Poster sessions



Poster boards and push pins will be provided. The boards measure 4' by 4'. All boards will be well-marked with the poster numbers. Two poster awards will be presented, sponsored by AUM LifeTech and Charles River Laboratories


Delineation of the molecular consequences of combining erlotinib with photodynamic therapy in non-small cell lung cancer
Shannon M. Gallagher-Colombo, Rensa Chen, Joann Miller, and Theresa M. Busch
Department of Radiation Oncology, Division of Radiation Biology, Perelman School of Medicine, University of Pennsylvania

Numerous aggressive cancers are characterized by overexpression or overactivation of the epidermal growth factor receptor (EGFR), making this receptor a favorite target for therapeutic intervention. Inhibition of EGFR is most commonly achieved through the use of anti-EGFR antibodies (i.e. cetuximab) or small molecule inhibitors (i.e. erlotinib); however, the efficacy of these molecular targeting agents is often limited by innate and acquired resistance to these drugs. Recent work from our lab has highlighted the benefits of combining photodynamic therapy (PDT) with the small molecule inhibitor of the EGFR, erlotinib. Notably, combination therapy results in improved tumor responses, vascular shutdown, and increased tumor cell death in H460 human tumor xenografts. Interestingly, these cells are reported to be resistant to EGFR inhibition. The goal of this work is to elucidate the molecular mechanisms underlying the improved therapeutic benefit observed in these tumors following combination treatment. We explored the effect of combination therapy on activation of signaling pathways induced by EGFR, including PI3K/Akt and MAPK/Erk. We also evaluated the impact of combination therapy on expression of other transcription factors that are important for tumor progression. The results from this work will provide important insights into the molecular mechanisms that underlie the enhanced therapeutic benefit observed when combining erlotinib with PDT, and may potentially provide new therapeutic targets to improve treatment efficacy in other tumor types.


back to the top



COX-2 as a mediator of oncogenic PKCε in prostate cancer
Rachana Garg1, Mahlet Abera1, Priti Lal2, Jorge Blando3, Fernando J. Benavides4, Michael D. Feldman2, Emer M. Smyth1, and Marcelo G. Kazanietz1
1Department of Pharmacology, Perelman School of Medicine, University of Pennsylvania; 2Pathology and Laboratory Medicine, Hospital of the University of Pennsylvania; 3The University of Texas, Austin, TX; 4Department of Molecular Carcinogenesis, The University of Texas MD Anderson Cancer Center, Smithville, TX

Protein kinase C epsilon (PKCε), a member of the PKC family of phorbol ester/diacylglycerol receptors, is upregulated in human cancers including prostate cancer. COX-2, a well-known NF-κB target gene, is also upregulated in prostate cancer. As PKCε plays an important role in prostate cancer cell survival and cooperates with oncogenic insults, we aim to determine if PKCε regulates COX-2 activation during prostate tumorigenesis. PKCε depletion diminishes TNFα- or LPS-induced COX-2 mRNA expression in LNCaP cells and constitutive COX-2 levels from PC3 cells. Conversely, PKCε overexpression potentiates TNFα or LPS-induced COX-2 expression in LNCaP cells. Notably, transgenic overexpression of PKCε in the mouse prostate causes preneoplastic lesions with elevated COX-2 levels. Interestingly, when we intercrossed the prostate-specific PKCe transgenic mice with mice that are haploinsufficient for Pten, the resulting compound mutant mice developed fully invasive adenocarcinoma with NF-kB hyperactivation and high COX-2 levels. Likewise, stable overexpression of PKCε in mouse prostate epithelial cell lines that are heterozygous (P8) or homozygous (CaP8) for Pten deletions led to significant enhancements in cell proliferation, motility, and invasiveness as well as in LPS-induced COX-2 mRNA expression compared to controls; this effect was more pronounced in CaP8 cells. Studies using human prostate tumors reveal a co-existence of PKCε overexpression, NF-κB hyperactivation, and COX-2 up-regulation. Lastly, treatment of PKCe overexpressing P8 or CaP8 cells with the COX-2 inhibitor NS398 caused a pronounced growth inhibition. Overall, our results suggest COX-2 as a potential mediator of PKCε oncogenesis in prostate cancer, particularly in the context of Pten loss.


back to the top


A novel BRISC-SHMT complex deubiquitinates IFNAR1 and regulates interferon responses
Hui Zheng1#, Vibhor Gupta2#, Jeffrey Patterson-Fortin2#, Sabyasachi Bhattacharya1#, Kanstantsin Katlinski1, Junmin Wu2, Bentley Varghese1, Christopher J. Carbone1, Bernadette Aressy2, Serge Y. Fuchs1, and Roger A. Greenberg2
1Department of Animal Biology and Mari Lowe Comparative Oncology Center, School of Veterinary Medicine, and 2Department of Cancer
Biology, Abramson Family Cancer Research Institute, Basser Research Center for BRCA, Perelman School of Medicine, University of Pennsylvania; #Equal contribution

Lysine63-linked ubiquitin (K63-Ub) chains represent a particular ubiquitin topology that mediates proteasome-independent signaling events. The deubiquitinating enzyme (DUB) BRCC36 segregates into distinct nuclear and cytoplasmic complexes that are specific for K63-Ub hydrolysis. RAP80 targets the five-member nuclear BRCC36 complex to K63-Ub chains at DNA double-strand breaks. The alternative four-member BRCC36 containing complex (BRISC) lacks a known targeting moiety. Here we identify a serine hydroxymethyltransferase (SHMT) as a heretofore-unappreciated component that fulfills this function. SHMT directs BRISC activity at K63-Ub chains conjugated to the type 1 interferon (IFN) receptor chain 1 (IFNAR1). BRISC-SHMT2 complexes localize to and deubiquitinate actively engaged IFNAR1, thus limiting its K63-Ub mediated internalization and lysosomal degradation. BRISC deficient cells and mice exhibit attenuated responses to IFN and are protected from IFN-associated immunopathology. These studies reveal a novel mechanism of DUB regulation, and suggest a therapeutic use of BRISC inhibitors for treating pathophysiologic processes driven by elevated IFN responses.


back to the top


RAG1/2 drives lymphomagenesis by off-target recognition of cryptic recombination signal sequences
Martina Mijuskovic, Yi-Fan Chou, Susanna M. Lewis, Olga Shestova, and David B. Roth
Department of Pathology and Laboratory Medicine, and Abramson Family Cancer Research Institute, Perelman School of Medicine, University of Pennsylvania

It has long been speculated that RAG1/2 recombinase, a site-specific endonuclease essential for immunoglobulin (Ig) and T-cell receptor (TCR) gene assembly, is involved in the pathogenesis of lymphoid malignancies. RAG1/2 recognizes conserved recombination signal sequences (RSS) that flank V, D and J segments of Ig and TCR genes, creating double strand breaks. These breaks are repaired so that corresponding coding gene segments are put together as “coding joints.” The main proposed mechanisms of RAG1/2 contribution to genomic instability include 1) premature DNA end release leading to chromosome translocations that deregulate oncogene expression and 2) off-target recognition of sequences similar to RSS (cryptic RSS) at non Ig/TCR loci, leading to ectopic recombination events. To investigate the contribution of the above mechanisms to RAG1/2-induced genomic instability, we applied whole genome sequencing to examine the genomes of thymic lymphomas from p53-deficient mice that carry either a wild type or a C-terminally truncated “core” RAG2 subunit of RAG recombinase. Our analysis revealed a large number of structural rearrangements, many of them recurrent and in genes already identified as drivers in cancer. We show that the predominant mechanism for RAG1/2-induced genomic instability is cryptic RSS-driven “coding end” formation reminiscent of canonical V(D)J recombination and that translocations originating from V(D)J loci contribute little to genomic instability. Furthermore, we identify Notch1 and cytokine receptor signaling as the main mutated pathways. Detailed analysis of the cryptic RSS involved in these rearrangements refines our understanding of aberrant RAG1/2 activity and will help develop tools to predict off-target sites in both human and mouse genomes.


back to the top


Metabolic advantages of urea cycle misregulation in renal cell carcinoma
Joshua D. Ochocki1, Nan Lin2, Bo Qiu1, and M. Celeste Simon1,3,4
1Abramson Family Cancer Research Institute, 2Department of Cancer Biology, 3Department of Cell and Developmental Biology, and 4Howard Hughes Medical Institute, Perelman School of Medicine, University of Pennsylvania

Renal cell carcinoma is the most common type of kidney malignancy, and the main subtype, clear cell renal cell carcinoma (ccRCC), makes up 75% of these cancers. ccRCC is distinguished by both radio- and chemo-resistance, and treatment options for this disease are limited. The urea cycle is involved in the detoxification of free ammonia in mammals, and it takes place mainly in the kidney. Argininosuccinate synthase 1 (ASS1) and arginase II (ARG2) are two key enzymes in the urea cycle, the former being responsible for the synthesis of arginine, while the latter hydrolyzes arginine to ornithine in the last step of the cycle. We have shown that both normal kidney and ccRCC cell lines require exogenous arginine for growth, and this suggests that increased arginine availability contributes to the pathogenesis of ccRCC. In this study, we measured expression changes of ASS1 and ARG2 as well as other enzymes involved in arginine metabolism in ccRCC versus normal kidney tissue to determine the metabolic advantages of an altered urea cycle. The expression levels of ASS1 and ARG2 were decreased 4-fold and 7-fold, respectively, based on ccRCC cancer genome atlas data. Stable knockdown of either ASS1 or ARG2 in normal kidney cells conferred a growth advantage, indicating a potential mechanism by which ccRCC tumors sustain increased proliferation. Moreover, the arginine transporter CAT-1 is 10-fold over-expressed in ccRCC. Cumulatively, our data demonstrate that arginine metabolism is altered in ccRCC tumors and that this increased arginine availability likely enables sustained proliferation of the tumor tissue. This study highlights the applicability of targeting cancer metabolism of a specific amino acid, arginine, as a possible therapeutic intervention for ccRCC.


back to the top


Diagnostic application of high-resolution SNP arrays for children with brain tumors
Jacquelyn J. Roth1, Mariarita Santi2,3, Lucy B. Rorke-Adams2,3, Brian N. Harding2,3, Laura S. Tooke2, and Jaclyn A. Biegel1,2,3
1Department of Pediatrics and 2Department of Pathology and Laboratory Medicine, The Children’s Hospital of Philadelphia, 3Perelman School of Medicine, University of Pennsylvania

The clinical utility of high-resolution single nucleotide polymorphism (SNP) array analysis for pediatric hematologic malignancies and solid tumors has been previously demonstrated by our laboratory. Since the introduction of this test in 2008, over 100 pediatric brain tumor specimens have been analyzed using the Illumina Human610-Quad or the HumanOmni1-Quad Bead Chip arrays. The most frequent tumor type studied was low-grade glioma, followed by medulloblastoma (MB), and less frequently supratentorial primitive neuroectodermal tumors, high-grade gliomas, ependymomas and meningiomas. Copy number alterations (CNAs) were detected in 90% of the tumors analyzed and varied based on the classification of the brain tumor. The most frequently detected CNA was a duplication at 7q34, which was observed in half of the low grade gliomas and results in a fusion product between KIAA1549 and BRAF. The second most frequently identified CNA was an isochromosome 17q detected in one third of the MBs analyzed. In a variety of tumors, SNP array analysis revealed deletions and regions of homozygosity encompassing known cancer associated genes including TP53, RB1, CDKN2A/B, CHEK2, NF1, and NF2, which represent potential germline alterations. In conjunction with SNP array testing, sequence analysis revealed BRAF mutations in 5 gliomas and no TP53 mutations in MBs. In our experience, SNP array analysis of pediatric brain tumors can be used in conjunction with pathological examination and molecular analyses to further refine diagnoses, offer more accurate prognostic assessments, identify potential germline alterations, and highlight possible therapeutic options for these patients.


back to the top


Targeting metabolic pathways to inhibit histone acetylation in castrate resistant prostate cancer
Supriya Shah, Ellen Jackson, Sarah Tse, and Kathryn Wellen

Department of Cancer Biology, Perelman School of Medicine, University of Pennsylvania

Androgen deprivation therapy (ADT) targeting androgen receptor (AR) function is currently used as a first line therapy for metastatic prostate cancer. Although successful initially, this treatment fails eventually. Deviations in several mechanisms cause inappropriate activation of AR, thus rendering the hormone depletion therapy ineffective. Previously, we showed that inhibiting histone-acetylation counteracts inappropriate AR activation by inhibiting its recruitment on chromatin, thus increasing the efficacy of the hormone therapy. Incidentally, it is reported that Akt activates the metabolic enzyme ATP-citrate lyase (ACL), and these molecules could impinge on global histone acetylation levels by limiting the acetylation substrate, acetyl-coA. We hypothesized that histone acetylation might be a function of metabolic reprogramming mediated by Akt/ACL pathway activation in prostate cancer. Preliminary analysis of human prostate tumors revealed that, indeed, pAkt levels correlate significantly with global levels of several histone acetylation marks, suggesting that activation of the PI3K-Akt pathway is a key determinant of tumor histone acetylation levels. Moreover, inhibition of either Akt or ACL results in loss of histone acetylation in PCa cells. Targeting ACL or AR function alone showed only moderate effects on cell survival, thus indicating possibility of molecular cross-talk. In keeping with this, when a combination therapy targeting AR function and ACL inhibition was used concurrently, a synergistic inhibition of castrate resistant prostate cancer (CRPC) cancer cells was observed. In conclusion, the combination therapy discovered here impinges on crosstalk between AR and chromatin, and promotes synergistic suppression of CRPC tumors.


back to the top


Identification of a new class of potential oncolytic small molecule compounds targeting EBV-positive cancers
Nadia Tikhmyanova1, David C. Schultz1, Theresa Lee1, Joseph M. Salvino2, and Paul M. Lieberman1
1The Wistar Institute; 2Department of Pharmacology, Drexel University College of Medicine

Epstein-Barr Virus (EBV) persists as a latent infection in many lymphoid and epithelial malignancies, including Burkitt’s lymphomas, nasopharyngeal carcinomas, and gastric carcinomas. Current chemotherapeutic treatments of EBV-positive cancers include broad-spectrum cytotoxic drugs that ignore the EBV-positive status of tumors. An alternative strategy, referred to as oncolytic therapy, utilizes drugs that stimulate reactivation of latent EBV to enhance the selective killing of EBV positive tumors, especially in combination with existing inhibitors of herpesvirus lytic replication, like Ganciclovir (GCV). At present, no small molecule, including histone deacetylase (HDAC) inhibitors, has proven safe or effective in clinical trials for treatment of EBV positive cancers. To identify new chemical entities that induce the EBV lytic cycle, we have developed a robust high-throughput cell-based assay to screen 66,840 small molecule compounds. Five structurally related tetrahydrocarboline derivatives were identified, two of which had EC50 measurements in the range of 150-170 nM. We show that these compounds reactivate EBV lytic markers ZTA and EA-D in all EBV-positive cell lines that we have tested, independent of the type of latency. The compounds reactivate a higher percentage of latently infected cells than HDAC inhibitors or phorbol esters in many cell types. The most active compounds showed low toxicity to EBV-negative cells, but were highly effective at selective cell killing of EBV-positive tumor and in vivo immortalized cells when combined with GCV. We conclude that we have identified a class of small molecule compounds that are highly effective at reactivating latent EBV infection in a variety of cell types, and show promise for lytic therapy in combination with GCV.


back to the top


Mitochondria-targeted autophagy facilitates cancer stem cell generation via regulation of oxidative stress
Kelly Whelan1, Sanders Chang1, Shingo Kagawa1, Hideaki Kinugasa1, Mitsuteru Natsuizaka1, Quentin McAfee2, Ravi Amaravadi2, Anil Rustgi1, and Hiroshi Nakagawa1
1Division of Gastroenterology, and 2Department of Medicine, Perelman School of Medicine, University of Pennsylvania

Cancer stem cells (CSCs) are subsets of tumor cells possessing self-renewal and differentiation capabilities, thereby facilitating tumor initiation, metastasis and renewal. CSC populations have been identified in esophageal squamous cell carcinoma (ESCC); however, mechanisms regulating expansion and maintenance of these CD44 High (CD44H) subpopulations have yet to be elucidated. Autophagy is a catabolic process through which intracellular components are degraded in conjunction with lysosomal machinery. While the role of autophagy in cancer is controversial, it has been suggested to promote tumor cell survival in the tumor microenvironment. What role, if any, this process plays in CSC biology remains largely unknown. Here, we report that mitochondria-targeted autophagy (i.e. mitophagy) facilitates conversion of non-CSCs to CSCs by relieving oxidative stress. Using the ESCC-relevant cytokine TGF-β to drive non-CSC to CSC conversion, we report activation of mitophagy concomitant with decreased mitochondrial DNA content during generation of CD44H CSCs in vitro. Additionally, interference with autophagic or mitophagic pathways suppresses CSC expansion in vitro and in vivo and inhibits CSC-mediated tumorigenicity in serial xenotransplantation studies. Such suppression of autophagy/mitophagy further reveals accumulation of reactive oxygen species, suggesting a role for mitophagy in regulating CD44H cell dynamics via modulation of oxidative stress in the tumor microenvironment. Interestingly, analysis of human ESCC tissue samples suggests that activation of autophagy correlates with poor clinical outcome in ESCC. Taken together, these results identify mitophagy as a critical regulator of CSC biology and highlight the potential benefit of targeting autophagy in ESCC.


