Funded Grants

Analyzing the role of Astrocyte Elevated Gene-1 (AEG-1) in malignant glioma

Brain tumors are the most prevalent solid neoplasms of childhood and the second leading cancerrelated cause of death in adults between the ages of 15-34 years. In the US in 2009, estimated new cases of brain cancer were 22,070 out of which 12,920 cases were estimated to die. The brain is composed of nerve cells or neurons that are supported by glial cells that include astrocytes, oligodendrocytes and microglia. Glial cells account for more than 90% of brain cells. Most of the brain tumors originate from glial cells and are known as gliomas. Astrocytomas, composed predominantly of neoplastic astrocytes, account for 80-85% of all gliomas and are staged as low grade (grade I) to high grade (grade IV) based on their aggressive behavior. Glioblastoma multiforme (grade IV astrocytoma; GBM) is an extremely aggressive, invasive and destructive malignancy with a 2-5 times faster proliferation rate than grade III tumors. GBM patients have a median survival of only 10 to 12 months despite aggressive therapy. Although GBM does not spread to distant organs, i.e., it does not metastasize, it has highly invasive capacity into surrounding normal brain. As such extensive surgical resection is not curative since remnants of invasive GBM cells are always left behind in the normal brain. Moreover, GBM is largely resistant to current treatments, such as chemotherapy or radiotherapy. Even after aggressive therapies, recurrence of GBM is a common event and accounts for the fatal outcome of the disease. As such, understanding the molecular mechanism of GBM pathogenesis, especially its invasive properties, is of vital importance for developing a targeted therapy to eradicate this fatal malady and promote disease-free survival.

Our search for molecules playing important role in regulating development and progression of GBM identified Astrocyte elevated gene-1 (AEG-1) as a key molecule in this process. Interestingly, AEG-1 has been identified to promote development and progression of not only GBM but also a variety of other cancers that include cancers of liver, breast, prostate, esophagus and lungs, melanoma and neuroblastoma. In all these diseases AEG-1 expression was significantly and markedly higher in cancer patients compared to normal individuals and the higher expression level of AEG-1 correlated with poorer prognosis of the disease. Our findings demonstrate that AEG-1 expression is significantly increased, compared to normal brain, in more than 90% of ~100 brain cancer patients analyzed, the majority of these patients suffering from GBM. Overexpression of a gene in such a high percentage of patients has not been demonstrated for any other gene. GBM is notorious for its invasion into the surrounding normal brain that results in recurrence of the disease after surgical resection. We demonstrated that AEG-1 plays a key role in mediating invasion of GBM cells by activating the intracellular signaling molecule NF-κB. NF-κB is known to regulate the expression of many genes necessary for tumor development, progression, invasion and metastasis and we documented that AEG-1 activates NF-κB by directly interacting with this molecule. When AEG-1 expression was inhibited by small inhibitory RNAs (siRNA), there was marked decrease in invading capacity of GBM cells. Additionally, inhibition of AEG-1 by siRNA strongly inhibited growth of primary human GBM cells implanted into the brain of athymic nude mice. In addition to facilitating invasion and metastasis, end-stage events in cancer, AEG-1 also promotes earlier events in carcinogenesis facilitating transformation of normal cells into cancer cells. In addition, AEG-1 also promotes new blood vessel formation or angiogenesis and confers resistance to chemotherapeutics, two important hallmarks of aggressive cancers. All these findings strongly suggest that AEG-1 plays a pivotal role in the development and progression of cancer, including GBM, and AEG-1 inhibition might be an effective therapeutic strategy to cure the patients of this fatal disease.

