Highlight HEALTH

A large-scale, multi-dimensional analysis of the genomic characteristics of glioblastoma, the most common primary brain tumor in adults, provides new insights into the roles of several genes and defines core biological pathways altered in tumor development [1]. The new Cancer Genome Atlas study, published in the September 4th advanced online edition of the journal Nature, also reveals a link between the DNA repair enzyme MGMT and a hypermutation phenotype, and has potential implications for the diagnosis and treatment of glioblastoma.

Glioblastoma is the most common and aggressive type of brain cancer. Patients newly diagnosed with glioblastoma have a median survival of approximately one year with generally poor response to therapy [2]. Gene expression profiling studies suggest multiple subtypes of glioblastoma that, when fully defined, may allow for more personalized therapeutic approaches [3-4].

The Cancer Genome Atlas (TCGA) is an integrated network of clinical sites, core resources and specialized genome characterization and genome sequencing centers that work together to accelerate our understanding of the molecular basis of cancer. The TCGA was launched in December 2005 as a pilot program to determine the feasibility of a large-scale effort to systematically explore genomic changes in all types of human cancer [5]. TCGA utilizes genome analysis technologies to catalog and discover major cancer causing genome alterations in large groups of human tumors through integrated multi-dimensional analyses. Glioblastoma is the first type of cancer to be studied in the TCGA pilot.

Investigators from seven cancer centers and research institutions across the U.S. integrated multiple types of data, including genetic mutations, gene expression, large-scale changes in chromosome number (amplification or deletion), epigenomics and clinical treatment. The scientists evaluated 206 biospecimens for DNA copy number, gene expression and DNA methylation (a chemical modification of DNA that reduces gene expression). Of these, 143 samples had matched normal peripheral blood DNA; 91 were selected for detection of somatic (meaning cells that differentiate into various tissues and organs, as opposed to germline cells (e.g. sperm and ova)) mutation in 601 selected genes. Eight genes were identified as significantly mutated, three of which were not previously reported for glioblastoma:

• Neurofibromatosis type 1 (NF1), the tumor suppressor responsible for the peripheral nerve disorder neurofibromatosis type 1.
• Avian erythroblastic leukemia viral oncogene homolog 2 (ERBB2), a gene that functions as a co-receptor for connective tissue-derived growth factors and is frequently amplified in breast cancer.
• Phosphatidylinositol 3-kinase, regulatory subunit 1 (PIK3R1), a gene that influences the activity of an enzyme called PI3 kinase, which regulates cell proliferation and survival.

Researchers then mapped the sequencing data with additional genome characterization information onto major biological pathways and identified a highly interconnected network of alterations. By copy number data alone, three critical biological signaling pathways were identified: the Receptor Tyrosine Kinase/Ras/Phosphatidylinositol 3-Kinase pathway (a.k.a. RTK/Ras/PI3K pathway), which controls cell proliferation, cell survival and RNA translation; the p53 signaling pathway, which controls senescence (aging) and apoptosis (cell death); and the Retinoblastoma (RB) signaling pathway, which controls cell cycle progression and cell division. In a given tumor sample, it was likely that there was at least one aberrant gene from each of the three pathways. In fact, 74% of the samples had mutations in all three pathways, suggesting that deregulation of the three pathways is a requirement for glioblastoma pathogenesis.

Oncologists already know glioblastomas that have a methylated MGMT gene (DNA methylation reduces gene expression) respond better to temozolomide, an alkylating chemotherapy drug that is the current standard of care for glioblastoma patients. By integrating methylation data, somatic mutation data and clinical treatment data, scientists identified a relationship between MGMT methylation and a hypermutator phenotype described previously [6]. In patients with MGMT methylation, temozolomide treatment introduces a strong selective pressure to mutate genes that are essential for DNA repair. Thus, patients who initially respond to temozolomide may evolve not only treatment resistance but also a hypermutator phenotype (since DNA repair genes have been mutated). Future selective therapies may therefore require targeting both DNA-repair-deficient cells and an alkylating agent.

National Institutes of Health (NIH) Director Elias A. Zerhouni, M.D. said [7]:

These impressive results from TCGA provide the most comprehensive view to date of the complicated genomic landscape of this deadly cancer. The more we learn about the molecular basis of glioblastoma, the more swiftly we can develop better ways of helping patients with this terrible disease. Clearly, it is time to move ahead and apply the power of large-scale, genomic research to many other types of cancer.

The power of this study lies in the statistically robust number of samples evaluated, allowing for the identification of molecular subtypes that may otherwise be undetectable. Additionally, multiple technologies were employed to identify genomic copy number alterations, which were used to validate the results from any one platform. These approaches highlight the power of comprehensive integrative analyses.

