Electrophoretic study of tartrate-resistant acid phosphatase isoforms in endstage renal disease and rheumatoid arthritis

Clinica Chimica Acta 301 (2000) 147–158 www.elsevier.com / locate / clinchim

Electrophoretic study of tartrate-resistant acid phosphatase isoforms in endstage renal disease and rheumatoid arthritis b a,c , b Karen Takahashi , Anthony J. Janckila *, Susan Z. Sun , b b a,b,c Eleanor D. Lederer , Prasun C. Ray , Lung T. Yam a

Special Hematology Laboratory, US Department of Veterans Affairs Medical Center, 800 Zorn Avenue, Louisville, KY 40206, USA b Department of Medicine, University of Louisville, Louisville, KY, USA c Department of Microbiology and Immunology, University of Louisville, Louisville, KY, USA Received 12 April 2000; received in revised form 15 June 2000; accepted 30 June 2000

Abstract The objective of this study was to identify the isoform, type-5a or type-5b, responsible for increased tartrate-resistant acid phosphatase (TRAP) activity in endstage renal disease (ESRD) and TRAP protein in rheumatoid arthritis (RA). We studied 24 sera each from healthy, ESRD and RA subjects. Type-5 TRAP activity and protein were quantitated by immunoassays. Isoform expression was determined by computerized imaging of non-denaturing polyacrylamide gels (PAGE) stained for TRAP activity. Other biochemical markers included: intact parathyroid hormone (iPTH), total and bone-specific alkaline phosphatase (TAP, BAP), N-telopeptides of type-I collagen (NTx), and free pyridinoline (Pyd). Isoform 5a was normal in both ESRD and RA. Isoform 5b was elevated in ESRD only. Serum TRAP activity correlated with both isoforms 5a and 5b in RA, but only with 5b in ESRD. TRAP protein assays did not correlate with PAGE assays for 5a or 5b. TRAP activity, but not protein, correlated with BAP and NTx in RA sera. Both TRAP activity and protein correlated with iPTH, TAP and Pyd in ESRD sera. Increased TRAP activity in ESRD was due to increased osteoclastic isoform 5b and related to bone turnover. Increased TRAP protein in RA was suspected, but not proven, to be isoform 5a and not related to bone turnover. Heterogeneity of serum TRAP and preferential expression of isoforms has clinical

Abbreviations: TRAP, tartrate-resistant acid phosphatase; ESRD, endstage renal disease; HD, hemodialysis; RA, rheumatoid arthritis; PAGE, polyacrylamide gel electrophoresis; EDTA, ethylenediaminetetraacetate; TAP, total alkaline phosphatase; BAP, bone alkaline phosphatase; iPTH, intact parathyroid hormone; Pyd, pyridinoline; NTx, N-terminal telopeptides of type-I collagen *Corresponding author. Tel.: 1 1-502-895-3401; fax: 1 1-502-894-6155. E-mail address: [email protected] (A.J. Janckila). 0009-8981 / 00 / $ – see front matter  2000 Published by Elsevier Science B.V. PII: S0009-8981( 00 )00338-7

148

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

significance in different diseases including ESRD and RA. Science B.V.

 2000 Published by Elsevier

Keywords: Tartrate-resistant acid phosphatase; Immunoassay; Gel electrophoresis; Endstage renal disease; Rheumatoid arthritis.

1. Introduction Type-5 tartrate-resistant acid phosphatase (TRAP) activity is increased in the serum of patients with a variety of diseases involving increased bone resorption [1–5]. The principal source of TRAP in these conditions is the bone resorbing osteoclasts. Immunoassays specific for the type-5 TRAP have been developed as surrogate biomarkers of bone resorption for patients at risk for accelerated bone loss [6–11]. Although these assays yield encouraging results they are not yet used routinely in the clinical setting. We developed two such immunoassays, one to measure TRAP activity by an enzymatic capture assay, and the other to measure total TRAP protein by a two-site ‘sandwich’ assay [12]. In that study of two groups of patients at risk for increased bone resorption, we found that a significant percentage of patients with endstage renal disease (ESRD) had significantly increased TRAP activity and protein. On the other hand, many patients with rheumatic diseases, particularly those with rheumatoid arthritis (RA), had increased total TRAP protein while TRAP activity was normal. Circulating type-5 TRAP exists as two isoforms; 5a and 5b [13]. These isoforms were originally identified by their different electrophoretic mobility in acid conditions whereby TRAP 5a migrates slower than 5b [14]. The faster migrating TRAP 5b is derived from osteoclasts [15]. The source and significance of TRAP 5a remains to be determined. We postulate that the difference in serum TRAP activity in ESRD and RA is due to a difference in the proportional expression of TRAP isoforms. Since TRAP 5a and 5b are antigenically related, current immunoassays fail to discriminate between them. Therefore, we used non-denaturing polyacrylamide gel electrophoresis (PAGE) in combination with TRAP immunoassays and other biochemical markers of bone metabolism to test our hypothesis.

