Molecular and Metabolic Mechanisms of Mechanical Postconditioning in an Isolated Rat Heart Model of Donation After Circulatory Death
1Department of Cardiovascular Surgery, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department of Pharmacology and Toxicology, Comenius University, Bratislava, Slovakia
Introduction: Donor heart availability could be substantially improved with donation after circulatory death (DCD). However, DCD hearts inevitably undergo warm ischemia, which raises concerns about graft function. Mechanical postconditioning (MPC) may promote graft recovery, but underlying mechanisms are not fully understood. Therefore, we investigated the roles of glucose metabolism and key signaling pathways in MPC using a rat heart model of DCD.
Methods: Isolated, working rat hearts underwent 20’ perfusion, either 0’ or 30’ global ischemia, and reperfusion without (control) or with MPC (2 cycles of 30s ischemia, 30s reperfusion). Contractile function (left ventricular (LV) work; developed pressure * heart rate), glycolysis (GLY), glucose oxidation (GO) were monitored over 60’ reperfusion. ATP and phosphocreatine (PCr) content, phosphorylation of key signaling proteins, oxygen efficiency (O2E; LV work/oxygen consumption) and cytochrome c (Cyt c) release were assessed at early reperfusion.
Results: Percentage recovery of LV work was either significantly improved (high recovery=HiR; 57±8%; p<0.05), or decreased (low recovery=LoR; 31±4%; p<0.05) by MPC compared with control (46±9%; n=9-12/group).
GLY rate was higher in HiR vs control (p<0.05) and LoR hearts (p<0.05), but no change in GO rates, ATP or PCr content was observed. Although no differences in Akt pathway activation were detected, phosphorylation of one downstream target, AS160 appeared to be increased in HiR vs LoR hearts. O2E was significantly higher (p<0.05) and Cyt c release tended to be lower in HiR vs LoR hearts.
Conclusion: MPC affects positively, but also negatively, post-ischemic contractile function under these experimental conditions. GLY rates seem to be relevant for a good recovery as does the ability of the heart to efficiently handle available oxygen at early reperfusion. These findings should help to establish effective reperfusion strategies and facilitate DCD heart transplantation.
Interactions of Nanoparticles with the Blood-Brain Barrier
1Division of Veterinary Pharmacology and Toxicology, Vetsuisse Faculty, University of Bern, Switzerland; 2Department of Clinical Chemistry, Inselspital, Bern University Hospital, Switzerland; 3Department of Neurosurgery, Inselspital, Bern University Hospital, Switzerland; 4Institute of Applied Physics, University of Bern, Switzerland
Nanotechnology is used widely in many fields and offers promising possibilities in medicine, e.g. as drug carriers, for personalized cancer therapy or for biodegradable implants. Hence, the effects of nanomaterials have to be carefully characterized in relevant in vitro models prior to use in a clinical setting.
In this study, we examined the effects of nanoparticles (NPs) on endothelial cells used as an in vitro model of the blood-brain barrier (BBB). We assessed the uptake of two types of silica-NPs, i.e. coated with polylactic acid (PLLA) or with poly-ε-caprolactone (PCL), and gold-NPs (all 80 nm) and investigated their effect on cell viability. In addition, the influence on mitochondrial function was explored. After careful evaluation of BBB characteristics of our model, the impact of NPs on BBB-permeability and interference with various pathways relevant for BBB-integrity was studied.
Despite the different surface characteristics, all three types of NPs showed a similar cellular uptake pattern with 85% (PCL, PLLA) and 95% (gold-NPs) after 24 hours of exposure. We observed a concentration-dependent decrease of cell viability (1010 NPs/ml: 40%, 106-102 NPs/ml: less than 20%). Investigations on mitochondrial function demonstrated that NP-exposure increased basal respiration, ATP-synthesis and leak respiration.
Interestingly, an increased BBB-permeability was found after exposure to the lower PCL-concentration. Evaluation of various pathways and proteins important for BBB-integrity showed no involvement of inflammation or disruption of tight and adherens junctions, respectively.
The impact of NPs on mitochondrial function will be further investigated with specific inhibitors.
Established human 2D and 3D BBB models that are closer to the in vivo situation will be needed to test NP-effects on the integrity and function of the BBB and putative related effects on co-cultured brain cells.
LIM Protein Ajuba Promotes Cancer Cell Proliferation and Survival in Hepatocellular Carcinoma
Departement for Biomedical Research, University of Bern, Switzerland
Background: The Lim domain protein Ajuba is a structural protein with a role in the maintenance of cell junctions, cell migration, differentiation and proliferation. We are investigating the role of Ajuba in hepatocellular carcinoma (HCC) and hypothesized that Ajuba expression is increased in HCC cells and that loss of Ajuba has a biological impact on HCC cell growth and survival.
Methods: Ajuba mRNA and protein expression were examined in mouse liver cancer cell lines BNL and RIL-175 and were compared to quiescent mouse liver tissue. Human liver cancer cell lines (HepG2, Hep3B, HLE, HLF, skHep1, Hepkk1, Huh7) were compared to primary human hepatocytes. The biological function of Ajuba was investigated using lentiviruses expressing shRNA sequences targeting Ajuba. Biological and functional assays were then performed with transduced cells in vitro and in vivo.
