Peanut protein extraction conditions strongly influence yield of allergens Ara h 1 and 2 and sensitivity of immunoassays(1)
Extraction is the basis for all subsequent measurements of the allergen content of peanuts, yet extraction conditions differ vastly in the literature. Some studies have demonstrated that the yield of crude peanut protein is highly dependent on extraction procedure and choice of buffer.
Unless stated otherwise, following defatting with n-hexane, 1.350 ml of extraction buffer was added to exactly 45 mg raw peanut flour and vortexed for 30 s. Proteins were extracted under constant agitation on a shaker. The extraction conditions were varied to determine the optimum protein yield and better conditions adopted in the experimental process. Samples used to test different centrifugation parameters and extraction buffers were extracted at 40°C for 1 h. Samples used to test different extraction temperatures were extracted for 1 h at 4, 21, 40 and 60°C. Since an extraction at 40°C was less effective, the effect of extraction times (30 min, 1 and 2 h) was tested at 21°C. Furthermore, one set of samples was vortexed for 1 min (instead of 30 s) without further extraction on a shaker. Because the extraction at 30 min was most effective, all subsequent experiments(exhaustive extraction, thermally processed peanuts and defatting reagents) were performed at 21°C, for 30 min. After protein extraction, all extracts were centrifuged three times at 12,600 × 90 g for 5 min. The supernatant containing the proteins was aliquoted and stored at -80°C. Three biological replicates were used for each extraction.
All buffers at pH 8.0 or 8.5 gave 27–45% more crude protein than their counterparts at pH 6.7–7.8 or pH 2.1. At pH 7.2, Tris gave an average of 36% more crude protein than TBS, and 25% more than ammonium bicarbonate at pH 7.8. At pH 8.5, these effects were even stronger, with Tris being 52% more effective in extracting crude protein than TBS and 46% more effective than ammonium bicarbonate. Adding reducing and denaturating agents to Tris did not increase the crude protein yield. The yield of Ara h 1 was highly susceptible to buffers of low pH, such as PBS (pH 2.1), and denaturants. For Ara h 2, buffers at pH 2.1 and in the pH range 7.8 to 10.6 gave different and substantially greater yield than buffers of pH 4.5 to 7.2. All protein extracts prepared with the 20 buffers extracts, except those made with PBS (pH 2.1) and citrate pH (4.5), showed essentially the same 1D-gel profile. TBS extracts gave Western bands at 42 and 50kDa that were absent with Tris. Extraction with PBS (pH 2.1) gave a distinct Western profile, with 15-and 23-kDa bands being unique to this extract and a band at 18 kDa having a particularly high intensity.
The Westerns prepared with TBS extracts contained 18 high-volume spots, identified as fragments of Ara h 1 and/or 3, Ara h 7 precursor and thioredoxin. Overall, the TBS extracts resulted in more sensitive Western blots, with 10 additional spots compared to PBS and Tris extracts.
Although the 1D-gel profiles of the tested extraction times did not differ, only 80% of crude protein, including around 40% of Ara h 1, was present after extraction for 1 min. Most Ara h 2 was extracted in the first minute and, while increasing slightly with time, the values were also associated with a large statistical variance. Defatting with n-hexane allowed extraction of 24% more crude protein, 90% more Ara h 1 and 20% more Ara h 2 than defatting with diethyl ether, although 1D-gel electrophoresis showed very similar profiles. Approximately half as much crude protein was solubilised from roasted or boiled peanuts compared to raw peanuts. For roasted peanuts, while the content of Ara h 2 was 76% less than for raw peanuts, the content of Ara h 1 was 40% more than in raw samples. The content of Ara h 1 and 2 for boiled peanuts was 70–87% less than in raw peanuts, but all major bands were present with lower intensity in 1D-gels.
When seed coats were defatted and extracted with various buffers, negligible crude protein and no Ara h 1 was detected, while Ara h 2 was less than 1% of that in the respective peanut kernel extracts. The fastest and most effective extraction method for extracting crude protein and Ara h 1 and 2 simultaneously involved using 20 mM Tris (pH 8.5) for 30 min at 21°C. These conditions resulted on average in the solubilization of 582 mg crude protein, 13.5 mg Ara h 1 (2.3%) and 41.3.
At similar pH values, Tris gave around 50% more protein than TBS under various conditions , suggesting that NaCl interfered with the extraction. Urea and SDS increased Ara h1 extraction. This may increase IgE-binding epitopes by changing the structure of a homotrimer of Ara h1. Extraction with sodium carbonate (pH 10.6) led to a high rate of crude protein and was effective in extraction of Ara h 2 but Ara h 1 must have aggregated, leading to an altered sensitivity of the ELISA. There are 10 known IgE-binding epitopes in Ara h 2, distributed throughout the molecule. Presumably DTT affected the Ara h 2 structure and disrupted the epitope for the monoclonal antibody in the Ara h 2 ELISA. Since SDS and urea are used in the Western blot protocol, Ara h 1 and 3 are present in the denaturated conformation. The addition of DTT to the Western assay would also have altered the confirmation of Ara h 2. Given that peanuts contain around 50% lipid (www.pca.com.au) and the peanut flour used was defatted, the total protein content in the peanut kernels measured is estimated to be around 29% peanut protein in relation to the whole kernels.