Researchers from Agilent and SISCAPA Assay Technologies have developed a targeted mass spec workflow that allows for multiple-reaction monitoring-based quantitation of proteins with sample cycle times as short as seven seconds.
The workflow, detailed in a paper published last week in the Journal of Proteome Research, achieves these cycle times by replacing traditional liquid chromatography with Agilent's RapidFire solid phase extraction system, enabling a dramatic increase in the throughput of targeted protein assays.
The publication isn't "the final word on this, obviously, but it does at least put a stake in the ground with respect to the kind of throughput that could ultimately be obtained in clinical [proteomic] assays," Leigh Anderson, CEO of SISCAPA Assay Technologies and author on the paper, told ProteoMonitor. "And from that point of view I think it's significant."
Throughput has been one of the most significant challenges facing proteomics researchers as they've worked to move mass spec-based tests into the clinic. Traditionally, extensive chromatography upfront of mass spec analysis has been required to accurately and reproducibly quantify proteins in complex matrices like plasma. Such chromatography runs have typically taken as much as 30 to 40 minutes, however, greatly limiting the number of samples researchers are able to run.
In turn, this limitation has hindered the development of mass spec-based proteomics as a viable clinical tool.
As Anderson noted during a presentation at the 2011 Mass Spectrometry's Applications to the Clinical Lab meeting (PM 2/11/2011), "it's well known in the diagnostics community that if you can't run a few thousand samples, you can't know if a biomarker is clinically relevant. So this is a huge limitation that we need to overcome."
With Anderson and his SAT colleagues, Agilent has emerged as one of the primary mass spec vendors tackling this problem. Recently, the collaborators have managed to bring the chromatography runs used in their SISCAPA workflows -- which combine antibody-based peptide enrichment with MRM mass spec – down to around three minutes per sample, Anderson said. Replacing that chromatography step with the RapidFire device brings this down to around seven seconds per sample.
Originally developed by Biocius Life Sciences, the RapidFire is a solid phase extraction device that uses proprietary switching valves and specially optimized cartridges to process samples at very high speed. The technology has traditionally been popular with pharmaceutical firms for ADME work but, upon purchasing Biocius in March 2011, Agilent said the system could prove key to its clinical proteomics efforts (PM 3/4/2011).
The JPR paper is the first publication the company has released demonstrating the technology as part of a targeted proteomics workflow.
In the paper, the researchers describe how they used the RapidFire system in combination with an Agilent 6490 triple quadrupole instrument to quantify levels in plasma of two proteins – LPS binding protein and mesothelin, at concentrations of 5,000 ng/mL and 10 ng/mL, respectively, achieving coefficients of variation in the 6 percent to 8 percent range.
Using the method, "a plate of 96 samples can be analyzed in less than 15 minutes and a 1,500-sample validation study can be performed in approximately 3 hours," the authors wrote, calling the system a "high-throughput, multiplexible, biomarker validation platform that we believe is ready for wide-scale application."
The workflow is able to achieve reliable quantitation even in the absence of chromatography due to the high purity of sample resulting from the upfront antibody enrichment, Anderson said, noting that in the case of analytes for which very high quality antibodies are not available, a chromatography-based approach would likely still be necessary.
"What we can say at this point is that we've been trying really hard to make good antibodies, and they are good enough," he said. "We don't know yet how far we can relax that criteria, but it's clear that the less specific the antibodies are, the more other stuff there will be in [the sample]. When the antibodies aren't perfect, then I think the approach will be to use some chromatography" instead of the RapidFire system.
Beyond the antibody question, there are other potential limitations to the technique, perhaps most significantly its multiplexing capabilities.
"Multiplexing is one of the biggest drawbacks of the system," Can Ozbal, director of RapidFire Operations at Agilent, told ProteoMonitor via email. "We routinely run multiplexed experiments with 10 MRMs – five peptides plus five isotope internal standards – and are experimenting with higher levels of multiplexing, and performance begins to suffer as multiplexing is increased."
