four monoclonal
M. Cox,2 Alison
antibodies
(mAbs) to the purple (“tartrate-resistant,” band 5) acid phosphatase of the human osteoclast (TRAP) and used these to develop a specific serum immunoassay. All four mAbs are of high affinity (Ka 1-5 x 108 lJmol) with a very fast Kasc (0.2-2.0 x 1O L moL1 _1) and a moderate KdISSOC (1-3 x iO s). Two of the mAbs were selected to develop a time-resolved fluorescence immunoassay to measure serum concentrations of TRAP. The mean serum immunoreactive TRAP in a group of healthy premenopausal women and men was 3.7 ± 1.8 pg/L (mean ± SD) and 3.5 ± 1.6 g/L, respectively. Significantly higher concentrations of TRAP were found in postmenopausal women (6.3 ± 2.3 j.tg/L) and in eight patients with Gaucher disease (19.3 ± 4.7 g/L). Further studies are required to investigate the value of serum TRAP as a marker of bone resorption. =
Indexing Terms:
bone disease/bone
resorption
The iron-containing purple acid phosphatase (band 5 acid phosphatase or “tartrate resistant” acid phosphatase; TRAP) of osteoclasts represents one putative marker for monitoring bone resorption.4 This enzyme (EC 3.1.3.2) is expressed in differentiated cells of the monohistiocytic system, notably in osteoclasts (1), Gaucher cells (2), and the phagocytic B lymphocytes of hairy cell leukemia (3). Although the intracellular and extracellular functions of TRAP are unknown, the recent demonstration of its capacity for free radical generation (4) suggests that it may participate directly in bone resorption-a process in which local production of reactive oxygen species has been implicated (4-6). The availability of simple assays of serum markers that reflect boneresorptive activity would greatly facilitate the investigation and monitoring of metabolic diseases that affect bone. Serum TRAP has been suggested as a potential ‘SmithKline
Beecham
Pharmaceuticals,
Great
Burgh,
Epsom,
Surrey KT18 5XQ, UK. 2 Department of Medicine, University of Cambridge Clinical School, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK. 3Author for correspondence. Fax +44 1737 364268; e-mail stephen&holmes%
[email protected]@INET. 4Nonstandard abbreviations:
acid phosphatase;
resistant” surface buffered
TRAP,
human
mAb, monoclonal
type-5
‘tartrate-
antibody;
SPR,
plasmon resonance; RU, resonance unit; HBS, HEPESsaline (see text); and MES, 2-(N-morpholino)ethanesul-
fonic acid. Received
March
2, 1995;
accepted
July
26, 1995.
marker
R. Hayman,2
Robert
C. Imrie,’
Karen
Reynolds,’
for bone
resorption, since the enzyme is inin which bone resorption is increased, such as immobilization, Paget disease, and hyperparathyroidism Previous methods for TRAP measurement in serum lack sensitivity and sufficient specificity for accurate determination. For example, assays that rely on the resistance of TRAP to inhibition by d-tartrate cannot be accurate, since tartrate is only a competitive inhibitor of the unrelated lysosomal and prostatic acid phosphatases (13) and itself produces slight inhibition of the band 5 human enzyme (14). Moreover, the widely distributed acid phosphatase found in erythrocytes is also tartrate-resistant (15). The use of anti-porcine uteroferrin antibodies (14, 16) has been described for the assay of TRAP but, although this offered the opportunity for increased sensitivity (analyte concentration), quantitative recovery of enzyme activity was not possible (16). To achieve sufficient specificity for development of an immunoassay for TRAP, it is necessary to obtain a pure source of antigen. With the recent isolation and sequencing of a cDNA clone for human TRAP it has been possible to obtain milligram quantities of the pure human band 5 acid phosphatase expressed in insect cells infected with recombinant baculovirus (4). We have used this purified recombinant TRAP to generate specific mouse monoclonal antibodies (mAb) for the development of a sandwich immunoassay for TRAP in serum, studying both healthy subjects and a small group of patients with Gaucher disease.
creased in conditions
(7-12).
