| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Endocrinology and Metabolism |
1 US Department of Agriculture/Agricultural Research Service Children’s Nutrition Research Center, and 2 Section of Neonatology, Department of Pediatrics, Baylor College of Medicine, Houston, TX 77030.
aAddress correspondence to this author at: USDA/ARS Children’s Nutrition Research Center, 1100 Bates St., Houston TX 77030. Fax 713-798-7119; e-mail zchen1{at}bcm.tmc.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Serum was collected from humans who were administered 42Ca, and 20 µL of serum was mixed with 2 mL of 0.22–0.67 mol/L HNO3 at room temperature for between 1 min and 16 h. The 42Ca/43Ca ratio in the supernatant was measured by a magnetic sector inductively coupled plasma mass spectrometer (ICP-MS). Calcium isotope ratios from these equilibration solutions were compared with data from oxalate-precipitated serum samples to determine the optimum equilibrium time and the effect of acid concentration on equilibrium.
Results: Various amounts of aggregated particles developed in different acid-serum mixtures. These affected the time required for isotope equilibration in the mixture. The shortest equilibrium time needed for the calcium isotopes varied from 1 to 6 h for samples acidified with 0.22–0.45 mol/L HNO3. Data obtained from these solutions were consistent with data from oxalate-precipitated calcium. The precision of 42Ca/43Ca ratio measurements was better than 0.5%.
Conclusions: We have developed a simple, rapid sample preparation technique for ICP-MS analysis in which 20 µL of serum can be used for accurate measurement of the calcium isotope distribution in a sample with good precision and a rapid analysis time.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Application of this type of study to situations where sample volume is limited would be very beneficial in increasing the ability to perform calcium kinetic studies. Additionally, the relatively large number of samples needed for determination of kinetic parameters (typically 10–15 blood samples/patient) requires improvement in the speed and simplicity of sample preparation and analysis.
Stable calcium isotope tracers have been used extensively for 
30 years for metabolic studies of calcium homeostasis in humans in a variety of clinical areas (1)(2)(3)(4)(5)(6)(7). Stable isotopes are safe for use in humans of all ages and do not have problems related to the disposal of contaminated waste, a major advantage over radioisotopes. Stable isotope tracers, highly enriched in naturally low-abundance isotopes, are administered intravenously or orally. Isotope ratio data are collected from blood, urine, or fecal samples to identify time-dependent tracer variation patterns. Such studies have a variety of applications ranging from simple measurement of mineral absorption to measurement of mineral excretion into the gut and stable-isotope-based multicompartmental modeling (8).
This report focuses on the development of a novel serum sample preparation methodology for calcium isotope analysis that would allow the use of a smaller sample size. It also describes a simple method of preparing samples using low volumes and analysis by inductively coupled plasma mass spectrometry (ICP-MS; ThermoFinnigan Element 2). We hypothesized that this approach would give results similar to those obtained with conventional sample preparation methods and subsequent TIMS analysis. The accuracy of the calcium isotope data that we obtain with TIMS and ICP-MS has been documented previously (5)(9).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
methods
Study design.
Samples were obtained from clinical studies of calcium metabolism in children. Informed written consent was obtained from the families of all children who participated in these studies. These studies were approved by the Institutional Review Board of Baylor College of Medicine (Houston, TX).
Patients were admitted to the General Clinical Research Center of Texas Children’s Hospital at 0730 in the morning and had a topical anaesthetic cream applied over the vein of interest. One hour later, a heparin lock catheter was inserted. Each child was subsequently infused with 5 mg of 42Ca over 1–2 min. Samples of whole blood (1 mL) were obtained for calcium isotope ratio measurement at 6, 12, 20, 40, 120, 180, 240, and 480 min after the infusion in nonanticoagulant red-top tubes. The samples of whole blood were then centrifuged, and the serum was separated from other blood components.
Sample preparation by acid-serum equilibration.
Various concentrations of HNO3 were selected for the analytical experiments to assess the effect of acid concentration on calcium isotope ratios in sample solutions. We varied the exchange time to determine the minimum time for reaching equilibrium and to ascertain the existence of any time-dependent isotope fractionation during the sample preparation process.
We mixed 20 µL of serum sample with 2 mL of HNO3 of a selected concentration at room temperature in a conical tube. The sample was vigorously mixed three times (15 s each time) in a Vortex-mix Genie (Fisher Scientific), once at the initial mixing of the serum and acid, once at the middle of the interaction period, and once close to the end of the interaction experiment. The mixture was then centrifuged at 530g for 12 min at room temperature in a Fisher Marathon 8K tabletop centrifuge. The supernatant was extracted and transferred to an acid-washed glass tube for ICP-MS stable calcium isotope analysis. If there were visible particulates present in the solution, the supernatant was centrifuged a second time to avoid blocking the uptake capillary of the ICP-MS inlet system.