Results: Steady state levels of Ajuba mRNA were significantly higher in mouse and human cancer cell lines compared to quiescent mouse liver tissue and isolated primary human hepatocytes. ShRNA expressing lentiviruses effectively knocked-down Ajuba protein and resulted in a decreased ability of the cells to migrate and to form colonies. In addition, Ajuba knockdown increased sensitivity to radiation-induced cell damage. In vivo injection of transduced cells show a significantly reduced tumor volume for shAjuba compared to the control RIL-175 using a syngeneic tumor model in C57BL/6 mice.
Conclusion: Our results suggest that Ajuba is highly expressed in proliferating cells and may have a crucial role in cell proliferation and survival. Also in vivo experiments showed that Ajuba knock down has a direct effect on tumor volume. We now aim to define the mechanistic role of Ajuba in cell proliferation and survival in HCC.
CD146 Positive Mesenchymal Stem Cells Possess a Superior Migration Potential Towards Induced Degenerative Intervertebral Discs
1Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 2AO Research Institute Davos, Switzerland; 3Department of Orthopaedic Surgery and Traumatology, Inselspital, Bern University Hospital, Switzerland
Introduction: Recent research is targeting at homing ability of human mesenchymal stem cells (MSCs) for regeneration of degenerative intervertebral discs (IVDs) . While a general homing process could be observed, the migrating MSC sub-population remains unknown. MSC migration described in other musculoskeletal tissues suggests an involvement of CD146 . Therefore, we hypothesized that CD146+ MSCs play a role in cell homing towards the degenerative IVD.
Aims: (1) Quantify migration of CD146+ vs. CD146- MSC towards conditioned medium (CM) from induced-degenerative IVDs; (2) Evaluate migration of CD146+ vs. CD146- MSC populations into induced-degenerative IVDs.
Materials and Methods: MSCs (n=6, 44±21.9y): isolated from bone marrow (KEK 2016-00141), FACS sorted in CD146+ and CD146- populations. Bovine caudal IVDs (n=24, 6-8m): 7 days degenerative loading . Disc culture medium collected daily=CM. Aim1: In-vitro migration assay: top chamber, 146+ or - MSCs (6x104cells/cm2) -> 8um membrane -> bottom chamber, CM=chemoattractant. After 16 h, nr. of migrated cells was counted. Aim2: CD146+ (PKH26) and CD146- (PKH67) MSCs were placed on separate degenerative bovine IVD (1x106cells/IVD). After 5 days, nr. of migrated cells was counted. Data analyzation: paired t-test.
Results: (1) Proportion of MSCs migrating towards CM was significantly higher for CD146+ (22.5 %) compared to CD146- (15.7 %; n=6 p<0.05) sorted MSCs. (2) Organ culture experiment confirmed: CD146+ MSCs with superior migration potential toward induced-degenerative IVD (179.6±29.6 cells/cm2) compared to CD146- (99.2±4.3 cells/cm2; n=3 p<0.05).
Conclusion: CD146+ MSCs showed a superior migration potential compared to CD146- MSCs both in vitro and in whole organ culture. Further studies will assess if the enhanced expression of CD146 is associated with an increased regenerative potential of this sub-population.
 Pattappa G. et al. Eur Cell Mater 2014
 Chang H. L. et al. The Journal of Clinical Investigation 2013
The Role of Iron in the Development and Activity of Osteoclasts
1Department for BioMedical Research, University of Bern, Switzerland; 2Swiss National Center of Competence in Research, NCCR TransCure; 3Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
Iron is the most abundant trace metal in humans. Membrane iron transporters are critical to maintain systemic/cellular iron homeostasis and to avoid iron metabolism disorders such as anemia and hemochromatosis. Recently, we identified an increase in transcripts encoding the Divalent Metal Transporter 1 (DMT1) during osteoclastogenesis in vitro. DMT1 is a membrane transporter for the uptake of ferrous iron, expressed either on the plasma membrane (in enterocytes) or on the endosomal membrane (in other cell types). Most dietary iron is in the ferric form like also the absorbed iron, which is transported through the blood bound to Transferrin. Thus, iron has to be reduced both on the enterocyte apical membrane and in the endosome of other cells type by a ferric reductase to go through DMT1. To understand the roles of DMT1 in osteoclast lineage cells, we generated osteoclast-specific DMT1-ko mice, using the cre/lox recombination system. Cre expression was limited to osteoclast lineage cells (OC) by placing cre under the control of TRAP promoter. The effects of osteoclast-specific DMT1 ko on the phenotype of bone and on the osteoclast lineage cells were analyzed. Bone density and structure were evaluated on lumbar vertebrae (L3-L5) by microCT. Vertebral trabecular thickness was significantly decreased in DMT1(OC)fl/flcre+ mice when compared to the cre- controls. Next, we will examine whether this effect on bone architecture is due to changes in osteoclastic bone resorption or osteoblastic bone formation using histomorphometry. Cell viability and osteoclastogenesis were assessed and DMT1 transcript levels determined by qRT-PCR. No changes in cell viability and osteoclastogenesis were detected despite the 70% decrease in DMT1 transcript levels. In further experiments the subcellular localization of DMT1 and the role of DMT1 in the cellular uptake of iron in vitro as well as the effects of DMT1 ko on OC activity will be analyzed.