(17),
Materials and Methods TRAP production. Human TRAP was produced from 50-100-L cultures of Sf cells inoculated with a high titer of baculovirus containing the TRAP expression vector as described by Hayman and Cox (4). The TRAP was purified by HPLC on S-Sepharose (Pharmacia, Uppsala, Sweden), protein capture step; Sephadex G25 (Pharmacia), desalting; Toya Pearl 650S (Tosohaas, Japan), ion exchange; and Superose-12 (Pharmacia), size exclusion. The method was adapted from Hayman and Cox (4) for large-scale purification, and a homogeneous (98% pure) band of 35 kDa was observed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and staining with Coomassie Blue. Recombinant TRAP (-80 mg) was purified from a 100-L culture; recovery was 54%, giving a specific activity of 202 .tmoIJmin per milligram, or 202 kU/g (4). CLINICAL CHEMISTRY, Vol. 41, No. 10, 1995
1495
mAb generation. Mice were immunized with 50 ig of TRAP in Freund’s complete adjuvant followed by booster doses of 50 jg in Freund’s incomplete adjuvant after 4 weeks and two doses of TRAP (50 j.g) intraperitoneally in saline. Spleen cells were harvested 1 day after the last dose and fused with myeloma cells according to the method of Zola (18). Positive hybridomas were cloned twice by the limiting dilution method and the clones expanded to 1-L spinner flasks for large-scale production. mAbs (50-100 mg) were purified from the tissue culture medium by Protein A chromatography (Prosep-A; Bioprocessmg, Consett, UK) and isotyped with a kit produced by Amersham (Amersham, UK). Epitope stoichiometry. We used surface plasmon resonance (SPR) technology (Pharmacia BlAcore) to map the epitope binding patterns of the four mAbs. A rabbit anti-mouse Fc antibody (RAMFc; Pharmacia) was immobilized to the sensor chip [6500 resonance units (RU); amine coupling kit from Pharmacia], and pairwise binding of mAbs was tested. With HBS buffer (10 mmol/L HEPES, pH 7.4, containing 150 mmol NaC1, 3.4 mmol EDTA, and 0.5 g of surfactant P20 per liter) and a flow rate of 5 Umin, mAb 1 was first bound to the RAMFc, followed by sequential injections of nonspecific mAbs (100 mgfL, 2 x 4 j.L), TRAP (5 mg/L, 10 ML), and mAb 2, and then regeneration with 100 mmol/L phosphoric acid. Affinity analysis. SPR measurement was used as above, with the RAMFc immobilized to the sensor chip surface and a flow rate of 5 L/min with HBS buffer. The mAb was first bound to the RAMFc (-1000 RU), followed by injections of TRAP (0-4 mg/L, 20 ML), buffer flow for 360 s, and regeneration of the sensor chip surface with 100 mmolJL phosphoric acid. Ln (dR /dT) vs t was used for association-phase analysis and Ln (Ri/Rn) vs t for dissociation-phase analysis with BlAcore software. Reagents for immunoassay. Microtiter plates, plate washer, and fluorometer were from Wallac (Turku, Finland). The plate-shaker incubator was purchased from Amersham. The coating buffer consisted of 50 mmol/L Na2HPO4, pH 7.4, containing, per liter, 150 mmol of NaCl and 0.2 g of Kathon (Rohm and Haas, London, UK). The blocking buffer was 50 mmolfL Tris-HC1 adjusted to pH 7.4 and containing, per liter, 150 mmol of NaC1, 10 g of bovine serum albumin, and 0.2 g of Kathon, pH 7.4. The wash solution contained 10 mmol Tris-HC1, pH 7.4, with, per liter, 150 mmol of NaC1, 0.5 g of Tween 20, and 0.2 g of Kathon. The assay buffer was 50 mmol/L Tris-HC1, pH 7.4, containing, per liter, 150 mmol of NaC1, 5 g of bovine serum albumin, 0.5 g of bovine gamma globulin, 10 mL of mouse serum, 0.1 g of Tween 40, 20 mmol of diethylenetriaminepentaacetic acid, and 0.2 g of Kathon, pH 7.4. The serum used for the calibrators was defibrinated human plasma from Scantibodies (Santee, CA; 3SH339). Biotinylated anti-TRAP mAb 4E6 was prepared with a kit supplied by Amersham. 1496
CLINICAL
CHEMISTRY,
Vol. 41, No. 10, 1995
TRAP immunoassay procedure. We coated microtiter plate wells with mAb 2H1 (2 mg/L, 100 FL/well) overnight at 4 #{176}C, after which the contents of the wells were aspirated and 250 Uwell of blocking buffer was added for 60 mm at 37 #{176}C. The wells were washed four times; 50 L of biotin-4E6 (2 mg/L in assay buffer) and 50 L of calibrators or serum sample were added and the wells were incubated for 60 mm at 22 #{176}C. The plate wells were washed four times and 100 j.Jlwell of Eu-streptavidin (200 g/L, Wallac) was added and incubated for 30 mm. After the immunoreaction was complete, the wells were washed four times and 200 L/well of enhancement solution (Wallac) was added; after a 5-mm incubation the fluorescence was measured. The fluorometer software was used for curve fitting and result calculations. TRAP serum enzyme assay. Serum was added (1 mL) to a 5-mL polypropylene tube, followed by 3 mL of 10 mmolJL 