New Page 1

Projects

1. Protein-protein interaction theory. Developing a comprehensive protein-protein interaction theory is important for our understanding of the cell, disease mechanisms, and to facilitate drug design. The theory behind protein-protein interactions is based on two major areas of research. 1) First principle theory of molecular interactions describes the forces and thermodynamics behind binding of different chemical moieties, and 2) the identification of an ever growing number of short peptide motifs (less than 15 amino acids) that can bind to, or be acted upon by protein domains. These short motifs impart some degree of specificity to protein-protein interactions. However, other than those interactions mediated through short motifs we have virtually no ability to predict protein-protein interactions.

With the long-term goal of developing a comprehensive theory, we are first focusing short peptide motifs. My lab has built Minimotif Miner, a new bioinformatics tool that is the most comprehensive database of short functional sequence motifs containing ~650 unique motifs (Balla et al, 2006). This website (http://mnm.engr.uconn.edu/) was featured in Science and the Journal of Proteome Research (Balla, et al 2006; Kaiser, 2006). We have validated the approach of motif prediction by analyzing annotated motif data for ~3000 motifs in the SWISS-Prot database and experimentally confirming several predicted motifs in Kalirin and EFF-1 {Balla, et al., 2006}. Since its publication approximately one year ago, MnM has been used in several publications and is featured as a links at several Bioinformatics websites. External users from 73 different United States universities and in 34 different countries have used MnM for more than 15,000 searches. Current projects are aimed at completing this database and enhancing the specificity of motif definitions.

2. Medical applications of Minimotif Miner. Several drugs target short functional motifs, thus analysis of disease-related proteins can be used to identify new potential drug targets. Using this approach, Minimotif Miner can be used by any scientist to generate new hypotheses about the function of any protein and postulate mechanisms by which coding region mutations cause any human disease {Schiller, 2007}.

As an example of the application of Minimotif Miner we have recently analyzed the HIV proteome for short functional motifs that are located on the protein surface and completely conserved in 100s of different HIV isolates. Several motifs identified may serve as new drug targets. Through collaborations with several HIV virologist (Drs. Maarit Suomalainen (University of Finland), Thomas Smithgail (University of Pittsburgh), Mark Muesing (Rockefeller University), Mark Marsh (University College London) Tomozumi Imamichi (National Institute of Allergy and Infectious Diseases), and Prasert Auewarakul (Mahidol University, Thailand), we are currently validating whether any of these motifs are necessary for HIV replication and/or infection. Other Investigators/collaborators are now using MnM to study Malaria, Alzheimer's Disease, Parkinson's disease, Herpes Infection, Cancer, etc.

3. Experimental analysis of motifs. Several MnM predictions have been experimentally validated and published. In my laboratory, MnM analysis of Kalirin revealed SH3 and Crk binding motifs that are important for regulation of its guanine nucleotide exchange activity {Schiller, 2006}. In collaboration with Dr. William Mohler (UConn Health Center; letter enclosed), we identified the role of 14-3-3 motifs in cell fusion in C. elegans {Balla et al., 2006}. A number of different motifs were validated in different isozymes of Glyceraldehyde-3-phosphoatase Dehydrogenase {Kravsky and Muronetz, 2007}. MnM was used to study the Dusty family of protein kinases and PKC phosphorylation sites in the Relaxin family peptide receptors {Peng et al., 2006, van der Westhuizen et al., 2007}. Although not referenced in their paper, Cotteret and Chernoff used MnM to study another kinase, Pak5 {Peng, 2006; Cotteret, 2006}.These results demonstrate that MnM analysis is useful and can lead to the discovery of important biological functions.

We are now experimentally identifying novel motifs. We have developed an algorithm (SMS) to identify novel short motifs enriched in proteomes (Rajasekaran et al., 2005,2006). Using the SMS algorithm to analyze human, rat and mouse proteomes, we have identified 75 novel motifs on the C-termini of proteins that are conserved among these species and present on the C-termini of approximately ½ the human genes. We have selected five motifs and expressed them on the C-termini of GST for binding experiments. Krishna Kadaveru, a new student in my laboratory is using Mass Spectrometry to analyze proteins from cell extracts that bind to these motifs, but not controls containing a mutated motif.

