| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Proteomics and Protein Markers |
Institut für Laboratoriums- und Transfusionsmedizin, Herz- und Diabeteszentrum Nordrhein–Westfalen, Universitätsklinik der Ruhr-Universität Bochum, Bad Oeynhausen, Germany.
aAddress correspondence to this author at: Institut für Laboratoriums- und Transfusionsmedizin, Herz- und Diabeteszentrum Nordrhein–Westfalen, Universitätsklinik der Ruhr-Universität Bochum, Georgstraße 11, 32545 Bad Oeynhausen, Germany. Fax 49-5731-972013; e-mail jkuhn{at}hdz-nrw.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We developed an HPLC electrospray ionization tandem mass spectrometry method to assay XT-I activity in serum by use of a synthetic peptide (Bio–BIK-F) as the XT-I substrate. On the basis of XT-I-mediated transfer of D-xylose from UDP-D-xylose to the synthetic peptide to form Bio-BIK-F-Xyl, we determined XT-I activity in human serum samples.
Results: Multiple calibration curves for the analysis of Bio-BIK-F-Xyl exhibited consistent linearity and reproducibility in the range of 0.20–20 mg/L, corresponding to XT-I activity of 1.14–114 mU/L under assay conditions. The mean (SD, range) XT-I activity values in 30 blood donor sera were 18.4 (3.0, 8.7–24.8) mU/L. The limit of detection and lower limit of quantification were 8.5 µg/L (0.05 mU/L) and 163 µg/L Bio-BIK-F-Xyl (0.93 mU/L XT-I activity), respectively. Interassay imprecision (CV) was 5.4%–26.1% in the range of 0.64 to 129 mU/L, and mean recovery was 107% (range, 96%–129%). Method comparison with the radiochemical assay showed a moderate correlation (r = 0.79). The Passing–Bablok regression line was: radiochemical assay = 0.045 LC-MS/MS + 0.061 mU/L, Sy|x = 0.186.
Conclusions: This simple and robust LC-MS/MS assay permits the rapid and accurate determination of XT-I activity in human serum.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
XT-I activity is increased in the serum of patients with connective tissue diseases such as systemic sclerosis (SSC), osteoarthritis, or pseudoxanthoma elasticum. Longitudinal studies found that XT-I activity remained increased in the serum of scleroderma and SSC patients (6), a finding that corresponds to an increase of proteoglycan metabolism in sclerotic organs and the theory that XT-I is secreted simultaneously into the extracellular matrix with chondroitin sulfate containing proteoglycans (6)(7)(8). Increased glycosaminoglycans have been found in patients with SSC, and the increased chondroitin sulfate and dermatan sulfate content in affected skin correlates with the severity of sclerotic skin (9)(10)(11).
To date, XT-I activity has been measured by the incorporation of radioactively labeled xylose into proteins or peptides containing the XT-I recognition sequence G-S-G (1)(12)(13)(14). Initial XT-I activity assays used deglycosylated cartilage proteoglycans or solutions of silk fibroin consisting of 
60% of the sequence repetition (SGAGA)n as a xylose acceptor (12). Different preparations of deglycosylated proteoglycans and varieties of silk composition showed considerably varying acceptor activities. Finally, the use of recombinant bikunin as a xylose acceptor enabled an accurate and sensitive determination of XT-I activity, even in human serum (13)(14). These XT-I assays have disadvantages, including time-consuming drying and washing steps and the use of expensive isotope-labeled UDP-[14C]D-xylose.
A novel matrix-assisted laser desorption/ionization time-of-flight mass spectrometry–coupled XT assay was recently described within the context of characterizing Drosophila melanogaster peptide O-xylosyltransferase (OXT), which is also a proteoglycan xylosyltransferase (15). The assay was used to demonstrate the transfer of xylose to the peptide DDDSIEGSGGR.