back to the top


Transcriptional repression of Gata3 is essential for early B cell commitment
Anupam Banerjee and David M. Allman
Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania

The mechanisms underlying the silencing of alternative fate potentials in very early B cell precursors remain unclear. Using gain- and loss-of-function approaches together with a synthetic Zinc-finger polypeptide (ZFP) engineered to prevent transcription factor binding to a defined cis element, we show that the transcription factor EBF1 promotes B-lineage commitment by directly repressing expression of the T-lineage requisite Gata3 gene. EBF1-deficient lymphoid progenitors exhibited increased T-lineage potential and elevated Gata3 transcript levels, whereas enforced EBF1 expression inhibited early T cell differentiation and caused rapid loss of Gata3 mRNA. Notably, ZFP-mediated perturbation of EBF1 binding to a Gata3 regulatory region restored Gata3 expression, abrogated EBF1-driven suppression of T cell differentiation, and prevented B cell differentiation. Furthermore, EBF1 binding to this region induced repressive histone modifications across this region. These data identify a novel transcriptional circuit critical for preventing T cell differentiation and adopting the B cell fate.


back to the top


Enhanced T cell function in a mouse model of human glycosylation
George Buchlis1,2, Pamela Odorizzi1, Paula C. Soto3, Oliver M.T. Pearce3, Daniel J. Hui2, Martha S. Jordan1, Ajit Varki3, E. John Wherry1, and Katherine A. High1,2,4
1Perelman School of Medicine, University of Pennsylvania; 2Department of Pediatrics-Hematology, The Children’s Hospital of Philadelphia;
3Departments of Medicine and Cellular and Molecular Medicine, University of California at San Diego, La Jolla, CA; 4Howard Hughes Medical Institute

Animal models of rAAV-mediated gene transfer have not been predictive of the human response to the vector capsid. Interestingly, human T-cells differ from even their closest primate relatives due to a mutation of the sialic acid (Sia) converting enzyme CMAH, which causes the loss of expression of the Sia Neu5Gc. This change also alters the expression and binding of inhibitory receptors (Siglecs) that bind these Sias. Compared to primate T cells, human T cells are hyperactive and responses could be lowered upon Siglec expression. Here we report that Cmah-/- mouse T cells proliferate faster and have greater expression of activation markers than wild-type mouse T cells. Metabolically re-introducing Neu5Gc diminishes the proliferation and activation of both human and murine Cmah-/- T cells. Importantly, Cmah-/- mice mount greater T cell responses to an Adenovirus encoding an Adeno-associated Virus capsid transgene (Ad-AAV). When infected with an acute form of Lymphocytic Choriomeningitis Virus (LCMV), Cmah-/- mice made higher frequencies of total CD8 T cells as well as more tetramer positive cells than C57BL/6 mice. At day 42, there were more tetramer positive cells in the Cmah-/- mice transitioning to a memory T cell phenotype, expressing low levels of KLRG1 and high levels of CD127. Upon LCMV peptide stimulation of splenocytes in vitro for 5 hours, Cmah-/- mice had more CD107a+IFNg+ CD8+ T cells compared to C57BL/6 mice. In addition, these CD107a+IFNg+ CD8+ T Cells were more polyfunctional in Cmah-/- mice, with higher frequencies of these cells secreting both Mip1a and IL-2, both Mip1a and TNFα, and both TNFα and IL-2. Mice that more accurately model the human immune system may serve as a better pre-clinical animal model for such studies.


back to the top


Functional and phenotypic properties of mast cells that develop in humanized c-kit-NSG mice
Kshitij Gupta, Irene Rahman, Hariharan Subramanian, and Hydar Ali
Department of Pathology, School of Dental Medicine, University of Pennsylvania

Mast cells are important players in host defense but also promote allergic diseases. Rodent models employed to study allergic diseases do not reflect all aspects of human diseases, possibly due to functional and phenotypic differences between human and murine mast cells. Tanaka et al. showed that transplantation of human CD34+/CD38- cord blood hematopoietic stem cells into NSG mice led to development of human myeloid cells and mast cells after 4-6 months. Moreover, transgenic expression of the membrane-bound human stem cell factor in NSG mice (c-kit-NSG) promotes accelerated human hematopoietic engraftment and myeloid differentiation relative to non-transgenic NSG mice. These mice develop tryptase-positive human mast cells. However, the functional properties of these mast cells have not been determined. The purpose of this study was to utilize humanized c-kit-NSG mice to determine the functional and phenotypic properties of human mast cells that develop in humanized mice. C-kit-NSG mice were preconditioned with busulfan or sublethally irradiated and transplanted with human fetal liver CD34+ cells. After 4-6 months, we found that both busulfan-treated and irradiated animals produced >85% human CD45+ cells in peripheral blood with similar levels in bone marrow and spleen. A subpopulation of c-kit+ cells were observed in bone marrow and spleen of c-kit-NSG mice, but these cells were FceRI negative. IHC revealed tryptase and chymase staining in spleen, lungs and skin. Surprisingly, these mast cells did not respond to GPCR ligands or human IgE, but a Ca2+ ionophore induced robust degranulation. These findings indicate that c-kit-NSG mice develop mast cells containing tryptase and chymase (MCTC) and have the capacity to undergo degranulation. However, they do not express FcεRI and are resistant to activation via receptor-mediated pathways.


back to the top


The dynamic microbiomes associated with acute and healing open fractures
Brendan P. Hodkinson1, Geoffrey D. Hannigan1, Kelly McGinnis2, Amanda S. Tyldsley1, Jason B. Anari2, Annamarie D. Horan2, Elizabeth A. Grice1, and Samir Mehta1
1Department of Dermatology, and 2Department of Orthopedic Surgery, Perelman School of Medicine, University of Pennsylvania

Traumatic open fracture wounds represent a significant public health and economic burden, and characterizing the microbes associated with these wounds could aid significantly in their management and treatment. Most previous studies have relied on culture-based identification, which is subject to bias and the results may be suboptimal. For the present study, we used culture-independent, high throughput sequencing of the bacterial 16S ribosomal RNA gene to longitudinally characterize the bacterial microbiota colonizing the open fractures of 30 patients and to find possible correlations between aspects of those communities and infectious complications, mode of injury, severity, and wound site. Results indicate that the microbiota colonizing open fracture wounds became increasingly similar to adjacent skin microbiota as wounds healed with regard to alpha (within sample) diversity, beta (between samples or shared) diversity, and the types of bacteria present. Alpha diversity of open fracture microbiota was significantly lower than that of the adjacent skin only upon presentation to the ER, and overall community dissimilarity (measured using beta diversity metrics) between paired wound center and adjacent skin samples decreased overall as healing progressed. Regarding the association between the microbiome and specific clinical factors, upper extremity wounds were lower in bacterial alpha diversity compared to those of the lower extremities, and a handful of taxa were found to be associated with mode of injury, severity, and wound location. Validation of these results in larger cohorts is warranted, and could lead to the identification of microbiome-based biomarkers that aid in management and treatment of open fractures.


back to the top


Specific LPS structures of non-typeable Haemophilus influenzae prevent IgM binding and phagocytosis by neutrophils
Jeroen D. Langereis1, Marien I. de Jonge2, and Jeffrey N. Weiser1
1Department of Microbiology, Perelman School of Medicine, University of Pennsylvania; 2Department of Pediatrics, Laboratory of Pediatric Infectious Diseases, Radboud University Medical Centre, Nijmegen, The Netherlands

Chronic obstructive pulmonary disease (COPD) incidence is increasing worldwide and causes huge healthcare burden with almost $24 billion spent every year in the USA alone. Non-typeable Haemophilus influenzae (NTHi) is a major pathogen causing exacerbations of COPD and there is a clear relationship between NTHi colonization of the lungs and disease severity. Usually, NTHi colonizes the nasal mucosa without obvious signs of inflammation, but the presence of NTHi in the lungs attracts immune cells, including large quantities of neutrophils, that cause acute inflammation. Although neutrophils are equipped with various mechanisms to kill bacteria, NTHi is not cleared effectively from the lungs of COPD patients. The objective of this study is to identify NTHi genes essential for growth and survival in the presence of neutrophils. A transposon mutant library was generated in NTHi and exposed to human neutrophils for 2h in the presence of human serum. Genes essential for survival in the presence of neutrophils were identified by next generation sequencing Tn-seq technology and their role was subsequently verified by in vitro neutrophil killing experiments. NTHi circumvents neutrophil-mediated killing by preventing opsonization by IgM. It does so by the incorporation of galactose residues into the lipooligosaccharide (LOS) structure, which is largely dependent on NTHi genes galE and lic2A. Directed gene deletion mutants showed an altered LOS structure, a dramatic increase in IgM binding and increased phagocytosis and killing by human neutrophils. Incorporation of galactose into the LOS of NTHi prevented IgM binding and subsequent phagocytosis by human neutrophils. Altering NTHi LOS biosynthesis might be an attractive and effective approach for novel antimicrobial drugs.


back to the top


A regulator of fatty acid metabolism, acetyl CoA carboxylase 1, controls T cell activation and survival

JangEun Lee1, Mattew Walsh1, Kyle L. Hoehn2, David E. James3, E. John Wherry4, and Yongwon Choi1
1Department of Pathology and Laboratory Medicine, and 4Institute for Immunology, Perelman School of Medicine, University of Pennsylvania;
Department of Pharmacology, University of Virginia Health System, Charlottesville, VA; 3Garvan Institute of Medical Research, Darlinghurst, NSW, Australia

Fatty acids (FAs) are essential constituents of cell membranes, signaling molecules, and substrates for energy metabolism. As lymphocytes undergo both dramatic functional and metabolic changes during activation and differentiation, it is likely that dynamic changes in FA metabolism also occur. However, the contribution of de novo lipogenesis on acquiring or maintaining T cell dynamics has not been well addressed. Here we demonstrated the essential role of FA synthesis on survival of lymphocytes under both homeostatic and inflammatory conditions. T cell-specific deletion of acetyl CoA carboxlyase 1 (ACC1), an enzyme that catalyzes production of malonyl CoA, a carbon donor for long chain FA synthesis, impaired T cell persistence in the periphery, and homeostatic proliferation in naïve mice. Loss of ACC1 did not compromise effector CD8+ T cell differentiation, as evidenced by expression of effector molecules such as IFN-γ; and granzyme B upon listeria infection, but did result in a severe defect in Ag-specific CD8+ T cell accumulation owing to defective survival of proliferating cells. Furthermore, in vitro mitogenic stimulation demonstrated that defective cell growth and survival caused by ACC1 deletion were rescued by exogenuous FA supplement. These results suggest the essential role of ACC1 as a regulator for optimal blastogenesis and survival of CD8+ T cells in controlling de novo lipogenesis.


back to the top


Transcriptional control of PD1 locus by the T-box transcription factors T-bet and Eomes in human CD8 T-cells

Laura M. McLane and Michael R. Betts
Department of Microbiology, Perelman School of Medicine, University of Pennsylvania

During chronic infection, constitutive antigen presentation results in gradual T-cell exhaustion characterized by a loss of proliferative capacity and T-cell functionality. Additionally, inhibitory receptor expression, including the expression of programmed cell death 1 (PD1), increases on exhausted T-cells. Recent genomic studies have revealed that exhausted T-cells have a unique transcriptional profile, suggesting a key step in understanding T-cell exhaustion is the characterization of the transcriptional regulators of inhibitory receptors. The T-box transcription factors Eomes and T-bet have been directly implicated in promoting and repressing CD8 T-cell exhaustion, respectively. In a preliminary ChIP seq screen, we found that both T-bet and Eomes bind to same region of the PD1 promoter, suggesting both factors regulate PD1 expression. Based on this finding we investigated the influence of these transcription factors on PD1 expression in human CD8 T-cells. Consistent with published murine data showing T-bet directly represses PD1, we find that as T-bet levels decrease, nuclear T-bet also declines and correlates with an increase in PD1 expression. In T-betlo/- cells, PD1 expression is high only if Eomes is present. We also show that stimulation of the calcium pathway, a positive regulator of PD1, results in increased Eomes, but not T-bet, expression. Taken together, our data suggest that Eomes plays a role in upregulating PD1 and/or promoting an exhausted phenotype through displacement or out-competition of T-bet at the PD1 promoter. Further studies of T-bet and Eomes at the PD1 promoter and in the context of chronic infection should provide novel insights into the role these factors play during T-cell exhaustion as well as other critical functions of CD8 T-cells.


back to the top


Starve a fever: Investigating the effects of diurnal changes in autophagy on the inflammatory response
Sarah C. McLoughlin and Garret A. FitzGerald

Institute for Translational Medicine and Therapeutics, Perelman School of Medicine, University of Pennsylvania

Autophagy is a cellular process whereby the cell digests its own components as a means of removing defective organelles, providing a source of amino acids or lipids in times of need, controlling inflammation, or defending against invading pathogens. Reduced autophagy is implicated in a variety of conditions ranging from neurodegenerative diseases to cancer. In vivo, autophagy is increased during the sleep phase and reduced during the active phase. The control of such diurnality is not fully understood and the relevance of a day/night pattern is unknown. Disrupted circadian rhythms affect a variety of cellular processes including metabolism and inflammation. Clock-disrupted mice show reduced expression of autophagy genes. Here, we investigate the effects of disrupted circadian rhythms on basal autophagy function and induced autophagic response in clock-deficient macrophages with the aim of understanding the mechanism by which circadian rhythm disruption leads to abnormal inflammatory responses.


back to the top


SEC61A and VCP regulate alphavirus infection and iron transport by regulating the trafficking of entry receptor NRAMP2
Debasis Panda1, Patrick P. Rose1, Sheri L. Hanna1, Kaycie C. Hopkins1, Beth Gold1, and Sara Cherry1,2
1Department of Microbiology, Perelman School of Medicine, and 2PENN Genome Frontiers Institute, University of Pennsylvania

Alphaviruses are a large class of insect-borne human pathogens that exhibit a broad host range in nature. These viruses infect more than one million people annually, with increasing morbitity and mortality, in part due to lack of vaccines or antiviral agents. As arthropod-borne viruses, the life-cycle of most alphaviruses is dependent on a mosquito vector and a vertebrate reservoir with humans as incidental hosts. How these alphaviruses, including the prototypic alphavirus Sindbis virus (SINV), exploit the host proteins to replicate is unknown. To identify such cellular factors, we performed a genome-wide RNAi screen using the model insect Drosophila and validated 96 genes that impacted infection of SINV. Previously, we found that the Natural Resistance-Associated Macrophage Protein (NRAMP2, Divalent Metal Transporter (DMT1)) acts as an entry receptor for SINV in insects and vertebrates. Furthermore, we identified a conserved role for SEC61A, an ER transport protein, and VCP, a vesicle transport protein, in facilitating SINV entry in insects and mammals. Detailed mechanistic investigations revealed that SEC61A and VCP selectively regulate the trafficking of the SINV entry receptor NRAMP2. Depletion of these proteins by siRNA or pharmacological inhibition lead to altered trafficking of NRAMP2 to lysosomal compartments and proteolytic degradation of NRAMP2 within lysosomes. Because NRAMP2 is also the major iron transporter in cells, this decrease in NRAMP2 levels attenuated iron transport. Taken together, this study revealed new genes and pathways involved in both infection and iron homeostasis that may serve as targets for antiviral therapeutics or for iron imbalance disorders.


back to the top


Host factors affecting transmission of Streptococcus pneumoniae
Aimee L. Richard and Jeffrey N. Weiser
Department of Microbiology, Perelman School of Medicine, University of Pennsylvania

Streptococcus pneumoniae robustly colonizes the nasopharynx and is a major cause of bacterial pneumonia. Nasal carriage of this microbe is common in humans, especially children, and is a prerequisite for the development of disease and transmission among hosts. While the importance of transmission of pathogens is widely accepted, there is currently little mechanistic understanding of this process. We have adapted an infant mouse model to find host and bacterial determinants of transmission of S. pneumoniae from inoculated index mice to uninfected contact mice. We observed that in the context of co-infection with influenza A virus, the pneumococcus is transmitted among wild-type littermates (~50% of contact mice acquire colonization). Flow cytometric analysis of nasal lavage fluid from mice infected with influenza alone and in combination with S. pneumoniae revealed a massive influx of neutrophils into the nasopharynx; these cells were present at much lower levels in mice that do not transmit the bacterium. Additionally, mice deficient for Tlr2 are colonized to a similar density but transmit S. pneumoniae more efficiently (100% transmission) than wild-type animals and display a greater proportion of neutrophils in the lavage fluid (91% vs. 75% of total cells). This difference depends on the genotype of the index mouse: Tlr2-/- mice transmit the pneumococcus to wild-type contacts at a higher frequency than in the converse scenario. We hypothesize that an influx of neutrophils into the nasopharynx of infected animals alters the dynamics of bacterial shedding. Current studies focus on defining the mechanism by which this heightened innate immune response precipitates transmission, which will further our understanding of how bacterial pathogens are spread from host to host.