Mechanosensitivity of Impulse Conduction in Cultured Cardiomyocyte Strands as Assessed with a Piezo Based Stretching Device
Department of Physiology, University of Bern, Switzerland
Introduction: Mechanosensitive ion channels (MSCs) change their conductance as a function of strain. In the heart, they are suspected to modulate action potential propagation (APP) in a strain dependent manner but data reported are controversial and systematic investigations are missing because of a lack of experimental approaches exhibiting appropriate dynamics.
Methods and Results: To systematically investigate the dependence of cardiac APP on strain over extended time periods, we developed an experimental platform that allows the application of defined levels of strain on cultured strands of cardiomyocytes (CMCs). The system is built around piezo actuators that permit to apply fast stretch protocols to a flexible micro electrode array (FMEA). The FMEA is composed of a thin silicone sheet (50 µm) bonded to glass carriers. Electrodes are implemented using low-energy gold ion implantation. Strands of neonatal rat CMCs are generated on the silicone sheets using lift-off patterning techniques and APP is recorded with in house developed electronics. The system permits to expose CMC strands to highly uniaxial strain of up to ± 12.5% at rates of 150% s-1 thereby covering the range of strains encountered by CMCs in vivo under both physiological and pathophysiological conditions. Using this experimental system, we found that APP in CMC strands exposed to ±12.5% strain (rise time: 80 ms) are only moderately sensitive to strain.
Conclusions: The novel piezo stretcher permits to establish highly reproducible strain-conduction relationships. While contradicting some earlier findings, the observation that physiological levels of stretch/relaxation have only moderate effects on APP supports the concept that changes in strain of the heart during the pump cycle should ideally not overly affect APP.
Ephaptic Coupling in the Heart Is Potentiated by the Distribution of Sodium Channels in Clusters in the Intercalated Disc
1Department of Physiology, University of Bern, Switzerland; 2Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland
Background: Numerous cardiac arrhythmias result from conduction disorders in the myocardium. Recent studies proposed that ephaptic coupling is involved in cardiac conduction and guarantees impulse propagation from one cell to the next when gap junctional coupling is reduced. Ephaptic interactions are mediated by large changes in the extracellular potential in narrow intercellular clefts. Furthermore, Na+ channels form clusters, suggesting that their distribution in intercalated discs may modulate ephaptic interactions. However, the ephaptic coupling mechanism is still controversial since there is no direct proof or refutation of it.
Aim: Our aim was to provide a proof of ephaptic interactions using patch clamp experiments, and to investigate in a high-resolution computer model the influence of Na+ channel distribution on the interactions between the Na+ current (INa) and the extracellular potential in a narrow extracellular cleft.
Results: In patch clamp experiments, approaching a HEK293 cell stably expressing Na+ channels to a non-conducting obstacle systematically increased peak INa at voltage steps around the activation threshold of Na+ channels and decreased peak INa at voltage steps far above the threshold (n=7 cells, p=0.015). We observed the same effects in computer simulations taking ephaptic interactions into account. In the computer model, ephaptic interactions were greatly modulated by the distribution of Na+ channels. Redistributing Na+ channels in clusters in the intercalated disc enhanced ephaptic coupling, especially when gap junctional coupling was reduced and when both cell membranes presented clusters facing each other across the intercellular cleft.
Conclusions: In conclusion, we demonstrated experimentally that cardiac INa is modulated by a restriction of extracellular space as predicted by the model, and our simulations reveal the functional role of the clustering of Na+ channels in the intercalated disc.
Involvement of Müller Glia Cells in Retinal Degeneration/Regeneration in Zebrafish and Mouse
1Department of Ophthalmology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland; 3Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
The aim of my project is to identify and compare the role of Müller glia cells (MCs) during retinal degeneration/regeneration in two different animal models: one with high regeneration capacity (zebrafish) and the other with low regeneration capacity (mouse). Even if it might be difficult to compare MC role between species as mouse and fish inhabit different environment, understanding the mechanisms by which zebrafish can regenerate a damaged retina may suggest strategies for stimulating retinal regeneration in mammals and, in particular, in humans. Furthermore, modulation of endogenous repair mechanism will minimize adverse effects including rejection or tumor formation seen after transplantation of retinal or stem cells. Therefore, activation, proliferation and differentiation of MCs and the pathways involved in these processes will be investigated.
