Since 1995, the Scientific Directors and the Office of Research on Women's Health have funded the Fellows Award for Research Excellence (FARE), a competition to recognize intramural fellows’ noteworthy research. Each winner of the FARE competition receives a $1000 travel stipend to present his or her work at an upcoming scientific meeting, the chance to display a poster at the FARE awards presentation ceremony, and the opportunity to serve as a judge for the following year's FARE competition.
Congratulations to all of our NICHD FARE winners. Check out some of their projects here!
Chromosomal neighborhood matters in Polycomb group repression
By Sandip De, PhD (Kassis lab)
DNA sequence alone is not enough to govern the complex growth and development of multi-cellular organisms. Epigenetics, the chemical modification to some DNA base pairs and associated proteins, is an important factor too. The Polycomb group proteins (PcGs) are critical players in epigenetic regulation. PcGs, first identified in fruit flies, help remodel chromatin to enable epigenetic silencing of several hundred genes where required. In Drosophila, cis-regulatory DNA elements known as Polycomb response elements (PREs) recruit PcG proteins. Two developmental genes that utilize the chromatin remodeling effects of PcGs and PREs are engrailed (en) and invected (inv). Given that epigenetic regulation influences development, stem cell regeneration, and even disease, we hope to use this model system to better understand PcG regulation.
The genes en and inv form a co-regulated gene complex containing four strong PREs in the region—two at inv and two at en. Fruit flies lacking en do not survive. Surprisingly, deletion of either the inv or en PREs fails to cause over expression of inv or en. Even more surprisingly, flies that have a deletion of all four PREs survive and are fertile. Genome-wide experiments uncovered six potential weak PREs present in the en/inv domain and strong PREs in the Enhancer of Polycomb [e(Pc)] and toutatis (tou) flanking genes. Further analysis using a high-throughput technique to analyze chromosome organization in the cell shows that these weak PREs, along with PREs in the neighboring regions, interact with each other to maintain the repression and three-dimensional structure of the en/inv domain in the absence of major PREs. These data show that chromosomal neighborhood is important for PcG repression and that PREs from flanking genes can regulate en and inv expression.
Small molecule therapeutics for Huntington’s disease
By Lori Griner, PhD (DePamphilis lab)
Huntington’s disease (HD) is a neurodegenerative disorder characterized by the build-up of toxic mutant huntingtin protein aggregates in neurons. The huntingtin aggregates cause neuronal cellular dysfunction and eventual cell death. Loss of neurons in HD patients leads to the patient suffering from cognitive, behavioral, and motor difficulties, contributing to a poor quality of life and shortened life expectancy. There is a dire need for therapeutics for HD patients, for there is no cure, and current therapies only attempt to control clinical symptoms.
A prevailing goal in HD therapy is to keep toxic mutant huntingtin protein aggregates to a minimum, allowing the neurons to survive. The DePamphilis laboratory has recently identified a family of small molecule therapeutics that promote autophagy, a biological process in which cells digest intracellular components, such as proteins, organelles, and pathogens. Initiating autophagy has been shown to promote the digestion and clearance of protein aggregates. These new small molecule therapeutics are more effective at inducing autophagy than the gold standard drug, rapamycin. In our studies, we have shown this family of novel small molecules to be effective at reducing mutant huntingtin protein aggregates and improving the survival of cells in an HD cell model system. Given that rapamycin is too toxic for long-term treatment in HD patients, the discovery of new autophagy-inducing therapies is of the upmost importance.
Exploiting dual host machineries
By Yi-Han Lin, PhD (Machner lab)
Intracellular bacterial pathogens utilize various strategies to infect and replicate within their host. These pathogens can alter specific host cell processes, such as signaling pathways and innate immune response, by translocating effector proteins into the host cell during the course of infection. These effector proteins target different cellular pathways and execute their function by mimicking eukaryotic host proteins.
