Development of immunoassays for serum tartrate-resistant acid phosphatase isoform 5a

Clinica Chimica Acta 359 (2005) 132 – 140 www.elsevier.com/locate/clinchim

Development of immunoassays for serum tartrate-resistant acid phosphatase isoform 5aB Tsu-Yi Chaoa, Su-Huei Leea, Mary M. Chena, David H. Neustadtb, Uzma A. Chaudhryc, Lung T. Yamb,c, Anthony J. Janckilac,d,* a

Division of Hematology and Oncology, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b Department of Medicine, University of Louisville, School of Medicine, Louisville, KY, USA c Special Hematology Laboratory, Department of Veterans Affairs Medical Center, Louisville, KY, USA d Department of Microbiology and Immunology, University of Louisville, School of Medicine, Louisville, KY, USA Received 28 October 2004; received in revised form 15 March 2005; accepted 21 March 2005 Available online 1 July 2005

Abstract Background: Serum tartrate-resistant acid phosphatase (TRACP) consists of 2 structurally related isoforms, TRACP 5a and 5b. TRACP 5b is from bone-resorbing osteoclasts. TRACP 5a may be a macrophage product of inflammation. We used a novel antibody to TRACP 5a to standardize immunoassays for serum TRACP 5a activity and protein. Methods: Biotinylated anti-TRACP antibodies were used to immobilize serum TRACP isoforms. TRACP activity was measured using 4-nitrophenyl phosphate as substrate. TRACP 5a protein was measured with an independent peroxidaseconjugated anti-TRACP antibody. Immunoassays were standardized for linearity of serum dose response, sensitivity and precision. Reference ranges for TRACP 5a were established from serum of 50 healthy males and 50 healthy age-matched females. Serum TRACP 5a activity and protein were determined in 29 cases of rheumatoid arthritis. Results: Serum matrix interference in both TRACP 5a assays required dilution to 10% serum to approach linearity. Intra-assay and inter-assay CV% were b10%. Mean serum TRACP 5a activity and protein were significantly higher in healthy men than women. There was a slight, but significant age related increase in both serum TRACP 5a and 5b among females, but not males, from age 20 to 70 years. TRACP 5a activity was positively correlated to TRACP 5a protein in healthy sera. Neither TRACP 5a activity nor protein was correlated strongly to TRACP-5b activity. TRACP 5a protein was significantly increased in 8/29 RA sera, whereas TRACP 5a and 5b activities were not. TRACP 5a activity and protein were not significantly correlated in RA sera.

B Supported by grants from the National Science Council of Taiwan (NSC-92-2314-B-016-024; T-YC), the Research Service of the Department of Veterans Affairs (AJJ) and the Clinical Research Foundation, Inc. (LTY). * Corresponding author. Veterans Affairs Medical Center, 800 Zorn Ave., Louisville, KY 40206, United States. Tel.: +1 502 895 3401; fax: +1 502 287 6205. E-mail address: [email protected] (A.J. Janckila).

0009-8981/$ - see front matter. Published by Elsevier B.V. doi:10.1016/j.cccn.2005.03.039

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

133

Conclusions: Although TRACP 5a and 5b are related biosynthetically, their circulating levels in healthy humans were independent, suggesting differential regulation of expression. In chronic diseases, increased TRACP 5a may represent pathological processes of inflammation unrelated to bone metabolism. Published by Elsevier B.V. Keywords: Acid phosphatase; Immunoassay; Inflammation; Bone resorption

