NEW YORK (GenomeWeb) – A team led by researchers from Uppsala University has developed an assay for protein detection using proximity-dependent initiation of hybridization chain reaction.
The method, named proxHCR, is a variation on the proximity ligation assay (PLA) technology that was also developed by Uppsalaresearchers. Unlike PLA, however, the proxHCR assay does not require enzymes, making it less expensive and more accessible, Ola Söderberg, an Uppsala researcher and leader of the development effort, told GenomeWeb.
Rights to the assay are owned by Uppsala-based Olink Bioscience, which also holds the rights to the PLA technology.
Both PLA and proxHCR use pairs of antibodies attached to unique DNA sequences to detect proteins of interest. When the antibodies bind their targets, the attached DNA strands are brought into proximity and ligate, forming a new DNA amplicon that can then be quantified via techniques including real-time PCR and hybridization of fluorophore-containing oligos. The quantity of the DNA corresponds to the quantity of the target protein.
Because the DNA labels can be constructed to hybridize only to their specific partner, these techniques eliminate the common immunoassay problem of antibody cross-reactivity. Compared to conventional immunoassay methods like ELISA, they offer higher multiplexing capabilities, lower sample demands, and very high sensitivity.
The proxHCR method differs from PLA primarily in its use of hybridization chain reaction to generate the combined pieces of DNA that serve as the readout for the protein detection. HCR uses kinetically trapped hairpin DNA structures that remain separate in the absence of an initiator nucleic acid but, in the presence of this initiator, combine to generate a double-stranded piece of DNA that will continue to grow until all the hairpin structures have been used.
The reaction, described in a paper published last week in Nature Communications, takes place in several steps. First, two antibodies attached to the hairpin structures bind their common protein target, bringing the hairpins into proximity. The initiator oligo is then added, which begins the chain reaction by binding to one hairpin structure which then binds to the next which then binds to another, with the reaction continuing until all available hairpin molecules have hybridized.
In this way, the hybridization of the DNA probes is able to proceed without the enzymes required by the PLA approach. This is notable in that these enzymes make up more than half the cost of a typical PLA, said Söderberg, who was also among the developers of the PLA approach. He estimated that the proxHCR approach would bring costs down from around $20 per assay to the $5 to $6 range.
"That means that labs that don't have the financial resources [for PLA] can still do this kind of experiment," he said, adding that thus far his group had found proxHCR to be comparable to PLA in terms of assay performance.
He and his co-authors noted that the reduced cost makes the assay potentially better suited than PLA to purposes like high-throughput screening of protein interactions or point-of-care testing for infectious diseases in developing nations. Additionally, the elimination of enzymes from the assay means researchers don't need a facility for refrigerating those enzymes — another advantage from the perspective of point-of-care testing in resource-constrained environments, he said.
"If you want to try to develop some point-of-care device, you have here an assay that is very cheap and not so complicated," he said, adding that the assay's two-antibody design would — as with PLA — offer potential gains in specificity and sensitivity over conventional immunoassay methods used for infectious disease testing.
Söderberg said that this was the application he was most interested in pursuing with the technology, but noted that due to the difficulty of lining up sufficient funding for such research, it will probably be several years before he and his colleagues can start work in this direction.
In the meantime, Söderberg said, he and his colleagues are developing the method along several different lines.
For instance, they hope to develop an ELISA-style version of the approach capable of capturing proteins in solution and doing the readout in microtiter plates or microfluidic systems. In the Nature Communications paper, they presented versions of the assay for detecting protein-protein interactions and modified proteins in situ, including assays for measuring E-cadherin/β-catenin and phosphorylation of proteins including platelet-derived growth factor receptor-β, Akt, and Syk.
They also demonstrated its use with a flow cytometry readout, developing an assay that detected the binding of epidermal growth factor to EGF receptor.
Söderberg and his colleagues are also looking to combine proxHCR protein detection with RNA readout, he said.
Additionally, they are working on multiplexing the assay. Multiplexing immunoassays is typically challenging due to issues of antibody cross-reactivity. However, because proxHCR uses oligonucleotides as a readout it should be more amenable to multiplexing, Söderberg said.
In fact, the Nature Communications authors noted that the structure of the hairpin molecules used in the approach "makes them very sensitive to changes in oligonucleotide sequence," which suggests their suitability for multiplexing.
"Oligos are more predictable [than antibodies]. You just have to be careful with the oligo design so you don't get cross reactivity," he said. "You can use software that I think works really well to predict the hybridization [of different oligos] and what will occur at certain temperatures and concentrations. For antibodies it is unknown where exactly [on a protein] the antibody binds in many cases, and they are not as predictable. So they are much more complicated [to multiplex.]"