back to the top


Antibody blocks acquisition of pneumococcal colonization through agglutination
Aoife M. Roche1, Aimee L. Richard1, Jeremy Rahkola2, Edward Janoff2, and Jeffrey N. Weiser1
1Department of Microbiology, Perelman School of Medicine, University of Pennsylvania; 2Division of Infectious Disease, University of Colorado, Denver, CO

The major benefit of conjugate polysaccharide vaccines has been attributed to herd immunity, resulting from their ability to decrease transmission by blocking the acquisition of colonization. This mucosal protection correlates with increased serum anti-capsular antibody (Ab) titers; however, the mechanism for this protection against colonization is unknown. A murine model was used to study the dynamics of this protection against Streptococcus pneumoniae. Mice passively immunized with rabbit anti-capsular Ab were protected from acquisition of nasal colonization. This protection was serotype-specific and correlated with the agglutinating effect of anti-sera. IgG was detected on the mucosa at the time of nasal challenge and was sufficient to block acquisition. The effectiveness of IgG was Fc- and complement-independent, suggesting an effect of Ab not involving opsono-phagocytosis. Immunization with F(ab’)2 fragments, which are bivalent and retain agglutinating activity, were more protective than Fab fragments, which are monovalent and no longer agglutinate their target. Additional evidence for the importance of agglutination in mucosal protection was provided by the use of human (Hu) anti-capsular mAbs. IgA1 is the major immunoglobulin on the nasal mucosa, but it is cleaved in its hinge region by a pneumococcal protease, eliminating its ability to agglutinate. Serotype-specific HuIgA1 mAb blocked colonization of the IgA1-protease mutant (agglutinated), but not the wild-type parent (not agglutinated). Finally, agglutinating anti-capsular Abs blocked transmission between littermates in an infant model using influenza co-infection. Our findings highlight the critical role of the agglutinating effect of Ab in mucosal protection.


back to the top


G protein coupled receptor kinase 2 (GRK2) is essential for FcεRI signaling in mast cells
Hariharan Subramanian, Kshitij Gupta, and Hydar Ali
Department of Pathology, School of Dental Medicine, University of Pennsylvania

Allergic diseases such as asthma are caused by an overzealous immune response to allergens in which immunoglobulin E (IgE) and mast cells play critical roles. Thus, aggregation of the high affinity IgE receptor (FcεRI) on mast cells by allergen results in rapid release of mast cell granules (degranulation) and generation of inflammatory cytokines that are responsible for the manifestations of the disease. Mast cells also express a G-protein coupled receptor (GPCR, C3aR) for the complement component C3a. Agonist-induced GPCR phosphorylation by G protein coupled receptor kinase 2 (GRK2) results in receptor desensitization. Accordingly, we have recently shown that GRK2 desensitizes C3aR and inhibits mast cell responses. However, the role of GRK2 in modulating FceRI signaling in mast cells was unknown. To examine this we used shRNA knockdown or GRK2flox/flox/Cre-recombinase techniques to silence GRK2 in mast cells. GRK2 knock down resulted in a significant decrease in human mast cell degranulation to IgE/antigen. Similar to human cells, we found that GRK2 promotes degranulation in mouse mast cells. Importantly, we demonstrate that GRK2 is essential for cytokine (IL-6 and IL-13) production following mouse mast cell activation. Mechanistically, our data reveal that the catalytic activity of GRK2 is not required for cytokine production and that GRK2 mediates activation of signaling molecules such as p38, AKT and NFkB. Additionally, GRK2-induced regulation of cytokines occurs at the transcriptional level and does not involve mRNA stability. Thus we have identified GRK2 as a novel regulator of the early (degranulation) and the delayed (cytokine production) phases of mast cell activation and propose that this study may have a direct impact on mast cell-specific allergic diseases.


back to the top


TSLP-elicited basophil responses mediate the pathogenesis of eosinophilic esophagitis
Elia D. Tait Wojno1,2#, Mario Noti1,2#, Brian S. Kim1-3, Mark C. Siracusa1,2, Paul R. Giacomin1,2,4, Meera G. Nair1,2,5, Alain J. Benitez6, Kathryn R. Ruymann7, Amanda B. Muir6, David A. Hill1,2,7, Kudakwashe R. Chikwava8, Amin E. Moghaddam9, Quentin J. Sattentau9, Aneesh Alex10-12, Chao Zhou10-12, Jennifer H. Yearley13, Paul Menard-Katcher14, Masato Kubo15,16, Kazushige Obata-Ninomiya17,18, Hajime Karasuyama17,18, Michael R. Comeau19, Terri Brown-Whitehorn7, Rene de Waal Malefyt20, Patrick M. Sleiman21-23, Hakon Hakonarson21-23, Antonella
Cianferoni7, Gary W. Falk14,24,25, Mei-Lun Wang6,24,25, Jonathan M. Spergel2,7,24,25, and David Artis1,2,24-26
1Department of Microbiology, 2Institute for Immunology, 3Department of Dermatology, 14Division of Gastroenterology, 23Department of Pediatrics, and 25Center for Molecular Studies in Digestive and Liver Diseases, Department of Medicine, Division of Gastroenterology, Perelman School of Medicine, University of Pennsylvania; 4Centre for Biodiscovery and Molecular Development of Therapeutics, James Cook University, Cairns, Queensland, Australia; 5Division of Biomedical Sciences, School of Medicine, University of California-Riverside, Riverside, California; 6Division of Gastroenterology, Hepatology, and Nutrition, 7Department of Pediatrics, Division of Allergy and Immunology,
Department of Pathology and Laboratory Medicine, 21Center for Applied Genomics, and 22Division of Human Genetics, Abramson Research Center, The Children’s Hospital of Philadelphia; 9The Sir William Dunn School of Pathology, The University of Oxford, Oxford, UK;
Department of Electrical and Computer Engineering, 11Center for Photonics and Nanoelectronics, and 12Bioengineering Program, Lehigh University, Bethlehem, Pennsylvania; 13Department of Pathology, Merck Research Laboratories, Palo Alto, California; 15Laboratory for Cytokine Regulation, Research Center for Integrative Medical Science, RIKEN Yokohama Institute, Kanagawa, Japan; 16Division of Molecular Pathology, Research Institute for Biomedical Science, Tokyo University of Science, Chiba, Japan; 17Department of Immune Regulation, Tokyo Medical and Dental University Graduate School, Tokyo, Japan; 18JST, CREST, Tokyo Medical and Dental University Graduate School, Tokyo, Japan;
Inflammation Research, Amgen, Seattle, Washington; 20Therapeutic Area Biology and Pharmacology, Merck Research Laboratories, Palo Alto, California; 24Joint Penn-Children’s Hospital of Philadelphia Center for Digestive, Liver and Pancreatic Medicine, Perelman School of Medicine, University of Pennsylvania and The Children’s Hospital of Philadelphia; 26Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania; #Equal contribution

Eosinophilic esophagitis (EoE) is a food allergy-associated disease characterized by esophageal eosinophilia and associated inflammation. EoE has become increasingly common in industrialized countries. However, current management strategies, such as treatment with swallowed steroids and dietary restrictions, are nonspecific or negatively impact the quality of life of EoE patients. Thus, there is an urgent need to identify novel, specific immunological pathways that underlie the pathogenesis of EoE that could be targeted to treat this disease. Recently, a genome-wide association study identified that EoE is associated with a gain-of-function polymorphism in the gene that encodes thymic stromal lymphopoietin (TSLP). However, whether TSLP directly promotes allergic inflammation in the esophagus and the mechanisms by which TSLP might contribute to the pathogenesis of EoE remain unknown. Here, we describe a new murine model of experimental EoE-like disease that was used to investigate the role of TSLP in allergic inflammation of the esophagus. Murine experimental EoE-like disease developed independently of IgE but was dependent on TSLP-elicited basophils. Critically, therapeutic antibody-mediated neutralization of TSLP or depletion of basophil populations ameliorated established EoE-like disease in mice. Finally, elevated TSLP levels and exaggerated basophil responses were observed in esophageal biopsies from EoE patients, and a gain-of-function polymorphism in TSLP correlated with increased basophil responses in patients with EoE. Together, these data indicate that TSLP-elicited basophil responses may play a key role in mediating the pathogenesis of EoE, suggesting that targeting the TSLP-basophil axis could represent a new therapeutic target in the treatment of EoE.


back to the top


Protective immunity against invasive Streptococcus pneumoniae infections by tissue-specific Th17 cells
Yan Wang, Bin Jiang, Jeffery N. Weiser, and Hao Shen
Department of Microbiology, Perelman School of Medicine, University of Pennsylvania

Pneumonia caused by Streptococcus pneumoniae (Sp) remains a leading cause of serious illness and death in children and elderly worldwide. Current vaccines are highly effective in preventing colonization by inducing serotype-specific antibodies. However, there is an increasing prevalence of infection by serotype strains not included in the vaccine; this highlights the need for a universal vaccine that protects against all serotypes. The first step in developing a universal vaccine is to elucidate the immune mechanism that can provide broad protection against different serotype Sp strains. We found that intranasal immunization of mice with Sp resulted in a strong CD4+T cell response in the lung that consisted of mostly Th17 cells but also IFNγ producing Th1 cells. These immunized mice were protected against lethal challenge with a different serotype strain of Sp and cleared bacteria from the lung by day 3 post-challenge. Adoptive transfer of T cells from immunized mice also provided protection against a heterologous challenge, with a dominant Th17 recall response in the lung from donor memory T cells. Furthermore, immunization of mice with Listeria monocytogenes (LM) provided protection against a recombinant Sp strain expressing a CD4 epitope from LM (LLO190), as shown by decreased CFU in the lung, and elevated LLO190-specific Th17 response. Our results suggest that memory Th17 cells may play a key role in providing broad protective immunity against invasive Sp infection in a serotype-independent manner.


back to the top


Buprenorphine is an effective antidepressant in a genetic model of depression, the Wistar-Kyoto rat
Caroline A. Browne and Irwin Lucki

Behavioral Psychopharmacology, Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania

Major depressive disorder (MDD) is a debilitating psychiatric disorder, with a lifetime prevalence of 17% in the United States. Despite the wide range of antidepressants available to treat MDD patients, approximately 50% of individuals are resistant to conventional antidepressants. Therefore, there is a pressing medical need to develop novel antidepressants. Recent clinical studies support the use of ketamine, an NMDA antagonist, as a rapid-acting antidepressant. However, the psychotomimetic and addictive properties of ketamine limit its potential as the gold standard antidepressant. The activation of kappa opioid receptors (κ-OR) is implicated in mediating an aversive response to stressful stimuli and the subsequent development of depression. Therefore, this study sought to characterize the potential antidepressant-like effects of the κ-OR antagonist buprenorphine (BPN) in Wistar-Kyoto (WKY) rats, a genetic rodent model of depression. One-way ANOVA indicated a significant effect of BPN treatment on the immobility of WKY rats in the forced swim test at 24 h (F3, 23 = 7.728, p = 0.0013), 48 h (F3, 23 = 30.79, p < 0.0001), 72 h (F3, 23 = 22.94, p < 0.0001), and 96 h (F3, 23 = 7.931, p = 0.0011). BPN produced protracted effects that lasted up to 96 h for the 5 mg/kg group. The decrease in immobility indicates an antidepressant-like effect of BPN. Although two-way ANOVA did not reveal a time*treatment interaction, there was a significant effect of time (F3, 80 = 7.851, p = 0.0001) and treatment (F3, 80 = 4.568, p = 0.0053) on the latency-to-emerge in the emergence test, indicating an anxiolytic effect of BPN. In conclusion, the rapid acting and protracted antidepressant-like behavioral effects of a single dose of BPN substantiates κ-OR modulation in the treatment of depression.


back to the top


Sleep deprivation impairs protein synthesis initiation due to attenuated mTORC1 signaling
Jennifer H. K. Choi1, Emily J. Davis1, Lucia Peixoto1, Mathieu Wimmer1, Pepe J. Hernandez1, Robbert Havekes1, Philippe Pierre2, and Ted Abel1
1Department of Biology, School of Arts & Sciences, University of Pennsylvania; 2Centre d’Immunologie de Marseille-Luminy, Institut National de la Santé et de la Recherche Médicale (INSERM) U631, Marseille, France

Sleep deprivation produces deficits in hippocampal synaptic plasticity and hippocampus-dependent memory storage. However, the molecular and cellular mechanisms that underlie these effects remain unclear. Several studies have suggested that signaling pathways associated with translation are altered during sleep and after periods of sleep deprivation. In addition, although sleep deprivation greatly increased mRNA levels of Arc/Arg3.1 and Hspa5/BiP, we found no change in corresponding protein levels, suggesting sleep deprivation reduces translation. Using microarray analyses, we show that a prominent effect of sleep loss is the down-regulation of genes associated with translation in the hippocampus. In addition, we observe that five hours of sleep deprivation is sufficient to reduce translation in the hippocampus in vivo when measured using SUnSET. Consistent with these results, brief sleep deprivation specifically reduces mammalian target of rapamycin complex 1 (mTORC1) in the hippocampus. Furthermore, hippocampal levels of the translation repressor phosphorylated-4EBP2, a target of mTORC1 signaling, are also reduced after brief sleep deprivation. Both mTORC1 and phosphorylated-4EBP2 levels returned to baseline non-sleep deprivation levels after 2.5 hours of recovery sleep post-five hours of sleep deprivation. Together our findings suggest that the detrimental effects of sleep deprivation are mediated by reduced translation initiation via down-regulation of mTORC1 signaling. Because translation and mTORC1 activation are required for long-term memory formation, our study sheds light on the underlying molecular mechanisms of the memory impairment induced by sleep deprivation.


back to the top


Are the brains of the blind the same or different? A multimodal MR imaging study
Ritobrato Datta1, Noah Benson1,2, Alexandra Daina1, Omar H. Butt1, and Geoffrey K. Aguirre1
1Department of Neurology, Perelman School of Medicine, and 2Department of Psychology, School of Arts & Sciences, University of Pennsylvania

Blindness alters brain structure and function, but the relationships between these specific changes across blinds have not yet been examined. In this study, we investigated whether there are single or multiple modes of brain alterations in a clinically heterogeneous population of blinds. To address this question, we collected multimodal MRI data from 68 subjects (34 blinds and 34 sighted). The blind subjects varied in their age of onset, age of severity of blindness and degree of visual deprivation. Gray matter thickness, surface area, mean cerebral blood flow and cross-modal activation to auditory sentences (% BOLD), were measured within the primary visual cortex. Volume and fractional anisotropy of optic chiasm, optic radiation, pericalcarine white matter and splenium were obtained. Lateral geniculate nucleus size was estimated. We replicated prior claims of differences between the sighted and blinds in almost every measure and found correlations across various measures in blinds but not in the sighted. A classifier, trained on the extracted measures, classified 66 out of the 68 subjects accurately as blind or sighted. We then calculated a ‘brain alteration score’ for each subject by integrating the measures based on their contribution weights to overall classification accuracy. Our results show significant differences brain alteration scores in the early blinds, late blinds and sighted subjects. In conclusion, our results provide direct evidence that early blind, late blind and sighted subjects vary in their degree of brain alteration, though there is overall a single mode of brain alteration across these groups. Our method integrating structural and functional alterations can be used as an imaging-based biomarker for early detection of neurological diseases prior to onset of clinical symptoms.


back to the top


The circadian clock controlled brain fatty-acid binding protein FABP7 regulates sleep quality in mammals
Jason R. Gerstner1, Isaac J. Perron1,2, Georgios Paschos3, Alison Kochersberger1, Heather Ballance2, Michael Grandner1, Jennifer Jager4, Takeo Yoshikawa6, Hiroshi Kadotani7, Yuji Owada8, Mitchell A. Lazar4, John Hogenesch1,3,5, Garret FitzGerald3,5, and Allan I. Pack1
1Center for Sleep and Circadian Neurobiology, 2Neuroscience Graduate Group, 3Institute for Translational Medicine and Therapeutics,
Institute for Diabetes, Obesity and Metabolism, and 5Department of Pharmacology, Perelman School of Medicine, University of Pennsylvania; 6RIKEN Brain Science Institute, Wako, Saitama-ken, Japan; 7Department of Psychiatry, Shiga University of Medical Science, Otsu City, Japan;
8Department of Organ Anatomy, Yamaguchi University Graduate School of Medicine, Ube Yamaguchi, Japan

The astrocyte brain-type fatty acid binding protein (Fabp7) gene expression cycles globally throughout mammalian brain based on time-of-day, however, mechanisms that contribute to this synchronized diurnal pattern of expression are not known. Further, the influence of Fabp7 expression on circadian locomotor rhythms and sleep/wake states in mammals has not been reported. In this study, we examined whether disrupting the clock or behavioral state can affect Fabp7 levels, and whether Fabp7 is necessary for normal circadian rhythms and sleep/wake behavior. We observed that the circadian oscillation in Fabp7 mRNA levels is absent in BMAL1 knockout (KO) mice, while overall Fabp7 mRNA levels are significantly higher. This pattern is mimicked in the Rev-erbα KO mouse, suggesting that Fabp7 transcription is indirectly regulated by BMAL1 through Rev-erbα. A timecourse of sleep deprivation showed reduced Fabp7 mRNA levels in mouse cortex at 9 hours, with a shift in the poly(A) tail at 3 hours, implicating differential effects of sleep loss on Fabp7 transcriptional and post-transcriptional processing. Fabp7 KO mice showed a reduction in diurnal locomotor running wheel activity, with an increase in circadian period length under free running conditions. Fabp7 KO mice also have decreased average sleep bout duration and an increase in the number of sleep bouts. Humans that carry a naturally occurring Fabp7 Thr61Met missense mutation showed fragmented sleep compared to non-carriers, suggesting Fabp7 regulates sleep consolidation. These results provide novel evidence for a molecular pathway linking lipid-signaling cascades in astrocytes with known clock mechanisms and sleep behavior in mammals.


