CAMBRIDGE, Mass.—Cell biology has been a sticking point inpharma, biotech and other aspects of life sciences, because to get a truepicture of what's happening inside live cells, researchers have to know thelocations of thousands of proteins and other molecules—a complex landscape forwhich there is a mostly incomplete map. However, steps are being made to fillin the overwhelming blank space on that map, with one of the latestbreakthroughs being the development of a technique by chemists at theMassachusetts Institute of Technology (MIT) that can tag all of the proteins ina particular region of a cell, allowing them to more accurately map thoseproteins.
The work has been carried out with the help of researchersat Harvard Medical School and the Broad Institute of MIT and Harvard, and thenew method coming out of that work combines the strengths of two existingtechniques: microscopic imaging and mass spectrometry.
The work started two or three years ago, recalls Dr. AliceY. Ting, the Ellen Swallow Richards Associate Professor of Chemistry at MIT.
"Ihad been thinking about methods for analyzing endogenous, rather thanrecombinant, proteins in living cells," she says. "A major limitation oflive-cell microscopy methods is that they almost always visualize recombinantproteins, which come with overexpression and tag-based artifacts. Mass spectrometryis one of the most powerful tools for studying endogenous proteins, and Iwanted to find a way to apply it to living cells."
She says there are other subcellular regions that she andher colleagues are very interested in mapping using this new proteomic method,because "the biology there is very rich," and she sees great value in this newmethod, which she says bridges the strengths of microscopy on the one hand andmass spectrometry on the other hand.
"Two major methods exist for studying proteins inside thecell: mass spectrometry and imaging of fluorescently tagged proteins. Thesemethods complement each other in terms of strengths and weaknesses," explainsDr. Peng Zou of MIT, who works with Ting at MIT. "On the one hand, massspectrometry is capable of analyzing thousands of proteins in parallel but doesnot provide any information about the spatial arrangement of these proteins inthe cells; on the other hand, imaging reveals the detailed localizationinformation of proteins, but it analyzes only a handful of proteins—less than10—at a time. We sought to combine the strengths of these two methods, so as touncover the protein contents at specific locations inside the cell. Thisrequires tagging proteins with a promiscuous labeling enzyme. We went ahead toengineer a peroxidase to achieve this goal."
Looking at next steps, Zou says, now that they havedemonstrated this method in the mitochondrial matrix, they are interested inextending its applications to such cellular compartments as the endoplasmicreticulum and the mitochondrial intermembrane space. These experiments are inprogress, but Zou stresses that as promising as this new technique is, it isstill very much in its infancy. To obtain a true proteomic map of the cellwould requires 100-percent coverage of the underlying proteome, without biastoward amino acid residues and complete control of labeling radius, which areamong the goals Ting, Zou and the other researchers are working toward.
The mitochondrial matrix that is the topic of the paperpublished by Ting's team in the Jan. 31 online edition of Science can be purified by density centrifugation, althoughthe quality of the preparation is poor, Ting notes.
"I think where this new method will have the greatest impactis in studying the proteomes of cellular regions that are impossible topurify—for example, non-membrane bounded cellular compartments like thesynaptic cleft," she says, adding, "There are potential applications for drugdiscovery and understanding disease mechanisms. For example, one could envisionusing the mitochondrial mapping scheme we report in the paper to analyzepatient-derived cells. What are the proteome-wide differences between cellsfrom healthy people versus people with mitochondrial disease? How do theseproteomes change in response to therapeutic intervention? Our method uses verylittle material and is simple to implement and therefore may be a practicaltool for studying disease."
In terms of additional applications and breakthroughs, Zounotes that so far, their method has revealed that the subcellular location of aheme-biosynthesis enzyme has been incorrectly assigned in the past.
"This brought up the question of the presence of anunexpected transporter that might have escaped notice in the previous model.This information is potentially useful for understanding genetic diseasesrelated to heme," Zou says. "Although our method does not impact thepharmaceutical industry directly, it helps reveal the protein composition atspecific cellular locations. Such information could be useful for diagnosis."