Sample preparation by calcium precipitation.
Approximately 2 mL of saturated ammonium oxalate solution was added to a 500-µL serum sample, and the pH of the solution was adjusted to 10 for a 12-h period of calcium precipitation (4)(10). A white calcium oxalate pellet was obtained after centrifugation. The supernatant was decanted, and the pellet was washed with an ammonium oxalate solution (10 g/L) and centrifuged again before being heated at 500 °C for 1 h. The ashed product was resuspended in 0.45–0.67 mol/L highly purified HNO3 and diluted for ICP-MS isotope ratio analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 mg/L was sufficient to generate isotope ratio data with a precision better than 0.5% relative SD (RSD) (9), and a 2 mL solution was adequate for 14 min of analysis. Initially, 50 µL of 42Ca-enriched serum was mixed with 100 µL of concentrated HNO3 in a clean 12-mL disposable glass test tube, and a clear solution was obtained. Deionized water was then added to the mixture to dilute the acid to 0.90 mol/L ( 1.8 mL total volume) to make the solution suitable for ICP-MS analysis. Light-yellow particulates developed in the solution after the addition of deionized water. Heating the mixture slightly to 70 °C for 10 min did not help to redissolve the particulates. The solution was therefore centrifuged for 10 min at 420g, and the supernatant was extracted. Test results showed that this solution had a calcium concentration of 2.5 mg/L.
determination of equilibrium time
A 42Ca-enriched serum sample collected 20 min after dosing was selected for a series of 10 acid–serum equilibrium tests with different mixing times: 1, 15, 30, 45, 60, 75, 90, 120, 240, and 960 min. We mixed 2 mL of 0.67 mol/L HNO3 with 25 µL of serum sample in each of the 10 test tubes. The mixtures were vortex-mixed several times during the interaction period, except for the sample with a 1-min interaction time, before being centrifuged to separate the aggregated materials from the solution. After centrifugation, the supernatant from the equilibration solution was collected and transferred to a clean glass tube for ICP-MS calcium isotope ratio (42Ca/43Ca) analysis (Table 1
). The isotope data collected from the equilibration solutions were similar for samples from the 1- to 16-h interaction times (Fig. 1
). Apparently the solutions achieved a maximum 
% excess [% Excess = 100 x (ratio in sample - ratio in natural abundance)/(ratio in natural abundance); 37% excess for this sample] when the interaction time was 
1 h. If the maximum % excess from the equilibration solution at a 240-min (6-h) interaction time was assumed to be in calcium isotope equilibrium and its degree of equilibration was 100%, the degrees of equilibration for other solutions at different interaction times could be calculated as shown in Table 1
. The degree of equilibration reached 90% at an interaction time of 1 min and increased to 95% at 30 min and 99.5% at 1h. This showed that interaction time played an important role in the acid-serum calcium isotope equilibration process and suggested that calcium isotope equilibrium might be established in a serum-diluted HNO3 mixture in a period of as short as 1 h.
|
|
determination of data accuracy
A complete series of eight serum samples from a study participant (6, 12, 20, 40, 120, 180, 240, and 480 min after intravenous infusion of the 42Ca isotope) was selected to prepare calcium isotope ICP-MS analytical samples by both the conventional precipitation methodology and the new acid-serum equilibration methodology presented here. Because 0.22–0.67 mol/L HNO3 is commonly used as sample solvent for ICP-MS analysis, 0.45 mol/L HNO3 was selected for this acid-serum interaction experiment. We added 2 mL of 0.45 mol/L HNO3 to 20 µL of each serum sample in a test tube and vortex-mixed the mixture for 15 s. The mixture was centrifuged immediately after vortex-mixing, and the supernatant was extracted for ICP-MS calcium isotope analysis.
A resuspended solution was also prepared from the product of the precipitation method. The solution was diluted with 0.45 mol/L HNO3 to make its calcium concentration nearly identical to the counterpart solution from the acid-serum equilibration. Both sets of sample solutions were then analyzed side by side by ICP-MS to determine the 42Ca/43Ca ratios.