1. Introduction

2. Materials and methods

Serum tartrate-resistant acid phosphatase (TRACP) consists of 2 isoforms, 5a and 5b, derived from a common gene product by differential post-translational modification. Serum TRACP 5a is an intact glycoprotein of 33 kDa with sialic acid in the asparagine-linked oligosaccharides [1]. Serum TRACP 5b is a proteolytically cleaved form with disulfide-linked polypeptide subunits of 16 kDa and 23 kDa, which has no sialic acid in its oligosaccharides [2]. Isoform 5a has a pH optimum of 5.2, while isoform 5b has a pH optimum of 5.8. TRACP 5b is derived from osteoclasts [3]. The source of TRACP 5a is not entirely clear, but it may derive from macrophages or dendritic cells [4]. Immunoassays selective for osteoclastic TRACP 5b are now well established and clinically relevant [5–8]. Serum TRACP 5b activity correlates to the number of osteoclasts and responds to anti-resorptive treatment in post-menopausal women [5,9]. Recently, it has also been determined that elevated TRACP 5b activity has predictive value for future fracture in elderly women [10]. Therefore it may be a useful biochemical marker for clinical assessment of many patients with various metabolic bone diseases. The clinical significance of serum TRACP 5a has not been determined, however about one-third of patients with established rheumatoid arthritis (RA) display significantly increased levels of total serum TRACP protein of low activity [4]. We believe this to represent mostly TRACP 5a because evidence from our laboratories has shown that 80–90% of circulating TRACP protein is isoform 5a. Because of the potential for TRACP 5a as a clinical disease marker, we set out to establish TRACP 5a assays using a newly developed antibody, (Ab 220), which reacts specifically with the trypsin sensitive epitope in the loop peptide of uncleaved TRACP isoform 5a [11].

2.1. Human subjects and TRACP sources To standardize assay conditions and validate assay performance, purified recombinant TRACP 5a produced by CHO cells was used as analyte [12], diluted into either normal serum as assay matrix or into assay buffer containing 1% bovine serum albumin (BSA). To establish a reference range and assess the effects of age and gender on serum TRACP 5a, a cohort of sera from 100 healthy consenting adults aged 20 to 70 years (50 males and 50 age-matched females) was used. Healthy study subjects were recruited at the Tri-Service General Hospital (TSGH), Taipei, Taiwan during the course of other related studies of TRACP. Waste specimens of serum from 29 informed, consenting adult patients with established rheumatoid arthritis (RA) were obtained after they had served their purpose for clinical evaluation by one of us (DHN). As control for the sera obtained in Louisville, a blood specimen was drawn from 25 informed, consenting healthy adult subjects among staff of the VA Medical Center, Louisville, KY. These studies were approved by the Human Studies Internal Review Boards of the VAMC, Louisville, KY and the TSGH, Taipei, Taiwan. 2.2. Antibodies Anti-TRACP antibody 14G6 (Ab 14G6) is a welldescribed antibody used for immunoassays of TRACP activity and protein [13]. It reacts with both isoforms 5a and 5b and is used as a biotinylated conjugate to capture total TRACP protein. Anti-TRACP antibody 220 (Ab 220) is a recently described antibody developed against intact recombinant TRACP secreted by CHO cells [11]. It is specific for the trypsin-sensitive epitope within the repression loop peptide of the intact TRACP molecule and reacts only with serum isoform

