Recent advances in technology are now enabling researchers to study cancer proteomes in unprecedented detail. As well as deepening our understanding of the cancer phenotype, proteomics holds a key advantage as it’s looking at a level that’s directly actionable, with the proteins themselves potential clinical biomarkers or druggable targets.
“You can sequence as many genomes as you like but, in the end, you still don’t really know which proteins are actually responsible for the onset and generation of disease,” says Dr Clive D’Santos, Head of Proteomics at the Cancer Research UK Cambridge Institute.
But a new revolution is taking place. Next generation technologies are now enabling researchers to carry out large-scale proteomic analyses, opening up similar opportunities to what’s possible in genomics.
“The sensitivity of the instruments nowadays is enabling us to look at almost what’s calculated to be a full proteome,” says D’Santos. “And it’s also giving us the opportunity to quantify these measurements, and importantly, it’s taking just days or even hours to do an analysis.”
Studying cancer genomes has advanced our knowledge of cancer biology and led to new methods to diagnose and treat the many different types of the disease. However, the potential rewards from proteomics could be even greater.
“The active agents that are carrying out the roles that give us the phenotype and that are druggable are the products of the genes, which are the proteins,” explains Dr Jyoti Choudhary, Head of Proteomics and Metabolomics Laboratory at The Institute of Cancer Research, London, UK.
Proteomics technologiesResearchers have a choice of two main approaches for studying proteomes, which are either based on mass spectrometry or antibody-based proteomics.
Many proteomics laboratories will opt for mass spectrometry, with a choice of techniques for analysis and quantification. The most common option is using a ‘bottom-up’ approach, which is based on fragmenting the proteins into peptides and sequencing them.
“We then infer the presence of a particular protein in a sample depending on how unique those peptide sequences are,” explains D’Santos.
The most frequently used method is known as Data-Dependent Acquisition (DDA). This prioritizes the profiling of the most abundant proteins and peptides, with the total number that can be analyzed in a sample being dependent on the speed and sensitivity of the mass spectrometer.
In recent years, a new family of methods known as Data-Independent Acquisition (DIA), are also gaining momentum. These aim to generate an extremely complete, unbiased picture of the samples’ proteomes.
“Rather than profiling down, what these instead are trying to do is to capture everything within that sample, and then you have some sophisticated software at the end to try to deconvolute everything and identify the peptides,” explains D’Santos.
Faster, more sensitive instruments“We can now get resolving power in some instruments that allow us to identify peaks that are very close together –meaning we can separate out peptides that have a difference of only millidaltons,” says D’Santos.
It is generally thought that is around half of the 23,000 genes in the genome are expressed in a cell at any one time, with some laboratories now able to identify 10 or 11,000 proteins in one run.
“It’s exciting that we’re now in a position where we can say we can identify and quantify full proteomes,” says D’Santos. “But on the other hand, we’re still missing an awful lot of information about the multiple isoforms of each individual gene product.”
To overcome this, there is a great deal of excitement about the potential of using a ‘top down’ approach to mass spectrometry. Rather than analyzing fragmented peptides, this approach instead looks at intact proteins.
“I think the reason that’s important is post-translational modifications,” enthuses D’Santos. “But the hardest part will be developing ways to do intact protein analysis on a high-throughput scale.”
Despite all the recent advances in technology, single cell analysis remains out of reach for proteomics researchers. But there is hope for the future from mass cytometry, a hybrid technique that couples mass spectrometry with flow cytometry.
“If you’re looking long-term, I’m excited about where that’s going especially where you need to carry out proteomics on small numbers of cells,” says D’Santos.
Cancer proteomicsResearchers studying cancer proteomics face unique challenges, not least with analyzing their data. The standard approach is by searching databases based on the reference genome sequence.
“One of the big problems with cancer is that it’s caused by mutations, so it doesn’t look like the reference genome,” explains Choudhary.
The response to this has spawned the approach of proteogenomics where researchers generate customized protein sequence databases using transcriptomic or genomics information, interrogating these instead.
Advances in mass spectrometry are also enabling researchers to carry out much larger scale analyses to characterize the collateral damage of mutations in cancer cells on the proteome.
“In one recent study, we looked at 50 colorectal cancer cell lines, mapping all the changes in the proteins against their genomes to build up networks of genetic mutations that were impacting on protein function,” says Choudhary.
“We’re now starting to create them for different cancer types,” she adds. “On one hand, this approach will help us to understand more about cancer biology, identifying some of the genes we can think of as critical – the ‘car crash’ sites – and on the other, we’re aiming to tie up this information into drug response, resistance, and so on.”
Others are using proteomics to interrogate the mechanisms that drive key processes in cancer biology, such as metastasis.
“A team at our institute recently discovered that asparagine influences the metastatic potential of cells in a mouse model of breast cancer,” says D’Santos. “So, we used proteomics to look for proteins that were changing their expression levels in cells from a mouse mammary tumor cell line deficient in certain enzymes that metabolize this amino acid.”
Other new proteomics techniques have more direct potential for clinical application. For example, imaging using mass spectrometry could lead to innovative new tools for cancer diagnostics or surgery. And the powerful new PROTACs technique that allows direct modulation of protein levels in a cell has big potential for the development of new therapeutics.
Another exciting new approach is using nanosensors that can seek out tumors and then sheds, in response to tumor-specific proteases, peptides that can then be detected in the urine – that could potentially transform our ability to detect cancer early.
A golden age for proteomics researchSince the word proteomics was first coined around 25 years ago, there has been a lot of progress.
“I think it’s been an amazing time, we’ve had such an amazing development of new generation technology,” enthuses Choudhary. “But a ‘Big Bang’ has always eluded the field, not because anything we’re doing wrong, but just because of the comparisons with genomics.”
But researchers believe that now is the time for proteomics to realize its full potential in revolutionizing the diagnosis and treatment of cancer.
“It’s very similar to the genomics revolution, but it will be a more powerful one because it’s the layer that’s closest to being able to do something about your discoveries,” says Choudhary.