September 1, 2014 (Vol. 34, No. 15)
Richard A. A. Stein M.D., Ph.D.
“We have not reached solutions; we have only begun to discover how to ask questions”—these words, set down decades ago by Lewis Thomas, remain as true as ever, inspiring ever more profound (and varied) scientific pursuits.
One such pursuit is protein profiling, or the analysis of proteins at the proteomic scale.
While it provides comprehensive information about protein structures and functions, protein profiling is not an end in itself. It is, rather, a means of visualizing cellular pathways and networks. And even this layer of information is not yet the end. Yes, the proteome and the signalome are interesting, but what really matters is their role in health and disease.
Protein profiling is a way to expose the processes that orchestrate development and maintain (or disturb) tissue homeostasis. Accordingly, protein profiling may drive the development of novel therapeutics. But even if it were to become a commonplace drug development tool, protein profiling would hardly exhaust its potential.
Protein profiling is not merely factual. It is also conceptual. And the concepts engendered by protein profiling, like all concepts, are developed, refined, and reshaped over time. One of the key lessons that science has provided is that even topics thought to be thoroughly understood may need to be revisited frequently. So it may prove with protein profiling. If so, the subject’s conceptual progress will likely rely on technological advances—but these are already emerging.
Summing up the legacy of decades of research, Steven R. Danielson, Ph.D., chemist at Thermo Fisher Scientific, stated, “The details of how biology works are more complex than we ever imagined.” To make sense of these details, Dr. Danielson insisted, it is “vital to develop high-performance instruments and software.”
These comments were made, appropriately, after the recent American Society for Mass Spectrometry (ASMS) meeting in Baltimore. At this meeting, Dr. Danielson and colleagues presented work performed in collaboration with colleagues from the University of Minnesota using the Orbitrap Fusion Tribrid mass spectrometer, an instrument that was released just one year earlier. “The Orbitrap Fusion provides investigators the opportunity to collect very high-coverage proteomic data,” said Dr. Danielson.
This work stemmed from the acute need to develop a sensitive and rapid approach to survey post-translational changes at the proteomic scale. Focusing on lysine acetylation, a key post-translational modification that shapes many cellular processes, including gene expression, transcription regulation, and the function of cytoskeletal proteins, Dr. Danielson and colleagues characterized peptides from Escherichia coli and Caenorhabditis elegans tryptic lysates using a combination of rapid immunoaffinity enrichment and mass spectrometry.
This approach helped identify 3,452 unique proteins from the C. elegans lysate, representing at least 17% of the organism’s proteome, and 1,666 acetyl-lysine-modified peptides from the E. coli lysate. These data were obtained after 55-minute and 45-minute liquid chromatography runs, respectively. These figures suggest that the approach is both sensitive and speedy.
A research effort at Thermo Fisher Scientific was launched recently to develop top-down proteomics approaches. The idea is to use the Orbitrap Fusion platform to characterize the spectrum of individual proteoforms of a specific protein.
“The global pattern of proteoforms is relevant to specific biological states and diseases,” explained Dr. Daneilson. “It involves not just determining that a specific modification is present, but also exploring its relevance to get a more comprehensive look at what happens to a specific protein rather than an isolated view of modifications that are present on specific amino acids.”
In our presentation at the ASMS meeting, we discussed two highly complementary phosphopeptide enrichment strategies and contrasted their differences,” said Charles L. Farnsworth, Ph.D., scientist at Cell Signaling Technology (CST). Post-translational modifications are key regulators of protein function, and they play fundamental roles in development, disease, and homeostasis, but because they often exist at low levels, their identification and the analysis of the cellular processes that they regulate is challenging.
In this approach, Dr. Farnsworth and colleagues used two enrichment strategies. The first was immunoaffinity enrichment using several classes of antibodies directed against phosphorylated amino acid sequence motifs; the second, immobilized metal affinity chromatography (IMAC), which is a charge-based metal affinity approach.
