Researchers from the Institute for Systems Biology in Seattle, including ISB’s founder and president Leroy Hood, have devised a fast and high-throughput method for the quantitative analysis of serum proteins using surface plasmon resonance imaging.
Described in an article on the online version of Molecular & Cellular Proteomics, the method is meant to address a major challenge to proteomics: detecting and quantifying specific proteins in complex biological mixtures.
Using serum as their biological fluid, the team developed methods to print antibody microarrays and used them to relatively and absolutely quantitate proteins in the serum.
In the article, the authors say that microarrays offer “good specificity and sensitivity” and that the use of antibody microarrays has grown recently as researchers “look for high-throughput measurements to complement their transcriptomic capabilities.”
However, most protein-profiling microarray methods require samples to be fluorescently labeled, often by using secondary antibodies, which results in high costs and intense labor requirements.
By comparison, SPR imaging monitors protein-binding microarrays in a label-free, real-time manner, making the method fast, economical, and free of label biases. The real-time aspect “allows for binding kinetics to be observed and for the quality of the antibody-protein interaction to be evaluated,” according to the authors. And SPR imaging is reusable, allowing large numbers of samples to be processed on the same sensor chip “efficiently.”
Still, though SPR imaging has been used to monitor molecular interactions such as carbohydrate-protein, peptide-protein, and protein-protein biding, as well as surface enzyme kinetics, its use for serum screening is relatively new, primarily due to the fluid’s complexity, according to the authors.
“The objective of this study was to determine if SPR imaging provides quantitative proteomic data from a complex biological material such as serum,” the authors said.
To that end, the authors fabricated a microarray with standard pin-spotting on bare gold substrates. Samples were applied for binding analysis with a camera-based SPR system. After determining that regeneration of the microarray surface was “highly successful,” the researchers set about validating their method.
“The objective of this study was to determine if SPR imaging provides quantitative proteomic data from a complex biological material such as serum.”
They measured the concentration of four serum proteins using part of a 792-feature microarray and found that in human serum, the transferring level was found to be 2.0 milligrams per milliliter, “which closely matched the 2.6 mg/mL obtained by ELISA analysis,” the researchers wrote. Transferrin levels in mouse serum using both the SPR platform and ELISA were found to be 1.2 mg/mL.
Human and mouse albumin levels were measured at 24.3 mg/mL and 23.6 mg/mL respectively, also in agreement with expected levels, the scientists reported.
They then tested their method to profile serum for liver cancer. The team used liver cancer serum due to the availability of large number of antibodies to liver-specific proteins and human serum samples from healthy, liver cancer, and other cancer subjects, as well as a good protein biomarker, alpha fetoprotein, to serve as a positive control.
Hierarchical clustering divided the serum samples into two distinct groups, and 39 distinct protein changes were observed. The levels of 35 proteins increased while the levels of four decreased. As they said they expected, their system found increased levels of alpha fetoprotein. In addition, it detected increased levels in eight other proteins that had been previously reported to increase in the serum of liver cancer patients: dipeptidyl-peptidase; complement C4; tumor necrosis factor; interleukin 4; interleukin 6; interleukin 10; catalase; transferrin; and carbamoyl-phosphate synthetase.
The system also detected a decrease in fructose-biphosphate aldolase, which had been previously observed in other studies. Of the 39 protein changes, 26 correspond to genes previously observed to show differential mRNA regulation in hepatic cell carcinoma.
In aggregate, the results suggest that SPR imaging analysis of microarrays is an “efficient means of serum profiling,” the authors said. Sample preparation was fast, convenient, and required no labeling “so the associated effort, expense, and artifacts [from labeling] were avoided,” they said.
Microarray fabrication was also easy and hundreds of microarrays could be printed with a 10-microliter volume of antibody, with many samples able to be analyzed by each array.
Another advantage of their method is that because the arrays can be regenerated, “inter-array variability can be eliminated and more experimental replicates can be conducted,” according to the researchers.
SPR microarrays also have the benefit of real-time analysis, providing information about the probe-target interaction. Most importantly, real-time analysis allows researchers to calculate kinetic parameters, which can be used to determine the affinity constant of an antibody and for absolute quantification on other microarrays.
When the parameters are unavailable, quantification requires a standard curve for each array “since each varies in the amount of immobilized antibody [but] when kinetic parameters are known, concentration can be calculated from binding rates,” they said.
However, there are drawbacks to using SPR imaging to read antibody microarrays, compared to label-based detection methods, they said. Sensitivity can be limited in some instances, though that can be improved by using a secondary antibody to amplify the signal, and “even without amplification, many disease-related serum proteins normally occur at measurable levels,” according to the researchers.
Another challenge for proteomic applications is “off-target” binding as single-antibody assays require highly specific binding properties while a sandwich assay can use two less-specific antibodies, which in combination will provide sufficient specificity.
Finally, the dynamic range of SPR imaging is limited by the width of “the linear region of the surface plasmon resonance curve and the data resolution of its camera detector,” the researchers note.
Even with the limitations, their results “demonstrate the feasibility of this high-throughput approach for both absolute and relative protein expression profiling,” the authors wrote.