2-(N-morpholino)ethanesulfonic acid (MES), pH 6.5, and 50 L of S-Sepharose FF (Pharmacia, 500 g/L suspension in MES buffer). The tube was rotated for 10 mm, centrifuged, and the supernatant fluid aspirated. The S-Sepharose FF was then washed twice by centrifugation with 4 mL of MES buffer. Thereafter, 100 L of 50 mmolIL Tris-HC1, pH 7.4, containing 500 mmol/L NaC1, was added to the S-Sepharose FF, and the mixture was vortex-mixed vigorously. After centrifugation, enzymatic activity was determined in 30 L of supernate, which was added to 200 j.L of 1 g/L 4-nitrophenyl phosphate (Sigma, Poole, UK) in 0.1 mol/L disodium tartrate and 0.2 mol/L sodium acetate (pH 5.5). The absorbance at 405 nm was determined every 10 s at 37 #{176}C by using a plate reader (Molecular Devices, Menlo Park, CA) and the results were expressed as mAlmin. The recovery of recombinant TRAP added to buffer or serum averaged 90% over the range 1-100 g/L (202 kU/g) by this method. For a comparison with immunoreactivity, samples were processed as above in the TRAP immunoassay. Patients. Venous blood (5 mL) was obtained, after verbal consent, from the following groups of volunteers: 31 premenopausal women (ages 22-61 years), 12 postmenopausal women (51-62 years), and 20 men (21-52 years). Blood was also obtained from eight patients (six women), ages 16-49 years (mean 34), with Gaucher disease. Results and Discussion Four erated;
mAbs were selected from the hybridomas gen1F1 (IgG1), 4E6 (IgG1), 2H1 (IgG1), and 5C1 (IgG2). The BlAcore complementary binding analysis permitted the assessment of the relative binding stoichiometries of each pair of mAbs in combination with TRAP. The binding response of the second mAb was directly correlated with the distance of its epitope from the epitope recognized by the first mAb (Table 1 and Fig. 1). From the 4 x 4 matrix, we concluded that mAbs 5C1 and 2Hi have overlapping epitopes and that mAbs 5C1 and 4E6 also had overlapping epitopes. Complementary binding pairs are lFi with 2H1 or 5C1; 4E6
assay reproducibility, parallelism with different matrices, linearity over the calibration curve, and recovery. Second mAb A typical calibration curve is depicted in Fig. 2. The 4E6 2111 5C1 First mAb 1FI lowest calibrator (0.5 g/L) has counts approximately 155b 90 528#{176}four times above background, 1F1 and the dose-response 2 4E6 -7 -2 207b curve was linear over the whole range. The within366#{176} 55 591#{176} 2H1 542#{176} assay CV was 3% and the between-assay CV was 55 77b 136#{176} 5C1 427#{176} 9), an S-Sepharose FF extraction of TRAP from serum can be used to measure TRAP enzymatic activity without interference from other phosphatases. The detection limit of this assay is 1 g/L, and the correlation coefficient between enzymatic and immunoreactive TRAP by this method is 0.7 (P C) (0 C)
Cs
E C
a) aI2
0
4
8
6
12
10
TRAP immUnoreactivity, ug/L Fig. 3. Correlation of TRAP immunoreactivity and enzymatic activity (202 kU/g) in serum by 5-Sepharose extraction (P 9, 202 kU/g), TRAP purified from bone (19), and hairy cell leukemia (7), and may reflect the existence of different isoforms of TRAP. As expected, serum concentrations of TRAP in Gaucher disease patients were greatly increased, and in this small series were about five times higher than the mean control value. This finding would be consistent 1498
CLINICAL CHEMISTRY, Vol. 41, No. 10, 1995
The purpose of this study was to generate and characterize mAbs to TRAP, to develop a serum immunoassay, and to carry out preliminary determinations of the concentration of immunoreactive TRAP in serum obtained from subjects with normal or increased bone turnover and in patients with Gaucher disease. In the present work we were not able to determine the concentration of other biochemical markers of bone formation (22) or resorption (23), but further studies, including the use of bone histomorphometry, are in progress to validate the use of immunoreactive TRAP as a marker for bone disease. The immunoassay reported here is specific, robust, and reproducible, and has sufficient sensitivity and range to measure TRAP in clinical samples of both normal and disease serum. We intend to determine whether measurement of immunoreactive TRAP alone or in combination with other biochemical markers will be generally applicable for monitoring bone disease. We acknowledge
the contribution of Ceri Lewis, Chris Mannix, for the production and purification of TRAP.
and Gary Pettman
References 1. Minkin C. Bone acid phosphatase as a marker function. Calcif Tissue mt 1982;34:285-90.
of osteoclast
2. Robinson DB, Glew RH. Acid phosphatase in Gaucher’s ease [Review]. Clin Chem 1980;26:371-82. 3. Yam LT, Li CY, Finkel HE. Leukemic reticuloendotheliosis.