Two Steps Fermentation Methods for Improving the Nutritive Value of Agro-Industrial By-Products (Rice Bran and Wheat Bran) for Poultry Through Anaerobic Fermentation Using Rumen Liquor
Institute of Animal Nutrition, Vetsuisse Faculty Zurich, University of Zurich, Switzerland
Rice bran and wheat bran are the most abundant agricultural by-products used as poultry feeds in many developing countries. However, their utilization in poultry is limited due to high fiber, low protein, and presence of anti-nutritional substances (e.g. ß-glucans, phytic acid). Literature survey shows that no systematic studies have been carried out so far to decrease the fiber content of these brans. Therefore, attempts were taken to develop a method to improve the nutritive value of those brans by reducing fiber and improving protein quality. To achieve these goals, two steps fermentation method was developed where the second fermentation of first fermented dry bran was carried out. In each step, fermentation was performed in an Erlenmeyer flask at a controlled pH (7.0), temperature (39°C) and an anaerobic environment as in the rumen for 3 and 6 hours. The composition of brans, buffer solution (McDougall buffer) and rumen liquor was maintained in a ratio of 1:2:3 respectively in both steps. The fermentation was conducted with 6 samples (n=6) at 6 different days. After fermentation, brans were dried in an oven at 100°C, then fresh brans, as well as the samples after each step of fermentation, were analyzed for DM, CP, CF, ADF, NDF and ADL content. Samples were compared by One way ANOVA followed by Tukey’s multiple comparison tests. Results showed that CP% increased (p<0.05), CF% and ADF% tended to decrease and NDF% (total fiber) was reduced significantly (p<0.05) in both steps. After step 1, NDF was reduced 10.7% (3h) and 17.0% (6h) in wheat bran and 2.3% (3h) and 7.5% (6h) in rice bran. However, after step 2, the decrease amounted to 9.1% (3h), 17.4% (6h) and 9.3 % (3h), 10.0% (6h) in wheat bran and rice bran respectively. It can be concluded from the results that two steps fermentation improved the quality of the brans where time is an important factor. However, further analysis is required to verify that the amino acid profile was improved for human nutrition.
Generation and Characterization of NHA1 (SLC9B1) KO Mice
1Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland; 2Division of Nephrology and Hypertension, University of Bern, Switzerland; 3Department for BioMedical Research, University of Bern, Switzerland
NHA1 and NHA2 are recently cloned sodium/hydrogen exchangers (NHEs) with poorly characterized functions. Together, the two NHAs constitute the SLC9B subfamily and are thus also known as SLC9B1 and SLC9B2, respectively. NHA1 was initially proposed to be a testis specific protein. However, unpublished data of our group indicate that NHA1 expression is not restricted to testis. NHA1 exhibits a wide tissue distribution in mice, including kidney, endocrine pancreas and mature osteoclasts. Furthermore, our analysis of NHA1 expression in tissues of NHA2 KO mice indicates that NHA1 is upregulated upon loss of NHA2 in mice, suggesting that there is a functional interaction between the two NHE isoforms.
To elucidate the biological function of NHA1 and to test for functional complementation between the two NHA isoforms, we generated several polyclonal NHA1 antibodies and NHA1 mutant mice. Antibodies were characterized and validated for IF, IB and IP applications and specificity was verified by peptide blocking experiments, overexpression in HEK cells as well as in tissue samples of WT and NHA1 KO animals. A detailed expression analysis of NHA1 on protein as well as mRNA level was performed in a large murine tissue array. Full body NHA1 mutant mice were generated by breeding mice with floxed exons 5 and 6 of the NHA1 gene locus with ZP3-Cre mutant mice.
Upregulation of NHA1 RANKL-induced osteoclast differentiation was verified by quantitative RT-PCR. We are currently studying osteoclast formation and resorption in vitro of osteoclasts obtained from NHA1 KO mice. Preliminary experiments with NHA1 KO osteoclasts surprisingly revealed that invitro osteoclast-mediated bone resorption is greatly reduced upon loss of NHA1. After obtaining NHA1/NHA2 double KO mice, detailed in vivo and in vitro experiments will be performed to study the role of the two SLC9B transporters in bone homeostasis.
Epicardium Formation in the Zebrafish
1Institute of Anatomy, University of Bern, Switzerland; 2Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain; 3Idiap Research Institute, Martigny, Switzerland
The zebrafish is a well-established model organism for developmental studies. Due to the transparency of the larvae, it is currently one of the best models to study vertebrate heart development in vivo. Similar to human hearts the zebrafish heart is mainly structured in three layers namely endocardium, myocardium and epicardium. Our interest is focused on the development of the epicardium, which plays an important role in nourishment, protection and regeneration of the myocardium. Nevertheless, the knowledge on the outermost layer of the heart is still quite scarce.
A transient structure, the proepicardium (PE), is located in the dorsal pericardial wall. We demonstrate a mechanism that releases PE cells from the pericardium and transfers them to the myocardial surface. These PE cells will start to populate the surface of the heart. We address the topic of layer formation by whole mount immunostainings, showing the localization, distribution and morphology of the cells. Recently, we established a pipeline of imaging and reconstructing the beating embryonic heart over an elongated period. In vivo imaging provides insight into the cell behaviors during epicardium formation.
Optogenetic Modulation of Sleep Slow Wave After Focal Ischemic Stroke
1Center for Experimental Neurology (ZEN), Department of Neurology, Inselspital, Bern University Hospital, Switzerland; 2Department of Neurology, Inselspital, Bern University Hospital, Switzerland; 3Graduate School for Cellular and Biomedical Science, University of Bern, Switzerland
Background: Clinical and experimental studies suggest a favourable effect of sleep on post-stroke neuroplasticity and functional recovery. We hypothesize that stroke causes an increase in sleep-like slow oscillations, characterized by ‘up’ and ‘down’ states in electroencephalogram recordings and that this so-called bistability is critical for recovery from stroke. To directly target the neuronal populations that underlie this bistability, here we combine cell-type specific optogenetic techniques with in vivo electrophysiological recordings to investigate the role of sleep oscillations on brain plasticity following stroke.
