Environmental Factor, June 2011, National Institute of Environmental Health Sciences
Intramural papers of the month
By Erin Hopper and Jeffrey Stumpf
- Innate immunity linked to DNA damage response
- Human PUMILIO proteins adopt multiple RNA binding modes
- Separation of function mutations identified in estrogen receptor alpha
- Estimating relative risk in epidemiological studies using imputed data sets
Innate immunity linked to DNA damage response
Tumor suppressor p53is a transcription factor that binds to response elements of greater than 200 genes, allowing cells to cope with DNA damage. NIEHS researchers have found that the innate immune system is yet another important target of the p53 network in a manner that is specific to primates.
Previous work at NIEHS showcased sequence motifs that predicted the location of p53 binding and the degree of activation of the downstream gene. These initial studies suggested that some Toll-like receptors (TLRs), proteins that recognize various molecules and trigger the inflammatory immune response, may be activated by p53.
This recent study reveals that in primary human lymphocytes and alveolar macrophages, expression of nearly all TLRs is altered by DNA damage, and most of the changes are dependent on p53. For example, TLR8 expression is controlled by p53, and a natural single nucleotide polymorphism (SNP) in the TLR8 response element changed p53 activation, suggesting the exact location of p53 binding.
The study also shows variability of p53 responsiveness among individuals. Because p53 is important in tumor suppression, this variability may help explain different genetic predispositions to cancer. These findings on individual differences in TLR induction by p53 activation and DNA damage should prove useful in the development of TLR-targeted vaccines and TLR-based cancer treatments.
Citation: Menendez D*, Shatz M*, Azzam K, Garantziotis S, Fessler MB, Resnick MA. (http://www.ncbi.nlm.nih.gov/pubmed/21483755) 2011. The Toll-like receptor gene family is integrated into human DNA damage and p53 networks. PLoS Genet 7(3):e1001360. (*co-first authors) Story (http://www.niehs.nih.gov/news/newsletter/2011/april/science-niehs-investigators/index.cfm)
Human PUMILIO proteins adopt multiple RNA binding modes
Investigators at NIEHS have demonstrated that human PUMILIO proteins can adopt multiple binding modes to interact with a variety of RNAs. PUMILIO proteins are members of the PUMILIO/FBF (PUF) family of proteins, which is a group of RNA-binding proteins that is responsible for post-transcriptional regulation of gene expression.
All PUF proteins contain an RNA-binding domain called the Pumilio homology domain (PUM-HD), in which eight PUM repeats and two pseudo-repeats come together to form a crescent-shaped structure. In a 1:1 binding mode, each PUM repeat recognizes and binds a single RNA base. However, recent studies have suggested that PUF proteins bind a wider variety of RNAs than would be predicted from a 1:1 binding mode.
In this study, the investigators crystallized the PUM-HD of human PUMILIO1 and PUMILIO2 in complex with four RNAs that varied in sequence. They found that these proteins exhibit three different binding modes around the fifth RNA base, including two 1:1 modes and one base-omission mode. These multiple binding modes allow for the recognition of a larger variety of RNA sequences. While these binding modes do not appear to vary the binding affinity of RNA, they may serve to expose alternate binding surfaces for various protein interaction partners.
Citation: Lu G, Hall TM. (http://www.ncbi.nlm.nih.gov/pubmed/21397187) 2011. Alternate modes of cognate RNA recognition by human PUMILIO proteins. Structure 19(3):361-367.
Separation of function mutations identified in estrogen receptor alpha
NIEHS researchers have identified critical sequences for each function of estrogen receptor alpha (ERalpha) by mutating the gene that encodes it. Because estrogen receptors (ERs) mostly determine the effects of estrogen, mapping receptor functions is important for understanding hormone regulation.
Ligands that bind to ERs have the ability to activate three distinct functions of this receptor: binding directly to DNA to regulate gene expression; binding to other transcription factors that bind DNA and alter gene expression; and facilitating rapid action signaling cascades in response to various ion exchange pathways. Of the three functions, crucial amino acids were only described for the direct DNA binding activity. The researchers focused on the D-domain of ERalpha and found sequences within this domain critical for interaction with other transcription factors. They also blocked nuclear localization of the receptor to block genomic responses to focus on the rapid action signaling events.
This study reported three distinct mutants that separately block recruitment of transcription factors, nuclear localization, and all ligand-mediated actions in the nucleus, but all three mutants maintained the ability to promote rapid action responses.
Separation of function mutants are critical for understanding the role of each function in vivo. Because targeting estrogen and its receptors is important in many therapeutic strategies, these studies promote the possibility of precisely changing a specific activity involved in estrogen regulation.
Citation: Burns KA, Li Y, Arao Y, Petrovich RM, Korach KS. (http://www.ncbi.nlm.nih.gov/pubmed/21285458) 2011. Selective mutations in estrogen receptor alpha D-domain alters nuclear translocation and non-estrogen response element gene regulatory mechanisms. J Biol Chem 286(14):12640-12649.
Estimating relative risk in epidemiological studies using imputed data sets
Researchers from the Biostatistics Branch and Epidemiology Branch at NIEHS recently developed a new method to estimate relative risk using imputed single nucleotide polymorphism (SNP) data from cases and their parents. Data imputation allows for the exploration of untyped SNPs and is required for meta-analyses of multiple studies that used different platforms for genotyping. SNP imputation involves using software to substitute in SNP data when gaps in the data are present. Before the development of this method, no technique for meta-analysis of imputed SNP data involving both family-based studies and case-control studies existed.
The researchers used a log-additive model to estimate the relative risk for a disease-related variant and ran a simulation to test the model. The simulation confirmed that the method was accurate at estimating risk using genotyped SNP data. Next, the researchers applied the method to a real data set using the Mexico City Childhood Asthma Study (1998-2003). This data set allowed for the comparison of relative risks calculated using genotyped SNP data and data imputed using the software package MACH.
The estimated relative risks calculated using the imputed data correlated very well with those calculated using the actual genotypes, confirming the effectiveness of this method for the analysis of imputed SNP data. This method will allow researchers to conduct meta-analyses using a mix of case-parent triad and case-control studies.
Citation: Shi M, London SJ, Chiu GY, Hancock DB, Zaykin D, Weinberg CR. (http://www.ncbi.nlm.nih.gov/pubmed/21296892) 2011. Using imputed genotypes for relative risk estimation in case-parent studies. Am J Epidemiol 173(5):553-559.
(Erin Hopper, Ph.D., is a postdoctoral fellow in the NIEHS Laboratory of Structural Biology Mass Spectrometry Group. Jeffrey Stumpf, Ph.D., is a postdoctoral fellow in the NIEHS Laboratory of Molecular Genetics Mitochondrial DNA Replication Group.)