The immunoaffinity LC-MS method was performed using PTMScan, CST’s proteomic technology. It allowed CST scientists, said Dr. Farnsworth, to enrich for and quantitate phosphopeptides that would not have been detectable by IMAC alone. Dr. Farnsworth and colleagues identified and profiled over 20,000 phosphopeptides from kinase inhibitor-treated gastric carcinoma cells, with a low false-positive discovery rate, demonstrating the strength of these two complementary approaches in capturing post-translational protein modifications for comprehensive phosphopeptide profiling.
These data helped generate a global view of the phosphorylation in cells, and allowed the annotation of an MAPK pathway and a tyrosine kinase pathway, key steps toward unveiling potential therapeutic targets related to the dysregulation of signaling in these pathways. The same strategy can also be used for other post-translational modifications, such as acetylation, succinylation, methylation, and ubiquitination. In addition, the strategy helped identify new drug targets, providing a platform to characterize substrates of specific signaling proteins, assess the effects of candidate therapeutic agents, and guide the design of subsequent studies.
Moreover, the use of multiple types of antibodies helps dissect the crosstalk between various types of post-translational modifications and clarify their interplay as they orchestrate various cellular processes. A comparative analysis revealed very little overlap between the phosphopeptides that were identified by PTMScan and the ones identified by IMAC, indicating that these two distinct methodologies enrich for different classes of phosphopeptides and present a high degree of complementarity to one another.
“Ideally, as we move ahead with this work, we would like to reduce sample quantity and develop a high-throughput platform to make our analysis more amenable to robotics, to profile many different types of tissues and gather information from larger cohorts and larger studies,” offered Jeffrey Silva, Ph.D., a group leader at CST.
In the course of these experiments, Dr. Farnsworth and colleagues also revealed that this approach can be used even when the nature of the post-translational modifications is not known prior to the experiment. In such a case, a panel of the PTM motif antibodies can be used to perform Western blot analyses to screen for the antibodies best suited for the particular project of interest.
In addition to the motif antibodies used for discovery-based PTM proteomics, CST has also developed other immunoaffinity reagents for LC-MS applications to monitor the activity of specific cellular pathways. “This is different from the motif antibody reagents described earlier because it allows investigators to focus solely on specific signaling pathways that are critical to the underlying biology of their research interests,” noted Dr. Silva.
“We started using activity-based protein profiling (ABPP) to measure the activity of enzymes directly in their native context and learn more about what the genome is doing,” said Enrique Saez, Ph.D., associate professor in the department of chemical physiology at the Scripps Research Institute. ABPP, which has recently emerged as a powerful chemical proteomics-based experimental strategy, represents a key functional proteomics technology.
In ABPP, the conserved structural or mechanistic characteristics of large enzyme families are exploited to design small molecules that could bind to and inhibit their active sites. Most ABPP probes contain a reactive group and a binding group, which target the active sites of enzymes, and a reporter tag, which helps characterize the cellular target. Because ABPP probes label only enzymes that are in the active state, they can be used to identify changes in enzyme activity that occur without alterations in protein levels.
“We have been interested in using probes that allow functional profiling in different conditions or treatments and that open up a new level of scrutiny, well beyond that of gene arrays or proteomics,” explained Dr. Saez.
In a study published December 22, 2013, in Nature Chemical Biology, a research team led by Dr. Saez and his Scripps colleague Ben Cravatt, Ph.D., a pioneer in the development of ABPP probes for various enzyme classes, showed that ABPP could identify targets that bind small molecules involved in adipocyte differentiation and lipid accumulation. In a phenotypic screen for adipogenesis and lipid storage, the investigators screened a library of serine hydrolase inhibitors.
This effort, which emerged from knowledge about the functions that members of this class of enzymes fulfill in lipid metabolism, unveiled several carbamates. Carbamates stimulate lipid accumulation in differentiating adipocytes.
Competitive ABPP was then used to identify the molecular targets of these carbamates, and this strategy identified several poorly described enzymes whose inhibition promotes adipocyte differentiation and lipid storage. A few of the carbamates identified in this screen inhibited the same serine hydrolase, Ces3, previously shown by genetic approaches to be involved in lipolysis.