Methods: CamkII-ChR2-EYFP(CamKII-ChR2), CamkII-ArchT-EYFP (CamkII-ArchT) and CamkII-mCherry (mCherry, control) adeno-associated viruses (AAV) were injected into the layer V of the forelimb somatosensory cortex of wild type mice. Animals were chronically implanted with optical fiber, electroencephalography (EEG) and electromyography (EMG) electrodes. Experimental ischemic stroke was induced by Middle Cerebral Artery Occlusion (MCAO). Electrophysiology recordings of brain cortical activity and optical stimulations to induce sleep-like down states of neuronal networks were conducted for ten days after stroke induction. In order to assess the effect of optic stimulation on functional outcome several behavioural tests were performed.
Results: Amongst all stimulation protocols tested, we found that both optical activation and silencing of cortical pyramidal cells evoked a down-state of the neuronal network. When the optic modulation of sleep-like slow oscillations was performed within a critical window of time, animals showed greater improvement in fine motor movements when compare to control animals.
Conclusion: Sleep-like slow oscillations have a key role in stroke recovery. Optogenetic modulation of sleep-like oscillation positively affects neuroplasticity and functional recovery, if the stimulation is conducted within a specific critical window of time.
Discovering the Role of Respiratory Supercomplexes in Zebrafish
1Institute of Anatomy, University of Bern, Switzerland; 2Spanish National Center for Cardiovascular Research (CNIC), Madrid, Spain
Mitochondrial activity plays a central role in heart function through its role as an energy and reactive oxygen species (ROS) source. These two mitochondrial functions rely mainly on the oxidative phosphorylation system (OXPHOS), a system extremely conserved along evolution. According to the Plasticity Model, OXPHOS is a dynamic system composed by 5 respiratory complexes (RCs), which can be associated in larger structures called respiratory supercomplexes (SCs) to increase the efficiency of electron flux according to cell demands.
While the zebrafish is a well-established animal model in CVD research given its capability to fully regenerate the heart after tissue damage, its use for mitochondrial biology has until now been limited.
As a first step, we characterized the pattern of RCs and SCs in zebrafish comparing to mouse finding different pattern of assembly between both spices. In general, RCs abound in their monomeric form and the biggest assembled structure, the Respirasome (cIcIII2cIV) seems to be almost absent in zebrafish.
Making use of Crispr/Cas9 technology, we have generated a Supercomplex Assembly Factor 1 (SCAF1) knockout, which allows us to understand the phenotypical consequences of the lack of cIII-cIV supercomplex assembly in zebrafish. We show the lack of two bands containing cIII and cIV in their RCs assembly profile.
Regarding phenotypical consequences of the lack of assembly of cIII and cIV, we show that SCAF1 KO animals are significantly smaller in size and weight than wild type animals. Moreover, females have differences in ovary maturation suggesting a growth retardation. In contrast, early embryo development is not altered as well as heart regeneration capacity.
In summary, we can conclude that SCs assembly is not necessary for basic developmental capacities but they could provide an evolutionary advantage regarding metabolic efficiency.
Exosomes Derived from Wharton’s Jelly Mesenchymal Stem/Stromal Cells Have Anti-Inflammatory Effects on Neuroglia
1Department of Obstetrics and Gynaecology, Inselspital, Bern University Hospital and Department for BioMedical Research, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
Wharton’s jelly mesenchymal stem/stromal cells (WJ-MSC) derived from umbilical cords have the capacity to reduce neuroinflammation in animal models of perinatal brain damage. The therapeutic function of WJ-MSC is mainly ascribed to their complex paracrine machinery involving the release of cell-derived extracellular vesicles such as exosomes. The aim of this study is to evaluate the anti-inflammatory effects of WJ-MSC-derived exosomes on neuroglia in vitro.
The microglia cell line BV-2 and primary mixed glial cells were activated by lipopolysaccharide (LPS). WJ-MSC-derived exosomes were isolated from cell culture supernatants using high-speed and ultracentrifugation. WJ-MSC-derived exosomes pre-stained with the fluorescent cell tracker dye CM-DiI were co-cultured with glial cells to evaluate their cellular uptake. Analysis was done by immunocytochemistry. Glial cells with or without co-cultured WJ-MSC-derived exosomes were evaluated for their production of pro-inflammatory markers by real-time PCR. The effects of WJ-MSC-derived exosomes on the LPS-induced toll like receptor 4 (TLR-4) signaling pathway were evaluated by Western blot methods.
WJ-MSC-derived exosomes co-localize with glial cells. Exosomes further suppressed TLR-4 signaling activation in LPS-stimulated BV-2 cells by preventing the degradation of NF-kB inhibitor alpha (IkBa) and the phosphorylation of extracellular signal-regulated kinases (ERK) (P<0.005). This led to a dampened upregulation of the pro-inflammatory genes tumor necrosis factor (TNF)-α and interleukin (IL)-6 and suppressed the TLR-4 responsive gene IkBa (P<0.05).
