Professor, Department of Biochemistry
1983 - PhD, McGill University
Distinguished Investigator of the CIHR.
Fellow of the Royal Society of Canada.
Cancer resistance to chemotherapy; genetic models of innate resistance or susceptibility to disease
Our laboratory uses a genetic approach in mouse to discover genes, proteins and pathways that play an important role in complex human diseases. Our long-term objectives are to translate knowledge obtained in laboratory mouse models, into clinical outcomes through the creation of novel diagnostic tools or new small molecules modulators with therapeutic value in the corresponding human disease. We are currently focusing on three major human diseases known to have clear genetic component: infectious diseases, cancer, and the birth defect spina bifida. Our genetic platform is based on the use of genetically diverse mouse inbred strains, recombinant congenic strains, and experimentally induced mutagenized mouse stocks (ENU mutants).
Our research on infectious diseases focuses on pathogens that invade and replicate inside host cells, and that use this strategy to evade the immune system and other normal host defenses. These pathogens include Mycobacterium tuberculosis (Tuberculosis), Legionella pneumophila (Legionaire’s disease), Candida albicans (Fungal pathogen), Plasmodium chabaudi (erythrocyte-stage malaria), and Plasmodium berghei (cerebral malaria). In these research programs, we have cloned major genes that regulate response to infections, and in the case of complex genetic control, we have isolated single gene effects in unique recombinant or mutant mouse strains. In the area of mycobacteria, we have shown that a transporter of metal ions (such as iron and manganese), which is expressed at the membrane of mycobacteria-containing phagosomes formed in macrophages, is critical to restrict intracellular replication of these microbes. The protein (known as Nramp1 or Slc11a1) acts as a biological chelator to remove from the phagosomal space metal ions that are required for metabolic activity and survival of the pathogen. In mice, Nramp1 mutations cause susceptibility to a number of infections including Mycobacteria, but also other pathogens such as Salmonella and Leishmania. In humans, it was demonstrated that sequence polymorphisms at NRAMP1 are associated with susceptibility to tuberculosis and leprosy in areas of endemic disease. We are currently studying the mechanism of transport of Nramp1 and are searching for bacterial proteins and biochemical pathways that activity is dependent on Nramp1 metal ions substrates, and that may be targeted for drug discovery.
In the case of Legionella, we have shown that mouse strains susceptible to this infection carry mutations in a protein known as Naip5/Birc1e. This cytoplasmic protein forms part of an ancient antigen-recognition system known as the NLR family, which includes both cell surface and intracellular receptors that recognize so-called PAMPs (pathogen-associated molecular patterns) in a manner distinct from that of the normal immune system. Upon infection of these cells, Naip5/Birc1e is essential for early intracellular sensing of Legionella products by macrophages. Sensing of bacterial products by Naip5/Birc1e results in stimulation of a multiprotein complex known as the inflammasome that amplifies the molecular signal. Activation of this pathway antagonizes the virulence determinants that allow Legionella to otherwise survive inside the phagosome. We have also observed that two members of the IRF transcription factors family (namely IRF1 and IRF8) play an important role in macrophage defenses against Legionella, acting either downstream or in parallel of Naip5/Birc1e signaling. We are currently using transcript profiling with microarrays and a proteomic approach to identify additional genes and proteins which expression and or phosphorylation status is modulated by Naip5/Birc1e signaling in response to Legionella flagellin.
Tuberculosis remains one of the major threats to global health, and there is an acknowledged role for genetic factors in disease susceptibility. In a mouse model of aerosol infection with Mycobacterium tuberculosis, we have shown that a locus on chromosome 7 is partly responsible for the complex genetic control of the inter-strain difference in susceptibility between C57BL/6J and DBA/2J. We have created special mouse strains in which the resistance-associated copy of chromosome 9 has been isolated on a genetically susceptible background. These strains are being used to characterize the molecular basis of the effect and to ultimately clone the gene responsible. Studying the mouse strain BXH2 in the same infection model, we have shown that mutations in the myeloid-specific transcription factor known as Icsbp/IRF8 completely impair natural defenses against TB, with the animals dying rapidly of fulminant infection. Deficiency in Icsbp/IRF8 also impairs innate and acquired immunity to a number of other infections, including typhoid and malaria. We are presently using genome wide approaches such as chromatin immunoprecipitation on tiling arrays (ChIP on chip), and transcript profiling in macrophages from mutant animals to identify and validate transcriptional targets of IRF1 and IRF8. Such genes constitute good candidates for possible alterations in humans suffering from innate susceptibility to mycobacterial infections.
We are currently studying the complex genetic control of susceptibility to malaria in mouse models. The malarial agent, Plasmodium, has a complex life cycle that involves both an insect vector and a mammalian host. In the latter, various forms of the Plasmodium parasites replicate in different cell types including hepatocytes, and erythrocytes where it induces a severe anemia. The subsequent accumulation of parasitized erythrocytes in brain capillaries is responsible for the severe and lethal cerebral pathology. Using Plasmodium chabaudi as a model for blood-stage replication of the parasite, we have mapped a number of loci that affect replication of the pathogen inside red cells. Furthermore, have shown that deficiency in the erythrocyte form of pyruvate kinase protects mice against malaria. More recently, we have also shown that homozygocity or heterozygosity for mutant PKLR alleles protects human erythrocytes against infection with Plasmodium falciparum. We have also shown that a mutation in the Vnn3 gene (that codes for a pantetheinase) causes susceptibility to blood-stage malaria. The pantetheinase product cysteamine is essential for early response to the parasite, and we can protect otherwise susceptible mice against malaria by pre-treating them with cysteamine. We are currently testing the clinical potential of cysteamine (used alone or in combination with artemisinin) as a novel prophylactic treatment against malaria in humans.