We have developed a novel, rapid, and sensitive liquid chromatography–tandem mass spectrometry (LC-MS/MS) method for the accurate determination of XT-I activity in serum without the use of radioactive reagents.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
serum samples
Venous blood samples were collected in serum monovettes from KABE Labortechnik GmbH. After clotting and centrifugation, the serum was separated and stored at –20 °C. Serum samples from 30 blood donors (15 males, age 18–65) and from 24 patients (12 males, age 18–65) who attended the hospital and the outpatient clinic for different reasons were selected by a random-access procedure. All specimens used in our study were reserve materials that were not needed for any other diagnostic analysis. No extra materials or increased sample volumes were obtained from the patients. The experimental design was approved by the local ethics committee, and all samples were completely anonymized before inclusion in the study.
cell culture
JAR choriocarcinoma cells, which release native XT-I in the cell culture supernatant, were cultured in a hybrid hollow-fiber bioreactor as described elsewhere (16). High Five/pCG255–1 insect cells, which produce cell culture supernatant containing recombinant XT-I (rXT-I-His), were cultured as described previously (17).
xt-i stock solutions
XT-I from human JAR cell culture supernatant was used to prepare a stock solution of native XT-I (10.0 mU/L, measured with the radiochemical XT-I activity assay). A further XT-I stock solution (1 166.7 mU/L) was prepared by ultrafiltration of High Five/pCG 255–1 insect cell culture supernatant containing rXT-I-His.
serum protein and radiochemical xt-i activity assay
Serum protein measurement was performed with the automated multianalyte analyzer system Architect ci8200 (Abbott).
The method for measuring XT-I activity is based on the binding of [14C]D-xylose to an XT-I acceptor, according to a previously described method that uses Bio-BIK-F as the XT-I acceptor (13)(14).
sample preparation for LC-MS/MS
Measurement of XT-I activity was based on the binding of D-xylose to Bio-BIK-F. The yellow-colored peptide was dissolved in HPLC-grade water, and the resulting acceptor solution used for the reaction mixture of the assay was adjusted to an absorbance at 450 nm of 0.400. The reaction mixture for the assay contained the following, in a total volume of 100 µL: 50 µL of sample, 20 µL of acceptor solution, 25 mmol/L MES buffer (pH 6.5), 25 mmol/L KF, 5 mmol/L MnCl2, 5 mmol/L MgCl2, and 30 µmol/L UDP-D-xylose. After incubation at 37 °C for 90 min, the reaction was stopped by heating the reaction mixture at 99 °C for 10 min. The samples were centrifuged at 10 000g, and then the clear supernatant was diluted 10-fold with water, transferred to an autosampler vial, and used for further analysis.
lc-ms/ms xt-i activity assay
For the measurement of XT-I activity, a 3 x 4-mm C18 (ODS, Octadecyl) cartridge (Phenomenx, Part No: AJ0–4287) maintained at 55 °C was used for separation by an HPLC system (Waters Alliance 2795XE) directly coupled to a Quattro LC tandem mass spectrometer fitted with a Z Spray ion source (Micromass). A 40-µL sample was injected at a flow rate of 0.3 mL/min. The gradient program was 100% aqueous formic acid (2 g/L) for 0.40 min, followed by a step gradient of 100% methanol containing formic acid (2 g/L). At 1.20 min, the mobile phase was reverted to 100% aqueous formic acid. The mass spectrometer was operated in electrospray positive ionization mode, and the system control and data acquisition were performed with MassLynx NT 4.0 software, with automated data processing by the MassLynx QuanLynx program provided with the instrument. Nitrogen was used as the nebulizing gas, and argon was used as the collision gas. Instrument settings were as follows: capillary voltage, 4.0 kV; source temperature, 120 °C; desolvation temperature, 400 °C; sample cone-voltage energy, 36 V; and collision energy, 35 eV. The collision gas pressure was 3.0 x 10–3 mbar. Sample analysis was performed in the multiple reaction monitoring (MRM) mode of the mass spectrometer, with a dwell time of 0.30 s and the mass transitions m/z 975.1
909.1. An electrically operated valve ensured that the eluate was introduced into the mass spectrometer for 1.8 to 2.8 min only.