back to the top


Serotonin transporter length polymorphism (5-HTTLPR) and mood symptoms in the context of hormonal fluctuation
Liisa Hantsoo1, Daniel Birmingham2, David Beversdorf3, William Malarkey2, C. Neill Epperson1, and Janice Kiecolt-Glaser2
1The Penn Center for Women’s Behavioral Wellness, and The Penn Center for the Study of Sex and Gender in Behavioral Health, Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania; 2Institute for Behavioral Medicine Research, Ohio State University Medical Center, Columbus, OH; 3Department of Neurology, University of Missouri, Columbia, MO

Research suggests that the short (s) allele of the serotonin transporter gene (5-HTTLPR) confers vulnerability for depression following adverse childhood events (ACEs). Ovarian hormones interact with 5-HTT and may increase likelihood of depression at points in the female lifespan including the premenstrual, perinatal, and menopausal periods. Few studies have examined this gene x environment interaction (GxE) in the context of hormonal fluctuation. Women (n=219) were genotyped as 5-HTTLPR s/s, s/l, or l/l; l carriers were combined for analysis. History of emotional changes premenstrually, perinatally or perimenopausally was determined by the Diagnostic Interview for Genetic Studies (DIGS). Logistic regression evaluated a model containing genotype, childhood emotional abuse (Childhood Trauma Questionnaire (CTQ)), and their interaction for emotional problems at each reproductive timepoint. Premenstrual, perinatal and menopausal mood changes were reported by 44%, 16%, and 12%. Perimenopausal emotional changes significantly correlated with premenstrual emotional changes (r=0.16, p=0.04). The ss genotype predicted premenstrual emotional changes (p=0.02) and emotional abuse predicted perinatal emotional changes (p=0.04). There was a trend toward GxE for perimenopausal emotional changes (p=0.06), which was significant when premenstrual emotional changes were included in the model (p=0.05); women with emotional abuse and ss genotype were more likely to report perimenopausal emotional changes than other women. Preliminary data suggest that 5-HTTLPR genotype contributes to mood disturbances at reproductive timepoints across the female lifespan. This relationship and the interaction with childhood adversity should be further explored.


back to the top


Amylin interacts with leptin in the ventral tegmental area to control food intake
Elizabeth G. Mietlicki-Baase, Brianne A. Jeffrey, Diana R. Olivos, and Matthew R. Hayes
Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania

Amylin is a neuropeptide produced by pancreatic b-cells that reduces food intake (FI) following central nervous system (CNS) activation. Systemic coadministration of amylin and the adipose-tissue derived hormone leptin suppresses FI beyond that produced by either peptide alone. CNS nuclei mediating this interaction have only been partially ascribed to processing by the hypothalamus and area postrema; however, other CNS structures are clearly required to mediate the interaction between leptin and amylin. The ventral tegmental area (VTA) is a likely site of action as both leptin and amylin receptor signaling in the VTA are physiologically relevant for the control of FI. To test whether VTA amylin and leptin signaling interact to control FI, we administered into the VTA moderately suprathreshold intraparenchymal doses of amylin (0.4 μg) and leptin (0.3 μg) alone or in combination. Both individually reduced FI, while the combination further suppressed FI compared to either drug alone. The FI suppression by leptin was primarily due to a decrease in meal size, whereas amylin induced early reductions in meal number and size, and later suppression of just meal size. In a separate experiment, intra-VTA doses of amylin (0.04 μg) and leptin (0.1 μg) that alone are each subthreshold for an effect on 24h FI were able to suppress feeding when given into the VTA in combination. Further evidence for an intra-VTA interaction of amylin and leptin signaling was provided by the finding that the FI and body weight suppression produced by intra-VTA leptin (0.6 μg) were attenuated by intra-VTA pretreatment with the amylin receptor antagonist AC187 (0.1 μg). These findings highlight the VTA as an important site mediating the cooperative effects of leptin and amylin.


back to the top


Modality-based segregation of ascending somatosensory axons in the direct dorsal column pathway
Jingwen Niu1, Long Ding2, Jian J. Li2, Hyukmin Kim3, Jiakun Liu1, Haipeng Li4, Andrew Moberly1, Tudor Badea5, Ian D. Duncan6, Young-Jin Son3, Steven S. Scherer2, and Wenqin Luo1
1Department of Neuroscience, and 2Department of Neurology, Perelman School of Medicine, University of Pennsylvania; 3Shriners Hospital Pediatric Research Center and Department of Anatomy and Cell Biology, Temple University School of Medicine; 4Department of Neurology, the First People’s Hospital of Chenzhou, Chenzhou, Hunan, People’s Republic of China; 5Retinal Circuit Development and Genetics Unit, National Eye Institute, Bethesda, MD; 6Department of Medical Sciences, School of Veterinary Medicine, University of Wisconsin, Madison, WI

The long-standing doctrine regarding the functional organization of the direct dorsal column (DDC) pathway is the “somatotopic map” model, which suggests that somatosensory afferents are primarily organized by receptive field instead of modality. Using modality-specific genetic tracing, here we show that ascending mechanosensory and proprioceptive axons, two main types of the DDC afferents, are segregated into a medial-lateral pattern in the mouse dorsal column and medulla. In addition, we found that this modality-based segregation is likely to be conserved in other mammalian species, including humans. Furthermore, we identified key morphological differences between these two types of afferents, which explains how modality segregation is formed and why a rough “somatotopic map” was previously detected. Collectively, our results establish a new functional organization model for the mammalian DDC pathway and provide insight into the long-standing question of how somatotopic and modality-based organization co-exist in the central somatosensory pathway.


back to the top


Anxiety vulnerability modulates perceptual, psychophysiological and neural responses to emotionally salient odors
Valentina Parma1,2, Fredrik Åhs3, Laura Hackl1, and Johan N. Lundström1,4,5
1Monell Chemical Senses Center; 2Department of General Psychology, University of Padova, Padova, Italy; 3Center for Cognitive Neuroscience, Duke University, Durham, NC; 4Department of Clinical Neuroscience, Karolinska Institute, Solna, Sweden; 5Department of Psychology, School of Arts & Sciences, University of Pennsylvania

Odors can impact our behavior by virtue of associations with behaviorally relevant stimuli, such as threats. Our personal dispositions – for instance, vulnerability towards anxiety – further modulate the quality of individual reactions. Fear olfactory learning is a vital mechanism allowing the autonomous and central nervous system to recognize the pattern of activity that represents an odor and to associate it with its meaning and the context in which it occurs. The study we conducted simultaneously registered perceptual ratings, arousal responses and functional magnetic resonance images of 25 young adults who had been classified having varying levels of “trait anxiety,” a tendency to experience anxiety across a range of everyday situations. All subjects were presented with an odor paired with an electrical shock and a “safe” odor, never associated with an irritating stimulus. Results indicate that subjects who presented a high level of trait anxiety perceived the odors as more intense, showed a higher psychophysiological reactivity, and were more likely to have enhanced responses in area of the fear network (amygdala, insula, cingulate cortex) when exposed to the odor paired with the shock as compared to the safe smell. Taken together, these results suggest that individual differences in anxiety vulnerability contribute to a differential modulation of perception, autonomic reactions and neural responses to emotionally salient odors. The present work contributes to the clarification of the neural and physiological sensory mechanisms underlying anxiety reactions, shedding light on the translational conceptualization of anxiety disorders.


back to the top


Loss of surface AMPA receptors and homeostatic changes in inhibitory synaptic transmission caused by anti-AMPA receptor
Xiaoyu Peng1, Ethan G. Hughes1, Emilia H. Moscato1, Thomas D. Parsons2, Josep Dalmau3,4, and Rita J. Balice-Gordon1

1Department of Neuroscience, and 3Department of Neurology, Perelman School of Medicine, University of Pennsylvania; 2Department of Clinical Studies-New Bolton Center, School of Veterinary Medicine, University of Pennsylvania; 4Department of Neurology, Hospital Clinic, Barcelona, Spain

Autoimmune mediated anti-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) encephalitis is a severe disease that results in short-term memory loss, seizures and other symptoms. To determine the effects of anti-AMPAR autoantibodies on synapses and neurons, hippocampal neurons cultured in vitro were treated with antibodies from patient CSF for hours to several days. Patient antibodies caused a decrease in the surface amount and synaptic localization of AMPARs, without affecting other excitatory synaptic proteins including NMDARs. The decrease in surface AMPARs resulted from increased internalization and degradation. While commercially available AMPAR antibodies directed against extracellular epitopes competed with patient antibodies for binding to surface AMPARs, treatment with commercial antibodies did not result in the loss of surface and synaptic receptor clusters, suggesting that patient antibody binding to a specific epitope causes internalization. Whole cell recording showed that patient antibodies decreased synaptic AMPAR-mediated transmission, while NMDAR-mediated currents were not affected. GABAaR-mediated spontaneous miniature inhibitory post-synaptic current amplitude and frequency, as well as inhibitory synapse size and density, were also decreased. The net effect of these changes resulted in the maintenance of the overall spontaneous firing activity of neurons. These results suggest that part of the pathophysiology of anti-AMPAR encephalitis is selective loss of surface and synaptic AMPAR clusters that results in excitatory synaptic silencing. Ongoing work is aimed at connecting the in vitro homeostatic compensation of inhibitory synaptic transmission to in vivo circuit plasticity and clinical symptoms in patients.


back to the top


Identification of Redeye, a new sleep-regulating protein whose expression is modulated by sleep amount
Mi Shi1, Zhifeng Yue1, Alexandre Kuryatov2, Jon Lindstrom2, and Amita Sehgal1,2
2Howard Hughes Medical Institute, and 2Department of Neuroscience, Perelman School of Medicine, University of Pennsylvania

Sleep is controlled by a homeostatic system characterized by the build-up of a drive to sleep during wakefulness and the decrease of this drive with increasing amounts of sleep. Electrophysiological correlates of sleep drive have been described, but very little is known about molecular components. In this study, we report a new protein involved in the homeostatic regulation of sleep in Drosophila. We conducted a forward genetic screen of chemically mutagenized flies to identify short-sleeping mutants and found one, redeye (RYE) that shows a severe reduction of sleep length. Cloning of RYE reveals that it encodes a nicotinic acetylcholine receptor α subunit required for Drosophila sleep. A previously identified sleep-regulating protein, SLEEPLESS (SSS), inhibits activity of RYE and thereby modulates the sleep phenotype of the RYE mutant. Levels of RYE oscillate in light-dark cycles and peak at times of daily sleep. Cycling of RYE is independent of a functional circadian clock, but rather depends upon the sleep homeostat, as protein levels are up-regulated in short sleeping mutants and also in wild-type animals following sleep deprivation. We propose that the homeostatic drive to sleep increases levels of RYE, which responds to this drive by promoting sleep.


back to the top


A spontaneously occurring mouse Rp1 missense mutation causing slowly progressive photoreceptor degeneration
Delu Song1, Steve Grieco1, Allan Hunter1, Yafeng Li1, Sally Chu1, Liangliang Zhao1,3, Ying Song1, Robert A. DeAngelis2, Lan-Ying Shi4, Qin Liu5, Eric A. Pierce5, Patsy M. Nishina4, John D. Lambris2, and Joshua L. Dunaief1
1F.M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, and 2Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania; 3Department of Ophthalmology, Second Hospital of Jilin University, Changchun, China; 4The Jackson Laboratory, Bar Harbor, ME; 5Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, MA

The Rp1 gene located on Chromosome 8q12 consists of four exons with an open reading frame of 6468 bp, which is primarily contained within exon 4 (788–6468 bp), and encodes a predicted protein of 2156 amino acids. Mutations in the Rp1 gene are a common cause of autosomal dominant retinitis pigmentosa. A novel L66P mutation caused by two adjacent point mutations was identified in exon 2 of the Rp1 gene. Mice homozygous, but not heterozygous, for the L66P mutation exhibited slow, progressive photoreceptor degeneration throughout their lifespan. Optical coherence tomography imaging demonstrated abnormal photoreceptor reflectivity at one month of age. Histology revealed shortening and disorganization of the photoreceptor inner and outer segments, as well as progressive thinning of the outer nuclear layer. Electroretinogram a- and b-wave amplitudes were significantly decreased with age. Western analysis showed that the quantity and size of the mutated Rp1 protein were normal. However, immunohistochemistry demonstrated that the protein was mislocalized. Axonemes were shortened in the L66P mice. In conclusion, the L66P mutation in the first doublecortin domain of the Rp1 gene is sufficient to impair Rp1 protein localization and function, leading to morphologic abnormalities in photoreceptor outer segment and progressive photoreceptor degeneration. This is the first missense mutation in Rp1 shown to be disease-causing. This study supports the concept that an Rp1 missense mutation can be pathogenic and provides a unique, slowly progressive photoreceptor degeneration model that mirrors the slow degeneration kinetics in most patients with retinitis pigmentosa.


back to the top


Roles of glucose transporters, metabolic sensors and intestinal brush border glycosidases in sweet taste
Sunil K. Sukumaran, Karen K. Yee, Ramana Kotha, and Robert F. Margolskee
Monell Chemical Senses Center

The GPCRs T1R2 and T1R3 encode the primary sweet taste receptor, responsive to a variety of sweet compounds ranging from sugars to non-caloric sweeteners and sweet proteins. Yet, there are strong indications that additional sweet sensing pathway(s) operate in the taste system. T1R3 knockout mice lose nerve responses to non-caloric sweeteners, but retain responses to some sugars. They also retain preference for sweet compounds in two bottle preference tests. We recently reported that T1R3 expressing taste cells express several glucose transporters and KATP channel subunits. We proposed that together they could constitute a sweet taste transduction pathway specific to monosaccharides like glucose and fructose. A major limitation of this hypothesis is that foods contain very little free monosaccharide sugars. Now, we demonstrate using RT-PCR, qPCR, in situ hybridization and immunohistochemistry that the disaccharidases Maltase-glucoamylase (Mgam) and Sucrase-isomaltase (Sis) are expressed in taste cells. In concert with the copious amounts of amylase originating from the main and accessory salivary glands, these enzymes can produce free monosaccharides that may activate a sugar-sensing pathway in T1R3-positive taste cells.


back to the top


Disruptions in plasticity and network activity in the hippocampus after diffuse brain injury in swine
Alexandra V. Ulyanova, Kevin D. Browne, Michael R. Grovola, Victoria E. Johnson, D. Kacy Cullen, and John A. Wolf
Department of Neurosurgery, Perelman School of Medicine, University of Pennsylvania

Functional and circuit level activity changes in the hippocampus induced by mild traumatic brain injury were studied using a swine model of closed-head rotational accelerations (200-300 rads/sec) that induce little or no loss of consciousness or subdural hemorrhage, yet exhibit axonal pathology. We used in vivo electrophysiological recordings to investigate the changes in hippocampal function in sham versus injured animals using high-density multi-electrode recording arrays and simultaneous afferent stimulation. Concentric bipolar stimulation was performed in Schaffer collaterals, perforant path, and the entorhinal cortex, using either paired-pulse or theta-burst paradigms. Input-output curves were generated and paired-pulse paradigms were utilized to examine changes in neurotransmitter release probabilities from injured versus sham animals. Changes in baseline network activity were visualized pre- and post-theta burst stimulation to examine changes in hippocampal excitability. Current source density analysis was utilized to examine changes in synaptic inputs post-injury to hippocampal layers. These experimental paradigms revealed a loss of paired pulse facilitation in area CA1 seven days post injury, potentially due to changes in neurotransmitter release probability. Changes in the oscillatory activity in the injured animals suggest a hyper-excitable network compared to sham recordings. These data indicate that mild traumatic brain injury in swine leads to dysfunction in various aspects of hippocampal circuitry post-injury, potentially underlying epileptogenesis and/or cognitive dysfunction. Further histopathological examination may indicate potential substrates for neuronal excitability changes and circuit disruption.


back to the top


Epigenetic inheritance of cocaine sensitization deficits
Mathieu Wimmer1, Fair Vassoler2, Samantha White1, Pavel Ortinski1, Heath Schmidt1, Ghazaleh Sadri-Vakili3, and Chris Pierce1
1Center for Neurobiology and Behavior, Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania; 2Biomedical Science, Tufts University, Medford, MA; 3Institute for Neurodegenerative Diseases, Mass General Hospital, Boston, MA

