Two research teams have separately developed new imaging mass spec technologies for targeted protein measurements in tissue.
The approaches, both of which use metal-conjugated antibodies to target proteins of interest, allow for highly multiplexed protein analyses at subcellular resolutions, potentially offering improved ability to delineate cell populations and investigate matters including cell-cell interactions and tumor heterogeneity.
Detailed in a pair of papers published last week, one in Nature Medicine and the other in Nature Methods, both efforts have their origins in the lab of Stanford University researcher Garry Nolan, with Nolan's lab leading the Nature Medicine study and the Nature Methods study led by University of Zurich researcher Bernd Bodenmiller, formerly a post-doc in Nolan's lab.
Nolan has been one of the earliest adopters of mass cytometry, a technique developed at the University of Toronto by researchers including Scott Tanner, who went on to found the mass cytometry firm DVS Sciences, which was purchased this January by Fluidigm.
DVS's mass cytometry instrument, the CyTOF, combines capabilities of flow cytometry and atomic mass spectrometry, allowing it to measure large numbers of proteins in single cells with high throughput. Atomic mass spectrometry detects proteins using antibodies linked to stable isotopes of elements, which can then be read with high resolution via mass spec.
While mass cytometry is most analogous to flow cytometry, the recent studies from Nolan's and Bodenmiller's groups have applied the underlying atomic mass spec approach to tissue imaging, creating workflows that might be viewed as advances over conventional immunohistochemistry.
Key to both efforts is the use of the metal-conjugated antibodies to tag proteins of interest, which allows for multiplexing in the range of 100 analytes. Where the two methods differ is in their approach to ionizing their samples for mass spec analysis, with the Nolan lab using multiplexed ion beam imaging (MIBI) and the Bodenmiller team using laser ablation. Additionally, while Bodenmiller and his colleagues have adapted DVS's (now Fluidigm's) CyTOF machine as their read-out platform, Nolan's team in the Nature Medicine study used secondary ion mass spectrometry on a magnetic sector mass spectrometer and plans to build a custom TOF instrument for future work, Michael Angelo, a post-doc in Nolan's lab and author on the study, told ProteoMonitor.
Both methods have their advantages and disadvantages, Angelo said, noting that perhaps the main advantage of the MIBI technique is its higher potential resolution.
Resolution is linked to the size of the beam spot used for ionizing the target tissue, and in the case of laser ablation methods, he said, "you bump up against a lot of thermodynamic and physical limits when you try to get the beam spot size small."
MIBI, though, uses a particle beam that "can focus the spot size down to an almost unlimitedly small diameter," he said, noting that while in the Nature Medicine paper he and his colleagues achieved resolution of around 200 nanometers, with other sources they could possibly get as low as 10 nanometers. The laser-ablation, CyTOF-based approach developed by Bodenmiller's lab, on the other hand, achieved resolution of 1 micrometer.
While this is a lower resolution than the MIBI approach, it represents "a big breakthrough" for laser ablation, Angelo said, noting that with laser ablation mass spec "the spot size of the laser has always been so large that it has been very hard to get subcellular resolution."
Indeed, Bodenmiller told ProteoMonitor, that while he first considered using the CyTOF for imaging studies while still a post-doc in Nolan's lab, it was not until he started his own lab at the University of Zurich and met Swiss Federal Institute of Technology Zurich researcher Detlef Günther, an expert in laser ablation and co-author on the Nature Methods study, that the idea came together.
One of Günther's students had developed a laser ablation cell that offered a significant improvement in resolution compared to existing devices, and "when I met [Günther], I saw that this technology could be couple to the mass cytometer for imaging," Bodenmiller said.
Combining the laser ablation cell with the CyTOF, Bodenmiller and his colleagues applied their technique to the study of breast cancer tumors, simultaneously imaging 32 proteins at subcellular resolution. Ultimately, he noted, the method could measure up to 100 proteins simultaneously.
Nolan and his team likewise applied their MIBI method to breast cancer tumors, however, with their current magnetic sector mass spec platform, the researchers could multiplex only around seven proteins, Angelo said. They are currently working to build a custom TOF instrument that will allow them to multiplex up to 100 targets while also significantly improving the system's acquisition speed, he said, noting that they aim to have this instrument completed within the next 12 months.
In both studies, the researchers analyzed formalin-fixed paraffin-embedded tissue, one of the most common storage formats for clinical samples. Both workflows are intended as improvements over conventional IHC, which is typically limited to multiplexing of between two and four analytes.
In each study, the researchers found good correspondence between their measurements and those obtained via conventional IHC assays. Additionally, in the Nature Methods work, Bodenmiller and his colleagues were able to identify interpatient tumor heterogeneity within classical breast cancer subtypes, suggesting, he said, that this information could lead to better targeted treatment within subtypes.
"Now that we have found a way to recognize that there are these differences in subpopulations ... knowing about the presence of a population [within an existing subtype] that does not respond well to treatment could then tell the clinician, for instance, that for this patient we need to give this additional drug that also takes care of this unique subpopulation," he said.
Bodenmiller said that his team is now using their approach to analyze several hundred additional breast cancer samples, with a focus on identifying protein profiles linked to metastasis and to treatment response.
"We have access to large tumor collections that have a wealth of clinical information associated with them, and we are now starting to analyze those with our approach with the hope that we will find some correlations to the clinical data," he said.
Angelo said that Nolan's lab is likewise looking to apply the technique to clinical questions, noting that it would allow them to "correlatively look at really large cohorts of tumors that have been archived over decades and match that up to the follow-up data that already exists to figure out what biomarkers correlate with what outcomes."
Another interesting possibility raised by being able to multiplex a number of markers in tissue is simultaneously examining "the immunophenotype [of tumor cells] and the immunophenotype of the infiltrating immune cells," he said. "So you can look at the interplay between cells in the tumor microenvironment like tumor-associated fibroblasts and tumor cells to see how those things key off one another."
Some of this work, however, will have to wait until the researchers build their higher-throughput, higher-multiplexing TOF system, Angelo said. "Larger scale studies looking at hundreds of tumors we will probably defer until we have that built. But we have a few smaller, targeted proof-of-principle studies underway."
Though still in the early stages of development, both systems could prove competitors to GE Healthcare's MultiOmyx protein detection system, which the company launched last year through its Clarient Diagnostic Services.
The MultiOmyx platform uses antibodies conjugated to fluorescent dyes to stain proteins of interest in batches of two to four at a time. Researchers then image the stained tissue and deactivate the fluorescent dyes via a proprietary process. They can then stain the tissue with the next round of antibodies, multiplexing in an iterative fashion.
In an interview last year, Michael Gerdes, a GE Global Research biologist and leader of the platform development effort, told ProteoMonitor that company researchers had measured as many as 65 proteins in a single sample and could potentially multiplex significantly more.
Last July the company launched its Hodgkin Lymphoma Profile diagnostic on the platform and has since published data from studies using the system for work in colorectal and breast cancer.
The MultiOmyx platform "is very impressive," Angelo said. "They've done a lot of amazing stuff there." However, he added, he believed the mass spec-based approaches presented by his and Bodenmiller's teams could ultimately provide advantages in terms of assay speed and multiplexing.
As owner of the CyTOF technology, Fluidigm could potentially offer some version of the imaging technique developed by Bodenmiller's lab on the device. Bodenmiller declined to comment, however, on whether he had discussed with the company any plans to do so.