Legionella pneumophila, the causative agent of Legionnaire’s disease, translocates over 300 effector proteins into its host during infection. We used bioinformatics to search for effector proteins that mimic E3 ubiquitin ligase, the enzyme that mediates the last step of protein ubiquitination and controls protein degradation in eukaryotic cells. After experimental verification, we found that the Legionella effector protein GobX exhibited robust E3 ligase activity, and in transfected fibroblast-like cells, displayed exclusive localization to the Golgi compartment. Truncation and mutagenesis studies revealed that a single cysteine residue at position 175 localizes GobX to the Golgi. We further demonstrated that a host-mediated lipid modification to the cysteine residue (called S-palmitoylation) is required for GobX’s subcellular localization and E3 ligase activity in host cells.
Taken together, we have identified a novel bacterial effector protein that can exploit both host ubiquitination and lipidation machineries. Such findings broaden our understanding of the strategies bacterial organisms evolved to interfere with their host and also provide insight into future designs for drug targets.
A novel tool for analyzing pooled biomarkers
By Emily M. Mitchell, PhD (Perkins lab, DIPHR)
Biomarkers are an increasingly valuable source of information, particularly as technology to measure biomarkers has improved. This increased capacity, however, conjures new challenges. The potentially high cost of analyzing biomarkers, along with corresponding measurement error from the lab assay, may prevent researchers from utilizing specimens to their full advantage. One method to simultaneously reduce lab costs and improve measurement precision is to pool specimens prior to performing lab tests. The subsequent measurement is representative of the average biomarker concentration from each of the individual specimens in the pool. Researchers may be hesitant to adopt a pooling strategy, however, since analysis of pools often requires specialized techniques. Statistical analysis can be particularly complicated when the distribution of a biomarker is skewed (i.e., not symmetric), as many measured biomarkers naturally are, since standard analyses may no longer apply.
In our study, we propose an analytical strategy to calculate the association between an observed exposure and a skewed biomarker that is measured in pools. We develop a straightforward and accessible method that is applicable to pools containing specimens from participants with identical predictor values, such as demographic information (e.g., age, BMI, socio-economic status). When pools are heterogeneous, i.e., specimens have different predictors, we propose an effective but more computationally complex alternative. By applying these estimation techniques to strategically formed pools, valid and efficient estimates of the association between an exposure and a biomarker of interest can be obtained at a fraction of the cost required to analyze all individual specimens. Our methodological contribution to the base of available statistical methods to analyze pooled specimens will empower researchers to more confidently consider pooling as a potential study design.
Additive effects of chromatin remodeling complexes on global chromatin structure
By Josefina Ocampo, PhD (Clark lab)
One of the most challenging aspects of eukaryotic cells is to understand how DNA is packaged into chromatin to fit the topological constraints imposed by the nucleus. The nucleosome is the basic subunit of chromatin structure and has inhibitory effects on transcription, DNA replication, and DNA repair. Thus, a balance between DNA packing and access must be achieved. Chromatin remodeling complexes are major actors in this dynamic process. They use the free energy obtained from ATP hydrolysis to assemble, eject, or slide nucleosomes. Mutations in components of human remodeling complexes were recently detected at high frequencies in human cancers.
We have addressed the roles of four different chromatin remodeling complexes in nucleosome organization in vivo in the yeast Saccharomyces cerevisiae, including ISW1, ISW2, CHD1, and the essential RSC complex. In yeast, nucleosomes are regularly spaced and show a global phasing (nucleosomes are bound to a particular DNA sequence that keeps them at a regular distance relative to the transcription start site).
We constructed strains with the essential RSC8 subunit under the control of the GAL promoter and isw1, isw2 or chd1 null mutations in all possible combinations in the same genetic background. In the absence of RSC, nucleosomes shift towards the transcription start site with consequent narrowing of the nucleosome-depleted region typically found at most promoters. Nucleosome spacing remained unchanged at every 165 base pairs (bp). In the mutants, the spacing in isw1 mutants reduced by 6 bp to about 159 bp. While the chd1 mutant showed loss of phasing with little change in spacing, the isw2 mutant did not show any obvious changes in global chromatin structure. The chromatin structures of the double, triple, and quadruple mutants represent the sum of the effects observed in the individual mutants, indicating that these remodeling complexes have distinct functions in chromatin organization.
