NEW YORK (GenomeWeb) – A team led by researchers at Saudi Arabia's King Abdullah University has developed a plasmonic nanostructure-based device capable of detecting single point mutations in peptides at picomolar concentrations.
In a study published today in Science Advances, the researchers demonstrated the device's ability to detect BRCA1 protein mutations linked to breast cancer.
The combination of plasmon resonance and Raman spectroscopy is commonly used in immunoassays, in which target molecules can be quantified by reading the shifts in plasmon resonance they cause upon binding to detection antibodies.
However, said Enzo Di Fabrizio, a King Abdullah researcher and senior author on the study, identifying single or small amounts of molecules in complex mixtures by their Raman spectra is challenging due to the difficulty of sorting out these spectra from one another.
"If you have a mixture of molecules and you want to try to see them through Raman spectroscopy it is difficult to distinguish them because the spectra can overlap and so on," he told GenomeWeb. "So you cannot put together too many different molecules."
Specifically, the authors noted, the method becomes ineffective when dealing with high molecular weight proteins or mixtures containing more than five components.
Approaches, therefore, are needed to simplify the samples being analyzed — a problem Di Fabrizio and his colleagues attacked from two different angles.
The first angle involved development of a plasmonic device with a hotspot small enough that only a limited number of molecules could be present in it at once.
This, Di Fabrizio noted, means that "if you have a complex mixture of molecules only a few of them can enter the hotspot" while "the others will be outside the hotspot and won't be excited and so you won't see them."
The researchers did this by using a design of self-similar chains, essentially a string of nanospheres of decreasing size and with decreasing distance between them.
Such nanostructures "perform very well because there are intrinsic hotspots between the smallest sphere and the previous one and if you are able to make the spheres very close together that hotspot is very small," Di Fabrizio said.
He noted that the researchers also discovered during the fabrication process that the inherent roughness and imperfections in the nanospheres serve to enhance the localization of the hotspot.
"This aspect that seems like it would be a problem is instead an advantage because the surface disorder makes for a stronger localization of the field," he said.
Having thus upped the sensitivity and specificity of their device, the researchers approached the problem from the sample side, as well, taking a page, in this case, from the mass spec-based proteomics world.
"If you want to see a single amino acid point mutation in a big protein using Raman spectroscopy it is almost impossible because in a big protein that signal will be masked," Di Fabrizio said. And so, instead of looking at intact BRCA1 proteins, he and his colleagues digested the BRCA1 proteins into peptides, which they then analyzed using the device.
"If you cut the protein and you save only the part where you suspect there is the mutation [of interest], you will end up with a mixture of small peptides" more amenable to analysis by Raman spectroscopy, he said.
In the Science Advances study, the researchers generated in vitro wild-type BRCA1 proteins and proteins with the missense mutation M1775R, which is linked to breast cancer. They then digested the proteins using the enzyme Glu-C and fractionated the resulting digesting via HPLC. This left them with two fractions, one containing the M1775 peptide along with 10 other peptides and the other containing the mutated M1775R peptide along with 13 other peptides. Using their self-similar chain-based device they were able to detect the mutated peptide at concentrations in the picomolar range.
The researchers benchmarked the approach by comparing it to a mass spec-based analysis. Although the two methods returned similar results, the Raman spectroscopy-based approach enabled measurements at concentrations three orders of magnitude lower than the mass spec method.
The method could be applied to actual clinical samples, Di Fabrizio suggested, using an approach similar to immuno-mass spec-based assays in which either the target protein or peptides could be pulled out of the sample using antibodies and then this simplified sample could be passed on to the device for detection.
Di Fabrizio said that he and his colleagues have patented their device and that they hope either to set up a company to develop it commercially or license it to an interested vendor. Mass production of the device would be relatively simple, he added, noting that the costs would likely be in the range of a few dollars each.
Di Fabrizio added that his lab also hopes to apply the device to better characterize the activity of DNA binding proteins. In particular, he said, Raman spectroscopy could provide spatial information that could help better understand such processes.
Techniques like CHiP-Seq can identify where in the genome a DNA-binding protein is active, but "very often you don't know how the protein is attached to the DNA," Di Fabrizio said.
"There are problems in biology where the spatial arrangement is very important," he noted. "So with this same technique we are trying to solve some biological problems where we are looking at the interaction between DNA and proteins and trying to better understand the spatial orientation of the protein with respect to the DNA."