Axonal Outgrowth

4. Understanding how neurons initiate axonal and dendritic outgrowth is important, not just for our basic understanding of development of the nervous system and neuronal connectivity, but also to aid in the development of new therapies for patients with neurodegenerative diseases, spinal cord injury, and head trauma. Axonal outgrowth requires the coordination of many cellular processes including, rearrangement of the cytoskeleton, targeted deposition and retrieval of membranes, and synthesis of new membrane and cytoskeletal constituents through signal transduction.

We are interested in Kalirin, a protein involved in several of these processes. Kalirin contains many different functional domains, is expressed in neurons in the developing and mature nervous system, and has properties that suggest a key role in neurite formation. Kalirin interacts both biochemically and genetically with many proteins studied in the context of neurite outgrowth, and screens for axonal pathfinding mutants in primitive animals have identified Kalirin orthologs. Antisense Kalirin oligonucleotides also block axonal outgrowth from sympathetic neurons grown in the presence of nerve growth factor (NGF).

Our laboratory is trying to better understand the molecular mechanisms by which Kalirin affects axonal outgrowth. Kalirin has two guanine nucleotide exchange factor (GEF) modules which activate small G-proteins of the Rho family. Since mutations in the N-terminal GEF module of Kalirin produce axonal path finding errors in invertebrates, we first focused on this module. Our initial experiments in collaboration with Dr. Victor May indicated that over expression of full-length or the isolated N-terminal GEF module is capable of inducing axonal outgrowth in sympathetic neurons (May et al, 2001). Kalirin induced axonal outgrowth was blocked by a dominant negative Rho protein (RhoG), suggesting that the GEF enzymatic activity was required for its effects on neurite outgrowth. Since Kalirin is expressed in adult neurons that do not readily regenerate, we studied the regulation of Kalirin's GEF activity. We found that three different domains are involved in regulating the GEF activity. In one example of regulation, Kalirin's GEF activity is modulated by intramolecular and intermolecular SH3-ligand interactions (Schiller, et al, 2005). This finding was enabled through determination of the SH3 domain 3D structure (PDB accession #1U3O) and identification of its intramolecular ligands containing a Pro-X-X-Pro motif by Minimotif Miner, a new Bioinformatics tool we developed to (Balla, et al 2006)(Kaiser, 2006). Minimotif Miner can be used by any scientist to generate new hypotheses about the function of any protein and mechanisms by which mutations cause any human disease (http://mnm.engr.uconn.edu).

Although dominant negative RhoG blocked Kalirin's effect on axonal outgrowth, a Kalirin mutant lacking GEF activity still induced neurite outgrowth. We concluded that the GEF domain induced neurite outgrowth, but surprisingly not through its enzymatic activity. We began to study non-enzymatic effects of the Kalirin GEF domain in a cell line model. These studies showed that Kalirin GEF could activate Pak kinase to induce lamellipodia formation (Schiller et al., 2005). Lamellipodia are flattened sheets found in axonal growth cones and Kalirin is present in growth cones (Rabiner and Eipper, unpublished observation).

In our investigation of non-catalytic effects of the Kalirin GEF domain, we also found that Kalirin also plays a significant role in activation of and signaling through the TrkA receptor for nerve growth factor, which is involved in axonal outgrowth (Chakrabarti et al., 2006). This provides and interesting connection and may explain several of the effects of Kalirin upon neurons. The connection between Kalirin and TrkA seems to be just one of a broad relationship of the RhoGEF family with receptor tyrosine kinase family which we observed in the literature and have summarized in a recent review (Schiller, 2006)