134

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

5a and not with serum isoform 5b. Antibody 220 was used as a biotin conjugate to capture isoform 5a in serum. Anti-TRACP antibody J1B (Ab J1B) is a welldescribed antibody that reacts with both isoforms 5a and 5b. Ab J1B was kindly provided by Dr. Jussi Halleen, Pharmatest, Oy, Finland. Ab J1B was used as a horseradish peroxidase (HRP) conjugate in assays for total TRACP protein and isoform 5a protein. 2.3. Calibrators TRACP activity assays were calibrated using a commercial solution of paranitrophenol (Sigma Chemical Co.) serially diluted to be equivalent to 10 IU to 0.3 IU defined as Amol paranitrophenol liberated/ minute/liter sample at the defined assay conditions. Immunoassays for TRACP protein were calibrated with the rTRACP 5a protein [11]. Serial dilutions of rTRACP calibrator were used from 5 ng/ml to 0.08 ng/ml. Quadratic curve fit was used for calibration of both assays. 2.4. Immunoassays for TRACP isoforms Immunoassays for TRACP isoform 5b activity were conducted according to published procedures [13] using Ab 14G6 to capture circulating TRACP. Bound activity was measured at pH 6.1 to maximize the contribution by isoform 5b and minimize the contribution by isoform 5a [5]. Immunoassays for total TRACP protein were conducted as described previously [13]. Antibody 14G6 was used to capture circulating TRACP isoforms. The bound TRACP protein was disclosed using Ab J1B-HRP conjugate. Immunoassay for TRACP 5a activity was conducted similar to that for TRACP 5b activity. Neutravidin-coated wells (Pierce Chemical Co.) were washed three times with phosphate-buffered saline. The wells were then coated with biotinylated antibody 220 at a concentration of 1 Ag/well in 100 Al of PBS for 2 h. Wells were then washed three times with PBS. Ninety microliters of immunoassay buffer (20 mmol/l TRIS pH 7.4 containing 150 mmol/l NaCl, 10 mmol/l EDTA, 2% glycerol, 0.05% Tween-20) were added to the wells. To this was added 10 Al of serum sample. Sample was incubated overnight at 4 8C. The next day, wells were washed four times with TRIS buffered saline containing 0.05% Tween-20. Two hundred

microliters of substrate solution (2 mg pNPP/ml of 100 mmol/l sodium acetate/50 mmol/l sodium tartrate buffer, pH 5.8) were then added. The reaction was allowed to proceed for 60 min at 37 8C after which the reaction was stopped by addition of 50 Al of 1 mol/l NaOH. The absorbance was read with a BioRad Microplate 550 Reader at A405 and serum TRACP 5a activity was calculated from the standard curve and expressed as U/l sample. Immunoassay for TRACP isoform 5a protein was conducted similar to that for total TRACP protein. NeutrAvidin wells were coated with biotinylated antibody 220 at 1 Ag/well as described above. Ninety microliters of immunoassay buffer were introduced into the wells to which was added 10 Al of serum sample. Calibrators were simultaneously added to provide a standard curve from 0.08 to 5 ng/ml. After overnight incubation at 4 8C, the wells were washed four times with phosphate buffered saline containing 0.05% Tween-20 (PBST). One hundred microliters of Ab J1B-HRP, diluted 1:2000, were introduced into each well and incubated 60 min at room temperature after which wells were again washed four times with PBST. Two hundred microliters of substrate consisting of 0.4 mg/ml ortho-phenylenediamine and 4 Al/ml 3% H2O2 in a buffer of 25 mmol/l citric acid/50 mmol/l sodium phosphate, pH 5.0 was then added. The reaction was allowed to proceed for exactly 15 min at room temperature. The reaction was stopped by the addition of 50 Al 2 mol/l H2SO4. Absorbance was measured in a BioRad 550 Microplate Reader at a wavelength of 490 nm and results calculated from the standard curve and expressed as Ag TRACP 5a/l sample. 2.5. Statistics Linear regression analysis was used to assess the change in serum TRACP isoform activity and protein with age for each gender. Student’s t-test was used to compare mean TRACP isoform activity and protein in the healthy male and female groups. Student’s t-test with Welch’s correction for unequal variance was used to compare mean TRACP protein levels in healthy and RA groups. Pearson’s product moment correlation for normally distributed data was used to assess the relationships among all TRACP isoform assays. The intra-assay imprecision (CV%) was estimated as the

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

Table 1 Performance characteristics of serum TRACP 5a immunoassays

mean CV% for duplicate measurements of 25 healthy sera and 29 RA sera. The inter-assay imprecision was estimated as the mean CV% for 6 daily singleton assays of 8 selected sera ranging in activity from 3.11 to 11.69 U/l and in protein from 3.11 to 15.19 Ag/l. The least detectable limit of TRACP 5a activity was defined as Amole/min/l paranitrophenylate at which the A405 was b3 S.D. above that generated by the zero standard. The least detectable limit of TRACP 5a protein was defined as the amount at which the A490 was greater than three S.D. above that generated by the zero standard.