2. Materials and methods

2.1. Patient samples Seventy-two sera in which TRAP activity and protein had been previously

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

149

determined by immunoassay [12] were used for study. These included 24 from healthy control subjects, 24 from ESRD patients, and 24 from RA patients. All sera were separated from blood and stored at 2 508C on the same day and had been kept at that temperature since the initial analyses. All patients were being treated routinely for their respective diseases at the time of study. To avoid bias, these sera were selected based on their activity and protein concentrations to represent the full range of TRAP present in each entire cohort. Thus, selected sera from healthy controls had a range in TRAP from 1.9 to 4.0 U / l and from 5.2 to 70.3 mg / l. Sera from selected ESRD patients had a range in TRAP from 2.0 to 9.3 U / l and from 32.0 to 128.6 mg / l. Finally, sera from selected RA patients had a range in TRAP from 1.5 to 4.9 U / l and from 26.0 to 167.5 mg / l.

2.2. Type-5 TRAP immunoassays Immunoassays for type-5 TRAP activity and protein were done simultaneously according to previously published methods [12]. Briefly, duplicate avidin-coated wells were coated with 0.5 mg of anti-TRAP 14G6-biotin and blocked with 3% gelatin. 50 ml of serum diluted in 50 ml of sample buffer (10 mmol / l Tris, 150 mmol / l NaCl, 0.05% Tween-20, 10 mmol / l EDTA, 2% glycerol, pH 7.5) were allowed to react with the 14G6 antibody for 16 h at 48C. After washing the wells, TRAP activity was disclosed in one set of wells by adding 200 ml p-nitrophenyl phosphate (pNPP) substrate in 100 mmol / l Na acetate / 50 mmol / l Na tartrate buffer, pH 5.5 and incubating for 60 min at 378C. The reaction was stopped and the yellow color developed by addition of 50 ml of 3 N NaOH. Absorbence was read at 405 nm. TRAP activity was expressed as mmol substrate hydrolyzed / min / l serum sample (U / l). TRAP protein was detected in the replicate wells by adding 100 ml of a second anti-TRAP monoclonal antibody (J1B conjugated to horseradish peroxidase kindly provided by Dr J. Halleen, University of Turku, Finland) at 25 ng / well and incubating for 60 min at room temperature. After washing the wells, peroxidase activity was disclosed by the addition of 200 ml of substrate (4 mg ortho-phenylenediamine dihydrochloride and 40 ml 3% H 2 O 2 in 10 ml of 25 mmol / l Na citrate / 50 mmol / l Na phosphate buffer, pH 5.0). The reaction was developed for 15 min at room temperature then stopped with 50 ml of 2 mol / l H 2 SO 4 . The absorbance was read at 490 nm. TRAP protein was expressed as mg / l serum sample. The specific activity of TRAP in each serum was calculated from the results of simultaneous immunoassay for activity and protein and expressed as Units / mg TRAP.