The second and most severe pathology associated with Plasmodium infection in humans is cerebral malaria (CM). This pathology is invariably lethal, and is caused by the accumulation of parasitized erythrocytes in the microvasculature, including the blood brain barrier. This accumulation causes a strong, host-driven, inflammatory response in situ that exacerbates pathology, and ultimately causes come and death. In mouse, CM can be induced by infection with Plasmodium berghei ANKA. We have mapped a novel locus on chromosome 19 (berr4) that controls susceptibility to CM in mouse strains BALB/c (resistant) and C57BL/6J (susceptible). We are currently studying the mechanism by which Berr5 exert its protective effect. We have also initiated a large genomic screen in ENU-mutagenized mice to identify novel genes and pathways involved in susceptibility to CM, in particular genes and proteins that cause the massive and lethal inflammation in P. berghei infected animals.
In our studies of the fungal pathogen Candida albicans, we have reported that susceptibility to infection in the A/J mouse strain is caused by a mutation in the complement component C5a. These studies showed that C5a is not only required for the destruction of C. albicans by mononuclear phagocytes, but have established that C5a acts as a potent early chemotactic factor to recruit neutrophils, and monocytes at the site of the primary infection. Recently, we have identified several inbred strains of mice that have normal C5a function, but yet are susceptible to C. albicans infection, and vice versa. The genetic determinants responsible for this susceptibility are currently being mapped out.
Our laboratory has worked for many years on the phenomenon of multidrug resistance in tumor cells, as caused by members of the P-glcoprotein (MDR) and multidrug resistance associated protein (MRP) family. More recently, we have studied resistance to two groups of drugs that are used in the treatment of multiple myeloma, namely the nuceloside analog arabinocytosine (cytarabine), and the proteasome inhibitor Bortezomib (Valcade). We are currently studying the mechanism by which Bortezomib induces cell death via formation of stress granules, and activation of programmed cell death, and how Bortezomib resistant cell lines developed in our laboratory can survive these effects of the drug.
More recently, we have turned our attention to genetic determinants involved in pre-disposition to cancer. In this new project, we have used chemical carcinogenesis with azoxymethane (AOM) to induce colorectal tumors in mice. We have screened several inbred strains of mice for differences in susceptibility to colorectal cancer induced by AOM, measured by the type, number and size of lesions detected in the colon. In this assay, A/J mice are susceptible while C57BL/6J mice are highly resistant. Using a unique set of AcB/BcA recombinant congenic mouse lines, we have determined that the genetic basis for this difference in susceptibility is caused by a single major locus, that we have named Ccs3. The locus maps on chromosome 3, and the minimal genetic and physical interval for this region contains an estimated ~50 genes. These genes are currently being investigated for their pattern of tissue and cell-specific expression, and for the presence of mutations in either normal tissues or in DNA from tumors emerging in susceptible animals. We are also interrogating human primary tumor samples for alterations in human homologs of the mouse genes present in the Ccs3 interval. Finally, we have developed a model of inflammation-induced colorectal cancer, induced by dextran sulfate and AOM. We are investigating the genetic basis of differential susceptibility of inbred mouse strains to colorectal tumors induced by this treatment.
Neural tube defects
Neural tube defects (NTD) such as spina bifida are very common in humans, and rank second only to cardiac defects in inborn errors. Our laboratory has been studying two mouse models of NTD, namely the splotch mouse, and more recently the looptail (Lp) mouse. Mice homozygote for the Lp mutation suffer from a very severe form of neural tube defect called craniorachischisis. This condition is lethal in utero. Using a positional cloning approach, we found that the Lp defect is caused by mutations in the membrane protein Vangl2. This protein is a mammalian homolog of the fly Van Ghog/strabismus, a gene known to be involved in regulating the development of certain structures in the fly, being part of a pathway known as the PCP pathway, standing for planar cell polarity. Flies with mutations in this pathway have defect in the organization of several appendages and structures (wings, eyes), which is caused by an inability of epithelial cells to properly organize in the planar polarity field. This pathway is highly conserved from flies to humans, and in mammals this pathway is also involved in certain types of morphogenic movements during embryogenesis known as convergent extension; Again, this pathway involves the dynamic rearrangement and migration of epithelial cells to form primordial structures, including the neural tube, the heart, the kidney and others.
In mammals, there are two Vangl genes, Vangl1 and Vangl2 that encode for highly similar membrane proteins. We recently detected mutations in the human VANGL1 gene in familial and sporadic cases of spina bifida. We have also recently determined that VANGL1 and VANGL2 genetically interact in the regulation of neurogenesis and in the appearance of neural tube defects in mice. These exciting findings suggest that the VANGL family may play a critical role in the etiology of this malformation. We are currently investigating large number of clinical cases to determine the frequency of such mutations, and are exploring the molecular basis for the loss of function in these patients. This is being investigated in mouse models, but also in zebra fish, where the human gene can correct the neural tube defect detected in fish embryos showing a defect in the corresponding fish homolog. Biochemically, we are investigating the multiprotein complex formed by Vangl proteins at the membrane and that is responsible for providing polarity information to the cells, and how alterations in this complex may cause spina bifida in humans.