As with so many advances in the healthcare realm, theemergence of each exciting new breakthrough gets everyone's hopes up that alarge swath of diseases will suddenly collapse at the hands of medical scienceand pharmaceuticals—followed by disappointment when no immediate results burstonto the scene. Perhaps the biggest example of that right now is "personalizedmedicine," which is the notion that therapies can be tailored specifically toindividual patients, or patients better matched to the best drugs for theircondition, or both. It's an area that has opened many possibilities in terms ofdiagnostics, drug discovery/development and companion diagnostics—made possiblein part by genomic and proteomic leaps in recent years—but it remains avirtually unrealized dream. The key to making that dream a reality lies, in largepart it seems, in cell biology.
That's not a simple road, though, and that even makes peoplewithin the healthcare community despair that they will see any progress intheir lifetime. As Keeley Wray wrote in June 2010 on Vector, the science andclinical innovation blog of Children's Hospital Boston, "We're supposedly inthe dawn of personalized medicine, where advances in molecular biology areproviding doctors the opportunity to optimize each patient's care. As atechnology marketing specialist in Children's Hospital Boston's Technology andInnovation Development Office, I should be enjoying the view. But I'm stillwaiting: how will it happen, when will it happen?"
However, while the "how" and "when" might be murky rightnow, the work has to start somewhere, and it largely began with the explosionin new sequencing technology and other genomics tools. But it is cell biologythat will carry it home to clinical settings and useful patient outcomes,maintains Dr. Stephen Murphy, the managing partner and president of thePersonalized Medicine Group in Greenwich, Conn., a clinical instructor of cellbiology and anatomy at New York Medical College and a member of the InformedCohort Oversight Board (ICOB) of the Coriell Personalized MedicineCollaborative.
"We can have all the sequences, SNPs or whatever else wewant from genomic research, but until we understand what A and B are doing downthe line in the cells themselves and how A and B might affect each other, we'reshooting in the dark," Murphy says. "Cellular biology is a critical piece ofpersonalized medicine."
Genomics is just thestart
Personalized medicine, by and large, gets people thinkingabout genomics, notes Dr. Michael A. Mancini, director of the integratedmicroscopy core and an associate professor of molecular and cellular biology atBaylor College of Medicine.
"You get cells from the patient and you sequence the wholegenome. That's been done a number of times and it's not far away from becomingaffordable, with so many people working toward a $1,000 genome or even lessexpensive perhaps," he says. "But that's not personalized medicine, no matterhow much the public thinks it is. Mapping a genome won't tell you the same kindof thing as if you take cells from patients and do the relevant physiology workand assays and say, 'Here's a tumor biopsy and let's try these 10 differentanti-tumor drugs and pick the best one or the best combination of a few ofthem.' That's the type of personalized medicine that is truly personal—not onlyimportant but made practical."
That is something that has only recently come to be possibleand achievable with available technology, he says, but it's a fact thatresearchers have known for quite some time.
As Andrea D. Weston and Leroy Hood of the Institute forSystems Biology in Seattle noted in a January 2004 Journal of Proteome Research article, "Systems Biology, Proteomicsand the Future of Health Care: Toward Predictive, Preventative and PersonalizedMedicine," the ability to predict and prevent disease will always be dictatedby how good researchers' and clinicians' fundamental knowledge is of the normaland diseased state of cells.
"Treating disease will require circumventing the limitationsof specific genetic or protein defects. To do this, these defects, whichinclude genetic mutations, inappropriate protein processing or folding,aberrant protein-protein or protein-DNA interactions and proteinmislocalizations, must first be accurately placed within the context ofdisease," they wrote.
When genome-wide association studies started coming out,Murphy says, people got very excited about the possibilities that the causes ofdisease and their cures might come simply by finding the right candidate genes.
"But once you find those genes, you're still left with theneed to say what else is going on. So it's more of a push-pull thing withgenomics and cell biology in personalized medicine," he says. "Genome-wideassociation studies create new targets for molecular biologists to explore, andit becomes an accelerative process. Although cell biology is critical tobringing the genetic data into perspective, it's tough to really separate thegenetic and cell biology parts into totally distinct units with personalizedmedicine, because they have to dance together for it all to work."
Still, personalized medicine will rely on some kind oftherapeutic interventions, whether drugs or gene therapies or something else,and cellular and molecular biology are key to that because it is the cell thatis the basic unit of life, Murphy emphasizes, not the genes.
"The great limiting step in any kind of medicine,personalized or not, is therapeutics and drug development," Murphy points out."You can only describe and develop so many potential new medicines without newpathways. Cellular biology is what helps you really identify and characterizethose pathways. So no matter how many genes you identify, you have to go backto cell biology to have it make sense."
"Genetics is critical, and it's a great tool, but genes areessentially two-dimensional, whereas cells are three-dimensional," addsMancini. "They're much more complex stories themselves even at the single-celllevel, much less the tissue level."
Boggled by biomarkers
That story becomes even more convoluted as genomics andproteomics advances keep revealing more twists and turns in the genome thatfurther complicate the already challenging environment of the cell. Biomarkers,which are a huge part of the future of personalized medicine—as they can helpclarify susceptibility to disease as well as disease vulnerability to variousdrugs—are a good example of this.
"Biomarkers are important for diagnostics, companiondiagnostics and therapies that will make personalized medicine a reality. Butbecause of shortcomings in cell biology knowledge, biomarkers can be just asconfusing as the disease states themselves were before we even had thebiomarkers for them," Murphy laments.
