| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Drug Monitoring and Toxicology |
1 Department of Renal Medicine C, Research Laboratory, Aarhus University Hospital, Skejby Sygehus, Brendstrupgaardsvej, DK-8200 Aarhus N, Denmark.
aAuthor for correspondence. Fax 45-89496003; e-mail bundgaardpernille{at}dadlnet.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Whole blood samples were obtained from 10 patients treated with calcineurin inhibitors and 20 healthy volunteers and frozen at –80 °C. CaN activity was measured by its ability to dephosphorylate a 19-amino acid peptide previously phosphorylated with [
-32P]ATP. Radioactivity was quantified by liquid scintillation, and results were converted from cpm to U of CaN. Validation of the assay included enzyme kinetics, linearity, precision (at low and normal CaN activities), analytical recovery, and limit of detection.
Results: The enzyme followed simple Michaelis-Menten-type kinetics: Vmax was estimated as 240 nmol 32P · L–1 · min–1 and Km as 70 µmol/L. The assay was linear within the concentration range examined. Analytical recovery varied from 68% to 72%. The total analytical SD was 0.059 and 0.053 U of CaN for high and low CaN activity, respectively. The within-day SD for high and low activity was 0.032 and 0.039 U of CaN, respectively. The limit of detection was 0.04 U of CaN, which is far below the values measured in patients treated with CaN inhibitors.
Conclusions: In addition to the pharmacokinetic monitoring applied today, the CaN assay can be used to monitor patients treated with calcineurin inhibitors, hopefully leading to prolonged graft survival.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(2)(3)(4). Although reports on CaN activity measurements in CsA-treated renal transplanted patients have been published, clinical validation data have been limited (5). CaN activity measurements using whole blood samples have been shown by Caruso et al. (6) to be superior to activity measured in human peripheral blood mononuclear cells. These authors compared total whole blood CaN activity in five healthy individuals with CaN activity in each blood component and reported that the total CaN activity recovered by adding each individual cell and plasma fraction accounted for only 61.9% of the enzyme activity measured in unfractionated whole blood. Measuring CaN activity in whole blood samples has several other advantages: smaller blood sample volumes are needed for pharmacodynamic assays, loss of drug effect caused by dissociation during cell isolation is prevented, and whole blood analysis is less cumbersome than using human peripheral blood mononuclear cells (7).
Various calculations have been used for the expression of CaN activity. Halloran et al. (8) expressed CaN activity as the percentage of the value for the 0-h pre-CsA control, whereas Piccinini et al. (9) calculated results as 32P released per minute per milligram of lysate protein. The possibility of expressing results by means of a calibration curve prepared with CaN has not been reported. To be able to compare between-assay results, we decided to express results in units of CaN. To implement this, we constructed calibration curves using bovine CaN, calmodulin, and buffer on each day of experimentation. We have previously published results of measurements of CaN activity in a FK-treated group of renal transplant patients on days 3 and 14 after transplantation (10). In this study, we investigated CaN activity in both healthy, nonmedicated volunteers and in renal transplant patients treated with CaN inhibitors.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The samples from the FK-treated patients were drawn 3 h after oral intake of drug, and the mean (SD) FK concentration was 17.1 (8.6) µg/L. The patients treated with CsA had blood samples drawn 2 h after oral ingestion of drug, and the mean CsA concentration was 1161 (524) µg/L.
The transplanted patients were all outpatients at the Department of Renal Medicine C, Skejby Sygehus, Aarhus University Hospital, Denmark, and all had stable renal graft function. Whole blood samples were collected in EDTA tubes and were exposed to N2 before freezing to avoid destruction of enzyme activity by oxygen. Samples were kept frozen at –80 °C for no longer than 2 weeks before analyses were performed.
This pharmacodynamic assay is a 2-day procedure, with phosphorylation of peptide on day 1 and analysis of whole blood samples with the CaN assay on day 2. If not stated otherwise, the buffers were purchased from Sigma.