In conclusion, we demonstrate that WJ-MSC-derived exosomes have anti-inflammatory effects on neuroglia. Hence not only WJ-MSC, but also WJ-MSC-derived exosomes are able to reduce neuroinflammation. As a result, WJ-MSC-derived exosomes represent a potential cell-free approach to treat perinatal brain damage.
Dysregulation of Lysosomal Compartment Potentiates the Anti-Tumor Effect of Sorafenib by Inhibiting Hepatocellular Carcinoma Progression
1Department for BioMedical Research, Visceral and Transplantation Surgery, University of Bern, Switzerland; 2Novartis Institutes for Biomedical Research, Basel ,Switzerland
Hepatocellular carcinoma (HCC) is one of the most common malignant cancers worldwide, and Sorafenib (SF) is still the first-line treatment for patients with advanced HCC. Alteration in the autophagic flux and in lysosomal compartment and trafficking are often linked to carcinogenesis and tumor progression. Similarly, lysosomal drug sequestration has been shown to induce chemoresistance development. Here we study the effect of the photosensitizer Verteporfin (VP) on HCC growth in a preclinical model without prior light activation, focusing particularly on its mechanism of action alone or in combination with SF. Combining in vitro VP with SF led to a reduction of tumor cell proliferation. It also decreased SF-induced autophagic flux, which was distinctly and differently activated between the two cell lines. Immunoblot analysis showed that VP inhibits the formation of newly forming autophagosomes (LC3-I) and induces an accumulation of high-molecular weight complexes of proteins (HMW-p62). Furthermore, VP co-localizes within lysosomes, increasing their number, altering their shape, size and inducing a strong alkalinization of lysosomal pH (assessed by FACS), which finally leads to lysosomal membrane permeabilization (LMP). Similar results were obtained in a subcutaneous HCC-cell line and patient-derived xenograft (PDX) mouse model. Here, VP potentiated the antitumor effect of SF, by synergistically decreasing tumor cell proliferation (Ki67), tumor angiogenesis (CD31), and the SF-induce autophagic flux, followed by a decreased accumulation of lysosomes. Taken together, these findings suggest that VP, without prior light activation, can significantly potentiate the anti-tumor effect of SF in a HCC preclinical model. VP inhibits the formation of newly forming autophagososmes and—by specifically targeting the lysosomes—concurrently induces their dysregulation and permeabilization.
Effect of Placental Cell Differentiation on Materno-Fetal Nutrient Transport
1Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland; 2Swiss National Centre of Competence in Research (NCCR) TransCure, University of Bern, Switzerland; 3Department of Obstetrics and Gynaecology, Inselspital, Bern University Hospital, Switzerland
During the differentiation of mononuclear cytotrophoblast cells (CTB) into a multinucleated syncytiotrophoblast (STB) layer a variety of morphological and functional changes occur. While proteins regulating the fusion process of CTB are well characterized, alterations in the expression and function of nutrient transporters are only scarcely documented. We hypothesized that syncytialization induces changes in the mRNA expression and functional activity of nutrient transporters. Thus we compared the mRNA expression and function of candidate membrane transporters involved in the transfer of lipids (ABCA1- ABCG1), amino acids (SLC3A2-SLC7A1-SLC7A5-SLC7A6-SLC7A7-SLC7A8-SLC7A11), and iron (SLC11A2-ferroportin-transferrin receptor) before and after syncytium formation.
CTBs isolated from 12 term healthy placentas were characterised by flow cytometry using cytokeratin (CK)-7, CK-18, e-cadherin and vimentin. Isolated CTBs were cultured for four days to form a syncytium and verified by increased expression of the syncytial hormones human chorionic gonadotropin (hCG) and placental lactogen (PL) and elevated hCG secretion into the medium as measured by ELISA. The gene expression of hCG, PL and membrane transporters was assessed by qPCR and normalized to YWHAZ. Pilot experiments were performed to study the functional activity by assessing the uptake of leucine and iron.
Isolated CTBs were highly pure (85-95% positive for CK-7, CK-18, E-cadherin; 3-10% for vimentin). The mRNA abundance and secretion of hCG and PL were significantly higher on day 4 than day 1, confirming the formation of a syncytium layer after four days in culture. The mRNA abundances of ABCA1, ABCG1, SLC3A2, SLC7A1 and SLC7A6 were upregulated while ferroportin and the transferrin receptor were significantly decreased in STB (p=<0.05). Pilot experiments on leucine and iron uptake in CTB and STB suggest different uptake capacities/kinetics and seem to reflect the mRNA data.
In conclusion, we demonstrated that the syncytialization process of CTBs influences the regulation of membrane transporters and consequently impacts on their functional activity.