2.3. Non-denaturing gel electrophoresis Serum samples (50 ml) were subjected to non-denaturing polyacrylamide gel electrophoresis (PAGE) in 5 3 75 mm tube gels according to previously

150

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

published methods [17]. To increase the separation of isoforms 5a and 5b, electrophoresis time was increased from 75 to 90 min. After electrophoresis in a buffer of 35 mmol / l b-alanine and 52 mmol / l acetic acid, pH 4.0, the gels were immediately stained for TRAP activity in a buffer of 100 mmol / l Na acetate, 50 mmol / l Na tartrate, pH 5.5 containing 0.25 mg / ml naphthol-ASBI phosphate and 0.025 mg / ml Fast Garnet GBC. Gels were stained at room temperature for 18 h after which the staining solution was replaced with 6% acetic acid to stop the reaction and preserve the gels. Gels were scanned with a ScanMaker III transilluminating scanner (Microtek Labs, Inc., Redondo Beach, CA) to generate digital images for densitometric quantitation of band intensities. Amounts of TRAP 5a and 5b were expressed as band volume; a product of optical density and cross-sectional band area. Quantitation was achieved using BioImage 2.1 software (B.I. Systems Corporation, Ann Arbor, MI) and a manufacturer supplied calibration step tablet.

2.4. Other biochemical tests Intact parathyroid hormone (iPTH) concentration and total alkaline phosphatase (TAP) activities were determined from separate serum samples drawn before hemodialysis as a part of the routine clinical evaluation of ESRD patients. These tests were performed by a reference laboratory (Laboratory Corporation of America, Louisville, KY) and interpreted according to their reference ranges defined as the 95% confidence interval about the geometric mean of the normal population aged 25 years and over. iPTH assay is a chemiluminescent two-site immunoassay (Immulite, Diagnostic Products Corp., Los Angeles, CA) with a reference range of 12–72 pg / ml. TAP assay is a colorimetric kinetic assay (Olympus America, Inc., Melville, NY) with a reference range of 25–165 IU. Bone-specific alkaline phosphatase (BAP) and serum free pyridinoline (Pyd) assays were performed by immunoassay kit procedures (Alkphase B and Serum Pyd, Metra Biosystems, Inc., Mountain View, CA) according to manufacturer directions. Normal mean and S.D. were determined from our control group to be 18.865.0 IU BAP and 1.3060.42 nmol / l Pyd. Serum N-telopeptides of type-I collagen (NTx) assays were performed by an immunoassay kit procedure (Osteomark, Ostex, International, Seattle, WA) according to manufacturer procedure. Normal mean and S.D. was determined from our control group to be 16.465.7 nmol / l BCE.

3. Results

3.1. TRAP concentrations in healthy, RA and ESRD sera ( Table 1) Type-5 TRAP activity was significantly increased only in the ESRD sera.

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

151

Table 1 Serum type-5 TRAP activity and protein concentrations in healthy controls and in rheumatoid arthritis (RA) and endstage renal disease (ESRD) patients Disease (n)

Healthy (24) RA (24) ESRD (24)

Immunoassay Activity a (U / l)

Protein a (mg / l)

Specific activity b (U / mg)

3.0560.55 2.6660.87 4.3862.33*

38.1616.7 77.8637.1*** 79.6627.9***

0.083 (0.04–0.37) 0.046 (0.02–0.10)*** 0.057 (0.02–0.84)*

a

Activity and protein concentrations expressed as mean6S.D. Specific activity expressed as median (range) (ESRD group was non-Gaussian). Comparisons among mean TRAP activity and protein performed by t-test with Welch’s correction for unequal variance. Comparisons among median specific activity done by Mann–Whitney tests. * P , 0.01; *** P , 0.0001 compared to healthy control group. b

Immunoassay for type-5 TRAP protein, on the other hand, revealed both ESRD and RA sera to have significantly increased TRAP. As a result, the calculated specific activities of TRAP defined by immunoassay results in ESRD and RA sera were significantly decreased compared to that of healthy sera.