peptide phosphorylation
A 19-amino acid peptide (Asp-Leu-Asp-Val-Pro-Ile-Pro-Gly-Arg-Phe-Asp-Arg-Arg-Val-Ser-Val-Ala-Ala-Glu) was used as substrate in the peptide phosphorylation (Schäfer-N). This sequence corresponds to the RII subunit of a cAMP-dependent protein kinase (11). Using a technique similar to that described by Hubbard and Klee (12), we phosphorylated the serine unit with [-32P]ATP (Amersham Pharmacia Biotech) by adding protein kinase A to a peptide phosphorylation buffer. The buffer consisted of 300 µmol/L ATP, 20 mmol/L MES, 200 µmol/L EGTA, 400 µmol/L EDTA, 2 mmol/L MgCl2, 50 µmol/L CaCl2, 50 µg/mL bovine serum albumin (BSA), doubly distilled water (DDW), [-32P]ATP (10 mCi/mL), and the 19-amino acid peptide (148.5 µmol/L). The reaction mixture incubated for 60 min at 30 °C and 10 min at room temperature. After incubation, a C18 extraction cartridge (Waters Corporation) was used to wash the peptide with 10 g/L trifluoroacetic acid. Finally, the peptide was eluted from the cartridge by 300 mL/L acetonitrile in 1 mL/L trifluoroacetic acid and evaporated to dryness overnight. Scintillation counting (Wallac 1410; LKB) was performed with 5 µL of the eluted peptide before evaporation to assure phosphorylation of the peptide and to quantify the radioactivity in cpm.
CAN assay
Using a method similar to that of Fruman et al. (11), we measured CaN activity as the release of radiolabeled phosphate from the 19-amino acid peptide described above. Whole blood (40 µL) was added to a lysis buffer (150 µL) containing 50 mmol/L Tris (pH 7.5), 1 mmol/L dithiothreitol (ICN Biomedicals Inc.), 50 mg/L phenylmethylsulfonyl fluoride (ICN Biomedicals), 50 mg/L soybean trypsin inhibitor, 10 mg/L leupeptin, 10 mg/L aprotinin, 1 mmol/L EDTA (pH 8.0), 100 µmol/L EGTA (pH 7.0), 1 mL/L ß-mercaptoethanol, and DDW. Samples were run in triplicate and subjected to three freeze/thaw cycles followed by centrifugation at 12 000g for 10 min at 4 °C.
A reaction mixture containing lysate (30 µL) and assay buffer (60 µL) was prepared. The assay buffer consisted of 20 mmol/L Tris (pH 8.0), 100 mmol/L NaCl, 6 mmol/L MgCl2, 500 µmol/L dithiothreitol, 100 µg/L BSA, 100 µmol/L CaCl2, 750 nmol/L okadaic acid, DDW, and 15 µmol/L radiolabeled peptide. The mixture was allowed to incubate for 15 min at 30 °C, and the reaction was terminated by the addition of 50 mL/L trichloroacetic acid in a potassium phosphate buffer. The assay mixture was applied to a 1.5-mL Dowex cation-exchange column and washed with 900 µL of DDW. The eluate was recovered, and radioactive free phosphate was quantified by liquid scintillation. Results were obtained in cpm.
phosphatases and CSA solution
Because there are four known phosphatases in blood that could interfere with our results (PP1, PP2A, PP2C, and CaN/PP2B), it was important for us to distinguish CaN from the others. To document that the activity measured as cpm reflected activity attributable to CaN and no other phosphatase, we added okadaic acid to the assay buffer to inhibit PP1 and PP2A. To determine the activity of PP2C, a phosphatase resistant to both okadaic acid and the CaN inhibitors CsA and FK, each sample was compared with the same blood sample maximally suppressed with a surplus of CsA. The remaining activity of the maximally suppressed sample reflected the PP2C activity. Results from the investigations with both healthy individuals and renal transplant patients treated with CaN inhibitors showed that the PP2C activity constituted 62% of the total activity measured (PP2C + CaN) in healthy individuals and 72% in the patient group.
The CsA solution used above was prepared by diluting 1.2 mg of CsA (Sigma Chemical) in 80 µL of 990 mL/L ethanol. The mixture incubated at room temperature for 10 min, after which 80 µL of Tween 80 was added (Sigma Chemical). To reach a final volume of 1 mL, 900 µL of Tris buffer (pH 7.5) was added. The final concentration in the whole blood samples was 10 µmol/L. The potential inhibitory effects of Tween and ethanol on CaN activity were investigated. The activities of samples containing Tween, ethanol, and Tris buffer (pH 7.5), but no CsA (n = 12) were compared with the activities in samples containing only Tris buffer (n = 12); the results were not significantly different (P = 0.62; data not shown).
