Peter F. McLean and Dr. Craig Thulin, Department of Chemistry and Biochemistry
Biomarkers in blood serum are the keys needed for the early detection of several types of cancer, complications of pregnancy, and potentially many other diseases, to allow time for successful treatment. Blood serum is ideal because it comes in contact with all parts of the body, its sampling is minimally invasive, and it is easily prepared with high reproducibility. There is a problem, however, that prevents the development and wide-scale utilization of this treasure chest of biological information. Pure blood serum is complex and cluttered with high-abundance non-target proteins which makes examination of the low-abundance and low-molecular weight potential biomarkers extremely difficult. Current protocol for blood serum preparation relies largely on incomplete and ineffective methods, making biomarker proteomics insufficiently reliable and cost effective for clinical application. The project which was carried out focused on the accuracy, reproducibility, simplicity, and speed of sample preparation in order to make biomarker analysis viable for wide scale application.
The methods for developing a separation protocol are summarized in three steps. First, determine the elution point of serum albumin via fractionation. Serum albumin is one of the largest proteins in blood (66 kDa) and is by far the most abundant, ranging between 60-80% of total protein mass in blood serum. For this reason, it was selected as a benchmark for the effectiveness of the separation. A step-wise gradient of increasing acetonitrile (ACN) concentration (20-80% by volume) was applied to 16.7 µL columns loaded with raw bovine serum and the resulting fractions were analyzed to determine at what percent organic serum albumin would be eluted from the column. The next step involved the screening of various eluents. The investigated buffers included mixtures of ACN and water with or without formic acid, as well as pH buffered solvents. Finally, after selecting a specific eluent, it was tested as the mobile-phase. In addition, binding capacity of the selected resin and loading concentration of raw serum were revisited under the modified mobile-phase conditions.
Much of the analysis for the first stages of method development were performed by SDS-PAGE. This allowed the visualization of the fraction components by mass, enabling identification of serum albumin and its relative concentration. By this same method of analysis, we narrowed the conditions of the mobile-phase down to one, which gave the most effective separation of serum albumin with sharp elution point conditions. The selected set of conditions, including the selected mobile-phase, was then performed and analyzed by High-Performance Liquid-Chromatography Mass-Spectrometry (HPLC-MS). The resulting comparison of treated and untreated serum (Figure 1) clearly shows an enrichment of many low molecular-weight compounds by up to two orders of magnitude (Figure 2). Furthermore, it shows nearly complete removal of our benchmark protein, serum albumin. Additional analysis of these spectra reveal that several hundred unique compounds can be identified in the treated serum sample that were not detectable in the untreated sample, although they were not characterized by this study.
In conclusion, this method of solid-phase extraction was successful in a number of ways. First, we showed that serum albumin, along with other major high molecular-weight proteins, can be removed from serum by solid-phase extraction. Second, an increase of over two orders of magnitude in the signal intensity of low-molecular-weight peptides was demonstrated, presenting options for further analysis. Finally, as a result of enriched low-molecular-weight species, many proteins and peptides that were undetectable in raw serum appear as prominent peaks.
On the other hand, this method does not claim to enrich 100% of the peptides present in blood serum, for a several reasons. First, the mechanism of separation employed by solid-phase extraction is not specific enough to ensure that no small molecules are binding to column. In fact, we expect that a number of very hydrophobic proteins and peptides will bind to the column, possibly irreversibly. However, it is well known that each method of serum purification or enrichment is specific to a particular subset of low molecular-weight compounds. So while the separation is effective only with a limited array of compounds, it can be a very useful or even preferred method of analysis for those compounds. Second, there was insufficient study to determine the quality of separation of the small peptides adherent to large carrier proteins like serum albumin. Third, the method we employed on the scale we used required us to combine the fractions of three separations to obtain sufficient protein to be analyzed by LC-MS. So while the method proved successful, its practicality was limited by the very small scale we developed. To address this issue, a brief investigation of the amenability of this method to larger columns and larger volumes was conducted, but remained inconclusive.
Still, we have shown this method’s ability to enrich a subset of the low-molecular-weight blood serum proteome, and optimization for specific analysis is a simple, straightforward process. In addition, this method of solid-phase extraction is an inexpensive, undemanding, and rapid method of enrichment which remains robust and amenable to automated systems with a great degree of versatility. Finally, I had the opportunity to present this work at the Spring Research Conference at BYU, and was selected as a semi-finalist in the 2006 Association of Biomolecular Resource Facilities (ABRF) National Conference Poster Competition in Long Beach, CA.