NEW YORK (GenomeWeb) – A team led by researchers at the Massachusetts Institute of Technology has developed a method for super-resolution imaging of the proteome of intact tissues.
Detailed in a paper published this week in Nature Biotechnology, the method combines expandable hydrogel-tissue hybrids with antibody labeling, allowing scientists to observe the three-dimensional structure of cell and tissue proteomes.
Compared to previous methods, this one is able to image much larger pieces of tissue and with higher throughput, said Kwanghun Chung, an MIT researcher and senior author of the paper. He and his colleagues hope to use the approach to generate comprehensive high-resolution maps of animal and human brains, he said.
Tissue-hydrogel hybridization is a tissue treatment that can be used to preserve a sample's three-dimensional structure and the architecture of its proteins while also making it transparent and permeable to reagents like antibodies. The tissue structure can then be investigated using methods like super-resolution imaging or expansion microscopy, which uses protein digestion to enable up to fourfold expansion of these hybrids, making them compatible with conventional microscopy techniques.
However, Chung and his co-authors noted, these approaches have limitations. For instance, the proteases used in expansion microscopy destroy some of the protein content. Another issue is the need to section portions of the tissue to allow antibodies to more rapidly penetrate areas with limited permeability. This, they noted, results in the loss of important information about intercellular structure.
To address these challenges, the MIT team devised a method that prevents protein crosslinking during the hydrogel-tissue hybridization step by increasing the concentration of acrylamide, which reacts with methyl groups that otherwise facilitate the crosslinking. They then treated the hybrids with detergent at high temperatures to denature and dissociate the proteins, allowing the hybrids to expand.
They tackled the permeability problem using stochastic electrotransport, a method Chung and his colleagues explored in a 2015 paper in the Proceedings of the National Academy of Sciences. Relying on normal diffusion processes, it could in some cases take weeks for antibodies to reach their protein targets in the tissue-hydrogel hybrid, while pressure-based and electrokinetic methods aimed at speeding the process can do damage to the sample. Using stochastic electrotransport, on the other hand, the researchers were able to label their samples with antibodies in one day.
The combination of easier labeling and expanding the hybrid without losing content allows for higher-throughput super-resolution imaging than was previously possible, Chung said.
"Using this technique [called MAP, for Magnified Analysis of the Proteome], we can get super-resolution images [similar to] using low-throughput super-resolution microscopy or electron microscopy," he said.
Like MAP, these latter techniques allow researchers to investigate antibody-labeled three-dimension structures, but, Chung said, they are slow and limited to much smaller sample sizes.
For instance, he said, super-resolution microscopy methods were typically limited to samples of around 100 micrometers, while with MAP, researchers are able to get super-resolution images of several millimeters of tissue. The method provides resolution of around 60 nanometers, which Chung said is comparable to super-resolution microscopy techniques.
He added that the apparent stability of the MAP samples should allow for many rounds of immunostaining, meaning researchers can look at high multiplexes of proteins. To date, Chung and his colleagues have done up to seven rounds of staining on a sample, using three to four antibodies per round. He noted, though, that the tissue was still fully intact after these seven rounds, suggesting that more rounds are feasible.
"We don't know the upper limit, but we think that we can do many more," he said.
One concern for the researchers was how well their antibodies would perform, given that they were targeting denatured proteins lacking the conformation information that could be necessary for some reagents. Thus far, this has not proven a significant problem, Chung said, noting that of the more than 120 antibodies the researchers have tested, more than 80 percent worked.
"That might be because many antibodies are made using denatured proteins or peptide fragments, or maybe when we move the tissue back to physiological conditions, these denatured proteins might renature," he said. "The bottom line is, this [MAP technique] is compatible with off-the-shelf antibodies."
The hybrid expansion the method enables is also reversible, allowing researchers to look at samples at different sizes, depending on how holistic a view they want. In the case of Chung's brain research, this means he and his colleagues can look both at individual fibers and the structure of the larger system.
"We want to map the brain's wiring, but in order to do that, we need to image large-scale brain tissue," he said. "At the same time, we need to have high enough resolution to resolve fibers that run next to each other. So in some cases, if you just want to image the bulk wiring, then you can shrink the [tissue] and image it, because once it is shrunk, the imaging time is smaller and it is easier to handle. And then after that imaging, if you want to get super-resolution images of areas of interest, you can expand it [and focus in on a specific portion]."
"So you can get the brain-wide perspective as well as a super-resolution [view] of a small region from a single tissue sample," he said.
Chung said a number of hurdles remain to their goal of generating comprehensive high-resolution brains map. One key challenge, he said, will be the informatics side of their work, given the vast amounts of data the technique generates.
"We are still at the very early stages," he said.