Use of CsA as a CaN inhibitor in this assay requires that the concentration used inhibits the enzyme activity efficiently. Experiments with various concentrations of CsA added to whole blood samples revealed that 0.5 µmol/L CsA inhibited the CaN activity by 69% and that increasing CsA concentration above 2 µmol/L did not increase the inhibition of CaN.
calibration curve
To convert activity measured in cpm to units of CaN, we daily constructed calibration curves with bovine CaN, using 0, 0.5, 1.0, and 2.0 U of CaN. CaN of unknown purity was purchased from Sigma Chemical, and each bottle contained 66.6 U of CaN. The CaN was diluted with a pH 8.0 buffer containing 20 mmol/L Tris and 3 q/L BSA. Each sample for constructing the calibration curve consisted of 4 µL of calmodulin (5 mol/L), 6 µL of Tris/BSA buffer, and 20 µL of the appropriate concentration of CaN. The purified bovine calmodulin used in preparing the calibration curve was purchased from BIOMOL Research Laboratories. Assay buffer (60 µL) was added to each sample, and the samples were assayed according to standard assay procedures. The result of the calibration curves were displayed graphically as units of CaN vs cpm.
conversion of results
From the test samples, the background phosphatase activity (0 U from the calibration curve) was subtracted, leading to the activity attributable to CaN and other okadaic acid-resistant phosphatases (PP2C). Subtracting the activity for the sample containing surplus amounts of CsA (PP2C activity) from the activity of the actual test sample assured that the activity measured was attributable solely to CaN. Thereafter, results were converted from cpm into units of CaN with the calibration curve.
determination of drug concentrations in blood
Blood FK concentrations were measured with an Abbott IMx method using monoclonal antibody (13). Blood CsA was measured with the Dade Emit assay on a Cobas Mira analyzer. The drug concentrations were measured to assure that patients had taken their medication.
enzyme kinetic analysis
To characterize the enzymatic assay, we performed steady-state kinetic experiments. We diluted substrate (substrate concentration 10x, 5x, 3x, 2x, 1.5x, 1x, 0.8x, 0.6x, 0.4x, 0.2x, and 0.1x) to determine Vmax and Km. Results were obtained in cpm, but Vmax is reported in nmol 32P · L–1 · min–1, and Km is reported in µmol/L.
validation
To investigate the optimum incubation time, we performed experiments varying the substrate concentration (10, 1, and 0.1 µmol/L) and the incubation time (2.5, 5, 10, and 15 min). Additionally, we performed experiments in which the incubation temperature, the ratio of whole blood to lysis buffer, the ratio of lysate to assay buffer, and the sizes of cation-exchange columns were varied to determine the optimum experimental conditions. To determine the limit of detection, the results from 19 calibration curves were used. Each calibration curve consisted of measurements with 0, 0.5, 1, and 2 U of CaN. All calibrators were run in triplicate. We defined the limit of detection as the upper limit of the 95% confidence interval (CI) for the 0 calibrator.
expression units
The measured CaN activity is expressed as units of CaN per 40 µL of EDTA-stabilized blood.
statistical analysis
The steady-state velocity plot and calculations were determined by curve fitting and analyzed by Sigma Plot, Ver. 8.0.2S. A linear equation with intercept, slope, and standard error of the estimate was obtained with Excel and Sigma Plot for the linearity experiments. ANOVA with repeated measurements was used for the precision studies. SPSS 10.0 was used to determine the 95% CI. The calibration curve was characterized by linear regression. We defined statistical significance as P <0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Calmodulin was shown to potentiate CaN activity by a factor of 20. The calmodulin dependency investigations demonstrated that 4 µL (5 mol/L) of calmodulin was required in the samples for the calibration curve.
|
enzyme kinetics
Within our standard experimental concentrations, the results of the steady-state kinetic experiments followed a simple Michaelis-Menten association (Fig. 2
). The results were fitted by nonlinear least-square fit to:
 |
|
As shown in Table 1
, the CaN activity increased with the incubation time for all substrate concentrations investigated, and there was no indication of substrate inhibition.
|
linearity
We assessed assay linearity using three different dilutions of enzyme and one concentrated sample. The samples contained 150%, 100%, 50%, and 25% of the normal enzyme content. Each dilution was made in six replicates, and the concentrated sample was run in triplicate. Samples were diluted with isotonic sodium chloride. Samples containing 150% of enzyme were prepared by centrifugation of 2 mL of whole blood at 2000g for 15 min. The blood samples had a hematocrit of 0.50, and removal of 666 µL of plasma after centrifugation gave an enzyme concentration of 150% of normal in the remaining volume. The results depicted in Fig. 3
demonstrate that the assay was linear at the concentrations investigated: y = 0.2211 (0.0208)x – 0.0051 (0.0174) U of CaN (SE of estimate, 0.0237; R2 = 0.96).