Assay

CV%a intra-assay

CV% inter-assay

Least detectable limit

5a activity 5a protein

2.28 3.94

9.51 7.34

0.125 U/l 0.063 Ag/l

a Intraassay CV% was estimated as the mean CV% for duplicate measurements of 29 RA sera and 24 healthy sera.

to 6.25% (2nd to 3rd dilution). The degree of serum interference was variable among sera. Attempts were made to diminish the interference at high serum concentrations by manipulating the sample dilution buffer or the sample to improve antigen antibody interaction (data not shown). Increasing EDTA up to 50 mmol/l, addition of 0.05% triton X-100, or increasing ionic strength up to 2 mol/l NaCl in the sample dilution buffer all caused reduced binding. Adjusting the pH of the dilution buffer from 4.0 to 9.0 did not improve antigen binding. Preheating the serum to 37 8C for 60 min or 65 8C for 15 min only reduced the sensitivity. Hereafter, to minimize serum matrix effect and to provide the best overall estimate, 10 Al serum was simply diluted into 90 Al assay buffer (10%) for serum TRACP 5a activity assay. The intra-assay error, interassay error and least detectable limit for the TRACP 5a activity assay are listed in Table 1.

3. Results 3.1. Standardization of TRACP 5a activity assay The amount of Ab 220 required to bind maximum amount of analyte was 0.5 to 1.0 Ag/well. To establish the linearity of dose response, 25 ng/ml rTRACP 5a was spiked into enzyme immunoassay (EIA) buffer and serial 2-fold dilutions were made into EIA buffer. Fifty microliters of samples of these dilutions were added to 50 Al EIA buffer and subjected to immunoassay for TRACP activity. Also 2-fold serial dilutions of 8 sera (4 from normal subjects and 4 from RA patients) were subjected to immunoassay. Fig. 1A shows that the dose response for pure rTRACP is linear throughout the dilution range. However dose responses for some sera approached linearity only after they had been diluted to approximately 12.5%

3.2. Standardization of TRACP 5a protein assay The amount of Ab J1B-HRP conjugate to give the greatest dynamic range of absorbance for the stan4

1.5

Nl 1

B TRACP 5a (µg/L)

A TRACP 5a (U/L)

135

1.0

0.5

Nl 2

3

Nl 3 Nl 4

2

RA 1 RA 2

1

RA 3 RA 4

0.0 0

10

20

Serum dose (µL/well)

30

0 0

10

20

30

rTRAP

Serum dose (µL/well)

Fig. 1. Dose response of serum TRACP 5a activity (A) and serum TRACP 5a protein (B) immunoassays. Eight sera, 4 healthy (Nl) and 4 rheumatoid arthritis (RA) were assayed using from 25 Al (1/4 dilution) serum to 0.78 Al (1/128 dilution) serum sample in a total of 100 Al/well. The dose response for both assays was not linear for some samples, especially at doses N12.5 Al. Purified rTRACP diluted in 1% BSA gave a linear dose response in both assays.

136

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

Table 2 Mean (FS.D.) serum TRACP isoforms in healthy males and females from Taipei cohort Subject (#) Activity (U/l) TRACP 5a

Table 3 Correlations between serum TRACP isoforms in healthy men and women (Taipei cohort)

Protein (Ag/l) TRACP 5b TRACP 5a

Total TRACP

Male (50) 6.85 F 1.27* 2.53 F 0.50 5.89 F 1.65* 11.25 F 3.45* Female (50) 5.68 F 0.90 2.46 F 0.68 4.66 F 1.30 7.98 F 2.27

Total TRACP

5a protein

5a activity

Males (50) 5a protein 5a activity 5b activity

0.937*** 0.760*** 0.336*

0.796*** 0.307*

0.118

Females (50) 5a protein 5a activity 5b activity

0.909*** 0.765*** 0.346*

0.796*** 0.170

0.338*

* p b 0.0001 compared to female.