3.2. Polyacrylamide gel electrophoresis Fig. 1 illustrates the electrophoretic separation of 5a from 5b isoforms stained for TRAP activity in representative sera from healthy controls, RA and ESRD cohorts. Typically, the activity of band 5a predominated in individual healthy and RA sera whereas activity of band 5b predominated in individual ESRD sera. Table 2 summarizes the quantitative data derived from image analysis of gels. There were no significant differences among the cohorts in mean isoform 5a band volumes. One serum each from normal and ESRD groups failed to show isoform 5b staining; five RA sera failed to show isoform 5b staining. Of the remaining sera in which isoform 5b was present, only the ESRD group had significantly increased band 5b volume. In healthy control sera, the mean ratio of 5a:5b was 3.18. In both RA and ESRD sera, the mean 5a:5b ratios were significantly reduced due to relative increases in the amounts of osteoclastic TRAP 5b. Furthermore, the 5a:5b ratio in ESRD sera was significantly lower than that in RA sera (P , 0.05). Levels of serum TRAP activity correlated significantly with both isoforms 5a and 5b volume by PAGE in control and RA sera, but only with osteoclastic 5b in ESRD sera (Table 3). Serum TRAP protein levels did not correlate with PAGE assays for isoforms 5a or 5b in any cohort. To determine if the proportion of isoform band volumes were disease related, sera were placed into tertiles based on calculated 5a:5b band volumes of . 2.0, # 2.0, # 1.0 to reflect increasing proportion of osteoclastic TRAP 5b (Table 4). Most normal sera had ratios of . 2 while none had ratios , 1. Most RA sera

152

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

Fig. 1. Non-denaturing polyacrylamide gel electrophoresis of sera from four representative cases of healthy controls, rheumatoid arthritis and endstage renal disease. Gels were stained for TRAP activity using naphthol ASBI-phosphate as substrate and fast garnet GBC as coupler at pH 5.5. Note that isoform 5a is predominant in normal and RA sera and the relative band intensities are similar. In contrast, isoform 5b is predominant in ESRD sera and shows increased intensity.

Table 2 TRAP isoform composition in sera from healthy subjects and in rheumatoid arthritis (RA) and endstage renal disease (ESRD) patients a Disease (n)b

Healthy (24) RA (24) ESRD (24) a

Band volume

5a:5b ratio

Isoform 5a

Isoform 5b

0.6860.22 0.5760.26 0.7660.25

0.2360.12 0.3160.30 0.7360.40*

3.1861.51 1.7760.78* 1.1960.64*

Band volumes and isoform ratios expressed as mean6S.D. * P , 0.001 compared to healthy control group. b While all 24 gels in each cohort showed band 5a staining, 23 from the normal and HD group showed band 5b staining and 19 from the RA group showed band 5b staining. Only these gels with 5b staining were used in calculations of band 5b volumes and 5a:5b ratios.

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

153

Table 3 Correlation of serum TRAP activity and protein to TRAP isoform band volumes a Control

TRAP activity TRAP protein a

RA

ESRD

5a

5b

5a

5b

5a

5b

0.44* 0.29

0.51* 0.31

0.46* 0.13

0.46* 0.00

0.35 0.36

0.46* 0.04

Pearson product moment correlation coefficients, r. * P , 0.05.

Table 4 Relationship between TRAP isoform 5a:5b ratio and disease a 5a:5b ratio

Disease

. 2.0 # 2.0 # 1.0 a

Normal

RA

ESRD

18 5 0

6 11 2

3 8 12

Numbers are cases within each tertile of calculated 5a:5b band volume ratio.

had ratios , 2 and . 1 indicating a proportional increase of osteoclastic 5b, but still a 5a predominance. In contrast, most ESRD sera had 5a:5b ratios of , 1 while only three had ratios . 2 indicating a predominance of 5b in this cohort. These findings indicated that since PAGE analysis of TRAP isoforms depends upon enzymatic activity, isoform band intensities reflect results of TRAP activity assays, but may not be proportionately increased in relation to TRAP protein assays.

3.3. Correlation of TRAP to markers of bone metabolism Mean BAP (31.869.7 IU) was significantly increased over control (P , 0.0001) whereas mean NTx (16.563.9 nmol / l BCE) was normal in RA sera. These bone markers correlated significantly with TRAP activity in both control and RA subjects (Table 5). TRAP protein concentrations in both control and RA Table 5 Correlation of serum TRAP activity and protein to bone metabolic markers in control and RA sera a Control

TRAP activity TRAP protein a

RA

BAP

NTx

BAP

NTx

0.50* 0.38

0.50* 0.16

0.62** 2 0.15

0.47* 0.30

Pearson product moment correlation coefficients, r. * P , 0.05; ** P , 0.01.