	Martin R. Schiller Lab Website
Home Affiliations Alumni Bibliography Collaborators Employment Frontiers in Bioscience *Funding* Minimotif Miner Marty's web tool page News People Projects Videos Contact	Projects 1. Protein-protein interaction theory. Developing a comprehensive protein-protein interaction theory is important for our understanding of the cell, disease mechanisms, and to facilitate drug design. The theory behind protein-protein interactions is based on two major areas of research. 1) First principle theory of molecular interactions describes the forces and thermodynamics behind binding of different chemical moieties, and 2) the identification of an ever growing number of short peptide motifs (less than 15 amino acids) that can bind to, or be acted upon by protein domains. These short motifs impart some degree of specificity to protein-protein interactions. However, other than those interactions mediated through short motifs we have virtually no ability to predict protein-protein interactions. With the long-term goal of developing a comprehensive theory, we are first focusing short peptide motifs. My lab has built Minimotif Miner, a new bioinformatics tool that is the most comprehensive database of short functional sequence motifs containing ~650 unique motifs (Balla et al, 2006). This website (http://mnm.engr.uconn.edu/) was featured in Science and the Journal of Proteome Research (Balla, et al 2006; Kaiser, 2006). We have validated the approach of motif prediction by analyzing annotated motif data for ~3000 motifs in the SWISS-Prot database and experimentally confirming several predicted motifs in Kalirin and EFF-1 {Balla, et al., 2006}. Since its publication approximately one year ago, MnM has been used in several publications and is featured as a links at several Bioinformatics websites. External users from 73 different United States universities and in 34 different countries have used MnM for more than 15,000 searches. Current projects are aimed at completing this database and enhancing the specificity of motif definitions. 2. Medical applications of Minimotif Miner. Several drugs target short functional motifs, thus analysis of disease-related proteins can be used to identify new potential drug targets. Using this approach, Minimotif Miner can be used by any scientist to generate new hypotheses about the function of any protein and postulate mechanisms by which coding region mutations cause any human disease {Schiller, 2007}. As an example of the application of Minimotif Miner we have recently analyzed the HIV proteome for short functional motifs that are located on the protein surface and completely conserved in 100s of different HIV isolates. Several motifs identified may serve as new drug targets. Through collaborations with several HIV virologist (Drs. Maarit Suomalainen (University of Finland), Thomas Smithgail (University of Pittsburgh), Mark Muesing (Rockefeller University), Mark Marsh (University College London) Tomozumi Imamichi (National Institute of Allergy and Infectious Diseases), and Prasert Auewarakul (Mahidol University, Thailand), we are currently validating whether any of these motifs are necessary for HIV replication and/or infection. Other Investigators/collaborators are now using MnM to study Malaria, Alzheimer's Disease, Parkinson's disease, Herpes Infection, Cancer, etc. 3. Experimental analysis of motifs. Several MnM predictions have been experimentally validated and published. In my laboratory, MnM analysis of Kalirin revealed SH3 and Crk binding motifs that are important for regulation of its guanine nucleotide exchange activity {Schiller, 2006}. In collaboration with Dr. William Mohler (UConn Health Center; letter enclosed), we identified the role of 14-3-3 motifs in cell fusion in C. elegans {Balla et al., 2006}. A number of different motifs were validated in different isozymes of Glyceraldehyde-3-phosphoatase Dehydrogenase {Kravsky and Muronetz, 2007}. MnM was used to study the Dusty family of protein kinases and PKC phosphorylation sites in the Relaxin family peptide receptors {Peng et al., 2006, van der Westhuizen et al., 2007}. Although not referenced in their paper, Cotteret and Chernoff used MnM to study another kinase, Pak5 {Peng, 2006; Cotteret, 2006}.These results demonstrate that MnM analysis is useful and can lead to the discovery of important biological functions. We are now experimentally identifying novel motifs. We have developed an algorithm (SMS) to identify novel short motifs enriched in proteomes (Rajasekaran et al., 2005,2006). Using