A growing body of evidence suggests that environmental information, such as cocaine exposure, can be inherited epigenetically. We established a rat model to examine the influence of paternal cocaine-taking on the behavior of the progeny. We combined behavioral and electrophysiological approaches to examine cocaine-induced behavioral and neuronal plasticity in the offspring of cocaine-experienced sires (CocSired) and controls that received saline (SalSired). Repeated exposure to cocaine produces behavioral sensitization, which is characterized by an augmented locomotor response to a subsequent psychostimulant challenge injection. Here, we show that male CocSired animals showed deficits in cocaine sensitization, indicating a resistance in the long-lasting behavioral responses following cocaine exposure. Expression of sensitization is accompanied by neuroadaptations in the nucleus accumbens (NAc), a brain area that is critically involved in the development of addiction. In particular, an increase in GluA2-lacking calcium-permeable AMPA receptors (CP-AMPARs) is known to accompany the expression of cocaine-induced behavioral sensitization. To determine the contribution of CP-AMPAR signaling in the NAc of CocSired and SalSired males, we measured the rectification of AMPAR-mediated currents using whole-cell patch clamp recordings. Following cocaine use, SalSired rats showed increased rectification, indicating increased CP-AMPARs in the NAc. However, this cocaine-induced plasticity was blunted in the offspring of cocaine-exposed rats. Taken together, these findings suggest that the offspring of cocaine-experienced rats show deficits in cocaine-induced behavioral and neuronal plasticity and that epigenetic mechanisms may contribute to this phenomenon.


back to the top


Structural and mechanical insights into the mechanochemical tuning of myosin
Michael J. Greenberg, Henry Shuman, Adam Zwolak, Tianming Lin, Roberto Dominguez, and E. Michael Ostap
Department of Physiology, and Pennsylvania Muscle Institute, Perelman School of Medicine, University of Pennsylvania

Myosin-Is are widely expressed, single-headed molecular motors that comprise the second largest myosin family in vertebrates with eight isoforms. Myosin-Is participate in a host of cellular processes including vesicular trafficking, pathogen response, regulation of actin dynamics, and nuclear transcription. The question thus arises, how can these motors with similar biochemical properties give rise to such diversity of function? The answer appears to lie in the force-sensing capabilities of the motors. Our single molecule biophysical studies demonstrate that the Myo1b isoform is exquisitely sensitive to tension, enabling it to act as a tension-sensitive anchor in the cell. Our studies also reveal that the closely related isoform, Myo1c, has a very different response to force despite its similar unloaded biochemical kinetics to Myo1b. To better understand the structural adaptations that tune the mechanochemical properties of Myo1b, we determined the crystal structure of Myo1b’s motor domain and first IQ-motif with bound calmodulin to 2.3 Å resolution. The structure reveals important features that differ from other characterized myosins. Notably, the N-terminal region is sandwiched between the motor domain and light-chain binding domain, such that it is positioned to communicate the lever arm position to the active site. Single molecule and ensemble biochemical assays show that these interactions play an important role in stabilizing the post-power stroke conformation of Myo1b and in tuning the rates of key biochemical transitions. We propose that the N-terminal region may have a role in tuning the mechanochemical properties of many isoforms.


back to the top


Defining Kapβ2 as a protein disaggregase for the ALS disease protein FUS
Lin Guo, Hejia Wang, Nikita Singh, and James Shorter
Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania

Amyotrophic lateral sclerosis (ALS, also called Lou Gehrig’s disease) is a debilitating, fatal neurodegenerative disease that is characterized by a severe and selective devastation of upper and lower motor neurons, progressive muscle weakness, paralysis, and death. Recently, ALS-linked mutations have been uncovered in the gene encoding the RNA-binding protein FUS (Fused in Sarcoma). Ordinarily, FUS is localized to the nucleus. However, in ALS, mutations in the PY-NLS (proline/tyrosine-nuclear localization signal) of FUS lead to impaired nuclear import, with mislocalization of the protein to the cytoplasm and formation of cytoplasmic inclusions in degenerating motor neurons. In neurons, FUS can be transported into the nucleus by interaction with the nuclear import receptor Karyopherinβ2 (Kapβ2). Interestingly, FUS mutants that have mutations in the PY-NLS display decreased affinity for Kapβ2 binding. In the current study, we investigated the interaction between FUS and Kapβ2. Our results indicate that Kapβ2 effectively inhibits and reverses WT FUS aggregation in vitro. The activity of Kapβ2 against FUS aggregates is slightly impaired in FUS disease mutants that have mutation in PY-NLS and is severely impaired in FUS disease mutants that lack the
PY-NLS. We also demonstrated in this study that Kapβ2 is effective against FUS aggregates that have different morphologies (i.e. amorphous, fibrils, and protein hydrogel). Moreover, Kapβ2 is shown to inhibit the aggregation of another ALS disease protein with PY-NLS: hnRNPA1. Although Kapβ2 is less effective in disaggregating hnRNPA1 fibrils, our results show that it has the potential to be developed as a general protein disaggregase for proteins bearing a PY-NLS.


back to the top


Potentiated Hsp104 variants antagonize diverse proteotoxic misfolding events
Meredith E. Jackrel1, Morgan E. DeSantis1,2, Bryan A. Martinez3, Laura M. Castellano14, Rachel M. Stewart1, Kim A. Caldwell3, Guy A. Caldwell3, and James Shorter1,2,4
1Department of Biochemistry and Biophysics, and 2Biochemistry and Molecular Biophysics Graduate Group, Perelman School of Medicine, University of Pennsylvania; 3Department of Biological Sciences, University of Alabama, Tuscaloosa, AL; 4Pharmacology Graduate Group, Perelman School of Medicine, University of Pennsylvania

There are no therapies that reverse the proteotoxic misfolding events that underpin fatal neurodegenerative diseases including amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD). Hsp104, a conserved hexameric AAA+ protein from yeast, solubilizes disordered aggregates and amyloid, but has no animal homologue and only limited activity against human neurodegenerative disease proteins. Here, we reprogrammed Hsp104 to rescue TDP-43, FUS, and α-synuclein proteotoxicity by mutating single residues throughout the middle domain of Hsp104. Potentiated Hsp104 variants enhanced aggregate dissolution, restored proper protein localization, suppressed proteotoxicity, and in a C. elegans PD model attenuated dopaminergic neurodegeneration. Potentiating mutations reconfigure how Hsp104 subunits collaborate, desensitize Hsp104 to inhibition, obviate any requirement for Hsp70, and enhance ATPase, translocation, and unfoldase activity. Our work establishes that disease-associated aggregates and amyloid are tractable targets and that enhanced disaggregases can restore proteostasis and mitigate neurodegeneration.


back to the top


Sequence-specific oxidative footprinting elucidates topology of amyloid-β 1-40 fibrils
Alexandra L. Klinger1, Janna Kiselar2, Serguei Ilchenko1, Hiro Komatsu1, Mark Chance2, and Paul H. Axelsen1
1Department of Pharmacology, Perelman School of Medicine, University of Pennsylvania; 2Center for Proteomics and Bioinformatics, Case Western Reserve University, Cleveland, OH

Amyloid-β1-40 peptide (Aβ40) is a principal component of the amyloid plaques associated with Alzheimer’s disease. Aβ40 spontaneously forms fibrils in vitro, but despite intense biophysical investigation important questions remain regarding its assembly. Herein, using synchrotron radiolysis coupled with mass spectrometry, we present resolution of hydroxyl radical modification rates for individual amino acid side chains in both fibrillar and prefibrillar forms of Aβ40. Side chain protection patterns in the fibrils validate an amyloid cross-β structure topology in which odd-numbered residues of the N-terminal β-sheet internally appose even-numbered residues of the C-terminal β-sheet. Further, our analyses reveal order in the N-terminus of the fibril peptides – not previously detectible by any other method. This study represents a significant advance in radiolytic protein footprinting methodology that allows detailed quantitative assessment to refine and expand current structural models.


back to the top


Human and canine bestrophinopathies share abnormal accumulation of cholesterol-associated lipofuscin in the RPE
Néstor Más Gómez1, Emily V. Dutrow1, Frank Stefano2, Robert F. Mullins3, Edwin M. Stone3, Kathleen Boesze-Battaglia2, Gustavo D. Aguirre1, and Karina E. Guziewicz1
1Department of Clinical Studies-Philadelphia, School of Veterinary Medicine, and 2Department of Biochemistry, School of Dental Medicine, University of Pennsylvania; 3Department of Ophthalmology, and Department of Visual Sciences, University of Iowa, Carver College of Medicine, Iowa City, IA

The BEST1 gene encodes bestrophin 1 protein that localizes to the retinal pigment epithelium (RPE). BEST1 mutations are responsible for a group of inherited retinal disorders known as bestrophinopathies. A hallmark of these diseases is an abnormal buildup of lipofuscin material within the RPE. Lipofuscin is undigested cellular waste and its fluorophore components were found to cause accumulation of cholesterol in cultured RPE cells. The large animal model for human bestrophinopathies, canine multifocal retinopathy 1 (cmr1), is a spontaneous bestrophin 1-/- mutation that demonstrates extensive accumulation of autofluorescent material (AF) in the RPE. Using hyperspectral autofluorescence imaging (HAI) analysis we characterized the AF of cmr1 versus control tissues by emission spectra in comparison to human patients’ lipofuscin spectra. We also investigated association of free cholesterol with the cmr1 AF and the human patients’ lipofuscin in the RPE. Cholesterol was detected by filipin binding and the percentage of colocalization was determined as a Pearson coefficient. Data analysis was performed with the 4.0 Elements software of the Nikon A1R laser scanning confocal microscopy. HAI showed that cmr1 and human bestrophinopathy RPE tissues share the same pattern of spectral emission. Higher levels of the emission intensity among control, cmr1, and human patients support higher levels of fluorophores in the RPE. Pearson coefficient analysis revealed colocalization by 70% and by 80 % in the cmr1 and the human patient RPE, respectively. As the cmr1 RPE pathology resembles that of human bestrophinopathies, our model will be important for characterizing the molecular mechanisms of BEST1-related retinopathies, and will benefit studies on AMD and Stargardt’s disease.


back to the top


Tie2 receptor dimerization and signaling is mediated by its extracellular FNIII domains
Jason O. Moore1,2, Kathryn M. Ferguson1, and Mark A. Lemmon1,2
1Department of Physiology, and 2Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania

The Tie family of receptor tyrosine kinases (RTKs) regulates a number of angiogenic processes that are critical in vascular development, as well as vascularization of tumor masses. The current signaling paradigm is that angiopoietin (Ang) 1 binds to Tie2, promoting Tie2 homodimerization and autophosphorylation. This in turn results in cell proliferation, vessel branching, and sprouting. The extracellular regions of the Tie receptors contain three each of immunoglobulin-like (Ig) domains, EGF-like domains, and fibronectin type III (FNIII) domains. The structure of the Ig/EGF domain region of Tie2 in complex with an Ang protein has been described. However, this structure does not fully explain receptor activation or dimerization. Focusing on the three membrane-proximal FNIII domains - missing from the previously reported structure - we have found that this region can independently drive Tie2 dimerization, indicating that the FNIII domains play an important role in defining the activated dimeric configuration of this receptor. To determine the molecular basis for this observation we have solved a 2.5 angstrom resolution crystal structure of the Tie2 FNIII domains, which reveals a domain architecture involving intermolecular interactions between the second and third FNIII domains that are highly reminiscent of those seen in ligand-induced dimers of the hGHR receptor. Guided by this crystal structure, we have generated mutations in the region that defines receptor dimerization. These mutations appear to reduce Ang1-stimulated phosphorylation of Tie2 in a cellular context.


back to the top


Mechanism of Drosophila EGFR inhibition by Kekkon-1
Camilla L. Oxley, Diego Alvarado, Neo Wu, and Mark A. Lemmon
Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania

Aberrant activation of receptor tyrosine kinases (RTKs) belonging to the epidermal growth factor receptor (EGFR) family drive human cancers, such as those in the lung and colon. Accordingly, the EGFR family has been intensely pursued as a therapeutic target, with five FDA-approved drugs to date. However, acquired resistance to these agents limits their long-term use, so it is imperative to develop additional therapeutic strategies to overcome resistance as well as to target newly identified oncogenic drivers. Signaling through the EGFR family members is activated by ligand-induced dimerization of the extracellular domains and is countered by several ligand or receptor antagonists. Antibody therapeutics such as cetuximab, which binds the ligand-binding site of EGFR, provide one model for extracellularly targeted inhibitors. Genetic and biochemical studies in D. melanogaster suggest others that have evolved naturally, and could be mimicked for use in human therapy. We have focused on Kekkon-1, a transmembrane glycoprotein that acts as an endogenous inhibitor EGFR, hypothesizing that understanding the molecular mechanism of this natural inhibitor will illuminate new avenues for EGFR inhibition. Using a range of biophysical techniques including analytical ultracentrifugation, X-ray scattering, and crystallography, we show that multiple domains in Drosophila EGFR contribute to Kekkon-1 binding and show that Kekkon-1 directly blocks EGFR dimerization (and thus signaling), but not ligand binding – apparently by forming a 1:1 complex that directly occludes the dimerization site in the EGFR extracellular region. These results may have important implications for the development of novel EGFR inhibitors.


back to the top


ADAMTS13 expressed in platelets offers systemic protection against arterial thrombosis and therapeutic benefits In TTP
Brandy Pickens1,2, Yingying Mao1, and X. Long Zheng1,2
1Department of Pathology and Laboratory Medicine, The Children’s Hospital of Philadelphia; 2Department of Pathology and Laboratory
Medicine, Perelman School of Medicine, The University of Pennsylvania

ADAMTS13, a plasma metalloprotease, cleaves von Willebrand factor (VWF) and inhibits arterial thrombosis and inflammatory response. Deficiency of plasma ADAMTS13 results in a potentially fatal syndrome, thrombotic thrombocytopenic purpura (TTP). Plasma infusion or exchange is the only effective therapy available to date. To develop novel therapeutics against TTP, we tested the hypothesis that recombinant ADAMTS13, expressed specifically in platelets, may offer systemic protection against arterial thrombosis and therefore provide therapeutic benefit for TTP. To test this hypothesis, we first generated transgenic mice (JAX B6SJL) carrying a human full-length ADAMTS13 gene under a murine platelet glycoprotein 1b (CD41) promoter. The transgenic mice were then bred with Adamts13-/- (CAST/Ei) mice for >4 generations. By Western blotting and activity assay, we show that a full-length, proteolytically active human ADAMTS13 protein is detected in platelet lysate obtained from transgenic (rA13-PltTG) mice but not in Adamts13-/- mice or wild-type mice. ADAMTS13 was releasable upon stimulation with thrombin (1 U/ml), collagen (10 µg/ml), and AYPGKF (0.5 mM). More significantly, rA13-PltTG mice had higher baseline platelet count than Adamts13-/- mice, but exhibited a dramatically reduced rate of thrombus formation in mesenteric arterioles after oxidative injury as compared with Adamts13-/- mice and wild-type mice. Finally, rA13-PltTG mice were protected from Shigatoxin-2 (Stx-2) or murine recombinant VWF (mVWF)-induced “TTP-like” syndrome, as demonstrated by fewer rA13-PltTG mice having TTP. In summary, the platelet-derived ADAMTS13 offers systemic protection against arterial thrombosis after oxidative injury and provide a therapeutic benefit to murine models of TTP.


back to the top


Lamin-A acts as a nuclear mechanostat, scaling with tissue stiffness and enhancing matrix-directed differentiation
Joe Swift1#, Irena L. Ivanovska1#, Amnon Buxboim1, Takamasa Harada1, P. C. Dave. P. Dingal1, Joel Pinter1, J. David Pajerowski1, Kyle R. Spinler1, Jae-Won Shin1, Manorama Tewari1, Florian Rehfeldt1, David W. Speicher2, and Dennis E. Discher1,2
1Molecular and Cell Biophysics Laboratory, School of Engineering and Applied Science, University of Pennsylvania; 2Center for Systems and Computational Biology, The Wistar Institute; # Equal contribution

Tissues can be soft like fat, which bears little stress, or stiff like bone, which sustains high stress, but any systematic relationship to specific proteins or to processes of differentiation is unknown. Proteomics analyses revealed that levels of the nucleoskeletal protein lamin-A showed characteristics of a ‘biological polymer’, scaling as a function of tissue elasticity, E, as did levels of collagens that directly determine E. Lamin-A was found to confer a viscous stiffness to nuclei in the high stress tissues where it predominates, thus providing physical protection to DNA. Evidence suggested a mechano-sensitive regulation of lamin-A levels, but the process of mechanical stimulation of molecular processes is poorly understood. Cysteine-shotgun mass spectrometry (CS-MS), a method capable of mapping the exposure of cysteine residues as proteins are stressed in complex biological systems, allowed us to identify stress-sensitive proteins. We found that lamin-A contains stress-sensitive domains with changes in protein conformation coupled to phosphorylation, a post-translational modification that increases protein solubility. Conformational changes in lamin-A under stress increased with mutations that cause disease in stiff tissue, but lamin-A levels in cultured cells also adjusted to matrix elasticity. Differentiation of stem cells to fat or bone was respectively enhanced by low or high lamin-A levels, and while transcription of lamin-A is regulated by the vitamin-A/retinoic acid pathway with broad roles in development and regeneration, nuclear entry of retinoic acid receptors was lamin-A-regulated. Tissue stiffness and stress thereby couple to lineage through a stress-sensitive regulation of lamin-A.


back to the top


A high-throughput inhibitor screen for the protein disaggregase Hsp104, a novel therapeutic target against fungi
Mariana P. Torrente and James Shorter
Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania

