While a large number of studies have been performed examining the toxicity of Qdots to cells in culture, the state of knowledge regarding their in vivo toxicity of Qdots is quite limited. Several studies have investigated the distribution and long-term consequences of Qdot exposure in mice and rats after intravenous injection and saw no signs of overt toxicity [2, 3]. However, in the study by Yang et al, [2] they did observe some evidence of sub clinical injury (renal damage observable at the ultra structural level). Several recently published studies have examined the ability of Qdots to cause inflammatory responses in vivo, after intraperitoneal injection [4] and intratracheal instillation [5]. In contrast to eariier studies, both of these studies concluded that Qdots are pro-inflammatory, but the reasons for this were different in the two studies. In the study by Hoshino and colleagues [4], inflammation was attributed to the presence of DNA conjugates on the surface of the Qdots, resulting in the activation of an innate immune response through the TLR9 receptor. Jacobson et al, [5] surmised that the lung inflammation caused by their Qdots was due to their unstable nature and the release of Cd, a known pulmonary toxicant. These studies underscore the need to fully characterize the physical and chemical characteristics of nanomaterials before inferences can be made regarding the extent of injury or the mechanisms of toxicity. In a recent publication by the National Academy of Sciences """"""""Toxicology in the 21^* Century"""""""" a major emphasis was placed on the importance of developing reliable in vitro screening assays for predicting the in vivo toxicity of chemicals. This link has been difficult to make, because ofthe acknowledged difficulties in replicating the complex biochemistry, cellular biology and physiology present in vivo with cell culture models. Nonetheless, the advent of modern """"""""omics"""""""" technologies, stem cell technology and the ever-increasing sophistication of in vitro culturing methods, gives reason to believe that this goal may be achievable. A core concept of this paradigm is the knowledge that transcriptomics, proteomics and metabolomics will be important technologies in this endeavor. If the differentiated state of cells in culture truly represents their in vivo condition, then toxicology testing, at least at the cellular level, will have come a long way towards achieving this goal. Until such time that this is possible, in vivo testing will still be necessary to accommodate all of the complicated cell-cell interactions and pathophysiological responses that are present in the whole animal. Importantly, by using these """"""""omics"""""""" technologies to define molecular targets and toxicity pathways, in both in vitro and in vivo, significant linkages can be made between these two model systems that can help in the understanding of mechanism of toxicity. Because of their size, their cost, and their reproductive biology, mice have been historically popular in toxicology studies. One distinct advantage of using mice for in vivo testing of chemical toxicity is their genetics. There are four reasons why their genetics make them attractive models for conducting toxicity studies. The first is that all individuals within a strain of inbred mice are genetically identical, and thus by examining the diversity in responses among different mouse strains, one can identify resistant and sensitive strains. Secondly, the mouse genome has been sequenced, and recently, 16 inbred mouse strains have been individually re-sequenced, with over 20,000 single nucleotide polymorphisms (SNPs) having been identified among these strains. Thirdly, because of advances in DNA microarray technologies, it is now possible to include global gene expression information as quantitative trait loci (expression QTLs or eQTLs) for genetic linkage studies. Finally, because the mouse and human genomes are highly conserved in both their primary sequence, as well as gene order, genes that affect function in mice can be readily mapped to syntenic regions on human chromosomes. These genes can then be further prioritized based on prior knowledge of QTLs in both rodents and humans, and by whether there is significant polymorphism associated with these genes in humans (see [6] and [7] for reviews). Another useful feature of the mouse, is the ability to modify its genome to test specific hypotheses regarding the role of particular genes/gene products in a phenotype. We have used this approach to investigate the role of the antioxidant intracellular thiol glutathione (GSH) in resistance to chemicals that cause oxidative stress. Many agents are known to exert oxidative stress, including many nanoparticles, and importantly, the Qdot core constituents cadmium, zinc, and mercury. These metals are also known to react with and at relatively low doses deplete GSH. An adaptive response that is commonly seen after exposing organisms to these agents is the induction of GSH biosynthetic enzymes, especially glutamate-cysteine ligase (GCL) - the rate-limiting enzyme in GSH biosynthesis. GCL is a heterodimer composed of two subunits, a catalytic subunit (GCLC) and a modifier (or regulatory) subunit (GCLM). GCLM acts to improve the catalytic efficiency of GCLC by lowering the Km for glutamate and increasing the Ki for GSH feedback inhibition of the holoenzyme. We, and others, have shown in previously published work, and in recently conducted studies, that Gclc and