This is an excellent example of how current genome characterization technologies can systematically explore the universe of genomic changes involved in cancer. The TCGA is also studying lung and ovarian cancer.

References

1. The Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature. 2008 Sep 4. [Epub ahead of print] DOI: 10.1038/nature07385
   View abstract
2. Mischel and Cloughesy. Targeted molecular therapy of GBM. Brain Pathol. 2003 Jan;13(1):52-61.
   View abstract
3. Mischel et al. Molecular analysis of glioblastoma: pathway profiling and its implications for patient therapy. Cancer Biol Ther. 2003 May-Jun;2(3):242-7.
   View abstract
4. Liang et al. Gene expression profiling reveals molecularly and clinically distinct subtypes of glioblastoma multiforme. Proc Natl Acad Sci U S A. 2005 Apr 19;102(16):5814-9. Epub 2005 Apr 12.
   View abstract
5. NIH Launches Comprehensive Effort to Explore Cancer Genomics. National Cancer Institute Office of Media Relations press release. 2005 Dec 13.
6. Cahill et al. Loss of the mismatch repair protein MSH6 in human glioblastomas is associated with tumor progression during temozolomide treatment. Clin Cancer Res. 2007 Apr 1;13(7):2038-45.
   View abstract
7. The Cancer Genome Atlas Reports First Results of Comprehensive Study of Brain Tumors: Large-Scale Effort Identifies New Genetic Mutations, Core Pathways. National Cancer Institute Office of Media Relations press release. 2008 Sept 4.

A number of recent advances in stem cell biology are poised to transform therapeutic approaches to a variety of cardiovascular diseases. In the July issue of the journal Cell Stem Cell, researchers report one such advance, demonstrating that they can direct mouse embryonic stem cells to develop into an embryonic cell layer called the mesoderm, which can differentiate (meaning become different in the process of development) into the heart, blood and other tissues [1].

Animal embryos differentiate into three cellular layers called germ layers: the ectoderm, mesoderm and endoderm. Tissues and organs of the body develop from these germ layers through further differentiation. The mesoderm germ layer differentiates into circulatory and urogenital systems, connective tissue, muscle and bone.

Embryonic stem cells have the potential to develop into almost any type of cell in the body. To be used therapeutically, however, scientists have to understand how to direct stem cells to become specialized cell types, such as skin or heart cells. Why use stem cells to repair the heart? Heart muscle cells from a patient generally won’t divide in a sufficient number to replace damaged areas. Taking a part of the heart muscle from another area only creates more damage. Embryonic stem cells thus provide an external and abundant source of cells for heart muscle repair.

Scientists at Washington University School of Medicine in St. Louis found that expression of the gene Mesp1 induced expression of mesodermal markers and genetic changes associated with the transition from embryonic to epithelial tissue (termed epithelial-mesenchymal transition).

A typical epithelium is a sheet of cells, often one cell thick, held tightly together by cell-cell junctions in a uniform array. Adhesions between neighboring epithelial cells inhibit movement of individual cells. In contrast, mesenchymal cells have neither organized structure nor tight intracellular adhesion, allowing for increased migratory capacity. Cells undergoing the epithelial-mesenchymal transition (EMT) experience transient structural changes that result in a loss of contact with neighboring cells and a gain in motility. This process is vital to movements that reorganize the embryonic germ layers and to the development of other migratory cell types [2]. Many of the changes associated with cells undergoing developmental EMTs are also observed in wound healing, fibrosis and cancer.

Using mouse embryonic stem cells, researchers showed that Mesp1 expression restricted the potential fates, generating progenitors or precursor cells with the potential to differentiate into cardiovascular cells but, importantly, not into hematopoietic cells (meaning blood-forming cells). They further demonstrated that Mesp1 induces expression of genes specific to cardiovascular development. The authors suggest that Mesp1 may selectively program the development of endothelial, cardiac and smooth muscle cells.

According to senior author Kenneth Murphy, M.D., Ph.D., senior author and Professor of Pathology and Immunology at Washington University School of Medicine [3]:

That’s the challenge to realizing the potential of stem cells. We know some things about how the early embryo develops, but we need to learn a great deal more about how factors like Mesp1 control the roles that stem cells assume.

This work has the potential to one day treat cardiovascular diseases using human stem cells. Scientists next plan to identify gene programs and map out the pathways that specify development of the three cardiac cell types: endothelial, cardiac and smooth muscle cells.

For more information on stem cells and the repair of a damaged heart, see Stem Cell Information at the NIH.