Efficiency of Preimplantation Genetic Diagnosis of Bovine in vitro Produced (IVP) Embryos Using Blastocoele Fluid or Embryonic Cells
1Clinic of Reproductive Medicine, Vetsuisse Faculty, University of Zurich, Switzerland; 2Center for Clinical Studies, Vetsuisse Faculty, University of Zurich, Switzerland
Preimplantation Genetic Diagnosis (PGD) involves collecting cells from an early embryo, which are used for genetic testing. Palini et al. were able to diagnose the sex human embryos using only the blastocoele fluid as a source of DNA [Palini et al. 2013]. The aim of our work was to compare the efficiency of PGD of bovine IVP embryos using blastocoele fluid or embryonic cells as a source of DNA for sexing the embryos by PCR. Bovine IVP blastocyst-stage embryos were randomly assigned to one of three experimental groups: 1) Collapsed Embryos (CE): blastocoele fluid was collected from embryos, 2) Biopsied Embryos (BE): cells were collected form embryos and 3) Intact Embryos (IE): embryos were left intact. In 1) and 2) the blastocoele fluid or embryonic cells were collected as described previously for equine embryos [Herrera et al. 2014; Herrera et al. 2015]. Collapsed, biopsied or intact blastocysts were vitrified, warmed, cultured in vitro and observed for 48 h to detect reexpansion and hatching. The DNA from the blastocoele fluid or from the cells was amplified by PCR. The survival rates after warming and IVC for 48 h and the efficiency of amplification after PCR were compared by ANOVA and Fisher's exact test between the experimental groups. The post-warming survival rates of blastocysts did not differ (P>0.05) between CE, BE or IE (93.1%, 96.8% and 95.6%, respectively). The hatching rates, did not differ (P>0.05) between BE or IE (75% vs. 47.8%) and between CE and BE, but was higher (P<0.05) for CE compared to IE (79.3% vs. 47.8%). The amplification rate after PCR was higher (P<0.05) for blastocoele fluid samples (41/41, 100%) than for biopsied cells (30/34, 80.3%). The present results demonstrate that blastocoele fluid can be collected from IVP blastocysts and used as a source of DNA for PCR, without impairing the viability of the embryo.
Anti-Fibrotic Effects of Azithromycin Are Enhanced in Human Lung Fibroblasts from IPF Compared to Controls
1Pulmonary Research Laboratory, Department for BioMedical Research, University of Bern, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland; 3Department of Pulmonary Medicine, Inselspital, Bern University Hospital and University of Bern, Switzerland
Introduction: Idiopathic pulmonary fibrosis (IPF) is a fatal lung disease with reduced survival. New anti-fibrotic treatments are approved, which slow down the disease progression, but no cure for IPF is available. Azithromycin (AZT) is an antibiotic agent that belongs to the group of macrolides and recently anti-fibrotic effects involving autophagy have been shown. The effect of AZT on fibroblasts from IPF compared to controls is unknown and the exact mechanisms need to be investigated.
Methods: Primary control human lung and IPF fibroblasts were exposed to TGF-β (5ng/ml), AZT (50μM) alone or combined prior to expression analysis of Collagen1α1 (Col1) by real-time qPCR. Col1 ELISA assay was used to determine Col1 secretion. Cytotoxicity was measured by LDH analysis. Microarray screening determined involved pathways.
Results: Col1 gene expression was significantly reduced after AZT in control and IPF fibroblasts (p-value≤0.0095, p-value≤0.001). After TGF-β, AZT induced a significant reduction in Col1 secretion compared to TGF-β alone (p-value≤0.01) in IPF, but not in control fibroblasts. The majority of IPF fibroblasts treated with AZT produced higher LDH levels compared to controls. Microarray analysis revealed increase of ATP6V1B2 gene expression after AZT treatment and suggests involvement of lysosomal ATPases.
Conclusion: AZT reduced Col1 gene expression after stimulation with TGF-β in IPF and in control fibroblasts. However, the effects were enhanced in IPF compared to control fibroblasts with regards to Col1 secretion. Increased cytotoxicity in IPF and lysosomal ATPase alterations might be involved.
Cellular and Molecular Mechanisms Directing T-Cell Migration Across the Blood-Brain Barrier
1Theodor Kocher Institute, University of Bern, Switzerland; 2Interfaculty Bioinformatics Unit, University of Bern, Switzerland
The endothelial blood-brain barrier (BBB) maintains central nervous system (CNS) homeostasis and strictly controls T-cell trafficking into the CNS. During multiple sclerosis (MS) or its animal model, experimental autoimmune encephalomyelitis (EAE), autoaggressive CD4+ interferon-g-producing Th1 or IL-17-producing Th17 cells enter the CNS causing neuroinflammation. Using live cell imaging and primary mouse brain microvascular endothelial cells (pMBMECs) as an in vitro BBB model, we previously characterised the multi-step extravasation of Th1 cells across the BBB. Two distinct migration pathways for T cells have been observed: paracellular diapedesis through the endothelial junctions and transcellular diapedesis through the endothelial cells. In particular, we identified ICAM-1 as an essential molecule for T-cell crawling over pMBMECs. Moreover, different surface levels of endothelial ICAM-1 direct Th1 cell to transcellular or paracellular sites for diapedesis. It has been suggested that Th17 cells may use mechanisms different from Th1 cells to cross the BBB. This prompted us to directly compare the cellular and molecular mechanisms used by these two CD4+ T-cell subsets to migrate across the BBB. To address this, we have compared the multi-step extravasation of in vitro polarized Th1 and Th17 cells under physiological flow by in vitro and in vivo live cell imaging. Preliminary observations indicate that Th1 cells have higher capacity to arrest on the BBB. Interestingly, both subsets rely on ICAM-1 for crawling on pMBMECs in vitro, and inflamed BBB in vivo. Both subsets can cross the BBB under inflammatory conditions, with Th1 cells being more efficient than Th17 in in vitro conditions. Identification of distinct molecular mechanisms mediating Th1 and Th17 cells’ migration across the BBB will allow us to identify novel therapeutic targets within the successful framework of therapeutic targeting of immune cell trafficking to the CNS for the MS treatment.