The enzyme activity was 1 mU = 1 nmol of incorporated xylose per min, which is equal to the synthesis of 175.5 µg/L Bio-BIK-F-Xyl under the assay conditions (incubation time = 90 min).
ion suppression
To measure the ion suppression, a continuous infusion of Bio-BIK-F-Xyl (4.0 mg/L) was introduced at a flow rate of 10 µL/min into the effluent from the HPLC column before introduction into the electrospray tandem mass spectrometer. Heat-inactivated pool serum (1 h at 60 °C) was injected after sample preparation (see above) into the LC-MS/MS system, and the MS/MS response of the MRM transition for the Bio-BIK-F-Xyl (m/z 975.1909.1) was recorded.
validation
The method was validated according to published criteria (18)(19)(20).
Standards, calibrators, and controls.
Standards, calibrators and controls for the XT-I activity were prepared with excessively xylosylated Bio-BIK-F. To generate a Bio-BIK-F-Xyl stock solution, we mixed 500 µL of 500 mmol/L MES, pH 6.5, 250 µL of 200 mmol/L MnCl2, 250 µL of 200 mmol/L MgCl2, 100 µL of 18.7 mmol/L UDP-D-xylose, 4.0 mL of 138 µmol/L Bio-BIK-F, and 4.9 mL of rXT-I-His with an XT-I activity of 1 166.7 mU/L. After incubation at 37 °C for 24 h, the reaction was stopped by heat inactivation (10 min at 99 °C). The mixture was centrifuged at 13 000g for 10 min, and the clear yellow supernatant was used to value the xylosylation of the acceptor peptide. Unxylosylated peptide was measured, and the amount of Bio-BIK-F-Xyl in the supernatant was calculated. An 80 mg/L Bio-BIK-F-Xyl stock solution was produced, and a series of calibrators, standards, and quality controls was prepared with native serum as diluent.
Linearity studies.
To construct a matrix-based calibration curve, 100 µL of the 80 mg/L Bio-BIK-F-Xyl stock solution was diluted with 1.9 mL of native serum. After mixing, 1.0 mL of the solution was diluted with 100 µL of 500 mmol/L MES, pH 6.5, 50 µL of 200 mmol/L MnCl2, 50 µL of 200 mmol/L MgCl2, and 800 µL of HPLC-grade water, and the supernatant was used as calibrator 9 after heat precipitation (10 min at 99 °C) and centrifugation (90 min at 13 000g). The other 1.0 mL of the solutions above was further diluted with 1.0 mL of native serum and mixed, and then 1.0 mL was used to produce calibrator 8, analogous to calibrator 9. The residual 1.0 mL of the mixture was further diluted with 1.0 mL native serum and mixed, and then 1.0 mL was used to prepare calibrator 7, continuing this procedure until calibrator 1 was prepared.
Limit of detection and lower limit of quantification.
The minimum of detectable concentration was assessed as 3 SD0, where SD0 is the value of the SD for which the concentration of the analyte approaches 0. We identified the limit of detection by performing 25 replicate measurements in a single LC-MS/MS XT-I activity assay with heat-inactivated pool serum (1 h at 60 °C) as a sample. Lower limit of quantification was the value for which the interassay CV reached <20%.
Imprecision.
The intraassay imprecision was measured by analyzing 10 replicates on the same day of native XT-I samples with activity that was extremely low (0.5 mU/L), very low (2.2 mU/L), low (7.6 mU/L), medium (15.0 mU/L), high (35.4 mU/L), very high (66.1 mU/L), and extremely high (127 mU/L), and the interassay imprecision measurement on 10 different days during a 1-month period of native XT-I samples with activity that was extremely low (0.6 mU/L), very low (2.4 mU/L), low (9.3 mU/L), medium (19.7 mU/L), high (42.0 mU/L), very high (87.3 mU/L), or extremely high (129 mU/L).
Stability.
The stability of XT-I in serum was investigated by measurement of the XT-I activity of a freshly prepared serum sample and the measurement the XT-I activity of the same sample at room temperature after 1 day, 1 week, and 1 month. In addition, the stability of the sample after preparation for LC-MS/MS (the Bio-BIK-F-Xyl containing supernatant) was determined after 1 day, 1 week, and 1 month of storage at room temperature. Furthermore, the XT-I activities of pool serum after several freeze-thaw cycles were studied. To investigate the heat stability of XT-I, we heated aliquots of pool serum to temperatures of 40 °C to 60 °C for 1 h; after incubation and centrifugation, the resulting supernatant was used to measure XT-I activity.