Part of the problem, as Mancini notes, is that researchershave barely even scratched the surface when it comes to biomarkers, noting thatwith some 20,000 genes present in the human genome, there are certainly farmore than 20,000 biomarkers because genes make multiple proteins.
"We don't even know how many translational modificationsexist in total," Mancini admits, "and we're probably going to need hundreds ofthousands of biomarkers to get at the answers. Those will take some time to bemade. This is an order of magnitude much larger than sequencing the genome."
Half of the genome is involved in transcriptionalregulation, for example, he adds, and of those 11,000 or so genes producingproteins for transcriptional regulation, there are probably 10 or more versionsof each protein.
"These permutations invite an enormous number ofcomplexities, and we don't even have biomarkers for those 11,000 genes, sowe're a long way away from true personalized medicine," Mancini says.
That transcriptional wrinkle alone would explain in part whythe O'Shea Lab at Harvard University, for example, which has a heavy focus oncell biology, has as one of its three major project areas "The Logic andEvolution of Transcriptional Control."
Dr. Erin O'Shea, the lab's eponymoushead, is the chair of the Coriell Personalized Medicine Collaborative ICOB onwhich Murphy serves.
"We seek to understand how regulatory regions of genestransform information about transcription factor input into quantitative geneexpression output," notes O'Shea, director of the Faculty of Arts and SciencesSystems Biology Institute at Harvard University, as well as a professor ofmolecular and cell biology at Harvard and a Howard Hughes Investigator, on herlab's website. "Our goal is to develop a quantitative model that describes howpromoter sequence influences the threshold for gene activation, maximumtranscriptional output and the sensitivity of the response. To achieve thisgoal, we are using model promoters to analyze the relationship betweentranscription factor input and gene expression output in single cells."
Her lab also is looking into encoding and decodinginformation in transcription factor dynamics and investigating the evolution oftranscriptional regulatory networks.
Beyond the cell
Given all the complexity involved with understanding theworkings of the genes and cells to make personalized medicine a reality, cellbiology and molecular biology aren't the end of the story either. What willlikely be needed is a more comprehensive and cross-disciplinary systems biologyapproach that integrates engineering, physics and mathematical approaches withbiologic and medical insights in what Dr. Ana Maria Gonzalez-Angulo, anassociate professor of breast medical oncology at the University of Texas MDAnderson Cancer Center, and her colleagues described in a June 2010 Journal of Clinical Oncology article as"an iterative process to visualize the interconnected events within a cell thatdetermines how inputs from the environment and the network rewiring that occursdue to the genomic aberrations acquired by patient tumors determines cellularbehavior and patient outcomes."
A multidisciplinary approach like this is needed, she says,because the massive amount of data generated by high-throughput technologieshas become too challenging to manage, visualize and convert to actionableknowledge otherwise.
As she notes in looking at her own oncology research, muchof the data used to explore the structure of signaling networks are contextualand cannot be generalized to a cancer cell in its microenvironment. Thisproblem is exacerbated, Gonzalez-Angulo says, by the fact that the developmentof high-throughput 'omics technologies has not been paralleled by equivalentimprovements in cell biology technologies and approaches that can helpresearchers understand the consequences of these 'omics changes on cellular andorgan-related outputs.
"Systems biology provides us with a common language for bothdescribing and modeling the integrated action of regulatory networks at manylevels of biological organization from the subcellular through cell, tissue andorgan right up to the whole organism," according to Prof. Jeremy K. Nicholson,chair in biological chemistry and head of the Department of Surgery and Cancerat Imperial College London.
Mancini notes that as a researcher, "you need any and everyway you can to connect the information, whether you're trying to kill cells orsave them," and working at the systems biology level, while it may take muchmore time and effort than sequencing a genome, running a genome-wideassociation study or running assays, "will make things more relevant thanwhat's been done up to now by focusing on the genes."
"Cell biology is hands-down in the top couple of things youneed to have in play for personalized medicine to truly work, but it's not theonly thing," Murphy adds. "To grasp all these genes, the proteins they create,the pathways at work and all the rest, the systems biology knowledge has to befiring on all cylinders. Because if you don't understand the biology, you don'thave anything."
The tools to get youthere
As noted in the main article by Dr. Ana MariaGonzalez-Angulo of the University of Texas MD Anderson Cancer Center, cellbiology tools haven't advanced at a speed parallel to those used in genomicsresearch and some other 'omics areas. However, they have come a long way andare enabling achievements in recent years that were not previously possible.Here, two cell biology experts share with us some quick thoughts about enablingtechnologies in their field.
Dr. Michael A. Mancini, director of the integratedmicroscopy core, Baylor College of Medicine:
"Cell biology and microscopy are somewhat synonymous overhistory, and now, being able to truly do quantitative and automatedmultiparameter microscopy work in real time with drug screening and cells fromdifferent patients—well, that's a big leap that's only been available in thelast five or six years. You combine current microscopy technology with theamazing wave of information you get from genomics, and it's like one plus one equals10. I mean, you can pick out any multiple you like, but I can tell you it'smuch more than just a one-plus-one-equals-two effect."
Dr. Stephen Murphy, managing partner and president, thePersonalized Medicine Group in Greenwich, Conn.:
"RNA expression arrays and new technologies for cell linestorage have been a couple of the key advancements that have let cell biologyreally shine in recent years. Devices that have allowed for the real-timeability to see what's going on at the molecular level have been key as well.Also, the willingness and ability to share cell lines throughout the researchcommunity instead of holding on to them and keeping them within a singleinstitution has been huge."