References

1. Lindsley et al. Mesp1 coordinately regulates cardiovascular fate restriction and epithelial-mesenchymal transition in differentiating ES cells. Cell Stem Cell, July 3, 2008 DOI: 10.1016/j.stem.2008.04.004
   View abstract
2. Shook and Keller. Mechanisms, mechanics and function of epithelial-mesenchymal transitions in early development. Mech Dev. 2003 Nov;120(11):1351-83.
   View abstract

3. Gene directs stem cells to build the heart. Washington University in St. Louis Medical News Release. 2008 Jul 2.

The first high-resolution structural connection map of the human cerebral cortex was published earlier this month in the journal PLoS Biology. The study reveals regions that are highly connected and central, forming a structural core network [1]. Intriguingly, this core network consists of many areas that are more active when we’re at rest than when we’re engaged in a task that requires concentration.

In the human brain, a complex network of fiber pathways link all regions of the cerebral cortex. This “brain wiring” is responsible for shaping neural activation patterns. To better understand the structural basis of functional connectivity patterns in the human brain, researchers at Indiana University and Harvard Medical School in the U.S., and at University Hospital Center and the University of Lausanne in Lausanne, Switzerland, used a brain imagining technique called diffusion spectrum imaging to map out axonal pathways in cortex white matter.

The outermost layer of the cerebrum is termed gray matter. Below the grey matter of the cortex, white matter, consisting of myelinated axons, interconnects different regions of the central nervous system. Axons or nerve fibers are long filaments that project from a nerve cell (called a neuron) and carry electrical impulses away from the cell body to other neurons. Myelinated axons are axons surrounded by an electrically-insulated phospholipid layer (meaning a fat-soluble molecule, think insulated brain wiring). The end of the axon comes close to but does not contact the next nerve cell; this gap between neurons is referred to as a synapse or synaptic junction.

Diffusion spectrum imaging is a type of MRI that identifies water molecules and monitors how they move. In nerve fibers, water molecules can be used to discern brain structural connectivity since they move along the length of fiber. The noninvasive technique offers researchers the ability to compare neural connection variability between participants and to relate it to differences in individual functional connectivity and behavior.

The scientists used network theory, usually applied in social network analysis, to measure network properties such as degree and strength (i.e. the extent to which the node is connected to the rest of the network), centrality and efficiency (i.e. how many short paths between other parts of the network pass through the node), and betweenness. Using measures from five healthy people (participants A — E), they calculated brain regions that were highly connected and contained numerous connector hubs.

The researchers found evidence for the existence of a structural core composed of densely connected regions at the top and back of the brain, straddling both hemispheres (figure). An earlier imaging study measured blood flow to different parts of the brain [2]. It positively correlated with the level of white brain matter axonal connectivity in individual participants.

Interestingly, regions that are in the most connected brain areas have been shown in various studies to have high levels of energy utilization and activation at rest [2], and significant deactivation during goal-directed tasks [2-4]. Termed the brain’s “default network”, it has been suggested to be involved with stimulus-independent thought (daydreaming) [5] and other internally focused thoughts and cognitions, such as remembering the past, envisioning future events and considering the thoughts and perspectives of other people [6]. These tasks all activate multiple regions within the default network.

According Dr. Olaf Sporns, co-author of the study and neuroscientist at Indiana University [7]:

This is one of the first steps necessary for building large-scale computational models of the human brain to help us understand processes that are difficult to observe, such as disease states and recovery processes to injuries.

Indeed, the default network is disrupted in several human disorders, including autism, schizophrenia and Alzheimer’s disease [6]. Using this technique to map out axonal pathways in the cortical network of patients with such diseases should help scientists identify default network regions that are disrupted. Researchers hope that they can better understand these disorders based on changes in the brain’s connectivity map.

For additional information on the human cortex and neurons, YouTube has a great video excerpt from the Discovery Channel on Neurons and How They Work.

References

1. Hagmann et al. Mapping the Structural Core of Human Cerebral Cortex. PLoS Biol. 2008 Jul 1;6(7):e159. [Epub ahead of print] DOI: 10.1371/journal.pbio.0060159
   View abstract
2. Raichle et al. A default mode of brain function. Proc Natl Acad Sci U S A. 2001 Jan 16;98(2):676-82.
   View abstract
3. Gordon et al. Common Blood Flow Changes across Visual Tasks: II. Decreases in Cerebral Cortex. J Cogn Neurosci. 1997;9(5):648-663.
4. Fox et al. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci U S A. 2005 Jul 5;102(27):9673-8. Epub 2005 Jun 23.
   View abstract
5. Mason et al. Wandering minds: the default network and stimulus-independent thought. Science. 2007 Jan 19;315(5810):393-5.
   View abstract
6. Buckner et al. The brain’s default network: anatomy, function, and relevance to disease. Ann N Y Acad Sci. 2008 Mar;1124:1-38.
   View abstract
7. New map IDs the core of the human brain. Indiana University Press Release. 2008 July 1.