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

154

Fig. 2. Sample distributions and mean TRAP activity and protein in RA sera that were either rheumatoid factor positive (RF 1 ) or rheumatoid factor negative (RF 2 ). No significant differences were obtained although a trend toward higher TRAP in RF 1 sera existed.

sera however, did not correlate with BAP or NTx. This suggests that the elevated TRAP protein in RA is not of osteoclastic origin and may not be related to bone turnover. RA is a chronic autoimmune disease with periodic exacerbations of acute inflammation. Neither TRAP activity nor protein concentrations correlated with the erythrocyte sedimentation rate (data not shown). When patients were divided according to rheumatoid factor (RF) status (Fig. 2), there was a tendency for TRAP to be higher in RF 1 sera compared to RF 2 sera, but statistical significance was not achieved in this small group of subjects. The activity of TAP, and concentrations of iPTH and Pyd were above the upper limit of normal in 5 / 23, 16 / 23 and 23 / 23 ESRD sera, respectively. Serum TRAP activity and protein concentrations were positively correlated to all of these markers of bone metabolism (Table 6) suggesting that the elevated TRAP in ESRD is osteoclast derived and associated with increased bone turnover. Although immunoassay of serum TRAP activity correlated with PAGE assay of TRAP 5b and, separately, with markers of bone metabolism in ESRD, direct correlations between TRAP 5b by PAGE assay and bone markers were not statistically significant.

4. Discussion Serum type-5 TRAP has long been considered a biomarker for osteoclastic Table 6 Correlation of TRAP activity and protein to bone metabolic markers in ESRD sera a

TRAP activity TRAP protein a

TAP

iPTH

Pyd

0.60* 0.49*

0.49** 0.45*

0.45* 0.44*

Pearson product moment correlation coefficients; r. * P , 0.05; ** P , 0.01.

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

155

activity. It is thought by many investigators to be a single entity. Actually, type-5 TRAP in serum is composed of two isoforms, 5a and 5b, defined by non-denaturing PAGE [13]. While only 5b is osteoclastic and related to bone resorption [15,16], immunoassays do not discriminate the isoforms, but instead measure both 5a and 5b. Furthermore, TRAP may circulate as active and inactive forms [18]. Failure to take these facts into consideration may undermine the specificity and sensitivity of TRAP immunoassays as measures of bone resorption rate. Using simultaneous immunoassays for TRAP activity and protein concentration, we previously determined that sera from many patients with ESRD and RA had elevated TRAP [12]. However, both TRAP activity and protein were high in ESRD, whereas only TRAP protein was high in RA. We postulate that differences in the nature of serum TRAP in these diseases is due to differences in the proportion of isoforms expressed. In this study, we compared TRAP activity and protein concentrations determined by immunoassay to isoform 5a and 5b band volume revealed by histochemical staining of PAGE gels to identify the predominant TRAP isoforms in sera of ESRD and RA patients. Also we sought to correlate serum TRAP activity and protein with isoform expression and with other markers of bone metabolism in these diseases. Lam et al. [19] had previously shown that biochemical assay of TRAP activity with pNPP correlated closely with densitometric assay of band 5b stain intensity in gels using a similar PAGE system employing a-naphthyl phosphate as the substrate. We modified the PAGE method slightly by increasing the electrophoresis time from 75 to 90 min thereby enhancing the separation of 5a from 5b. We found also that tube gels must be used to obtain consistent migration and resolution of activity bands. Polyacrylamide slab gels were unsatisfactory because there are no physical boundaries to separate the sample lanes. The large volume of whole serum needed for the technique caused severe lane distortions, inconsistent migration and poor resolution in slab gels. The increased TRAP activity in ESRD sera was attributed to increased osteoclastic isoform 5b as defined by PAGE. Furthermore, both TRAP activity and protein were correlated to other markers of bone turnover in ESRD. Nevertheless, the specific activity of serum TRAP, as defined by the ratio of activity / protein in immunoassays, was subnormal in our cohort of ESRD sera suggesting that some of the circulating TRAP 5b was inactive. Serum TRAP activity in RA correlated with BAP and NTx indicating its relationship to bone metabolism. Unfortunately, the isoform contributing to the increased TRAP protein in RA sera could not be identified with certainty because PAGE analysis depends upon enzymatic activity. Although equivalent amounts of TRAP protein were present in RA and ESRD, a proportional increase in TRAP activity and 5b band volume were only seen in ESRD. Our finding of increased osteoclastic TRAP 5b activity in ESRD is consistent with previous observations that serum