Gclm mRNA, protein levels, and GCL enzyme activity, are induced by exposure to Cd and MeHg. For instance, we have shown that MeHg can induce GCL expression in mouse tissues as adults [8, 9] or during development [10, 11]. It has been shown that Cd can induce both Gclc and Gclm [12-16]. This up-regulation of GCL and GSH is an important adaptive response to these forms of stress, and shows that GSH and GCL expression influence the susceptibility of mice to Cd and Hg toxicity and, by extension, to Qdots containing these metals. We have examined the ability the ability of TOPO-PMAT coated Qdots to induce oxidative stress and upregulate GCLC and GCLM in multiple celltypes in vitro. We observed that Qdots are capable of inducing GCLM in some cells, and this seems to be related to the degree of uptake of these particles (see Project 1 Preliminary Data section). Part of the explanation for the inducibility of GCL genes relates to the presence of antioxidant response elements (AREs) in the 5'promoter of GCLC and GCLM genes [17-20]. Interestingly, research has shown variability among humans with respect to induction of GCL after exposure to GSH-depleting drugs [21, 22]. Walsh and colleagues have published on a common trinucleotide repeat (GAG) polymorphism in the 5'-untranslated region (5'-UTR) of the human GCLC mRNA and have shown that this polymorphism is functionally important [23]. Work from our laboratory found a correlation between the number of GCLC GAG repeats and lung function in patients with cystic fibrosis (CF) [24]. Additionally, Nakamura and colleagues [25- 27] have shown an association between SNPs in both GCLC and GCLM have been associated with an increased incidence of coronary artery disease with deficits in maintenance of vascular tone. Finally, Custudio et al. [28] have found that polymorphisms in a GCL influence the retention of MeHg in humans. This raises the interesting possibility that GCL polymorphisms may influence susceptibility to the toxicity of Qdots containing these toxic metals. It is therefore important that animal models of GCL over expression or insufficiency be used to more thoroughly address the role of GSH synthesis in disease susceptibility, and variable sensitivity to environmental toxicants, including Cd and Hg, and nanoparticles such as Qdots. We have generated mice in which a portion of the Gclm gene has been deleted (GCLM null mice;[29]), resulting in a complete lack of expression of this GCL subunit in these mice. In addition, we have generated transgenic mice that inducibly over express both mouse GCL subunits [30]. We have also crossed Gclm inducible mice to Gclm null mice [31]. Using these genetic approaches we can modulate the level of GCL and GSH synthesis in these animals, providing a unique set of animal models to determine the role of GSH biosynthesis in modulating toxicant-induced oxidative stress and injury. We have found that over expression of GCL genes confers resistance to xenobiotic-induce oxidative stress in the liver [30]. Additionally, we have found that GCLM null mice show increased sensitivity to chemical stress in the liver [29] and regain resistance when Gclm is turned back on [31]. Cultured neurons from these GCLM null mice are hypersensitive to agents that induce oxidative stress, including domoic acid [32], organophosphate insecticides [32], and MeHg [1]. Several years ago, we proposed to evaluate the toxicity of Qdots to these GCL transgenic mice and were warded and ROl grant from the NIEHS to carry out this work. So far we have evaluated the toxicity of OPO-PMAT coated Qdots toward multiple cell types in vitro and examined the ability of these Qdots to elicit nflammatory responses in vitro (see Project 1 Preliminary Studies) ,and in vivo in mice exposed by nasal instillation (see below). In the current application, we seek to extend our current work with these mice, to include assessments of he toxicity of inhaled Qdots in multiple inbred strains of mice. The principle reason for doing so is that it will llow us to identify genes and molecular toxicity pathways associated with the physical and chemical characteristics of a variety of Qdots in an unbiased fashion. The approach we wish to use has been variously ermed """"""""genetical genomics"""""""" and """"""""systems genetics"""""""" but basically exploits advances in genetics, genomics, transcriptomics, epigenetics, high throughput DNA sequencing, bioinformatics and other data intensive technologies. An important advance in mouse genetics involves the resequencing of 16 different inbred mouse strains by eriegen Sciences as a contract with the NIEHS Center for Rodent Genetics (URL: httD://mouse.Derieaen.com/mouse/index.html). This very successful effort has yielded highly valuable nformation on the prevalence of neariy 20,000 SNPs among these strains. Strains used for this analysis ncluded C57BL/6J (reference strain), 129S1/SvlmJ, A/J, AKR/J, BALB/cByJ, C3H/HeJ, DBA/2J, FVB/NJ, OD/LtJ, (these strains account for approximately 85% of all inbred mouse research). Also included are Strains that derive from major taxonomic groups (CAST/EiJ, MOLF/EiJ, PWD/PhJ, WSB/EiJ), and strains that ave a special purpose in behavioral, immunological or genetic distance from other inbred strains (BTBR +tf/J, NZW/LacJ, KK/HiJ). Investigators have begun to use these strains and strains present in the mouse iversity panel to begin to map quantitative trait loci that are linked to disease phenotypes of interest. Examples of the kinds of studies