A Novel Bead-Based Immunoassay for the Measurement of Heat Shock Proteins 27 and 70

Abstract

Heat shock proteins (HSPs) play an essential role in protecting proteins from denaturation and are implicated in diverse pathophysiological conditions like cardiovascular diseases, cancer, infections, and neurodegenerative diseases. Scientific evidence indicates that if HSP expression falls below a certain level, cells become sensitive to oxidative damage that accelerates protein aggregation diseases. On the other hand, persistently enhanced levels of HSP can lead to inflammatory and oncogenic changes. To date, although techniques for measuring HSPs exist, these assays are limited for use in specific sample types or are time consuming. Therefore, in the present study, we developed a single-molecule assay digital ELISA technology (Single Molecule Array-SIMOA) for the measurement of HSPs, which is time effective and can be adapted to measure multiple analytes simultaneously from a single sample. This technique combines two distinct HSP-specific antibodies that recognize different epitopes on the HSP molecule. A recombinant human HSP protein was used as the standard material. The assay performance characteristics were evaluated by repeated testing of samples spiked with HSP peptide at different levels. The limit of detection was 0.16 and 2 ng/mL for HSP27 and HSP70, respectively. The inter- and intra-assay coefficients of variation were less than 20% in all tested conditions for both HSPs. The HSP levels assayed after serial dilution of samples portrayed dilutional linearity (on average 109%, R² = 0.998, p < 0.001, for HSP27 and 93%, R² = 0.994, p < 0.001, for HSP70). A high linear response was also demonstrated with admixtures of plasma exhibiting relatively very low and high levels of HSP70 (R² = 0.982, p < 0.001). Analyte spike recovery varied between 57% and 95%. Moreover, the relative HSP values obtained using Western blotting correlated significantly with HSP values obtained with the newly developed SIMOA assay (r = 0.815, p < 0.001 and r = 0,895, p < 0.001 for HSP70 and HSP27, respectively), indicating that our method is reliable. In conclusion, the assay demonstrates analytical performance for the accurate assessment of HSPs in various sample types and offers the advantage of a huge range of dilution linearity, indicating that samples with HSP concentration highly above the calibration range can be diluted into range without affecting the precision of the assay.

Keywords: cell lysates; heat shock proteins; plasma; single molecule assay.

Figure 1

The effect of sample dilution on quantification of HSP70. To determine matrix related effects of test samples, plasma samples were spiked with recombinant HSP70 protein and twofold serial dilutions performed using sample diluent. Samples were pre-diluted four-folds and run in duplicate alongside full calibration curve. All sample concentrations were determined using the calibration curve from the run. Spearman correlation between expected and observed concentrations (R² = 0.994, p < 0.001). The error bars are indicated.

Figure 2

The effect of sample dilution on quantification of HSP27. To determine matrix related effects of test samples, plasma samples were spiked with recombinant HSP27 protein and twofold serial dilutions performed using sample diluent. Samples were pre-diluted four-folds and run in duplicate alongside full calibration curve. All sample concentrations were determined using the calibration curve from the run. Spearman correlation between expected and observed concentrations (R² = 0.998, p < 0.001). The error bars are indicated.

Figure 3

Admixture Linearity curve. To determine the linearity of sample reads along the entire calibration curve and to ensure that measurements from all parts of the calibration curve are reliable, an admixture of two plasma samples (including a sample exhibiting a low HSP70 concentration and a sample spiked with 10,000 ng/mL of human recombinant HSP70) was performed. The concentrations of the mixture were determined using the calibration curve from the run. Spearman correlation between expected and observed concentrations (R² = 0.982, p < 0.001).

Figure 4

Quantification of HSP27 and 70 by Western blotting (A) and the newly developed SIMOA technique (B). Peripheral blood mononuclear cells from samples from 7 independent volunteers were lysed unheated or heat shocked at 42 °C for 1h before being lysed. S1 = sample 1 unheated; S1HS = sample 1 heat shocked; HSP = heat shock protein. For samples 1, 5 and 6, the water in the water bath—where samples were heat shocked—penetrated the Petri dishes. However, we decided not to discard them, but rather to use them—for comparison purpose—in both the SIMOA and Western blotting techniques. The signal intensity of HSP—relative to actin—obtained using Western blotting correlated significantly with HSP values using the present SIMOA technique: Spearman correlation test (r = 0.815, p < 0.001 and r = 0.895, p < 0.001 for HSP70 and HSP27, respectively).

Figure 5

Representative standard dose-response curve and precision profile for HSP70 measurement. HSP70 was measured by quantifying signals from bound analyte on beads in units of average enzymes per bead (AEB). The average AEB (n = 2), standard deviation (SD), and coefficient of variation (CV) of each calibration point are portrayed in the embedded table. The error bars are indicated.

Figure 6

Representative standard dose-response curve and precision profile for HSP27 measurement. HSP27 was measured by quantifying signals from bound analyte on beads in units of average enzymes per bead (AEB). The average AEB (n = 2), standard deviation (SD) and coefficient of variation (CV) of each calibration point are portrayed in the embedded table. The error bars are indicated.