156

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

TRAP activity is increased in chronic renal failure [4,5] and that increased bone resorption occurs in a select group of ESRD patients who develop secondary hyperparathyroidism [20]. While serum TRAP activity and protein concentrations in ESRD correlated with iPTH, TAP and Pyd, significant associations between these bone markers and the amount of TRAP 5b as determined by PAGE could not be established. Rheumatoid arthritis is a chronic inflammatory disease that can be associated with increased bone resorption locally and systemically [21]. Activated macrophages expressing TRAP are abundant in the synovial tissues of affected joints [22] which could be an additional rich source of circulating TRAP. The TRAP activity in RA correlated with both isoform 5a and 5b. Also, the TRAP activity, but not protein, correlated with BAP and NTx indicating that TRAP activity in RA is related to bone turnover. Much of the elevated TRAP protein in RA could be a low activity 5a isoform of a source other than osteoclasts, possibly derived from inflammatory macrophages. We were unable to find, however, a statistically significant relationship in these few cases between TRAP and ESR, a marker of acute inflammation, or RF status, a marker of chronic disease activity. Improved methods have been recently developed for specific quantitation of isoform 5b. The method of Halleen et al. [16], like the one described here, uses specific antibody to immobilize type-5 TRAP. By increasing the pH of the enzymic reaction to 6.1, near the optimum of 5b and well beyond the optimum of 5a, the method measures selectively 5b activity. This method reveals a significant negative correlation between TRAP and BMD in post-menopausal women [23] and a decline in TRAP activity in accordance with a clinical response to hormone replacement therapy. The method of Nakanishi et al. [24] is a kinetic assay in which fluoride is used as a specific inhibitor of types-5a and 5b TRAP and heparin as specific inhibitor of TRAP 5a. The difference between fluoride-resistant and fluoride-sensitive serum TRAP activity measured in the presence of heparin yields the calculated 5b activity. The method shows the predicted age-related differences in mean TRAP activity, but has not been tested for clinical sensitivity and specificity in metabolic bone diseases. A direct comparison of the two assays is warranted. Inactive or low-activity isoforms such as 5a in RA sera would be difficult to quantitate unequivocally by activity-based immunoassay or PAGE. Further study including isolation of TRAP 5a and 5b isoforms from normal and abnormal sera for independent analysis of their biochemical properties is needed to identify the nature of increased TRAP in disease. More detailed information on the properties and origin of the two closely related TRAP isoforms will help to identify specific clinical applications for immunoassay of TRAP protein and increase the specificity and sensitivity of TRAP activity as a marker of bone resorption.

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

157

Acknowledgements This work was supported by grants from the Research Service of the US Department of Veterans Affairs, Washington, DC (AJJ) and from the Research Foundation of Louisville (AJJ and LTY) and the Jewish Hospital Foundation (LTY), Louisville, KY.