that have been done in toxicology include work by Rusyn and colleagues to nvestigate the genetic determinants of acetaminophen toxicity [33]. Importantly, a genetic determinant iscovered in this mouse study (Cd44) was subsequently found to be susceptibility factor for acetaminophen verdose in humans [34]. However, probably the most important advance in the area of """"""""system genetics"""""""" was the commitment by the complex Trait Consortium members to begin to systematically intercross 8 inbred, genetically diverse, strains of mice (most of which are included in the Periegen resequencing effort) with the goal to create over 1000 ecombinant inbred lines from these mice [35, 36]. This resource is well on its way toward accomplishing this goal. As reported recently by Morahan and colleagues at The Western Australian Institute for Medical esearch in Perth, and the Australia Animal Resources Centre, Canning Vale, Western Australia, they have lready established over 1000 such inbred lines [35]. These lines will be quite useful for combined genomic and transcriptomic analysis to identify so called eQTLs. This approach has become recently possible because of refinements in bioinformatics platforms and pathway analysis tools. For example, Wu et al, [37] used gene set enrichment analysis to further define trans-eOTL ands and in so doing were able to predict and subsequently show that Cyclin H has a role in controlling oxidative phosphorylation. Investigating the relationship between gene mutations and eQTLs in yeast, Yeger- Lotem et al. [38] concluded that there is a bias for gene centered mutational data to favor genes which are associated with transcription and signal transduction;while transcriptional response data center mostly on shifts in metabolism. This group recently developed a tool (ResponseNet) to link transcriptional changes with gene mutations/polymorphisms by incorporating protein-protein and protein-gene interaction networks into the analysis. Doing so dramatically improved the interpretation of both of these data sets and led to the discovery of novel pathways associated with a-synuclein toxicity. In a similar approach Ideker and colleagues have taken the approach of combining eQTL analysis with protein interaction networks (eQTL electrical diagrams - eOED) which prioritizes the genes located within an eQTL area on the basis of known regulatory protein interactions [39]. Another recently published approach has been to compare the eQTLs discovered in recombinant inbred mouse lines with those discovered in the mouse Genome Diversity Panel inbred lines [40]. This comparative approach resulted in the verification of previous candidate trans-eOTLs on chromosome 12, in which many Serpine gene family members are located. In addition, Breitling and colleagues [41] have suggested using a strain permutation-based corrections to help classify those trans-eOTLs (so-called hotspots) that are more likely to be true, and those that are likely false positives. Furthermore, Gerrits et al. [42] have found that the differentiation state and variation in cell types can dramatically influence the eQTL profile, and so a well characterized and consistent culture conditions will be important to maximize the likelihood of detecting eQTLs that are related to Qdot exposure. On the other hand, if there are differences in response because there are differences in the differentiation state and/or relative cell types present that are genetically determined, then these differences may be uncovered using genetical genomics/genetics systems approaches. Another important goal of the proposed work would be to link eQTLs with chromatin condensation state, to uncover epigenetic influences on gene expression such as CpG island methylation and/or acetylation. It is important to consider that DNA polymorphisms present among the CC strains may not be as important as epigenetic differences brought about by environmental factors. Accordingly we will order such mice from the same source and house them under identical conditions. These approaches should allow for the identification of pathways that are perturbed by toxicant exposures. Information gathered from these studies will be valuable in helping to predict the role of Qdots, and their constituent metals on their toxicity, as well as in the design of cores and coatings that will optimize the desirable properties of these materials while minimizing their adverse health impacts. In conclusion, the above referenced literature leads us to hypothesize that differences in inbred mouse strains will be important determinants of susceptibility to Qdot-induced lung injury. The resequencing of the 16 inbred mouse strains by Periegen, and the availability of several relatively inexpensive whole mouse genome expression arrays now makes it feasible to screen for genes that may predispose certain strains of mice to Qdot-induced toxicity. Sensitive CC strains will be identified and using recombinant inbred strains, together with the recently acquired SNP data, we will map genes whose allelic differences may account for these changes in response to Qdots. Inten/al mapping analyses will be used to further investigate whether these factors might be acting in cis or trans, and the proposed interaction pathways will be further investigated using DNAse hypersensitivity site analysis by deep sequencing, and other techniques available at the Northwest Reference Epigenome Mapping Center.
Showing the most recent 10 out of 27 publications