Recovery.
We established the recovery efficiency of the assay by measuring the XT-I activity of inactivated pool serum before and after enrichment with different amounts of pool serum. In addition, the XT-I activity was measured in 10 different samples before and after enrichment with known amounts of native XT-I from JAR cells. Analytical recoveries were calculated as the measured concentrations divided by the expected concentrations and expressed as a percentage.
method comparison
We compared the LC-MS/MS XT-I activity assay with the radiochemical XT-I activity assay by measurement of the same donor samples (n = 30) and the same clinical samples (n = 24). Both assays were performed on the same day to decrease a loss of XT-I activity in the samples.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Optimal sensitivity for the detection of Bio-BIK-F-Xyl by mass spectrometry in the MRM mode was achieved at a collision energy of 35 eV.
|
Good sensitivity of the XT-I activity assay was reached with a Bio-BIK-F solution at an absorbance of 450 nm of >0.30 in the reaction mixture containing 50 µL of pooled serum, 20 µL of acceptor solution, 25 mmol/L MES (pH 6.5), 25 mmol/L KF, 5 mmol/L MnCl2, 5 mmol/L MgCl2, and 100 µmol/L UDP-D-xylose in a total volume of 100 µL and an incubation of 90 min at 37 °C. Therefore, in the test for optimal results, we selected as the acceptor solution a Bio-BIK-F solution with an absorbance at 450 nm of 0.40, which is equal to a peptide concentration of 60 µg/mL (33 µmol/L). Substrate saturation kinetics of UDP-D-xylose are presented in Fig. 2
in the online Data Supplement. As a result of this investigation, we selected a UDP-D-xylose concentration of 30 µmol/L. We tested the use of heat inactivation and denaturation (99 °C for 10 min) for stopping the XT-I enzyme reaction and removing proteins, respectively. We found that the enzyme reaction stopped immediately after the reaction mixture was heated, the resulting denatured protein could easily be removed after centrifugation of the sample, and the product peptide Bio-BIK-F-Xyl was completely dissolved in the supernatant and not precipitated. The Bio-BIK-F-Xyl–containing supernatant was analyzed as described in Materials and Methods.
|
validation
We calculated XT-I activity by integrating the area under the extracted MRM chromatogram for Bio-BIK-F-Xyl, the synthesis of which was catalyzed from the enzyme. As described above, loss of the xylose moiety from the protonated molecular ion of Bio-BIK-F-Xyl (m/z 975.1, doubly charged ion) yielded mainly the double-charged fragment ion at m/z 909.1 (Bio-BIK-F+2H)2+ under assay conditions. Therefore, maximum sensitivity was achieved by monitoring the transition: m/z 975.1909.1.
Ion suppression attributable to the sample matrix was investigated as described above. A typical ion chromatogram in which the response of the MRM transition of Bio-BIK-F-Xyl was continuously monitored is shown in Fig. 2A
. Heat-inactivated pooled serum sample preparation was injected at time point 0 min. We observed suppression of the MS/MS response at 0.3 to 0.5 min and 1.9 to 2.0 min, respectively, whereas the Bio-BIK-F-Xyl signal of a pooled serum preparation in the LC-MS/MS XT-I activity assay was detected at a retention time of 2.2 min (Fig. 2B
).
To generate a calibration curve, we prepared several calibrators with calibrator concentrations of 0.20–20 mg/L (corresponding to a XT-I activity of 1.14–114 mU/L under assay conditions) in the same matrix as the study samples, but the calibrator matrix contained a defined amount of Bio-BIK-F-Xyl instead of UDP-D-xylose or Bio-BIK-F. The set of calibrators was measured for 4 consecutive days. The resulting multiple calibrator curves were linear and reproducible with the linear regression equation: y = 557.3x + 89.4 peak area for Bio-BIK-F-Xyl (r = 0.999) (see Fig. 3A
in the online Data Supplement). Dilution linearity is an important assay characteristic for the ability to estimate out-of-range samples. The dilution linearity across the range of 0.01–64 mg/L Bio-BIK-F-Xyl (corresponding to a XT-I activity of 0.06–365 mU/L under assay conditions) is shown in Fig. 3B in the online Data Supplement.