He added that one potential workaround for this multiplexing limitation would be doing multiple injections, noting that "since a RapidFire experiment takes under 10 seconds, even two or three injections is still significantly faster than most HPLC-MS techniques."
Anderson said he thought that the RapidFire platform would be most useful for assays in the range of three- or four-plex. "It's not going to be running 50-plex assays," he said.
There are also questions regarding the platform's sensitivity. While the researchers successfully used it to quantify mesothelin in the 10 ng/mL range, many clinically relevant protein biomarkers are present in plasma at sub-ng/mL levels.
"The trade-off with eliminating LC, of course, is you limit your resolution, and that translates to limiting your sensitivity and selectivity," Mary Lopez, director of Thermo Fisher Scientific's BRIMS Center, told ProteoMonitor, adding that the commercially available ELISA for mesothelin has a range of .08 ng/mL to around 5 ng/mL, compared to the 10 ng/mL presented in the JPR study.
Anderson suggested, however, that sensitivity could be upped by increasing the amount of sample used in the process. "This was just using 10 microliter samples," he said. "The sensitivity can be increased pretty much linearly by just using more sample."
Lopez also noted the technique's multiplexing limitations. However, she added that the possibility of a mass spec cycle time of seven seconds was "very compelling."
"That's a huge step forward in protein and peptide measurement with mass spectrometry," she said. "I think it's the [direction] all of us are hoping to push this technology. We want to get to the point where small molecule folks have been for years now in doing these very rapid assays."
Lopez's employer Thermo Fisher is, of course, a primary competitor of Agilent's in the mass spec space and has also been making significant efforts to push mass spec-based proteomics into the clinic. Much of this work has centered around the company's Mass Spectrometric Immunoassay, or MSIA, technology – like SISCAPA an immunoenrichment technique, but one that enriches at the level of proteins as opposed to peptides.
Last month, Thermo Fisher announced a deal with German diagnostics firm Immundiagnostik to develop mass spec-based assays for protein and peptide quantitation, an agreement that will offer it a venue to refine and demonstrate its clinical mass spec workflows (PM 10/26/2012).
In an email to ProteoMonitor following the announcement of the deal, Immundiagnostik spokesperson Susanne Kuhlendahl said that the company, which has traditionally specialized in immunoassays, was "convinced that [mass spec] technology will more and more replace immunoassays in laboratory diagnostics."
She added that she expected that mass spec's higher specificity and the potential cost savings enabled by its multiplexing ability would drive the growth of clinical interest in the technique.
Despite these advantages and advances like those presented in the JRP paper, significant challenges still remain. In particular, Lopez suggested, sample prep workflows up front of the mass spec analyses must be refined and optimized.
"If you look at the overall workflows, the limiting factor in sample analysis is really not the time on the mass spec or the time taken for the LC separation," she said. "The limitations are really much more comprehensive in the areas of trypsin digestion and the immunobinding and washing steps."
The JPR study offered encouraging data on that front, as well, Anderson said, noting that the researchers had been able to reduce CVs for the trypsin digestion step to below 3 percent.
Key to this, he said, was making the digestion assay as simple as possible. "We've tried to take everything out except what is absolutely necessary and make the process one which involves only the addition of reagents. So we're not doing any cleanup of the digest – no filtration, nothing other than just adding things sequentially to a well. So from that point of view it's down really to the reproducibility of the automated pipetting and things like temperature control, which is relatively straightforward to manage."
Anderson noted, though, that it was still early days for optimization and standardization of these processes. "We need to be not only optimizing the protocols but running them day in and day out over months," he said. "Then we'll know what the real practical precision levels are. But so far the results have been very encouraging."
Ozbal noted that in addition to Agilent's work with Anderson and SAT, the company is also incorporating the RapidFire platform into other targeted proteomics collaborations, including research agreements with an unnamed academic institution and several biotechnology companies as well as ongoing work with Integrated Diagnostics on workflows for that firm's protein biomarker tests for lung cancer and Alzheimer's disease (PM 1/13/2012).