dard curve from 0.08–5 Ag/l TRACP 5a was a 1:2000 dilution of HRP conjugate. The linearity of the dose response was determined as for the TRACP 5a activity assay described above except that serial dilutions of rTRACP were started from 5 ng/ml. Fig. 1B shows that the dose responses were linear for some sera, but not others. In the latter instances approximate linearity was achieved only after the samples had been diluted to 12.5% to 6.25% (2nd to 3rd dilution). Therefore, as with the TRACP 5a activity assay, 10 Al of serum were added to 90 Al of EIA buffer for immunoassay of serum TRACP 5a protein. The intra-assay error, inter-assay error and

Pearson product moment correlation. ***p b 0.001, * p b 0.05.

least detectable limit for the TRACP 5a protein assay are listed in Table 1. 3.3. Serum TRACP isoforms in healthy men and women The mean (F S.D.) for TRACP isoforms in healthy men and women of the Taipei cohort are summarized

7. 5

30

5b; r = 0.52***; slope = 0.017 U/yr

5.0

2.5

0.0

20

30

40

5a; r = 0.37**; slope = 0.034 µg/L/yr

A TRACP (µg/L)

TRACP (U/L)

5a; r = 0.53***; slope = 0.025 U/yr

50

60

20

Total; r = 0.45**; slope = 0.072 µg/L/yr

10

0

70

20

30

Age (Female)

50

60

70

30

5a; r = 0.24 ns

TRACP (µg/L)

5.0

2.5

20

30

40

5a; r = 0.11 ns

C

5b; r = 0.10 ns

TRACP (U/L)

40

Age (Female)

7. 5

0.0

B

50

Age (Male)

60

70

D

Total; r = 0.11 ns 20

10

0

20

30

40

50

60

70

Age (Male)

Fig. 2. Influence of subject’s age on serum TRACP 5a and 5b activity (A, C) and TRACP 5a and total TRACP protein (B, D) in healthy agematched males and females from the Taipei cohort. ns; not significant, **p b 0.01, ***p b 0.0001.

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

in Table 2. Mean TRACP 5b activity was not significantly different between men and women, however mean TRACP 5a and total TRACP were significantly higher in men. Fig. 2 summarizes the relationships between age and serum TRACP isoforms in this group of men and women. There was a significant age related increase in both TRACP 5a and 5b in women (Fig. 2A, B), but not in men (Fig. 2C, D). Table 3 summarizes the calculated correlation coefficients among all TRACP isoform assays for normal male and female groups. From our previous experience, it was expected that TRACP 5a protein would correlate highly with total TRACP protein since TRACP 5a is the predominant isoform in serum. Interestingly, although the TRACP 5a protein levels correlated highly with total TRACP, the strength of the correlation between TRACP 5a activity and TRACP 5a protein was weaker than expected. This finding suggests that serum TRACP 5a may be activated to different degrees in individual subjects, either in vivo or in vitro. The correlations between TRACP 5b activity and TRACP 5a activity or protein were weak or non-significant in normal individuals, which is consistent with our hypothesis that TRACP 5b and TRACP 5a are derived from different cells and may reflect different metabolic processes. 3.4. Serum TRACP isoforms in RA Mean serum TRACP 5a and 5b activities in RA were no different than those of healthy sub-

Table 4 Correlations between serum TRACP isoforms in healthy and RA subjects (Louisville cohort) Total TRACP

5a protein

5a activity

Healthy (25) 5a protein 5a activity 5b activity

0.917*** 0.620*** 0.099

0.524** 0.200

0.066

RA (29) 5a protein 5a activity 5b activity

0.945*** 0.247 0.102

0.291 0.124

0.258

Pearson product moment correlation. ***p b 0.001, **p b 0.01, *p b 0.05.