|
The limit of detection of the LC-MS/MS XT-I activity assay was 8.5 µg/L, and the lower limit of quantification calculated from the Table 1
data was 163 µg/L Bio-BIK-F-Xyl, corresponding to 0.05 mU/L and 0.93 mU/L XT-I activity, respectively. Intra- and interassay imprecisions of the assay, assessed for 7 separated sample pools, are shown in Table 1
.
|
To estimate matrix effects, we added the same amount of Bio-BIK-F-Xyl to serum samples with various protein concentrations. Using the quantification of Bio-BIK-F-Xyl with the LC-MS/MS activity assay, we investigated the effects of the addition of low, medium, and high amounts of Bio-BIK-F-Xyl (see Fig. 4 in the online Data Supplement).
We found serum XT-I to be very stable in vitro with no measurable loss in enzyme activity in serum samples stored at room temperature for 1 month. The Bio-BIK-F-Xyl–containing supernatant obtained after sample preparation for the LC-MS/MS XT-I activity assay was also stable at room temperature for at least 1 month. In addition, XT-I activity was not decreased after 10 freeze-thaw cycles. Continuous increase of sample temperature did diminish XT-I activity (Fig. 3
).
The mean recovery of XT-I activity at a concentration of 2.2–50.2 mU/L was 107% (range, 96%–129%).
method comparison
We performed method comparison analyses with samples from 30 healthy blood donors and 24 clinical samples. The mean XT-I activities of the blood donor and clinical samples were 18.4 mU/L (range, 8.7–24.8 mU/L; SD = 3.0 mU/L; CV = 16.2%) and 18.1 mU/L (range, 2.7–24.7 mU/L; SD = 8.7 mU/L; CV = 48.3%), respectively. The correlation between the radiochemical assay and the LC-MS/MS method was moderate (r = 0.79). The Passing–Bablok regression line was: Radiochemical assay = 0.045 LC-MS/MS + 0.061 mU/L, Sy|x = 0.186 (Fig. 4
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The sample preparation for the LC-MS/MS XT-I activity assay, with its simple incubation, heat denaturation, and centrifugation steps, is very easy and relatively fast compared with the costly and time-consuming radiochemical assay (13)(14). The supernatant obtained with Bio-BIK-F-Xyl as an XT-I reaction product can be used directly for quantification in the LC-MS/MS system, but it is also stable enough to be stored at room temperature for 1 month.
XT-I is the central enzyme involved in connective tissue metabolism and catalyzes the rate-limiting step in proteoglycan biosynthesis. Patients with connective tissue or joint diseases show increased XT-I activity in their blood and other body fluids (24)(25)(26). Therefore, the measurement of XT-I activity in different body fluids is of great interest. Recent advances in mass spectrometry have made the introduction of LC-MS/MS into clinical laboratories possible (27)(28)(29)(30). We demonstrated that this technology enables rapid, precise, and sensitive measurement of XT-I activity in blood serum and other biological fluids.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  C. Gotting, J. Kuhn, and K. Kleesiek Serum Xylosyltransferase Activity in Diabetic Patients as a Possible Marker of Reduced Proteoglycan Biosynthesis Diabetes Care, October 1, 2008; 31(10): 2018 - 2019. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Ponighaus, M. Ambrosius, J. C. Casanova, C. Prante, J. Kuhn, J. D. Esko, K. Kleesiek, and C. Gotting Human Xylosyltransferase II Is Involved in the Biosynthesis of the Uniform Tetrasaccharide Linkage Region in Chondroitin Sulfate and Heparan Sulfate Proteoglycans J. Biol. Chem., February 23, 2007; 282(8): 5201 - 5206. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and clinical samples (
) were assessed in side-by-side assays on the same day and analyzed by Passing–Bablok regression: Radiochemical assay = 0.045 LC-MS/MS + 0.061 mU/L; r = 0.79; Sy|x = 0.186.