|
analytical recovery rate
We used whole blood samples from healthy nonmedicated individuals to determine analytical recovery. Samples were analyzed with standard assay procedures. Bovine CaN was added to a whole blood sample in two different concentrations, 0.25 and 0.50 U of CaN. All samples were run in triplicate. As shown in Table 2
, the recovery rate for added CaN varied from 68% to 72%.
|
precision
We determined the precision of the assay at two different CaN activities. Ten patients treated with CaN inhibitors and thereby having low CaN activity and 20 nonmedicated healthy volunteers with higher CaN activities were investigated. One blood sample was drawn from each participant. The blood sample was split into four identical tubes and frozen at –80 °C until analysis. All samples were run in triplicate. Two of the tubes were thawed and subjected to analysis on day 1 to determine the within-day variation. We estimated the between-day variation by analyzing of the two last samples on day 2. The mean value for the patients was 0.21 U of CaN and that for healthy volunteers was 0.34 U of CaN (Fig. 4
), which was significantly different (P <0.01).
|
Using ANOVA with repeated measurements, we determined the within-day SD as 0.039 U of CaN and the between-day SD as 0.036 U of CaN. For the 20 healthy individuals, the within- and between-day SD were estimated to be 0.032 and 0.050 U of CaN, respectively.
detection limit
The mean and 95% CI of 19 calibration curves are shown in Fig. 5
. The slope of the line is 8812, and the intercept is 3459 cpm. The limit of detection is 0.04 U of CaN, which is below the values we measured in patients treated with CaN inhibitors.
|
Using a limit of quantification of twice the detection limit (14), we obtained a limit of quantification of 0.08 U for the CaN assay. We have never measured CaN activity lower than 0.07 U in treated patients; we therefore feel that the method could be used in the range from 0.08 to 1.50 U of CaN. The upper limit of the range is defined from the highest measured CaN activity in a healthy nonmedicated volunteer.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), with analytical recovery rates of 68% and 72% after addition of 0.25 and 0.5 U of bovine calcineurin, respectively. The limit of detection of 0.04 U of CaN assured us of the practicability of measuring CaN activity in patients treated with CaN inhibitors because we have demonstrated that the CaN activity in a group of such patients is far higher than the observed detection limit. To date, we have not found other published studies presenting a detection limit for this pharmacodynamic assay. The within- and between-day CVs of 19% and 17%, respectively, are relatively high for patients treated with CaN inhibitors. This is somewhat higher than those found in other studies. Brunet et al. (15) reported inter- and intraassay CVs of 8% and 5%. The difference between the ways that those authors compensated for PP2C activity and the way that we did could explain the difference. However, the most important question is whether the CVs are comparable when they are obtained by different methods. Brunet et al. (15) did not describe in detail how they obtained the CV; therefore, a direct comparison is not possible. Caruso et al. (6) reported CVs of 4% in samples with high CaN activity and 12% in samples with low CaN activity. They determined the CaN activity in 10 replicates of each sample to assess the precision of the method, which is not directly comparable to the precision measurements we performed. However, the CV of 12% for the patients treated with CaN inhibitors is in the same range as those found in the present study. The use of whole blood samples in this pharmacodynamic assay provides numerous advantages compared with the use of lymphocytes, mainly because measuring CaN activity in whole blood is the best estimate for the in vivo equilibrium. That equilibrium will be disrupted when whole blood samples are subjected to Ficoll centrifugation or other forms of cell separation, leading to loss of CaN activity. Our results were confirmed by previous reports concerning measuring CaN activity in whole blood samples. Halloran et al. (8) found that measurements in whole blood samples were more precise than analysis done with lymphocytes and also discussed maintenance of the in vivo situation. Caruso et al. (6) reported lower CVs for measurements of CaN activity in whole blood compared with lymphocytes.