Fungi are a growing public health problem. These microorganisms can cause fatal infections in immunocompromised hosts such as transplant, AIDS, and cancer patients, as well as in premature babies and the elderly. Efficacious treatment of these infections is often difficult, as medications that eradicate the fungus also harm human cells, becoming toxic to the human host. Thus, more specific anti-fungals able to bypass these problems are needed. Hsp104 is a ring-shaped hexameric yeast AAA+ ATPase capable of restoring misfolded and aggregated proteins back to their normal structure and function. This protein is essential for cell viability in challenging conditions, when proteins tend to aggregate more readily. Interestingly, Hsp104 orthologues are found in all kingdoms of life except animals. Inhibition of Hsp104 would allow for therapies specifically targeted against fungi, which are dependent on Hsp104’s activity for survival, while producing little or no side effects. However, only one inhibitor of Hsp104’s activity is known to date. Here, by way of a high throughput colorimetric screen including over 16,000 compounds, we identified 20 small-molecule drugs capable of inhibiting Hsp104’s activity. These molecules inhibit Hsp104’s ATPase activity in a non-competitive manner. Out of these, we mined molecules that interact with Hsp104 by mechanisms other than non-specific colloidal effects, and inhibit Hsp104 function both in vitro and in vivo. It is our hope that the results of this study can ultimately open the door to a new, more selective generation of antifungals with fewer side effects.


back to the top


Simple ultrashort echo time MRI measure associated with cortical bone porosity
Mahdieh Bashoor-Zadeh, Chamith S. Rajapakse, Cheng Li, Wenli Sun, Alexander C. Wright, and Felix W. Wehrli
Department of Radiology, University of Pennsylvania

Because cortical bone (CB) porosity is a major contributor to bone fracture, assessment of CB quality is important in subjects at risk for diseases such as osteoporosis and renal osteodystrophy. Due to limitations in resolution and other factors, microstructure of CB pore spaces cannot be resolved by current clinical imaging modalities. The MRI approach described here could be used clinically to assess changes in CB porosity. Ultrashort echo time imaging (UTE) can be used to obtain proton signals from bone water that resides in pore spaces (free water) or is bound to collagen matrix (bound water) on clinical MRI systems. The purpose of this study was to derive a parameter associated with CB porosity using a UTE sequence that can be translated for clinical imaging. Sixteen cadaveric human cortical bone samples (9 female and 7 male, age range 27-97 y, 65 ± 22 y) 36 mm in thickness from the 38% site of the tibia diaphysis were imaged using UTE and μCT. In UTE imaging, two acquisitions were performed with short and long-TE. The short-TE value was acquired to capture signal from both free water (T2* > 1 ms) and bound water (T2*=300-400 μs). The long-TE image was acquired to capture signal essentially from free water in pore spaces only. Porosity index was defined as the ratio of mean CB intensity in the long-TE image to that of short-TE image. The UTE-porosity index was highly correlated with μCT derived porosity (R2=0.68, p < 0.0001), and showed a positive association with age (R2= 0.58, p < 0.001). Correlations between UTE-porosity index and reference porosity values obtained from high-resolution μCT images suggests that volumetric CB porosity can be assessed using a clinically practical 3D UTE protocol involving collection of only two echoes.


back to the top


Non-invasive measurement of cerebral hemodynamics during extracorporeal membrane oxygenation therapy
David R. Busch1,2, Ann L. McCarthy3, Jennifer M. Lynch2, Peter J. Schwab1, Madeline E. Winters1, Arjun G. Yodh2, Maryam Y. Naim4, and Daniel J. Licht1
1Department of Neurology, The Children’s Hospital of Philadelphia; 2Department of Physics, University of Pennsylvania; 3Temple University School of Medicine; 4Anesthesiology and Critical Care Medicine, The Children’s Hospital of Philadelphia

Extra corporal membrane oxygenation (ECMO) provides a crucial bridge to pediatric patients with eminently life-threatening pulmonary or cardiac disorders. During ECMO, the lungs and/or heart are bypassed through large-vessel cannulation to permit mechanical blood oxygenation and perfusion. Cerebral perfusion is critical for survival and long-term quality of life, but is not monitored. Perfusion is largely dependent on the mechanical pump rate, which is set first by patient weight, then adjusted based on clinical observations (arterial pressure, heart rate, peripheral tissue oxygenation, etc.). However, clinicians currently have no direct measurement of cerebral blood oxygenation and perfusion to optimize this mechanical circulation. Diffuse optical spectroscopy (DOS) and diffuse correlation spectroscopy (DCS) provide rapid, quantitative, and non-invasive measurements of tissue blood oxygenation, volume, and blood flow. We have applied DOS/DCS to pediatric ECMO patients during manipulation of the ECMO pump rate. Initial results suggest that DOS/DCS can ascertain the pump rate necessary to maintain mean arterial pressure at a point optimal for cerebral perfusion (i.e., autoregulated).


back to the top


Functionalization of endothelial cells with biodegradable magnetic nanoparticles for targeted delivery applications
Jillian E. Tengood, Ivan S. Alferiev, Ilia Fishbein, Robert J. Levy, and Michael Chorny
Department of Pediatrics, The Children’s Hospital of Philadelphia

Stent angioplasty is widely used clinically for treating vessel obstruction; however, this treatment causes arterial injury and endothelial denudation, which in turn can result in neointima formation, vessel renarrowing (restenosis) and/or thrombosis. Magnetically targeted cell delivery is a promising strategy for preventing restenosis. Our group has magnetically guided delivery of endothelial cells to a stented vessel. Cell delivery techniques take advantage of loading cells with MNPs in vitro prior to delivery, resulting in magnetically responsive cells, where intracellular MNP stability and degradation kinetics are critical for safe and efficient magnetic targeting. The optimization of MNP formulations for targeted delivery applications is facilitated by using fluorescently-labeled particles that enable tracking of MNP-cell interactions, uptake kinetics, intracellular compartmentalization, and fate. We address the challenge of evaluating complex MNP degradation and tracking using an approach based on real-time fluorimetric measurements on cells loaded with nanoparticles co-labeled by complementary fluorescent probes that form a good Forster Resonance Energy Transfer (FRET) donor-acceptor pair. Its use for studying intracellular degradation kinetics of biodegradable polymer-based MNP may elucidate the stability and fate of MNP in the biological milieu, and provide a sensitive and non-invasive approach for examining intracellular degradation of MNP. This work allows us to investigate and optimize a number of important variables controlling the functionality and performance of MNP-loaded endothelial cells as part of the targeted cell delivery approach.


back to the top


A computational model comparing the effect of brain stiffness and neck strength on a concussion following a head impact
Andrew G. Voyiadjis1, Vivek B. Shenoy2, and Douglas H. Smith1
1Department of Neurosurgery, and 2Department of Material Science and Engineering, Perelman School of Medicine, University of Pennsylvania

Multiple studies have shown that females have a higher rate of concussion than males while participating in the same sport activity such as soccer. A recent study has attributed significantly lower head-neck stability in females than males to this increased risk. Here, we examined the sex differences in neck strength as well as brain stiffness as contributing factors that influence the rate of concussion. Notably, recent studies have shown that females have significantly stiffer brains than age-equivalent males. We developed a computational finite element model of head rotational acceleration using real-world mechanical loading conditions of concussion. We applied a range of neck strength and brain stiffness parameters and examined their differential effects on brain tissue stresses and strains. We found that for stronger necks, brain stresses increase and strains decrease when the same rotational acceleration and brain stiffness is applied. In addition for stiffer brains, the stresses increase and strains decrease in the brain under the same neck strength and rotational acceleration. Contrary to a widely held assumption, our data suggest that greater neck strength actually increases the risk of brain damage at a given head acceleration by inducing larger tissue stresses. In contrast, we found that stiffer brains suffered greater tissue stresses at set accelerations, which may account for the higher incidence of concussion in females.


back to the top


Genomic architecture and treatment response in pediatric acute myeloid leukemia: the children’s oncology group
Marijana Vujkovic1, Edward F. Attiyeh1,2, Rhonda E. Ries3, Michelle Horn1, Elizabeth K. Goodman1, Todd A. Alonzo4, Robert B. Gerbing5, Alan S. Gamis5, Soheil Meshinchi3, and Richard Aplenc1,6
1Division of Oncology, The Children’s Hospital of Philadelphia; 2Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania; 3Fred Hutchinson Cancer Research Center, Seattle, WA; 4Keck School of Medicine, University of Southern California, Los Angeles, CA; 5Children’s Oncology Group, Arcadia, CA; 6Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania

Childhood acute myeloid leukemia (AML) is characterized by chromosomal instability and requires intensive therapy to cure. We present a large-scale study from 3 Children’s Oncology Group trials (AAML-0531, AAML-03P1, and CCG-2961) to define the genomic architectural profiles of pediatric AML and describe broad and focal gains and losses using SNP microarrays. A total of 452 matched tumor-remission samples from 459 children with de novo AML were genotyped and a matched allele-specific copy number analysis of tumors (ASCAT) was performed. The profiles were stratified into three risk categories: favorable, standard, or poor. Genomic areas with significant recurrent aberrations were identified with GISTIC 2.0. We investigated whether profiles of genomic instability in the AML genome are predictive of 3 year event-free survival (EFS) by using the total number of copy number alterations (CNAs) as a measure for allelic imbalance. On average, leukemic samples acquire 1.14 somatic CNAs. Known mutations in AML were enriched in recurrent focal CNA regions, as shown by amplifications on 17q24 (*TK1), 1q32, 3q28, 11q23 (*MLL), 6q27 (*DLL1), 2q32.1, and 4q35.2 (Fig 2). Focal CN-LOH regions were confined to 11p15.5 (*NUP98, *PICALM, *WT1), 1p36.3 (*RUNX3, *NRAS), 9p24.3 (*MLLT3), 3q25.3, 6p23 and 7q35 (*MLL3). Deletions include 7q36.1 (*MLL3, *EZH2), 16p13.11 (*MYH11), 9q21.32, 11p13 (*WT1), 2q37.1 (*IDH1, *DNMT3A), 10p12.31 (*MLLT10), 11q23.3 (*MLL), 16q22.1 (*CBFB), and 1p36.3 (*RUNX3). The association between CNA status and 3 year EFS was significantly lower in standard risk patients with CNAs (51% vs 34%, P 0.03). This study confirms established regions of CNA enrichment in pediatric AML and identifies novel regions that may involve driving mediators of tumor fitness and/or acquired resistance to targeted therapies.


back to the top


Kartogenin promotes mouse embryo limb joint formation and skeletal growth
Rebekah S. Decker1, Eiki Koyamac, Shoutian Zhu2, Peter G. Schultz3, and Maurizio Pacifici1
1The Children’s Hospital of Philadelphia; 2California Institute for Biomedical Research, La Jolla, CA; 3The Scripps Research Institute, La Jolla, CA

While much is known about the structure of synovial joints their susceptibility to common pathologies such as osteoarthritis, surprisingly little is known about the mechanisms that direct their development. Such basic information could be extremely useful in understanding pathogenesis and treatment of joint disease. Recently, one of our research groups discovered that the small molecule kartogenin (KGN) promotes chondrocyte differentiation from adult mesenchymal stem cells and enhances repair of injured articular cartilage. What has remained unknown, however, is whether KGN may be able to stimulate the development of synovial joints during embryogenesis. Here, we used an ex vivo system to evaluate the influence of KGN on synovial joint formation in reporter-expressing mouse embryo limb explants. Kartogenin treatment strongly stimulated the growth and elongation of the cartilaginous phalangeal elements. Strikingly, putative joint sites of KGN treated limbs also displayed evidence of accelerated joint formation, including condensation of joint interzone cells and organization of distinct joint tissues. Analysis revealed that the effects of KGN are likely realized through modulation of the transforming growth factor β (TGFβ) signaling pathway. Our data provide clear evidence that KGN is a strong stimulator of joint formation and sustains these complex processes in ex vivo mouse embryo limbs. The limb growth and elongation seen in response to KGN reiterates its pro-chondrogenic abilities first revealed by our previous studies, and also points to plausible additional properties in limb regeneration.


back to the top


SNARE dynamics during melanosome biogenesis
Megan K. Dennis1, Adriana R. Mantegazza1, Subbarao G. Setty2, Ricardo A. Linares Saldana3, Cedric Delevoye4,5, Elena V. Sviderskaya6, Dorothy C. Bennett6, Graca Raposo4,5, and Michael S. Marks1
1Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania; 2Microbiology and Cell Biology,

Indian Institute of Science, Bangalore, India; 3NIH National Human Genome Research Institute, Bethesda, MD; 4Institut Curie, Centre de Recherche, Paris, F-75248 France; 5Structure and Membrane Compartments, Centre National de la Recherche Scientifique UMR144, Paris, France; 6Centre for Molecular and Metabolic Signalling, Division of Biomedical Sciences, St. George’s, University of London, London, UK

Melanosomes are lysosome related organelles (LROs) that coexist in pigmented cells alongside late endosomes and lysosomes. Hermansky Pudlak syndrome (HPS) is a group of genetic disorders caused by impaired biogenesis of LROs such as melanosomes. Defects in two complexes, Biogenesis of Lysosome Related Organelles Complex 1 and 2 (BLOC1, BLOC2), are responsible for 6 of 9 HPS subtypes. The functions of BLOC1/2 are poorly understood, but they appear to act at discrete transport steps to deliver cargo from early endosomes to melanosomes. BLOC1- melanocytes are severely hypopigmented, and melanogenic cargoes are largely trapped in enlarged early endosomes. In contrast, BLOC2- melanocytes are modestly hypopigmented, and a cohort of melanogenic cargoes localize to small, weakly pigmented melanosomes. A number of data suggest that BLOC2 targets endosomal transport intermediates to melanosomes. We hypothesized that BLOC1/2 function in part to facilitate sorting and delivery of SNARE proteins that mediate fusion of these intermediates with melanosomes. BLOC1 binds to the endosomal SNARE protein syntaxin 13 (STX13), and we have observed STX13-positive tubules adjacent to melanosomes by immunoelectron microscopy and live cell imaging. We propose that STX13 plays a role in cargo delivery to melanosomes and that STX13 dynamics are altered in BLOC1/2 deficient melanocytes. Using melanocytes from wild type, BLOC1- and BLOC2- mice, we analyzed GFP-STX13 dynamics by live cell microscopy and found significantly shorter STX13-positive tubules in BLOC1- and BLOC2- cells. In BLOC2- cells, STX13 tubules make fewer, more transient contacts with melanosomes than in wild type melanocytes. Our studies will elucidate the roles of BLOC1/2 in SNARE regulation as required for cargo delivery to melanosomes.


back to the top


Improved efficiency of direct reprogramming to cardiomyocytes from fibroblasts with the use of small molecules
Jamie L. Ifkovits, Russell C. Addis, Jonathan A. Epstein, and John D. Gearhart
Institute for Regenerative Medicine, and Department of Cell and Developmental Biology, Perelman School of Medicine, University of

Recent studies have been successful at utilizing overexpression of transcription factors to generate cardiomyocyte-like cells from fibroblasts, albeit at a low frequency in vitro. Although improved efficiency is observed in vivo following induction of MI in rodent models, additional work needs to be completed to generate up to 100M cardiomyocytes necessary to repair the human heart post-injury. This work focuses on identifying small molecules to enhance direct reprogramming towards cardiomyocytes. Fibroblasts were transduced with lentiviral vectors inducing expression of five transcription factors and a reporter construct in which the genetically-encoded calcium indicator GCaMP is driven by the cardiomyocyte-specific Troponin T promoter. The small molecules were added, alone or in combination, at different time points. Reprogramming efficiency was evaluated at day 14 using both ICC of cardiomyocyte-specific structural proteins and the number of cells demonstrating robust calcium oscillations. Robust beating of single and groups of cells was observed in conditions treated with two different molecules as early as day 10. Upon GCaMP quantification, ~3-5 times the number of cells displaying calcium transients were observed in the groups treated with the same two molecules compared to the vehicle control. Current work aims to investigate the mechanism behind this improved efficiency to facilitate the development of alternative treatment strategies for the clinic.


back to the top


Functional screening identifies epigenetic factors promoting myocyte proliferation
Ying Liu1, Anthony Olarerin-George2, Kate Stewart1, Jun Kong1, Sara Cherry3,4, John Hogenesch2, and Edward Morrissey3,4
1Department of Medicine, 2Institute for Translational Medicine and Therapeutics, 3Department of Microbiology, and 4Penn Genome Frontiers Institute, Perelman School of Medicine, University of Pennsylvania

The ultimate goal of cardiovascular regenerative medicine is to regenerate lost myocytes. In mammals, embryonic cardiomyocytes (CMs) are highly proliferative, yet adult mammalian CMs are viewed as permanently withdrawn from the cell cycle. The cellular and molecular mechanisms for the transition of CMs from a proliferative state, characteristic of embryonic stages, to the differentiated, hypertrophic phenotype typical of adult cells remain elusive. Epigenetic modification has been implicated in numerous biological processes including cell proliferation and differentiation. A comparative transcription profile by RNA-Seq was performed between embryonic and adult murine CMs. We identified hundreds of epigenetic modifiers expressed at high levels in mouse embryonic heart, but are expressed at low levels or are silent in adult heart. We performed a high-throughput functional screen by over-expressing these targets in cultured neonatal mouse CMs to further identify the potential factors that boost adult myocyte proliferation. Out of 103 candidates, 20 factors showed strong effects on stimulating cardiomyocyte proliferation by promoting both karyokinesis and cytokinesis. These preliminary data suggest that adult myocyte proliferation can be greatly encouraged by manipulating epigenetic modifiers. Further mechanistic and in vivo studies are in progress.