jects (Fig. 3A), whereas serum TRACP 5a protein and total TRACP protein were significantly increased in RA (Fig. 3B). As with healthy subjects, there was no significant correlation between serum TRACP 5a and 5b activities in RA, but there was a strong correlation between measured total TRACP protein and TRACP 5a protein (Table 4). Unlike healthy subjects, however, measured TRACP 5a activity and 5a protein did not correlate significantly. When the 8 sera with high TRACP 5a protein were censored from the analysis, significant correlation between TRACP 5a activity and protein was achieved (r = 0.67; p b 0.001). Therefore, the abnormality in TRACP isoform expression in some RA subjects was due to elevated levels of inactive TRACP 5a protein.

10.0

40 ns

A

B TRACP (µg/L)

ns

TRACP (U/L)

137

7.5

5.0

2.5

30

p < 0.05

p < 0.01

20

10

0.0

0

5a Nl

5a RA 5b Nl 5b RA TRACP/Disease group

5a Nl

5a RA Tot Nl Tot RA TRACP/Disease group

Fig. 3. TRACP isoform 5a and 5b activity (A) and TRACP 5a and total TRACP protein (B) in healthy (Nl) and rheumatoid arthritis (RA) sera (Louisville cohort). Dashed line in B represents upper limit of normal (mean + 2S.D.) for TRACP 5a protein. Solid line in B represents upper limit of normal (mean + 2S.D.) for total TRACP protein.

138

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

4. Discussion We had previously advanced the hypothesis that serum TRACP 5a may be an inflammatory disease marker based on four previous experimental observations [4,6]. First, in a series of established RA patients, we found that approximately 30% of sera had increased levels of inactive TRACP protein, which did not correlate with TRACP 5b or other bone markers. Second, when a serum pool from RA patients was subjected to column chromatography, the area under the TRACP 5a peak was increased at least 2-fold over that of a normal serum pool, suggesting that increased TRACP protein in individual RA sera was due to isoform 5a with low activity. Third, we have shown that TRACP 5a with low specific activity is selectively secreted from human macrophages and dendritic cells in vitro. Finally, the serum TRACP protein assays in RA correlated weakly, but significantly, to C-reactive protein. Although we were able to determine that most of the circulating TRACP protein in RA and normal serum was due to low activity isoform 5a, there were no isoform 5a antibodies available to specifically measure TRACP 5a in individual sera to confirm our hypothesis. In this study, we describe 2 new immunoassays for TRACP 5a activity and protein using a unique antibody specific to an epitope in the loop peptide present only on serum TRACP 5a. Binding of TRACP 5a by Ab 220 causes an up-shift in the pH optimum and specific activity to that of TRACP 5b [11]. Both assays are near linear and have good performance characteristics with b10% analytical error, making them suitable for investigations of their clinical significance. The analytical error is relatively high due to the need to make 10-fold dilutions of serum to achieve a near linear dose response. Also, not all sera deviated from linearity to the same degree. All attempts to relieve high-dose serum matrix interference with antigen–antibody interaction were ineffective including increasing ionic strength, increasing or decreasing pH, increasing divalent cation chelator concentration, mild heat denaturation, or addition of nonionic detergent. Among normal adults, there was an age related increase in both serum TRACP isoforms 5a and 5b in women but there was no effect of age in men. An age-related increase in osteoclastic TRACP 5b activ-