Measured calcium isotope ratios from the two sets of solutions were similar (Table 2
). This was surprising because the interaction time for the equilibration solution was only a matter of 1 min before centrifugation. Both sets of data had the same trend in variability (Fig. 2
). The difference between the two sets of analytical data, however, was consistent. Calcium isotope ratios in the equilibration solutions were always slightly lower than the ratios in the precipitated calcium. The difference between the two sets of data apparently increased with increasing isotope enrichment in the sample. The difference could be up to 10% for a serum sample (Table 2
, sample s1) with >200% excess (not shown in Fig. 2
), the very first sample collected from the studied individual after isotope infusion. This indicated that data from the equilibration solutions were not accurate enough to be considered representative in this case. If the isotope excess of the precipitation calcium was used as the representative data, a degree of equilibration (as a percentage) for each equilibration solution could be calculated on the basis of the difference in isotope excess between the two sets of sample solutions (Table 2
). Our experiments suggested that these acid-serum interaction solutions could achieve 95–99% equilibration after a very short period of interaction.
|
|
Tests were further conducted on the serum sample (s1) that had the largest ( 10%) difference between the two sets of 42Ca/43Ca ratios (Table 2
) to better define the equilibrium time required for samples with extraordinarily enriched 42Ca. Equilibration solutions were prepared using 0.45 and 0.67 mol/L HNO3, with interaction times varying from 1 to 6 h. The supernatants were collected and analyzed for 42Ca/43Ca ratios, and the results are shown in Table 3
. We observed that with increasing interaction time, the degree of equilibration increased. The differences in isotope data obtained from the 0.45 and 0.67 mol/L HNO3 equilibration solutions were consistent (Table 3
, sample s1). The degrees of equilibration for the 0.67 and 0.45 mol/L HNO3 solutions at an interaction time of 1 h were 96% and 99%, respectively. At an interaction time of 6 h, they were 98% and 99.5%, respectively. These results suggest that 0.45 mol/L HNO3 was more effective than 0.67 mol/L HNO3 in the equilibration tests. The equilibration solution from the 6-h, 0.45 mol/L acid-serum mixture gave an isotope ratio virtually identical to the ratio from the precipitation solution (within the instrumental error of 0.5% RSD), with a degree of equilibration of 99.5%. It therefore appears that when 0.45 mol/L acid is used for acid-serum equilibration, the solution will achieve equilibrium in 6 h for samples with highly enriched calcium isotopes.
|
Further tests were conducted using both 0.22 mol/L HNO3 and deionized water solutions. Either 0.22 mol/L HNO3 or deionized water was mixed with a serum sample and allowed to equilibrate for 1 h. The results were compared with the results obtained with 0.67 mol/L acid for the same sample (Table 3
, samples f1 and f2). Deionized water in our laboratory is slightly acidic (pH 5.9–6.0), probably because of dissolved CO2 and the extensive use of acid in this working environment. Analytical results from the deionized water-serum solutions fluctuated because of unstable signals during ICP-MS analysis, with a precision of >1.5% RSD for calcium isotope ratios. As a result, data from the deionized water solutions were not precise enough to be used for our data comparison and evaluation purposes. This suggests that water is not the best sample matrix and that a weak acid is preferable for introduction of equilibration solutions into an ICP-MS inlet system.
Similar to the results from the comparison of 0.45 and 0.67 mol/L acid-sample solutions, there were systematic differences in isotope data from the 0.22 and 0.67 mol/L acid solutions (Table 3
). At an interaction time of 1 h, the degree of equilibration was <94.7% for the 0.67 mol/L acid and was >99.3% for the 0.22 mol/L acid. Apparently, 0.22 mol/L HNO3 induced acid-serum calcium isotope equilibration faster than 0.45 mol/L HNO3, and 0.45 mol/L HNO3 induced equilibration faster than 0.67 mol/L HNO3. This was based on the observation that at 1 h of interaction time, the degree of equilibration for the 0.67 mol/L HNO3 solution was lower than that for the 0.45 mol/L HNO3 solution and the degree of equilibration for the 0.45 mol/L HNO3 solution was lower than that for the 0.22 mol/L HNO3 solution. A 6-h interaction solution using 0.45 mol/L acid might barely reach a degree of equilibration equivalent to one achieved in 1 h by a 0.22 mol/L acid solution. This result confirmed our previous test results that a more diluted acid reagent might act more efficiently in calcium isotope equilibrium tests.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Before this study, an oxalate precipitation method was usually applied to extract calcium from 500 µL of a serum sample for TIMS analysis; total calcium recovery was usually >93% (4)(15). Measured calcium isotope ratios from the precipitation method agreed well with the ratio expected from calculations, with a difference of 0.2–3.7% and a mean difference of 1.3%, which was not significant (4). Isotope ratios from the precipitation calcium are taken as representative data in this report.