back to the top


EBF2 determines and maintains brown adipocyte identity
Sona Rajakumari1,2, Jun Wu4, Jeff Ishibashi1,2, Hee-Woong Lim1,3, An-Hoa Giang4, Kyoung-Jae Won1,3, Randall R. Reed5, and Patrick Seale1,1
1Institute for Diabetes, Obesity, and Metabolism, 2Department of Cell and Developmental Biology, and 3Department of Genetics, Perelman School of Medicine, University of Pennsylvania; 4Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA; 5Center for Sensory Biology, Department of Molecular Biology and Genetics, Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore MD

Both brown (heat-producing) and white (energy storing) adipocyte differentiation is regulated by the master transcription factor Pparγ. While Pparγ can explain the characteristics that are common to the two types of fat cells, it remains unclear whether Pparγ also controls lineage-specific gene programs. Here, we show that Early B-Cell Factor-2 (Ebf2) determines brown versus white adipocyte identity. An Ebf binding motif was highly enriched within brown adipose-specific Pparγ binding sites that we identified by genome-wide ChIP-Seq. Of the Ebf isoforms, Ebf2 was selectively expressed in brown relative to white adipocytes and was bound at brown fat-specific Pparγ target genes. Strikingly, Ebf2 expression in myoblasts or white pre-adipose cells recruited Pparγ to brown-selective binding sites and reprogrammed cells to a brown fat fate. Notably, Ebf2 and Pparγ cooperated to directly and powerfully activate transcription of Prdm16, a key regulator of thermogenic genes. In brown fat cells, Ebf2 expression was essential for establishing and maintaining a brown fat-specific gene program. Finally, brown adipose tissue from Ebf2-deficient mice displayed a complete and very specific loss of brown-specific characteristics. Taken together, these results show that Ebf2 determines brown adipocyte identity by recruiting Pparγ to brown fat-selective target genes.


back to the top


Ebf2 expression identifies brown adipose precursors in the somitic mesoderm
Wenshan Wang1,2, Li Huang1,2, Andranik Durgaryan4, Sona Rajakumari1,2, Hee-Woong Lim1,3, Kyoung-Jae Won1,2, Boris Kablar5, Hong Qian4, and Patrick Seale1,2
1Institute for Diabetes, Obesity & Metabolism, 2Department of Cell and Developmental Biology, and 3Department of Genetics, Perelman School of Medicine, University of Pennsylvania; 4Center for Hematology and Regenerative Medicine (HERM), Karolinska Institutet, Stockholm, Sweden;5Department of Anatomy & Neurobiology, Dalhousie University, Halifax, Canada

Brown adipocytes, skeletal muscle and some white adipocytes originate from Myf5-expressing cells in the somite. However, brown adipocyte precursor cells within this multipotent lineage had not been isolated or defined. Here, we examined the development of brown adipocytes arising from Myf5-expressing cells (Myf5 lineage +) through in vivo lineage tracing and by the prospective isolation of precursor populations. Myf5 lineage+ cells gave rise to brown fat as well as dorsal dermis and skeletal muscle in the interscapular region. Pparγ, the master adipogenic factor, was detected in clusters of Myf5 lineage+ cells at embryonic day 14 (E14) of mouse development. At this stage, fibroblastic Myf5 lineage+; PDGFRα+ cells preferentially underwent brown adipocyte differentiation, whereas Myf5 lineage+; PDGFRα- cells developed into skeletal muscle cells. RNAseq studies identified Ebf2, a known regulator of brown fat differentiation, as one of the most highly enriched genes in freshly sorted Myf5 lineage+; PDGFRα+ cells. Strikingly, Ebf2-expressing precursor cells purified from E14 Ebf2-GFP reporter mice uniformly differentiated into brown adipocytes. By contrast, the cell fraction lacking Ebf2 expression was devoid of brown adipogenic precursors and expressed master regulators of other lineages. EBF2 and MYOD, a specific marker of myoblasts, were expressed in distinct populations of Myf5 lineage+ cells within somites starting from E12. Altogether, these results establish EBF2 as a specific marker of committed brown adipose precursors during embryogenesis.


back to the top


Multiple modes of regulation of lineage-specific transcription factors by Wnt signaling in C. elegans embryogenesis
Amanda L. Zacharias1, Travis Walton1, Joshua T. Burdick2, Elicia Preston1, and John I. Murray1,2,3
1Department of Genetics, 2Graduate Group in Genomics and Computational Biology, and 3PENN Genome Frontiers Institute, Perelman School of Medicine, University of Pennsylvania

The invariant C. elegans lineage robustly generates diverse cell fates, with Wnt and other signals regulating expression of lineage-specific transcription factors (TFs). We previously used lineage tracing to map spatiotemporal expression of reporters for 127 embryonic TFs. This approach identified many “lineally repetitive” (LR) TFs expressed in multiple posterior branches of the lineage, suggesting that they are potential targets of Wnt signaling and that they may act combinatorially to specify cell fates. We tested whether 15 of these posterior LR genes, ceh-6, ceh-13, ceh-27, ceh-36, elt-6, mir-57, nhr-25, nhr-67, nob-1, pal-1, pax-3, tlp-1, tbx-11, unc-130, and vab-7 are regulated by Wnt signaling by analyzing expression patterns in embryos treated with RNAi against the Wnt effector TF pop-1, its export factor lit-1, and its coregulator and β-catenin homolog sys-1. We found that all 15 genes had altered expression patterns, indicating that they are regulated by Wnt signaling. The regulatory regions of these genes all harbor candidate POP-1 binding sites, suggesting that they could be direct targets, which we are working to test. While previously identified targets are both repressed by POP-1 in unsignaled anterior cells and activated by POP-1 in Wnt-targeted posterior cells, we instead observed two other types of responses. Some genes required POP-1 function only for repression in anterior cells, while others required POP-1 only for activation in posterior cells. We hypothesize that these differences result from the effect of different contexts of transcription factors that also contribute to the regulation of these genes. We plan to investigate the architecture of Wnt-regulated enhancers in the future.


back to the top


A knockout mouse model to define the roles of the epithelial splicing regulatory proteins in development
Thomas Bebee1, Katherine Sheridan1, Benjamin Cieply1, Sunder Sims-Lucas2, Daniel Bushnell2, Maximilian Reichert3, Carlton Bates2, Anil Rustgi3, and Russ Carstens1
1Department of Genetics, and 3Department of Gastroenterology, Perelman School of Medicine, University of Pennsylvania; 2Rangos Research Center, Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, PA

The Epithelial splicing regulatory proteins 1 and 2 (Esrp1 and Esrp2) coordinate a splicing regulatory network (SRN) of transcripts associated with processes necessary for epithelial differentiation. Esrp targets include regulators of cell polarity, adhesion, cytoskeletal organization, and motility such as FGFR2, CD44, and p120 Catenin. Esrp expression is epithelial cell-type-specific and is down-regulated during the Epithelial to Mesenchymal Transition (EMT), and up-regulated by the reverse process (MET). While signaling cascades and transcriptional regulation of EMT/MET events have been extensively evaluated during development, the contribution of tissue-specific splicing regulatory factors is a hitherto understudied consideration. We propose the Esrps regulate an SRN converging on pathways crucial for normal epithelial and mesenchymal interaction during embryonic development. To evaluate the requirement of the Esrps in development, we generated mouse knockout alleles for Esrp1 and Esrp2. Esrp1 knockout mice are neonatal lethal and present with an array of developmental defects that implicate the Esrps in the regulation of FGF and additional signaling pathways. Moreover, splicing analysis in Esrp1 knockout mouse tissues verify the loss of Esrp1 is sufficient to promote splicing switches of known ESRP targets in vivo. Esrp knockout mice provide the first evidence for a functional requirement of Esrp-regulated splicing in normal development, and provide a genetic tool to evaluate the complex regulation of epithelial/mesenchymal interaction at the level of alternative splicing. Furthermore, we have generated conditional knockout alleles for Esrp1 and Esrp2 that will afford the identification of Esrp targets in a variety of epithelial populations from various tissues.


back to the top


A comprehensive and unbiased screen of RNA binding proteins to identify novel splicing factors
Benjamin Cieply, Tatiana Karakasheva, and Russ Carstens
Department of Medicine, and Department of Genetics, Perelman School of Medicine, University of Pennsylvania

Pre-mRNA alternative splicing is a fundamental mechanism of regulating gene expression, and questions regarding its regulation remain to be addressed. Many gene alternative splicing isoforms are expressed in a tissue/cell specific manner during development and in adults. We are interested in the mechanisms by which these refined gene expression patterns are achieved. Over 400 RNA binding proteins (RBPs) with canonical RNA binding domains or motifs exist in the human genome, but the functions of many of these proteins, including splicing regulatory activity, have not been defined. We hypothesize that novel splicing factors, including tissue-specific examples, reside within this set of genes and our broad goal is to discover them. Greatly confounding this goal is determining which RBPs to test on which transcripts. To streamline our investigations, our first aim is to comprehensively screen the RBPs for their ability to generally regulate splicing by tethering them to custom EGFP-based splicing reporters via the lambda-N/BoxB system. We obtained over 350 RBP cDNAs in Gateway entry vectors from the Human Orfeome collection. These are being transferred into our Lambda-N fusion expression vector and then co-transfected with BoxB element splicing reporters. We have identified several previously known splicing factors including TIA-1, RBM11, and RBM39, thus validating our system, as well as many candidate novel splicing factors. Future work based on this screen will be aimed at: 1. Cross referencing and validating gene expression arrays to identify tissue specific expression patterns of novel splicing factors and 2. Identifying their functional significance and target genes in those tissues.


back to the top


RNA-seq to identify the causal gene for cone-ERG loss in a canine model of cone-rod dystrophy
Gautami Das1, Anand Swaroop2, Matthew Brooks2, Gustavo Aguirre1, and Keiko Miyadera1
1School of Veterinary Medicine, University of Pennsylvania; 2National Eye Institute, National Institutes of Health

Cone-rod dystrophy (CRD) is a progressive retinal degenerative disease affecting 1 in 40,000 individuals. It is characterized by primary cone dysfunction/loss, or, in some cases by concomitant loss of both cones and rods and frequently leading to blindness. A study on a research colony mapped a 14Mb locus for a recessive form of canine CRD (cord1) to chromosome 15 in dogs, and a homozygous exonic 44bp insertion (ins/ins) in RPGRIP1 was identified. The interval was subsequently narrowed down to 1.74Mb still containing RPGRIP1. In a recently re-established research colony at University of Pennsylvania, however, some
RPGRIP1ins/ins dogs retain normal cone-ERG, while others show severely diminished or no cone-ERG responses. Our aim was to identify the gene/mutation within the locus as a possible cause of cone-ERG loss in the colony. Illumina GAIIx 76bp single-end RNA-seq was performed on retinal cDNA libraries generated from cone-loss (n=2, 9 months) and cone-normal (n=2, 9-11 months) RPGRIP1ins/ins dogs. Genomatix and Partek Genomic Suite were used to identify single nucleotide variants (SNVs) for haplotype analysis, examine differential expression of genes/transcripts, and to identify novel genes/transcripts. Sequence reads were visualized using Integrative Genomics Viewer. Data from the identified SNVs showed both cone-loss and cone-normal dogs share the same haplotype across the 1.74Mb region. There was no difference in the overall expression of any of the genes (or any of their specific isoforms) across the locus between the two groups suggesting that neither RPGRIP1 nor any other gene in this locus is associated on its own with cone-ERG loss phenotype in the colony. Involvement of other factor(s) beyond the primary locus is suggested for cone function in these dogs.


back to the top


Rev-erbα coordinates multiple phases of circadian gene expression through alternate mechanisms of gene regulation
Logan J. Everett1,2, Ankur Roy1, Dan Feng1, Erika Briggs1, Zachary Gerhart-Hines1,3, Bin Fang1,3, Fenfen Wang1, Jennifer Jager1,3, Anne Bugge1,3, Zheng Sun1,3, and Mitchell A. Lazar1,3
1Institute for Diabetes, Obesity, and Metabolism, 2Department of Genetics, and 3Division of Endocrinology, Diabetes, and Metabolism, Perelman School of Medicine, University of Pennsylvania

Rev-erbα is an essential and functionally redundant component of the core clock and a critical regulator of circadian metabolism. While loss of both Rev-erbα and Rev-erbβ lead to complete loss of rhythm and severe metabolic dysfunction, loss of Rev-erbα alone has been shown to de-stabilize the molecular clock and cause dysregulation of lipid and cholesterol metabolism in the liver. Our group has previously demonstrated that Rev-erbα represses target genes by recruiting the co-repressors NCoR and HDAC3, thereby enforcing rhythmic expression with a trough at zeitgeber time 10 hours (ZT10) corresponding to the peak Rev-erbα protein levels. By integrating newly collected and previously published genome-scale data sets, we discovered an additional class of Rev-erbα targets that have reduced expression in Rev-erbα-/- mouse livers at ZT10 and are rhythmically transcribed in phase with Rev-erbα protein levels. The additional targets identified are also enriched for core clock components and key metabolic regulators, suggesting that physiological roles of Rev-erbα are mediated by both repressed and activated targets. Comparison of Rev-erbα binding sites associated with gene repression versus activation revealed additional genomic and epigenomic features delineating distinct regulatory functions. Repression-associated sites are enriched for canonical Rev-erb-binding motifs, while activation-associated sites occur in close proximity to recognition motifs for other liver-specific and circadian factors, suggesting that Rev-erbα-dependent activation is mediated by additional regulators. Thus, our results suggest Rev-erbα coordinates multiple phases of circadian expression, and its effect on target gene transcription is highly dependent on the genomic context of each binding site.


back to the top


Integrator regulates transcriptional elongation by recruiting super elongation complex
Alessandro Gardini1, David Baillat1, Matteo Cesaroni1, Deqing Hu2, Jill Marinis3, Mitchell A Lazar3, Ali Shilatifard2, and Ramin Shiekhattar1
1The Wistar Institute; 2Stowers Institute for Medical Research, Kansas City, MO; 3Division of Endocrinology, Diabetes, and Metabolism,
Department of Medicine, The Institute for Diabetes, Obesity, and Metabolism, Perelman School of Medicine, University of Pennsylvania

Integrator is a multi-protein complex that associates with the C-terminal domain (CTD) of RNA polymerase II (RNAPII) and is involved in the 3’-end processing of RNAPII-dependent small nuclear RNAs. To define the scope of Integrator’s function, we performed chromatin immunoprecipitation (ChIP) followed by high throughput sequencing using antibodies against Integrator. We show that, in addition to small nuclear RNA genes, Integrator occupies a large number of protein coding genes, including several Immediate Early Genes (IEGs) such as the FOS and JUN family members. Integrator is recruited to IEGs in a stimulus-dependent manner where it occupies the transcriptional start sites and extends into the body of the genes. Functional analysis revealed that Integrator, while exerting a small effect on basal transcription, plays a pivotal role in IEGs transcriptional activation by epidermal growth factor (EGF). Indeed, depletion of Integrator diminished RNAPII occupancy across the gene body and 3’ end of EGF-responsive genes concomitant with a block in transcriptional elongation. Importantly, Integrator interacts with the Super Elongation Complex (SEC) and is critical for SEC recruitment during transcriptional activation. We propose a role for Integrator as a stimulus-dependent mediator of RNAPII elongation during fast and transient transcriptional responses.


back to the top


Role of a steroid hormone receptor E75 in Drosophila circadian clock
Shailesh Kumar1, Dechun Chen1, Xiangzhong Zheng1, Christopher Jang1, Alexandra Nall1, and Amita Sehgal1,2
1Department of Neuroscience, and 2Howard Hughes Medical Institute, Perelman School of Medicine, University of Pennsylvania

Both circadian clocks and steroid hormone signaling are important for normal physiology, but little is known about molecular links between these processes. Through a gain-of-function screen for novel circadian rhythm genes, we identified a circadian function for a nuclear receptor, Ecdysone Induced Protein 75 (Eip75 or E75), which is induced by steroid signaling. We found that overexpression or knockdown of the E75 gene in clock neurons leads to arrhythmic rest: activity behavior and dampened oscillations of the period (PER) clock protein. Effects of E75 on the clock are mediated through its direct repression of the transcriptional activator, CLK. PER inhibits the activity of E75 on CLK, thereby providing a mechanism for a previously proposed de-repressor effect of PER on CLK. The ecdysone receptor is also expressed in central clock cells and manipulations of its activity parallel the effects of E75 on circadian rhythms. We find that E75 buffers the circadian clock against stressful conditions (nutritional and thermal), suggesting an important function for steroid signaling in the maintenance of clock function and rhythmic behavior.


back to the top


Biased gene conversion skews allele frequencies in human populations, increasing the disease burden of recessive alleles
Joseph Lachance and Sarah A. Tishkoff