The intriguing properties of AEG-1 promoting tumorigenesis have been elucidated by cell culture studies as well as studies using human GBM cells implanted in athymic nude mouse which lack functioning immune system thereby allowing growth of human cancer cells without immune-mediated rejection. These studies are vital to establish the tumorigenic properties of a new oncogene. However, these findings need to be further confirmed by more stringent animal models. If AEG-1 is really important for tumor development and progression, what will happen if AEG-1 is absent? Based on the strong data obtained so far, one would hypothesize that in the absence of AEG-1, cancer, if it develops at all, would significantly lose its aggressiveness. In case of GBM one would argue that lack of AEG-1 would abrogate invasion of GBM into surrounding normal brain so that the cancer might be completely removed by surgical resection resulting in cure. This hypothesis can be validated by an animal model in which AEG-1 expression is knocked out selectively in particular organs or cell types. Since GBMs arise from astrocytes, developing an animal model in which AEG-1 expression is knocked out exclusively in astrocytes would be a valuable model to analyze the hypothesis. Current technologies based on 'Cre-loxP' system facilitates this approach where a gene might be deleted in specific organs or cell types utilizing a tissue-specific or cell-type specific promoter. Indeed, using similar system, Dr. DePinho's laboratory has generated a mouse model of GBM where two tumor suppressor genes p53 and PTEN have been selectively knocked down in astrocytes. This mouse model develops aggressive GBM mimicking the human disease. It would be interesting to see what happens if AEG-1 is knocked down in this model. Will this GBM model lose its aggressive properties in the absence of AEG-1? We are currently in the process of generating a mouse line in which AEG-1 is selectively knocked down in astrocytes. We have already generated the necessary reagents to create such a mouse line which will be ready in very near future. This astrocyte-specific AEG-1 knock-out mouse (AEG- 1KO mouse) will be crossed with the astrocyte-specific p53 and PTEN knock-out model of GBM to establish the importance of AEG-1 in GBM progression. The animals will be subjected to a battery of histological, immunological and molecular analyses to establish our hypothesis. The AEG-1KO mouse will be used to identify novel AEG-1 downstream genes mediating its functions.

Although AEG-1 expression is profoundly higher in GBM, in normal brain AEG-1 is also expressed although at a low level. The presence of low level of AEG-1 in normal astrocytes indicates that AEG-1 might serve some physiological function. The brain functions are regulated by neurotransmitters which relay information in the brain and glutamate is a very important neurotransmitter. Glutamate accumulates in the synapses, where two neurons or a neuron or an astrocyte meet, and stimulates the neurons to activity. However, glutamate level needs to be maintained at a very critical level in synapses. Excessive glutamate accumulation in synapses results in too much excitation of neurons leading to their death, a phenomenon called excitotoxicity. Astrocytes contain glutamate transporters that remove glutamate from the synapses back into the astrocytes thereby maintaining critical glutamate concentration. The major glutamate transporter in astrocytes is EAAT2. We have identified that AEG-1 downregulates EAAT2 expression. In GBM, AEG-1 overexpression with associated decrease in EAAT2 results in toxic accumulation of glutamate in synapses leading to neuronal death. On the other hand in the absence of AEG-1, EAAT2 expression might be increased resulting in reduced levels of glutamate in the synapses leading to impairment of neuronal function. The AEG-1 KO model will thus provide insights into the physiological function of AEG-1 in normal brain. The animal would be subjected to analysis of motor and behavioral skills as well as learning and memory tests to check the role of AEG-1 in normal brain function in collaboration with Dr. Joseph Porter in the Department of Psychology.

AEG-1 was cloned as a novel gene only in 2002. However, studies from laboratories all over the world have already established that it is a key regulator of carcinogenic process including GBM. Development of a conditional AEG-1 knock-out mouse is thus extremely valuable to understand the biological function of AEG-1. Using our developed model, AEG-1 can be knocked out not only in astrocytes but in any other organ as well and thus might be a valuable tool for studies of cancers of other organs. Our proposed studies are therefore highly significant. Since AEG-1 knock-out mouse has not been generated yet, our proposed studies are also innovative. The James S. McDonnell Foundation encourages new ideas and approaches, early in their development, that are unlikely to be funded from traditional sources. Our proposed studies employing AEG-1KO mouse falls exactly in this category. The foundation also mandates that all proposals submitted to the foundation must clearly link the experimental models and questions to the realities of human disease. We have already established the importance of AEG-1 in human GBM. Additionally, the mouse GBM model developed by Dr. DePinho has already been accepted as an authentic model mimicking the human disease. In this context, our studies will provide insights corroborating to human GBM. The proposal, therefore, is directly appropriate to the quests of the Foundation.