Preterm birth in the context of increasing income inequality
By Maeve Wallace, PhD (Mendola lab, DIPHR)
Income inequality is the unequal distribution of income in a population. It is an issue of increasing concern in the United States (U.S.) where the current income gap between the wealthiest and the poorest Americans is the largest it has been since 1928. Income inequality is thought to have a harmful effect on population health. One metric of population health is the preterm birth rate, defined as birth of a baby before 37 weeks gestation. Preterm birth is a leading cause of infant illness and death in the U.S., and these births more often occur to poorer women compared to wealthier women.
We used statistical models to determine if rates of preterm birth were greater in states where income inequality was increasing compared to where it was decreasing using electronic medical records from 19 hospitals in 11 states and the District of Columbia from 2002 to 2008. We found that women living in states where inequality increased over the year leading up to delivery were about seven percent more likely to have a preterm birth. This was true regardless of the degree of initial inequality, and the amount of increase in inequality was less important than the increase itself. Our findings were not explained by differences in maternal characteristics including race, age, insurance status, smoking or alcohol use during pregnancy, chronic medical conditions, or state-level poverty or unemployment rates. Understanding the mechanisms through which increasing income inequality increases the risk of preterm birth and identifying modifiable risk factors should be priorities for future reproductive health research.
Studying blood vessel building blocks—one at a time
By Jianxin Alex Yu, PhD (Weinstein lab)
Angiogenesis, a process through which blood vessels sprout from existing vessels, is critical for vertebrate organogenesis and plays an essential role in pathological conditions such as cancer. Although vessel formation has been studied at the tissue level, limited in vivo imaging and a lack of genetic tools have hampered the study of individual cellular architecture and behaviors during angiogenesis. Basic questions, such as how individual endothelial cells (ECs, the building blocks of blood vessels) coordinate movements and shape changes during sprouting and lumenized tube formation still remain poorly understood.
We have developed endothelium-specific transgenic zebrafish and high-speed two-photon confocal imaging methods to examine in vivo endothelial morphological changes at single cell resolution. New fluorescent transgene tools simultaneously mark both EC nuclei and plasma membranes, or cell-cell junctions, to monitor the morphology and dynamic behaviors of individual ECs. Single cell analysis and three-dimensional reconstruction reveal the heterogeneity of EC morphology hinting at multiple cellular mechanisms governing endothelial tube formation.
These newly developed transgenic tools and imaging methods allow us to visualize complex cellular and subcellular dynamics during angiogenesis with an unprecedented level of resolution. Application of this approach with emerging single cell transcriptome sequencing technologies will help us understand not only the concerted EC behaviors during normal vessel development but also the underlying cellular mechanisms of abnormal endothelial formations in zebrafish models of human vascular disease.
Spindle superheroes and cell division
By Michael Shaofei Zhang, PhD (Dasso lab)
Every human life derives from a single fertilized egg that eventually divides to around 40 trillion cells in a normal adult body. The majority of these cells have identical genetic materials enclosed in chromosomes. Failure to maintain the integrity of chromosomes may cause severe disorders and diseases, including birth defects and cancer. We are trying to understand how normal cells manage to maintain chromosome integrity during the numerous cell divisions that occur during a lifetime.
When a cell is about to divide, each chromosome replicates itself. The next major task is to equally distribute the two copies of chromosomes into two daughter cells. A protein structure called spindles is responsible for chromosome allocation. It works like two Spidermen sitting on opposite sides of a mother cell. They are shooting webs (spindles) to catch one copy of each chromosome and then pull the chromosomes to either side.
Real spindles are not as magical as Spiderman’s webs. They require signals to find their chromosomes. A protein named RCC1 on chromosomes generates the required signals, but RCC1 also exists in other places. How do spindles know which RCC1 is the correct target? In our study, we demonstrate that an RCC1 inhibitor named RanBP1 specifically blocks signals from RCC1 proteins not located on chromosomes. When we removed RanBP1 from the system, spindles were unable to find their targets. We hypothesize that in the absence of RanBP1, spindles are puzzled by RCC1 signals from both on and off chromosomes. RanBP1 may help spindles target chromosomes by blocking non-chromosomal RCC1 signals.
Our study not only reveals a novel cellular function of RanBP1, but also suggests a potential therapeutic target to correct abnormal cell divisions.