the SMS algorithm to analyze human, rat and mouse proteomes, we have identified 75 novel motifs on the C-termini of proteins that are conserved among these species and present on the C-termini of approximately ½ the human genes. We have selected five motifs and expressed them on the C-termini of GST for binding experiments. Krishna Kadaveru, a new student in my laboratory is using Mass Spectrometry to analyze proteins from cell extracts that bind to these motifs, but not controls containing a mutated motif. Axonal Outgrowth 4. Understanding how neurons initiate axonal and dendritic outgrowth is important, not just for our basic understanding of development of the nervous system and neuronal connectivity, but also to aid in the development of new therapies for patients with neurodegenerative diseases, spinal cord injury, and head trauma. Axonal outgrowth requires the coordination of many cellular processes including, rearrangement of the cytoskeleton, targeted deposition and retrieval of membranes, and synthesis of new membrane and cytoskeletal constituents through signal transduction. We are interested in Kalirin, a protein involved in several of these processes. Kalirin contains many different functional domains, is expressed in neurons in the developing and mature nervous system, and has properties that suggest a key role in neurite formation. Kalirin interacts both biochemically and genetically with many proteins studied in the context of neurite outgrowth, and screens for axonal pathfinding mutants in primitive animals have identified Kalirin orthologs. Antisense Kalirin oligonucleotides also block axonal outgrowth from sympathetic neurons grown in the presence of nerve growth factor (NGF). Our laboratory is trying to better understand the molecular mechanisms by which Kalirin affects axonal outgrowth. Kalirin has two guanine nucleotide exchange factor (GEF) modules which activate small G-proteins of the Rho family. Since mutations in the N-terminal GEF module of Kalirin produce axonal path finding errors in invertebrates, we first focused on this module. Our initial experiments in collaboration with Dr. Victor May indicated that over expression of full-length or the isolated N-terminal GEF module is capable of inducing axonal outgrowth in sympathetic neurons (May et al, 2001). Kalirin induced axonal outgrowth was blocked by a dominant negative Rho protein (RhoG), suggesting that the GEF enzymatic activity was required for its effects on neurite outgrowth. Since Kalirin is expressed in adult neurons that do not readily regenerate, we studied the regulation of Kalirin's GEF activity. We found that three different domains are involved in regulating the GEF activity. In one example of regulation, Kalirin's GEF activity is modulated by intramolecular and intermolecular SH3-ligand interactions (Schiller, et al, 2005). This finding was enabled through determination of the SH3 domain 3D structure (PDB accession #1U3O) and identification of its intramolecular ligands containing a Pro-X-X-Pro motif by Minimotif Miner, a new Bioinformatics tool we developed to (Balla, et al 2006)(Kaiser, 2006). Minimotif Miner can be used by any scientist to generate new hypotheses about the function of any protein and mechanisms by which mutations cause any human disease (http://mnm.engr.uconn.edu). Although dominant negative RhoG blocked Kalirin's effect on axonal outgrowth, a Kalirin mutant lacking GEF activity still induced neurite outgrowth. We concluded that the GEF domain induced neurite outgrowth, but surprisingly not through its enzymatic activity. We began to study non-enzymatic effects of the Kalirin GEF domain in a cell line model. These studies showed that Kalirin GEF could activate Pak kinase to induce lamellipodia formation (Schiller et al., 2005). Lamellipodia are flattened sheets found in axonal growth cones and Kalirin is present in growth cones (Rabiner and Eipper, unpublished observation). In our investigation of non-catalytic effects of the Kalirin GEF domain, we also found that Kalirin also plays a significant role in activation of and signaling through the TrkA receptor for nerve growth factor, which is involved in axonal outgrowth (Chakrabarti et al., 2006). This provides and interesting connection and may explain several of the effects of Kalirin upon neurons. The connection between Kalirin and TrkA seems to be just one of a broad relationship of the RhoGEF family with receptor tyrosine kinase family which we observed in the literature and have summarized in a recent review (Schiller, 2006)