Technical note
Formulation of immunoassay calibrators in pasteurized albumin can significantly enhance their durability

https://doi.org/10.1016/j.jim.2009.12.001Get rights and content

Abstract

Calibrator matrix can have significant effects on the commutability of assay standards and on the maintenance of their integrity. We have observed marked instability in progastrin-releasing peptide (proGRP) assay standards traceable to the bovine serum albumin (BSA) used in matrix formulation. Attempts were made to improve calibrator stability using different albumin pretreatments. Observed analyte recoveries in calibrators prepared with untreated BSA were consistently less than 45% after 1 week of storage at 4 °C. Pre-treating the BSA by chromatography on immobilized heparin or benzamidine failed to improve calibrator durability with day 7 recoveries of less than 55%. In marked contrast, calibrators formulated with albumin pasteurized at pH 3.0 displayed remarkable stability. Recoveries of > 97% were observed after 4 weeks of storage at either 4 °C or room temperature. Even calibrators incubated for 4 weeks at 37 °C gave recoveries between 91–106%. This improvement was not seen with BSA pasteurized at neutral pH.

Albumin pretreatment is straightforward, easily scalable and dramatically improves calibrator stability. Matrix formulated with acid-pasteurized BSA may prove more generally useful when assays are plagued by poor calibrator durability.

Introduction

The successful implementation of an immunoassay is highly dependent on the availability of reliable standards. Degradation of calibrator performance over time can thus have a significant impact on any calibration strategy. Often assay standards have to be prepared in the absence of a fully appropriate analyte-free biological matrix. Thus, it is common practice to formulate calibrators in buffered Cohn fraction V bovine serum albumin. Although these artificial matrices offer certain benefits when compared to pooled normal sera, problems can arise due to the poorly defined nature of ethanol fractionated BSA. Of particular concern is the presence of variable levels of endogenous enzymes which can have unpredictable effects on analyte stability.

We have recently developed a fully automated time-resolved immunofluorometric assay (TR-IFMA) for proGRP (Nordlund et al., 2008b, Nordlund et al., 2008a). Due to the marked heterogeneity of proGRP in biological matrices we chose to calibrate using a recombinant proGRP peptide expressed in Escherichia coli. However, during development we noted a significant instability in these recombinant calibrators. With certain batches, storage at 4 °C for as little as 24 h resulted in proGRP recoveries of less than 90% which fell below 45% by day 7 (Nordlund et al., 2008a). Significantly, matrix buffer prepared with BSA marketed as “RIA grade” provided somewhat better stability. This observation strongly implicated the BSA component as contributing to the poor calibrator durability. It further indicated that pretreatment of the albumin used in matrix formulation could potentially improve calibrator durability. We thus tried to emulate the commercial “RIA grade” albumin by charcoal/dextran stripping fraction V BSA using a modification (Iscove et al., 1980) of the method described by Chen (Chen, 1967). In addition, we screened BSA that had been pre-treated by passage through two affinity resins selected for their ability to bind serum proteases.

Section snippets

Albumin pretreatments

Fraction V BSA (Sigma Chemical Co, St. Louis, MO) was heat-treated by titrating a 10% aqueous solution to pH 3.0 with 5 mol/L HCl and holding at 56 °C for 30 min. After cooling to room temperature the pH was adjusted to 6.8 with 5 mol/L NaOH and the solution passed through GFA/C glass fiber filters (Pall Corporation, East Hills, NY). Charcoal-extracted BSA was produced in a similar manner to the heat-treated BSA but with the inclusion of dextran-coated charcoal during the heating step (Iscove et

Results

Calibrators prepared with BSA treated by group-specific affinity chromatography did not show improved durability. In matrix containing 100 ng/L proGRP observed recoveries at day 7 were 38% and 52% for heparin-agarose and benzamidine-agarose pre-treated BSA respectively (Fig. 1). In marked contrast, the use of charcoal/dextran stripped BSA resulted in significant improvements in stability with day 7 recovery approximating 95%. However, control experiments, in which BSA was heat-treated without

Discussion

Serum and purified albumin fractions have long been known to contain appreciable levels of protease activity (Jobling et al., 1915, Peters, 1996). Inactivation of these proteases is most probably responsible for the improved calibrator stability seen in matrix prepared with pasteurized BSA. The requirement for an acid pH during the pasteurization process could possibly be related to the presence of heat-stable protease activity in many albumin preparations (Peters, 1996, Wilson and Foster, 1971

Conclusion

The replacement of calibrator matrix formulated with crude BSA with that containing pasteurized albumin extended the lifetime of proGRP assay calibrators from < 48 h to at least 4 weeks. Pre-treating BSA by acid pasteurization is simple, reproducible and avoids the use of highly toxic and unstable protease inhibitors. We believe that similarly treated albumins may prove a useful addition to the armentarium needed when preparing stable assay calibrators.

Acknowledgment

This work was supported by the Norwegian Cancer Society.

References (9)

There are more references available in the full text version of this article.

Cited by (2)

  • Biotechnology of plasma proteins

    2012, Biotechnology of Plasma Proteins
View full text