Department of Genetics, Perelman School of Medicine, and Department of Biology, School of Arts & Sciences, University of Pennsylvania

Gene conversion refers to the non-reciprocal transfer of genetic information between two recombining sequences, and there is evidence that this process is biased towards G and C alleles. Using high-coverage whole genome sequences of African hunter-gatherers, other global human populations, and primate outgroups we quantified the effects of GC-biased gene conversion (gBGC) on population genetic data. We found that genetic distances (described by FST and population branch statistics) are modified by gBGC. In addition, gene conversion results in modified allele frequency distributions: A/T → G/C SNPs are shifted towards high frequency derived alleles and G/C → A/T SNPs are shifted towards low frequency derived alleles. Summary statistics of site frequency spectra (Tajima’s D, Fay and Wu’s H, and mean derived allele frequency) depend strongly on whether alleles are favored by gBGC. These effects are strongest in high recombination regions of the human genome. By comparing the site frequency spectra of unbiased and biased sites, the strength of gene conversion was estimated to be on the order of Ne*b ≈ 0.009. We also found that shifted allele frequency distributions due to gBGC can result in large (41.0% to 62.8%) differences in the disease burden of recessive alleles. Furthermore, we found that recessive disease burdens are substantially greater for European and Hadza genomes compared to Yoruba genomes. Taken together, our findings reveal that molecular genetic phenomena like GC-biased gene conversion have important population genetic implications.


back to the top


DNA methylation is required for the control of stem cell differentiation in the small intestine
Karyn L. Sheaffer1, Rinho Kim1, Reina Aoki1, Ellen N. Elliott1, Jonathan Schug1, Lukas Burger2.3, Dirk Schübeler2,4, and Klaus H. Kaestner1
1Department of Genetics and Institute for Diabetes, Obesity and Metabolism, Perelman School of Medicine, University of Pennsylvania;
Friedrich Miescher Institute for Biomedical Research, 3Swiss Institute of Bioinformatics, and (4) Faculty of Science, University of Basel, Basel, Switzerland

The intestinal epithelium has a unique organization in which stem cells are harbored in crypts that produce progenitors and finally clonal populations of differentiated cells that are replaced every three to five days throughout adult life. Disrupted maintenance of the intricate balance of proliferation and differentiation leads to loss of epithelial integrity, barrier function, or cancer. There is a tight correlation between epigenetic status of genes and expression changes during differentiation; however, the mechanism of how changes in DNA methylation direct gene expression and the progression from stem cells to their differentiated descendants is unclear. Using conditional gene ablation of the maintenance methyltransferase Dnmt1, we show that reducing DNA methylation causes intestinal crypt expansion in vivo. We established the molecular mechanism underlying this loss of proliferative control through the determination of the DNA methylome in stem cells and their differentiated descendants. We found that DNA methylation is highly dynamic at intestinal enhancer regions during differentiation, and associates with genes important for both stem cell maintenance and differentiation.


back to the top


Defining the role of a single olfactory receptor in odorant perception
Casey Trimmer1, Jason R. Willer2, Andreas Keller3, Leslie B. Vosshall3, Nico Katsanis2, Hiroaki Matsunami2, and Joel D. Mainland1
1Monell Chemical Senses Center; 2Duke University, Durham, NC; 3The Rockefeller University, New York, NY

Approximately 8% of men and 0.5% of women of Northern European descent have a genetic variant in a visual receptor that leads to a perceptual deficit; subjects are “colorblind,” or unable to distinguish between red and green colors. Humans also have a large number of genetic variants in odorant receptors, suggesting that there may be olfactory phenomena analogous to color blindness, but the perceptual effects resulting from this variation are unclear. To address this issue, we asked 321 human subjects to rate the intensity and pleasantness of 68 odors and then sequenced the subjects’ olfactory receptors to examine genetic variation. We identified 35 polymorphisms found in 11 olfactory receptor gene clusters that are associated with alterations in odorant perception. After examining one of these clusters in detail, we found that genetic variation in OR10G4 accounts for a significant portion of the variance in perceived intensity and pleasantness of the “smoky” odor guaiacol. This finding suggests a mechanistic basis underlying the genotype/phenotype association and suggests that the absence of a single odorant receptor lowers the perceived intensity of the receptor’s ligand. By examining additional gene clusters we hope to develop a model relating receptor activation to odor perception.


back to the top


Proposing a technique to shorten scan time in digital breast tomosynthesis
Raymond J. Acciavatti, Predrag R. Bakic, and Andrew D. A. Maidment
Department of Radiology, Perelman School of Medicine, University of Pennsylvania

In digital breast tomosynthesis (DBT), a 3D image of the breast is generated from x-ray projections at various angles. Penn is one of a small number of medical centers in the world to convert its screening practices from 2D digital mammography to DBT. There are two schemes for acquiring projections in DBT: step-and-shoot motion (SSM) and continuous tube motion (CTM). The benefit of CTM is shorter scan time and thus less patient motion; the trade-off is greater focal spot blurring. To optimize the design of the system, we propose a different x-ray tube velocity for DBT. Unlike existing CTM systems with constant tube velocity, we consider a tube velocity that varies with time (e.g., sinusoidally). The tube velocity between projections is large and smoothly approaches zero during each projection to minimize focal spot blurring. To quantify the benefits of the newly proposed design, a contrast-detail array of calcifications was simulated in a computer breast phantom with 25 µm voxels. At a scan time used clinically (3.5 s), we found that existing CTM systems have comparable image quality to SSM systems. By reducing the scan time to 1.4 s, however, calcifications appeared blurry in the CTM design, and exhibited a considerable loss of contrast relative to SSM. The loss of contrast exceeds 10% for 400 µm or smaller objects, and exceeds 50% for 100 µm or smaller objects. Conversely, in the newly proposed design, the calcifications are sharply resolved at a 1.4 s scan time. Also, the loss of contrast is minimal (<8% for 275 µm or smaller objects, <3% for 275 µm or larger objects). In conclusion, this work proposes a technique for operating a DBT system at a short scan time comparable to 2D digital mammography (<2.0 s). The benefit of this design is to minimize the effect of patient motion.


back to the top


Multi-scale dispersal patterns and long-term control of a Chagas disease vector in an urban environment
Corentin M. Barbu1, Karthik Sethuraman1, Jen Manne2, Javier E. Quintanilla Calderón3, and Michael Z. Levy1
1Center for Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of Pennsylvania; 2Department of Global Health and Population, Harvard School of Public Health, Boston, MA; 3Facultad de Ciencias y Filosofia, Universidad Peruana Cayetano Heredia, Lima, Peru

Arequipa, with close to one million inhabitants, is Peru’s second largest city. It is currently undertaking a campaign to control Triatoma infestans, the main vector of Chagas disease. T. infestans mobility has long been linked to human movement, suggesting that separate vector populations in large, interconnected urban centers such as Arequipa may behave as a large connected population. Here, we develop a new spatial model-based methodology to estimate T. infestans migration patterns at the city-block, neighborhood, and city level. We apply this method to spatio-temporal infestation data collected during vector control activities. We estimate that an existing infested household will generate a secondary infestation in a completely susceptible population on average every 1.12 years [0.9-1.3]. We find that the rate of dispersal to neighboring city blocks is on the same order of magnitude as longer-distance dispersal and that both of these are much less common than dispersal within a city block. These estimates are compatible with previously observed auto-correlation patterns of infestation showing a strong barrier effect of streets and with genetic diversity patterns observed using microsatellite markers. The limited impact of distance on the probability of dispersal out of city-blocks suggests that dispersal is highly linked to human movements. In the context of low participation rates in the control campaign (60-85%) along with high infestation prevalence prior to control (10-30%), we discuss the impact of migration on long-term control efforts.


back to the top


Heterozygous inactivation of Gnas induces heterotopic ossification and impairs normal skeletal development
Girish Ramaswamy1,4, Deyu Zhang1,4, Frederick S Kaplan1,2,4, Robert J. Pignolo1,2,4, and Eileen M. Shore1,3,4
1Department of Orthopaedic Surgery, 2Department of Medicine, 3Department of Genetics, and the 4Center for Research in FOP and Related Disorders, Perelman School of Medicine, University of Pennsylvania

Progressive osseous heteroplasia (POH), Albright hereditary osteodystrophy (AHO), and pseudohypoparathyroidism 1a/1c (PHP) form a spectrum of disorders caused by heterozygous inactivating mutations in GNAS, a gene that encodes multiple transcripts including the α-subunit of the stimulatory G-protein (Gsα) of adenylyl cyclase. These disorders exhibit subcutaneous heterotopic ossification (HO); however, POH is the most severe form and is characterized by HO progression into deeper connective tissues including muscle and fascia. The GNAS gene shows genomic imprinting, and POH is associated with paternal inheritance of the mutation. Mice with paternal inheritance of a heterozygous deletion of Gnas exon 1 (KO) have lower body weight and length, and develop subcutaneous ossifications with age. We performed µCT and histology to examine skeletal development in these KO mice. At postnatal days 1 (P1) and 14 (P14), KO mice weighed less than wild-type (WT) littermates. Tibiae from P1 and P14 KO mice were shorter in length (15% ± 4). Trabecular bone, analyzed through µCT scans of the distal epiphyseal region in P14 KO mice, revealed dramatic reductions in bone volume (36% ± 11) and bone volume fraction (20% ± 12). Trabecular microarchitecture was altered with a significant decrease in trabecular thickness and a concomitant increase in the structure model index, suggesting that trabecular bone is more rod-like in the mutants. Femoral mid-diaphysis µCT analyses showed reduced cortical thickness (20% ± 10) and cortical bone volume (35% ± 8) in P14 mutants. P14 KO hindlimb histology showed a marked decrease in the length of the hypertrophic zone in growth plates. Heterozygous paternal allele inactivation of Gnas adversely affects normal skeletal development.


back to the top


Targeted proteomic analyses reveal altered synaptic composition in behaviorally depressed adult female monkeys
Stephanie L. Willard1, Karin E. Borgmann-Winter1,2, Matthew L. MacDonald1, Carol A. Shively3, and Chang-Gyu Hahn1
1Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania; 2Department of Child and Adolescent Psychiatry, The
Children’s Hospital of Philadelphia; 3Department of Pathology/Comparative Medicine, Wake Forest School of Medicine, Winston-Salem, NC

Altered synaptic functioning is implicated in the pathophysiology of depression. Notably, depression is twice as prevalent in women as in men, and studies of synaptic function report sex differences. We recently observed reduced neuropil, postsynaptic density (PSD)-95 and spinophilin in the anterior hippocampus of behaviorally depressed adult female cynomolgus macaques, suggesting that synaptic structure and function may be altered in female depression. Proteins within the PSD microdomain regulate synaptic structure and function, yet the degree to which alterations in the composition of the PSD contribute to depression remains unknown. Thus, we isolated the PSD from the anterior hippocampus of adult female monkeys characterized for behavioral depression. A quantitative analysis of 200+ synaptic proteins was conducted using liquid chromatography-selected reaction monitoring mass spectrometry (LC-SRM/MS) with a neuronal proteome standard prepared from a stable isotope-labeled mammalian (SILAM) brain, a method recently developed by our lab. Disrupted PSD organization was observed, with several proteins differing between depressed females and controls, including those related to cellular adhesion, intracellular signaling, scaffolding and glutamate receptors. The dysregulated proteins were less abundant in the PSD compared to the whole tissue, suggesting that protein availability and targeting to the PSD may be compromised in female depression. Deficits such as these could lead to altered receptor complex formation, and attenuated receptor activation and signaling pathways. These results identify specific proteins whose alterations may severely impact synaptic function, and will be considered as therapeutic targets, particularly for the treatment of depressed women.


back to the top


Inhibition of Peroxiredoxin 6 phospholipase A2 activity reduces lung damage resulting from lung exposure
Bavneet Benipal, Sheldon I. Feinstein, and Aron B. Fisher
Institute for Environmental Medicine, Perelman School of Medicine, University of Pennsylvania

Hyperoxia-induced lung injury results in increased pulmonary permeability, inflammation and respiratory failure. The lung damage is caused by an imbalance between ROS generation and ROS scavenging by antioxidant defense mechanisms. Peroxiredoxin 6 (Prdx6) plays a major role in protecting lung cells against oxidative stress and is the only antioxidant enzyme that has both peroxidase and phospholipase A2 (PLA2) activities. Cell protection by Prdx6 involves peroxidase and PLA2 activities, however, Prdx6 PLA2 activity also activates Nox2, leading to ROS generation. MJ33 is a potent inhibitor of Prdx6 PLA2 activity and blocks agonist-induced activation of Nox2. Studies suggest that unlike the cell model, in which Prdx6 PLA2 is protective, in the mouse model, Prdx6 PLA2 may promote lung damage due to activation of Nox2. Based upon this consideration, we hypothesize that blocking Prdx6 PLA2 activity using MJ33 or by mutation will be protective in a mouse model exposed to hyperoxia. To test this hypothesis, we evaluated the effect of blocking Nox2 in protection against hyperoxia and the effect of MJ33 on Nox2-mediated ROS production in the intact lung following hyperoxia. Three groups of mice were used. The first group was kept at room air and the second and third groups were kept in an O2 chamber (exposed to 100% O2). The second group was injected with PBS and the third group was injected with MJ33 (50 nmoles) at 0 and 48 hours. Survival study data showed that Nox2 KO mice survived longer than WT mice in hyperoxia, suggesting that blocking Nox2 activity is protective against hyperoxia. Lung-injury study data from BALF and lung tissue showed that treatment with MJ33 reduced lung inflammation and permeability and blocked lipid peroxidation induced by hyperoxia.


back to the top


Pathogenesis and therapy of retinal degeneration in a dense deposit disease mouse model
Imran Mohammed1, Delu Song2, Allison M. Lesher1, Lin Zhou1, Takashi Miwa1, Joshua L. Dunaief2, and Wen-Chao Song1
1Department of Pharmacology, and Institute for Translational Medicine and Therapeutics, Perelman School of Medicine, University of
Pennsylvania; 2F.M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, University of Pennsylvania

Age-related macular degeneration (AMD) is the leading cause of blindness in elderly people and is classified as ‘dry’ or ‘wet.’ Complement activation is implicated in dry AMD and individuals carrying single nucleotide polymorphism (SNP) in complement factor H (CFH; Y402H) are at higher risk. Moreover, dry AMD-like drusen deposits are seen in the retinas of patients with dense deposit disease (DDD), a complement-mediated kidney disorder in which patients are prone to blindness. The lack of a dry AMD mouse model has hampered the study of this disease thus far. Here, we have used a newly developed mouse DDD model with retinal injury, which shares some features of human dry AMD, to examine the role of complement in retinal degeneration and test potential anti-complement therapies for this condition. The mutant mouse fHmut/mut/fP-/- bears a nonsense mutation in CFH and is deficient in properdin (fP; a positive complement regulator). These mice develop a DDD-like phenotype and die from renal failure by 3 months of age. Ultra-structural histology of the retina revealed sub-RPE (retinal pigment epithelium) deposits and atrophy of RPE cells with >50% penetrance in male mice. Functional analysis (e.g. electroretinography and ophthalmoscopy) confirmed the retinal injury. Immunofluorescence showed complement deposition at the RPE-Bruch’s Membrane interface, Muller-cell activation and cone-photoreceptor degeneration when compared to age-matched wild type (WT), fP-/- and fHmut/mut mice. Treatment with anti-C5 mAb (1 mg/20 g, twice weekly for 12 weeks) rescued the renal failure in mutant mice (n=8), suggesting key role of complement C5 in DDD pathogenesis. Results from studies of anti-C5 therapy on retinal injury in this model will be presented.


back to the top


The role of collagen organization in determining function of fibrotic skeletal muscle
Lucas R. Smith and Elisabeth R. Barton
Department of Anatomy and Cell Biology, School of Dental Medicine, University of Pennsylvania

Fibrosis is the pathologic accumulation of extracellular matrix (ECM) components within a tissue leading to loss of function. Within skeletal muscle, the ECM not only provides a cell scaffold as in other tissues but also transmits forces produced by muscle fibers. Skeletal muscle undergoes fibrosis in response to many conditions, including muscular dystrophies, which lead to impaired function. The major component of skeletal muscle ECM is fibrillar collagen, the chief load bearing structure within the ECM that also provides passive stiffness critical to skeletal muscle. Here we have used standard and novel measures of fibrosis in skeletal muscle from dystrophic and aged mice to determine their relationship to active and passive muscle function. We tested extensor digitorum longus, soleus, and diaphragm muscles. Standard measures of fibrosis, including collagen area determination and protein content, show an increase of 188% and 101%, respectively, in 18 week old diaphragm muscles. However, the 29% increase in passive stiffness is only loosely correlated with standard fibrotic measures. Collagen fibril density determines the wavelength of birefringent light transmitted from sirius red-stained tissue using circularly polarized light. Using this method, we determined that dystrophic muscle had a higher percentage of loosely packed collagen. Importantly, we demonstrated much better correlation between the percentage of tightly packed collagen and passive stiffness (R2 = 0.54). Current anti-fibrotic therapies under investigation target collagen synthesis, but these targets have not yet been translated into therapies. This work provides targets of anti-fibrotic therapies in muscle to target collagen organization that is disrupted in skeletal muscle fibrosis and more closely linked with impaired function.


back to the top