Arch Intern Med 1972;130:248-56. 4. Hayman AR, Cox TM. Purple acid phosphatase macrophage
and
osteoclast.
ties and crystallization
Characterization,
of the recombinant
dis-
of the human
molecular
di-iron
proper-
oxo protein
secreted by baculovirus infected insect cells. J Biol Chem 1994; 269:1294-300. 5. Henley R, Worwood M. Luminol peroxidation catalyzed by human isoferritins. Arch Biochem Biophys 1991;286:238-43. 6. Key LL, Reis WL, Taylor RG, Hayes BD, Pitzer BR. Oxygen derived free radicals in osteoclasts: the specificity and location of the nitroblue tetrazolium reaction. Bone 1990;i1:115-9. 7. Kraenzlin ME, Lau K-H, Liang L, Freeman TK, Singer FR, Stepan J, Baylink DJ. Development of an immunoassay for human serum osteoclastic tartrate-resistant acid phosphatase. J Clin Endocrinol Metab 1990;71:442-51. 8. Lau K-H, Onishi T, Wergedal JE, Singer FR, Baylink DJ. Characterization and assay of tartrate-resistant acid phosphatase activity in serum: potential use to assess bone resorption. Clin Chem 1987;33:458-62. 9. Choy FYM. Gaucher disease: comparative study of acid phosphatase and glucocerebrosidase in normal and type-i Gaucher tissues. Am J Med Gent 1985;21:519-28.
10. Yam LT. Clinical significance of the human acid phosphateses. Am J Med i974;56:604-i6. 11. Rico H, Villa LF. Serum tartrate resistant acid phosphatase (TRAP) as a biochemical marker of bone remodeling. Calcif Tissue mt 1993;52:149-50. 12. Cheung CK, Panesar NS, Haines C, Masarei J,
R.
Immunoassay
of a tartrate-resistant
acid
serum. Clin Chem i995;41:679-86. 13. Jacobsson K. On the inhibition of prostatic tartrate. Scand J Clin Lab Invest i958;il:358-60.
14.
Whittaker
human
band
KB,
Cox
TM,
Moss
5 (“tartrate-resistant”)
DW. acid
phosphatase
in
phosphatase
by
An immunoassay phosphatase that
of in-
volves the use of anti-porcine 1989;35:86-9.
uteroferrin
antibodies.
Clin Chem
15. Abul-FadI MAM, King EJ. Properties of the acid phosphateses of erythrocytes and of the human prostate gland. Biochem J 1949;45:51-60. 16. Echetebu ZO, Cox TM, Moss DW. Antibodies to porcine uteroferrin used in measurement of human tartrate-resistant acid phosphatase. Clin Chem 1987;33:1832-6. 17. Lord DK, Cross NCP, Bevilacqua MA, Rider SH, Gorman PA, Groves AV, et al. Type 5 acid phosphatase: sequence, expression
and chromasomal localisation of a differentiation-associated protein of the human macrophage. Eur J Biochem i990;189:287-93. 18. Zola H, ed. Monoclonal antibodies. A manual of techniques. Boca Raton, FL: CRC Press, l987:2l4pp. 19. Crocker AC, Landing BH. Phosphatase disease.
Metabolism
1960;9:341-62.
studies
in Gaucher’s
20. de la Piedra C, Torres R, Rapado M, Diaz Curiel M, Castro N. Serum tartrate-resistant phosphatase and bone mineral content in postmenopausal osteoporosis. Calcif Tissue Int 1989;45:58-60. 21. Allen SH, Nuttleman PR, Ketcham CM, Roberts P.M. Purification and characterization of human bone tartrate-resistant acid phosphatase. J Bone Miner Res 1989;4:47-55. 22. Hyldstrup L, Clemmensen I, Jensen BA, Transbol I. Noninvasive evaluation of bone formation: measurements of serum alkaline phosphatase, whole body retention of diphosphonate and serum osteocalcin in metabolic bone disorders and thyroid disease. Scand J Clin Lab Invest 1988;48:611-19. 23. Uebelhart A, Schlemmer A, Johansen JS, Gineyts V, Christiansen C, Delmas PD. Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links. J Clin Endocrinol
Metab
1991;72:367-73.
CUNICAL CHEMISTRY, Vol. 41, No. 10, 1995
1499