The investigators went on to validate the therapeutic potential of inhibiting Ces3 in obesity-diabetes models, illustrating the power of ABPP to aid in the discovery of new drug targets. “We intend to exploit the full potential of this approach to uncover new enzyme functions and biological pathways,” remarked Dr. Saez, who acknowledges that doing so means that specific probes will have to be developed for additional enzyme classes.
Insights from Live-Cell Imaging
Aaron T. Wright, Ph.D., senior scientist at the Pacific Northwest National Laboratory, led a study that assessed organelle-specific activity-based protein profiling in living cells. According to Dr. Wright, this study, which appeared in March 10 in Angewandte Chemie, featured a chemical probe that was designed to “enter live cells, travel only to a particular organelle, and be specific for a particular class of enzymes.”
The study integrated high-resolution live-cell imaging with proteomics, and it showed that the activity-based probe was highly specific for cathepsins B and Z, two lysosomal cysteine proteases. Using this probe and mass spectrometry, the investigators confidently identified the probe’s targets. In addition, the investigators used structured illumination microscopy to confirm the probe’s localization to the lysosome.
This strategy allowed investigators to dynamically visualize the cellular fate of a probe that enters the cell and is targeted to the lysosome. Understanding lysosomal proteomics promises to fill a long-existing gap in dissecting the lysosomal biology that is relevant for the pathogenesis of several diseases, including cancer, neurodegenerative conditions, and mucolipidoses.
“In the short term, we would like to develop a similar approach for other organelles,” commented Dr. Wright. “In the long term, we intend to perform live-cell protein profiling to track probes that enter the cell.”
Other probes, which mimic nutrients, can provide information about their fate and disposition after cellular uptake. “Mass spectrometry-based proteomic analysis, together with real-time imaging, can help us identify which proteins interact with a specific probe molecule,” explained Dr. Wright.
In-Depth Structural Information
“In biotherapeutics, it is very important to put the post-translational modifications of proteins in context,” said Frank Sobott, Ph.D., professor of chemistry at the University of Antwerp. “These modifications not only impact three-dimensional protein structure, they also shape interactions with other proteins.”
In a recent collaborative effort with investigators at Waters, Dr. Sobott and colleagues developed an approach based on native mass spectrometry that is expected to incorporate size-exclusion chromatography, to analyze native protein structures and noncovalent protein complexes. “The aim of this work is primarily to generate a wealth of details about certain key proteins, as opposed to finding out about as many proteins as possible, which is the focus of classic proteomics,” explained Dr. Sobott.
At the recent ASMS meeting, Dr. Sobott and colleagues elaborated on the approach. They described how electron-transfer dissociation followed by activation was used to develop a new charge-reduction method. In addition, they discussed how top-down fragmentation coupled to ion mobility provided the tools to analyze noncovalent protein complexes.
The structural characterization of the target proteins by this approach allows protein conformations and characteristics (such as the relative abundance of different protein isoforms) to be dynamically visualized over time and under various conditions. This strategy, used by Dr. Sobott and colleagues to test known commercial or recombinant proteins, promises to fill an important gap—clinically relevant knowledge about post-translational protein modifications, such as the phosphorylation of kinases in certain disease states, is often not fully integrated with structural information, such as from biophysical studies or crystal structures.
“When a protein undergoes a covalent modification, it is important to understand how this affects its three-dimensional structure and its interactions with other proteins, and trying to establish a link between structural biology and survey-type proteomics types of information is one of the motivations for our work,” noted Dr. Sobott.
Filling this gap plays an important role in drug design, where substantial research efforts focus on inhibiting proteins that play key roles in pathogenesis, but understanding the global impact of changes in single proteins on protein interaction networks often lags behind. Learning about how changes in any given protein affect the stability of entire protein complexes is, therefore, a task of key importance for translating research findings into clinical applications.
“In the future, understanding the role of proteins in biological matrices and in their biological context will become more often a routine step in drug development,” concluded Dr. Sobott.