The CaN assay contains some obstacles that need attention before the results can be considered a true picture of CaN activity. CaN is highly vulnerable to oxygenation; therefore, all whole blood samples need to be exposed to N2 before freezing to avoid destruction of enzyme activity by oxygen. When not being handled, we closed all blood samples with caps to avoid unnecessary oxygenation. Although phosphatases other than CaN are capable of dephosphorylating the peptide, special conditions have been developed to inhibit these phosphatases. Okadaic acid at a final concentration of 750 nmol/L is used in the assay buffer to inhibit PP1 and PP2A. Okadaic acid has very little inhibitory effect on CaN activity, especially in the concentrations we used. Bialojan and Takai (16) found that the ID50 (concentration of okadaic acid required to obtain 50% inhibition) was 4–5 µmol/L. In additional experiments, we found that 151 nmol/L okadaic acid inhibits the two phosphatases as effectively as 750 nmol/L without inhibition of CaN activity (data not shown). These findings confirm a previous report by Fruman et al. (17). PP2C is both Ca2+ and CaN inhibitor resistant, which has been demonstrated by Lui et al. (18). We determined the magnitude of PP2C activity as the remaining activity of samples maximally suppressed by CsA. It is our impression that a major limitation of this assay is the solubility of the hydrophobic CsA molecule in the aqueous milieu. We have tried to overcome this by use of high concentrations of CsA and use of Tween 80. Use of Tween and ethanol in the CsA solutions raises the question of whether these substances themselves can inhibit the enzyme activity. Additional experiments, described earlier, indicated that this is not the case. Another potential problem related to the use of high CsA concentrations is precipitation of CsA in the solution. With the amounts described for our experiments, we have not experienced this problem but are very aware of its existence.
As described earlier, preliminary experiments demonstrated that there was no increase in CaN inhibition when the CsA concentration was increased over 2 µmol/L. This indicates that 10 µmol/L is surplus CsA.
To our knowledge, no other study has been published with kinetic constants obtained from whole blood samples with this enzymatic assay. We have expressed Vmax as nmol 32P · L–1 · min–1, which is not directly comparable to results presented as pmol 32P · min–1 · (mg protein)–1. As a consequence, we are only able to compare our data obtained for Km with results from other research groups. The enzymatic characterization gave a Km of 70 µmol/L, which is comparable to what has been found previously by others. Blumenthal et al. (19) reported that dephosphorylation of intact RII [cAMP-dependent protein kinase regulatory subunit (type II)] gave a Km of 20 µmol/L and Vmax of 2 µmol · min–1 · mg–1. To define the minimum primary structure required for recognition by CaN, they used a series of synthetic phosphopeptides as substrates. The dephosphorylation of a 19-residue peptide (Asp-Leu-Asp-Val-Pro-Ile-Pro-Gly-Arg-Phe-Asp-Arg-Arg-Val-Ser-Val-Ala-Ala-Glu) gave kinetics constants comparable to those obtained with the RII subunit, with a Km of 26 µmol/L and Vmax of 1.7 µmol · min–1 · mg–1. Hubbard and Klee (12) performed in vitro experiments including phosphopeptide, calcineurin, and calmodulin and reported that one should obtain a typical Vmax value of 1.3 µmol · min–1 · mg–1 and a Km of 40 µmol/L. Performing the steady-state kinetic experiments, we used the highest possible substrate concentration, which was 100 µmol/L. Although we did not reach a clear saturation, a simple Michaelis-Menten association is still the most plausible kinetic association. No experimental observations gave us reasons to consider more complicated kinetic models.
We have demonstrated the practicability of this enzymatic assay for measuring CaN activity. The present method shows good linearity, analytical recovery rate, and acceptable precision. The limit of detection is well below the CaN activity we measured in patients treated with CaN inhibitors. Using the present method, we measured the CaN activity in whole blood samples instead of peripheral blood mononuclear cells, and we introduced the use of a calibration curve, which enabled us to report results in units of CaN and allowed us to compare values obtained at different times. Finally, we investigated the enzyme activity in patients treated with CaN inhibitors and healthy nonmedicated individuals. The CVs of our PD assay and the overlap between values obtained in calcineurin-inhibited patients and healthy controls indicate that this assay will be inadequate for use as a sole monitoring tool for an individual patient. Further investigations must be performed before this could be the case. Nonetheless, the CaN assay can be a useful scientific tool toward a deeper understanding of the CaN inhibition in various groups of renal transplant patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  H. H. van Rossum, F. P.H.T.M. Romijn, K. J. Sellar, N. P.M. Smit, P. J.M. van der Boog, J. W. de Fijter, and J. van Pelt Variation in Leukocyte Subset Concentrations Affects Calcineurin Activity Measurement: Implications for Pharmacodynamic Monitoring Strategies Clin. Chem., March 1, 2008; 54(3): 517 - 524. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, calibration curve with calmodulin omitted. Calmodulin is shown to potentiate the effect of CaN by a factor 20.