References [1] Tavassoli M, Rizo M, Yam LT. Elevation of serum acid phosphatase in cancer with bone metastasis. Cancer 1980;45:2400–3. [2] Lau K-HW, 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. [3] Stepan JJ, Silinkova-Malkova E, Havranek T et al. Relationship of plasma tartrate resistant acid phosphatase to the bone isoenzyme of serum alkaline phosphatase in hyperparathyroidism. Clin Chim Acta 1983;133:189–200. [4] Scarnecchia L, Minisola S, Pacitti MT et al. Clinical usefulness of serum tartrate-resistant acid phosphatase activity determination to evaluate bone turnover. Scand J Clin Lab Invest 1991;51:517–24. [5] Gavilanes EL, Parra EG, de la Piedra C, Caramelo C, Rapado A. Clinical usefulness of serum carboxyterminal propeptide of procollagen I and tartrate-resistant acid phosphatase determinations to evaluate bone turnover in patients with chronic renal failure. Miner Electrolyte Metab 1994;20:259–64. [6] Lam K-W, Siemens M, Sun T, Li C-Y, Yam LT. Enzyme immunoassay of tartrate-resistant acid phosphatase. Clin Chem 1982;28:467–70. [7] Kraenzlin ME, Lau K-HW, Liang L et al. Development of an immunoassay for human serum osteoclastic tartrate-resistant acid phosphatase. J Clin Endocrinol Metab 1990;71:442–51. [8] Chamberlain P, Compston J, Cox TM et al. Generation and characterization of monoclonal antibodies to human type-5 tartrate resistant acid phosphatase: development of a specific immunoassay of the isoenzyme in serum. Clin Chem 1995;41:1495–9. [9] Cheung CK, Panesar NS, Haines C, Masarei J, Swaminathan R. Immunoassay of a tartrate-resistant acid phosphatase in serum. Clin Chem 1995;41:679–86. [10] Halleen J, Hentunen TA, Hellman J, Vaananen HK. Tartrate-resistant acid phosphatase from human bone: purification and development of an immunoassay. J Bone Miner Res 1996;11:1444–52. [11] Halleen JM, Hentunen TA, Karp M, Kakonen S-M, Pettersson K, Vaananen HK. Characterization of serum tartrate-resistant acid phosphatase and development of a direct two-site immunoassay. J Bone Miner Res 1998;13:683–7. [12] Nakasato YR, Janckila AJ, Halleen JM, Vaananen HK, Walton SP, Yam LT. Clinical significance of immunoassays for type-5 tartrate-resistant acid phosphatase. Clin Chem 1999;45:2150–7. [13] Lam K-W, Eastlund DT, Li C-Y, Yam LT. Biochemical properties of tartrate-resistant acid phosphatase in serum of adults and children. Clin Chem 1978;24:1105–8.

158

K. Takahashi et al. / Clinica Chimica Acta 301 (2000) 147 – 158

[14] Lam K-W, Li C-Y, Yam LT, Desnick RJ. Comparison of the tartrate-resistant acid phosphatase in Gaucher’s disease and leukemic reticuloendotheliosis. Clin Biochem 1981;14:177–81. [15] Lam K-W, Lee P, Li C-Y, Yam LT. Immunological and biochemical evidence for identity of tartrate-resistant isoenzymes from human serum and tissues. Clin Chem 1980;26:420–2. [16] Halleen JM, Alatalo SL, Suominen H, Cheng S, Janckila AJ, Vaananen HK. Tartrate-resistant acid phosphatase 5b, a novel serum marker of bone resorption. J Bone Miner Res 2000;15:1337–45. [17] Li C-Y, Yam LT, Lam K-W. Acid phosphatase isoenzymes in human leukocytes in normal and pathologic conditions. J Histochem Cytochem 1970;18:473–81. [18] Halleen JM, Kaija H, Stepan JJ, Vihko P, Vaananen HK. Studies on the protein tyrosine phosphatase activity of tartrate-resistant acid phosphatase. Arch Biochem Biophys 1998;352:97–102. [19] Lam WK-W, Lai LC, Yam LT. Tartrate-resistant (band 5) acid phosphatase activity measured by electrophoresis on acrylamide gel. Clin Chem 1978;24:309–12. [20] Hruska KA, Teitelbaum SL. Renal osteodystrophy. N Engl J Med 1995;333:166–74. [21] Joffe I, Epstein S. Osteoporosis associated with rheumatoid arthritis: pathogenesis and management. Semin Arthritis Rheum 1991;20:256–72. [22] Gravallese EM, Harada Y, Wang J-T, Gorn AH, Thornhill TS, Goldring SR. Identification of cell types responsible for bone resorption in rheumatoid arthritis and juvenile rheumatoid arthritis. Am J Pathol 1998;152:943–51. [23] Halleen J, Alatalo S, Suominen H, Cheng S, Vaananen K. Use of tartrate-resistant acid phosphatase 5b in minitoring antiresorptive treatment: comparison with other markers of bone turnover. Calc Tiss Int 2000;66(Suppl 1):585. [24] Nakanishi M, Yoh K, Miura T, Ohasi T, Rai SK, Uchida K. Development of a kinetic assay for band 5b tartrate-resistant acid phosphatase activity in serum. Clin Chem 2000;46:469–73.