Because of the relatively large number of blood samples collected per individual, a fast, simple method for the analysis of these samples would be highly advantageous. In this study, we developed a sample preparation method that uses 20 µL of human serum. Calcium isotope ratios obtained from the equilibrated solutions are consistent with those from the precipitation calcium. In fact, 10 µL of serum should be adequate to conduct a similar ICP-MS calcium isotope analysis if a microflow inlet system is used. The time required for sample preparation with this new method is only several hours, which is substantially shorter than time required for the oxalate precipitation method.
The various amounts of aggregated materials formed at different concentrations of acid solutions might be affected by pH and protein net charges, which are pH dependent (16). Studies have indicated that at pH >5 there is a high percentage of calcium bound to serum albumin and globulin mainly because of electrostatic attraction (16)(17). Studies have also shown that when human serum albumin is exposed to an increasing pH, albumin ionization increases, causing repulsion between molecules, which leads to decreased formation of aggregates (18). This corresponds to our observations that although aggregates formed in the testing tubes for 0.45 mol/L acid-serum solutions and that even more aggregates formed in the 0.67 mol/L acid-serum equilibration mixture, very few aggregates formed in the 0.22 mol/L acid-serum equilibration mixture.
Total calcium loss was expected to be proportional to the amount of aggregates that formed in the mixing solution. The loss of calcium in the aggregates, however, may not have noticeable effects on isotope ratios in the equilibrated solution. Very limited calcium isotope fractionation was reported in geological and biological systems, typically 0.1–0.2% in biological samples (19)(20). In our experiments, small 42Ca and 43Ca isotope fractionations were observed only in the short-interaction-time, nonequilibrated solutions (Tables 1–3
). The magnitude of fractionation was, again, directly related to the amount of aggregates formed in the solutions and the interaction time. This small degree of isotope fractionation may be attributable to the "isotope effect" and the difference in isotope concentrations in the solution. More 40Ca2+ and 42Ca2+ (lighter, more abundant) were expected to incorporate into the aggregates than 43Ca2+ (heavier, less abundant) at the initial stage of mixing. Hence, 42Ca/43Ca ratios obtained from the early mixing solutions were lower than 42Ca/43Ca ratios from the equilibrated solution, and small isotope fractionations observed (mostly <2%, but up to 10% in extreme cases; see Table 2
).
The problem of calcium isotope fractionation with this new sample preparation technique for ICP-MS may be overcome by giving the mixture enough time to equilibrate, as demonstrated in Tables 2
and 3
. The number of calcium ions bound to aggregates was expected to be exchangeable. Our experiments suggested that 0.22 mol/L HNO3 might reach calcium isotope equilibrium with serum in 1 h, faster than either 0.45 or 0.67 mol/L HNO3. The time length required to achieve acid-serum calcium isotope equilibrium is probably related to the amount of aggregate formed in the mixing solutions. A more reliable 6-h equilibration was recommended for unknown samples with exceptionally high % excess of calcium isotope. Vigorous periodic mixing may promote the equilibration process. On the other hand, although the 0.22 mol/L acid solutions required less time to reach isotope equilibrium, supernatants from the 0.45 mol/L solutions were cleaner than their 0.22 mol/L counterparts because more aggregated materials were separated from the testing solutions. Separation of more aggregated materials from the mixing solution might benefit the ICP-MS sample inlet and ion extraction systems in the long term.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  L. Ceglia, S. S. Harris, S. A. Abrams, H. M. Rasmussen, G. E. Dallal, and B. Dawson-Hughes Potassium Bicarbonate Attenuates the Urinary Nitrogen Excretion That Accompanies an Increase in Dietary Protein and May Promote Calcium Absorption J. Clin. Endocrinol. Metab., February 1, 2009; 94(2): 645 - 653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Morris, K. M. Hawthorne, T. Hotze, S. A. Abrams, and K. D. Hirschi Nutritional impact of elevated calcium transport activity in carrots PNAS, February 5, 2008; 105(5): 1431 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Abrams, K. M. Hawthorne, O. Aliu, P. D. Hicks, Z. Chen, and I. J. Griffin An Inulin-Type Fructan Enhances Calcium Absorption Primarily via an Effect on Colonic Absorption in Humans J. Nutr., October 1, 2007; 137(10): 2208 - 2212. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Chen, I. J. Griffin, L. M. Plumlee, and S. A. Abrams High Resolution Inductively Coupled Plasma Mass Spectrometry Allows Rapid Assessment of Iron Absorption in Infants and Children J. Nutr., July 1, 2005; 135(7): 1790 - 1795. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