ity in women is expected since more subjects will have osteoporosis in older age groups. Why TRACP 5a activity would increase with age in women is not entirely clear at present. It is possible that osteoclasts secrete some TRACP 5a as well as TRACP 5b, becoming detectable during increased rate of bone resorption only because it is activated artificially by binding to Ab 220 in immunoassay. When all ages were combined, mean TRACP 5a was significantly higher in men but mean TRACP 5b was not. TRACP isoforms 5a and 5b are derived from the same gene and probably the same precursor protein and they are both expressed by monocyte-derived cells, perhaps simultaneously. Yet they appear to be unrelated in their relative proportions in serum and in their disease associations. TRACP 5a activity and protein correlated only weakly or not at all with TRACP 5b activity in normal and RA sera suggesting that they may be derived from different cell types and might participate in different metabolic processes aside from the caveat noted above. As expected, TRACP 5a protein correlated strongly with total TRACP protein. We had previously reported that when TRACP isoforms in pooled sera were partially purified by chromatography and assayed with a common capture antibody, 85–90% of circulating TRACP was the isoform 5a [6]. In this study, the measured amount of TRACP 5a protein in the Taipei cohort accounted for an average of only 52% and 58% of the total TRACP measured in male and female sera respectively, whereas in the Louisville cohorts of healthy and RA sera, TRACP 5a protein accounted for 75% to 79% of the total TRACP protein. This discrepancy may be the result of the relatively large CV% and lack of perfect linearity for TRACP 5a protein assays. This error could contribute to inter-laboratory differences in measured TRACP. Although Ab 220 used in both assays captures the same entity at a defined epitope specific for isoform 5a, the correlation between TRACP 5a activity and protein was weaker than we expected. Indeed, the correlation between TRACP 5a activity and protein was not significant in RA sera. Only when the individual sera with elevated TRACP 5a protein were censored from the analysis, did the correlation become significant. This suggested to us that TRACP 5a itself might circulate as active and inactive forms, the proportions of which may be vari-

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

able. There is literature from several laboratories documenting an increase in specific activity when monomeric or uncleaved TRACP is cleaved with a variety of specific proteases [14–16]. This is not likely to be the sole mechanism for the difference between active and inactive TRACP 5a in our study since Ab 220 reacts with the trypsin sensitive loop peptide of intact monomeric TRACP. The valency state of the di-iron center in TRACP is another factor that contributes to the specific activity of the enzyme [17]. The purple diferric form is inactive, whereas the mixed valency pink form is active. We have previously separated two forms of TRACP 5b from hairy cell leukemia spleen, which had different specific activities, the former (peak 1) having about half as much activity as the latter (peak 2) [18]. Kaija et al. [19] reported similar results with osteoclastic TRACP 5b and showed that the earlier, less active peak had more purple diferric TRACP than the latter peak, and suggested that the difference in activity was due to different proportions of purple to pink enzyme. Therefore, differences in the relative proportions of purple to pink TRACP 5a may explain our findings. Other possible explanations are that inactive TRACP 5a could be an immature proenzyme that has not been fully processed post-translationally to manifest maximum activity. Alternatively, the inactive enzyme could be a senescent protein that has been irreversibly reduced and is undergoing the process of inactivation and catabolism [20].

Acknowledgements

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

The authors gratefully acknowledge Ms. Laura Wood for her dedicated technical assistance. [14]

References [1] Lam KW, Li CY, Yam LT, Desnick RJ. Comparison of the tartrate-resistant acid phosphatase in Gaucher’s disease and leukemic reticuloendotheliosis. Clin Biochem 1981;14: 177 – 81. [2] Janckila AJ, Nakasato YR, Neustadt DH, Yam LT. Diseasespecific expression of tartrate-resistant acid phosphatase isoforms. J Bone Miner Res 2003;18:1916 – 9. [3] Halleen JM, Ylipahkala H, Alatalo SL, Janckila AJ, Heikkinen JE, Suominen H, et al. Serum tartrate-resistant acid phospha-

[15]

[16]

[17]

