June 15, 2006 (Vol. 26, No. 12)

Effective Sample Prep Is the Key to Successful Proteomic Biomarker Discovery

The proteome may be a treasure trove for biomarker discovery, but it is clear that the enormous complexity and dynamic range in protein concentration (10(5)-10(11)) creates a significant challenge for today&#8217s analytical technologies, making it extraordinarily difficult to identify rare proteins of interest to scientists and clinicians. As a result, strategies to simplify and enrich target proteins in highly complex samples are being developed to dig deep enough into a proteome to characterize biologically important differences.

The key to successful biomarker discovery and identification and characterization of proteins important for diagnostics, disease processes, drug side effects, therapeutic responses, etc., is effective sample preparation. No matter how powerful the mass spectrometers used to characterize proteins, the old saying applies “garbage in, garbage out.”

Some of the factors to consider when comparing and contrasting tools for proteomic sample preparation include their capability to handle large mass loads; depth of proteome coverage (how low can you go?); breadth of proteome coverage (are classes of proteins missing?); specificity of separation/degree of resolution; reproducibility; degree of fractionation; extent of enrichment prior to analysis; convenience; throughput (samples per week); and cost (hardware, consumables, software, labor).

When the starting material for mass spectrometry analysis is plasma, serum, CSF, urine, or other biofluids, we have found an effective solutiona strategy that partitions the proteome by immunoaffinity capture prior to multidimensional fractionation. The combination of powerful partitioning and fractionation solutions gives researchers an effective approach to study target proteins and discover important biomarkers.

IgY Immunoaffinity Partitioning

Beckman Coulter (www.beckmancoulter.com) offers a unique line of low-abundance protein enrichment chemistries, offering results on a multimilligram scale. The ProteomeLab IgY-12 spin and LC column kits selectively partition 12 highly abundant proteins found in plasma or serum, enriching the sample up to 25-fold. Multiple formats allow sample processing from 20 to 250 microliters per cycle.

The IgY enrichment chemistry has been optimized for human/primate and rodent samples. But one of the key attributes of IgY is its ability to partition orthologous proteins across species, allowing animal models and human samples to be processed using the same chemistry. (Figure 1)

This partitioning approach differs from protein depletion methods because it allows the capture and subsequent analysis of highly abundant proteins, if desired. Depletion methods automatically discard the bound material. Soluble biofluids, such as serum, plasma, CSF, urine, nipple aspirate, bronchial alveolar lavage, lymphatic fluid, follicular, ascites, tears, saliva, etc., are studied more clearly when the 12 highly abundant proteins are partitioned and the remaining medium- to low-abundance protein targets are highly enriched.

The IgY partitioning method has been applied at an industrial scale for in-line processing of hundreds of samples1. Several single component IgY antibody spin column kits are also available for researchers analyzing isoforms or biomolecules physiologically bound to highly abundant proteins, such as human, rat, or bovine albumin, IgG, Transferrin, Fibrinogen, or HDL.


Fig. 1: Structure and advantages of IgY antibodies: Polyclongla, High avidity, Broad antigen binding, Clean capture

Combining Partitioning with 2-D Fractionation

As previously mentioned, the combination of proteome partitioning and fractionation has a concerted effectboth enriching low-abundance proteins, as well as removing the masking effect created by peptides derived from highly abundant proteins. Combining these two approaches permits identification of low-abundance serum proteins in the sub-nanogram/mL range by mass spectrometry, such as skeletal muscle fast twitch isoforms of Troponin T, which is present in human serum at levels from 500 pg/mL to 1.2 ng/mL 2 (Figure 2). Protein identification at this depth is a breakthrough in this industry, paving the way for the discovery of many relevant biomarkers.

Beckman Coulter&#8217s ProteomeLab PF 2D (Proteome Fractionation in 2 Dimensions) System fractionates a proteome in the first dimension by pH using chromatofocusing and then automatically injects each fraction into the second dimension for ultrahigh-resolution separation by hydrophobicity. The system generates liquid fractions of the proteome, as well as pH/hydrophobicity 2-D maps, which identify regions of differential protein expression3. With up to a five-mg mass load, the system allows the isolation of low-abundance proteins at levels that can readily be identified by mass spectrometryincreasing the working range of biomarker discovery in biofluids, tissues, cell lines, or microbes.

2-D liquid fractionation can resolve high and low molecular mass proteins and peptides, because the second dimension is not size-based. Proteins remain soluble throughout the partitioning and fractionation process, with no need to in situ digest and extract from a polyacrylamide gel matrix having attendant low yields. Membrane and other highly hydrophobic proteins remain accessible to analysis. The top-down approach preserves intact protein primary structure and reduces complexity during fractionation compared to bottom-up digested peptide strategies. In situ protease digestion of fractions is easily accomplished at high yields after intact protein fractionation.


Fig.2: Using partitioning and PF 2D fractionation, proteins have been identified by mass spectrometry along a concentration range of 10(8) from human serum.

Conclusion

Although any single protein preparation technology is unlikely to be ideal for all samples, the more one can simplify the sample, reduce the dynamic range of protein concentrations, enrich for medium- and low-abundance proteins, and remove interfering peptide mass fingerprints, the better the mass spectrometry results and the deeper one can dig into the proteome.

Additional partitioning strategies to simplify proteomes will emerge, using techniques orthogonal to antibody-based approaches, such as subcellular fractionation and affinity enrichment of phosphoproteins, glycoproteins, lipoproteins, and other classes of post-translational modifications. Fractionation tools will also get more powerful with capabilities of handling larger mass loads, more sophisticated graphical and imaging tools, more sensitive protein detection systems, and more automation linking fractionation to mass spectrometry analysis.

Many important biomarkers will emerge from these discovery efforts, having significant impact on the development of clinical diagnostic panels and the improvement of human health.

Previous articleLew Kontnik
Next articleRichard A. Gibbs, Ph.D.