139

tase 5b, but not 5a, correlates with other markers of bone turnover and bone mineral density. Calcif Tiss Int 2002;71: 20 – 5. Janckila AJ, Neustadt DH, Nakasato YR, Halleen JM, Hentunen T, Yam LT. Serum tartrate-resistant acid phosphatase isoforms in rheumatoid arthritis. Clin Chim Acta 2002; 320:49 – 58. Halleen JM, Alatalo SL, Suominen H, Cheng S, Janckila AJ, Va¨a¨na¨nen HK. Tartrate-resistant acid phosphatase 5b: a novel serum marker of bone resorption. J Bone Miner Res 2000;15: 1337 – 45. Janckila AJ, Takahashi K, Sum SZ, Yam LT. Tartrate-resistant acid phosphatase isoform 5b as serum marker for osteoclastic activity. Clin Chem 2001;47:74 – 80. Miyazaki S, Igarishi M, Nagata A, Tominaga Y, Onodera K, Komoda T. Development of immunoassays for type-5 tartrateresistant acid phosphatase in human serum. Clin Chim Acta 2003;329:109 – 15. Hannon RA, Clowes JA, Eagleton AC, Hadari AA, Eastell R, Blumsohn A. Clinical performance of immunoreactive tartrateresistant acid phosphatase isoform 5b as a marker of bone resorption. Bone 2004;34:187 – 94. Alatalo SL, Ivaska KK, Waguespack SG, Econs MJ, Va¨a¨na¨nen HK, Halleen JM. Osteoclast derived serum tartrate-resistant acid phosphatase 5b in Albers-Schonberg disease (type II autosomal dominant osteopetrosis (ADO2)). Clin Chem 2004;50:883 – 90. Gerdhem P, Ivaska KK, Alatalo SL, Halleen JM, Hellman J, Isaksson A, et al. Biochemical markers of bone metabolism and prediction of fracture in elderly women. J Bone Miner Res 2004;19:386 – 93. Janckila AJ, Parthasarathy RN, Parthasarathy LK, Seelan RS, Hsueh Y-C, Rissanen J, et al. Properties and expression of human tartrate-resistant acid phosphatase isoform 5a by monocyte-derived cells. J Leukoc Biol 2005;77:209 – 18. Janckila AJ, Parthasarathy RN, Parthasarathy LK, Seelan RS, Yam LT. Stable expression of human tartrate-resistant acid phosphatase isoforms by CHO cells. Clin Chim Acta 2002; 326:113 – 22. Nakasato YR, Janckila AJ, Halleen JM, Va¨a¨na¨nen HK, Walton SP, Yam LT. Clinical significance of immunoassays for type-5 tartrate-resistant acid phosphatase. Clin Chem 1999;45: 2150 – 7. Orlando JL, Zirino T, Quirk BJ, Averill BA. Purification and properties of the native form of the purple acid phosphatase from bovine spleen. Biochemistry 1993;32:8120 – 9. Marshall K, Nash K, Haussman G, Cassady I, Hume D, de Jersey J, et al. Recombinant human and mouse purple acid phosphatases: expression and characterization. Arch Biochem Biophys 1997;345:230 – 6. Ljusberg J, Ek-Rylander B, Andersson G. Tartrate-resistant purple acid phosphatase is synthesized as a latent proenzyme and activated by cysteine proteinases. Biochem J 1999; 343:63 – 9. Doi K, Antanaitis BC, Aisen P. The binuclear iron centers of uteroferrin and the purple acid phosphatases. Struct Bond 1988;70:1 – 26.

140

T.-Y. Chao et al. / Clinica Chimica Acta 359 (2005) 132–140

[18] Janckila AJ, Latham MD, Lam K-W, Chow K-C, Li C-Y, Yam LT. Heterogeneity of hairy cell tartrate-resistant acid phosphatase. Clin Biochem 1992;25:437 – 43. [19] Kaija H, Jia J, Lindqvist Y, Andersson G, Vihko P. Tartrateresistant bone acid phosphatase: large-scale production and purification of the recombinant enzyme, characteriza-

tion, and crystallization. J Bone Miner Res 1999;14: 424 – 30. [20] Buhi WC, Ducsay CA, Bazer FW, Roberts RM. Iron transfer between purple phosphatase uteroferrin and transferrin and its possible role in iron metabolism in the fetal pig. J Biol Chem 1982;257:1712 – 23.