An Integrated Analysis of microRNA and mRNA Expression Profiles from Horses with Severe Equine Asthma
1Swiss Institute of Equine Medicine, Department of Clinical Veterinary Medicine, Vetsuisse Faculty, University of Bern, Switzerland; 2Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland; 3Faculty of Veterinary Medicine, Université de Montréal, Canada
Severe equine asthma is an incurable obstructive respiratory condition affecting up to 20% of horses in temperate climates. Upon exposure to airborne antigens from hay feeding, affected horses show neutrophilic airway inflammation and bronchoconstriction, which leads to lower exercise tolerance and increased respiratory effort. Severe equine asthma has therefore far reaching implications, not only for the individual animal and its welfare but also for equestrian sports and horse breeding.
Our study aims at investigating potential regulatory networks underlying the pathophysiology of the disease by profiling microRNAs (miRNAs) and mRNAs in lung biopsies from affected horses. MiRNAs regulate post-transcriptional gene expression by binding mRNAs, which either leads to the degradation of the target or translational repression. MiRNAs are key regulators in most biological processes and their deregulation has been shown to play a role in the pathology of many diseases.
In order to perform miRNA and mRNA expression profiling, we extracted both small RNAs and mRNAs from lung tissue samples of symptomatic horses (n=7), affected asymptomatic horses (n=5) and healthy control horses (n=8). As a first step, we identified novel equine miRNAs with the tool miRDeep2. Afterwards, the combined set including novel and known miRNAs was subjected to differential miRNA expression analysis with DESeq2.
As a result, we identified 8 significantly differentially expressed miRNAs between symptomatic horses and controls and 20 significantly differentially expressed miRNAs between asymptomatic and healthy horses (adjusted P < 0.05). Differential mRNA expression analysis revealed 204 deregulated mRNAs between symptomatic animals and controls and 648 deregulated mRNAs between asymptomatic and healthy horses (adjusted P < 0.05). Currently, the miRNA and mRNA expression profiles are being integrated applying a negative correlation approach to identify regulatory networks of severe equine asthma.
Characterization and Implementation of Intra-Body Communication for Leadless Cardiac Multisite Pacemaker Systems
1Department of Cardiology, Inselspital, Bern University Hospital, Bern, Switzerland; 2ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland; 3Institute for Human Centered Engineering, Bern University of Applied Sciences, Biel, Switzerland
Recently introduced leadless cardiac pacemakers effectively overcome all lead-related limitations of conventional pacemaker systems. However, these devices only feature single-chamber pacing capability although dual-chamber pacing is highly desirable due to physiologic reasons. Implanting e.g. a leadless pacemaker into the right atrium and a second one into the right ventricle would enable leadless dual chamber pacing but requires wireless communication for device synchronization. A power-efficient wireless communication method is crucial to preserve the pacemaker’s longevity (modern pacemakers consume only 5-10 μW of power).
As communication method, we use intra-body communication. This method uses the electrical conductivity of blood and tissue: the data from one device is modulated and applied as a small alternating current signal to the myocardial tissue and blood via electrodes. The signal can be registered almost simultaneously by the other device. To characterize the electrical properties of the myocardium and the blood, we performed in-vitro and in-vivo measurements on porcine hearts. Additionally, we developed a finite element model that serves as a tool to compute e.g. impedance and transfer function for different geometrical dimensions and electrode arrangements in the heart. By the reproduction and measurement of simple model geometries on the bench with real tissue and blood, the model was validated.
The computed results for impedance and transfer function are in good accordance with the measured values. The findings of the tissue characterization were used to develop pacemaker modules, which can perform bidirectional communication in the heart. For device synchronization via wireless communication, less than 1 μW average power was applied to the tissue. Therefore, intra-body communication is a promising ultra-low power wireless communication method for multisite pacemakers.
Time Resolved OCT Images Evaluation for Online Control of Selective Retina Therapy
1ARTORG Center for Biomedical Engineering Research, Ophthalmic Technology Laboratory, University of Bern, Switzerland; 2Department of Ophthalmology, Inselspital, Bern University Hospital, Switzerland
Selective retina therapy (SRT) is a laser treatment targeting specific posterior retinal layers. It is focused on inducing damage to the retinal pigment epithelium (RPE) layer, while sparing other retinal tissue compared to traditional photocoagulation. The RPE layer is the main light absorbing structure of the retinal tissue, and is responsible for the nourishment of the retinal visual cells. However, the targeted RPE layer is invisible with most imaging modalities and induced SRT lesions can not be monitored. The aim of this work is to utilize time-resolved optical coherence tomography (OCT) images (M-scans) acquired during the SRT laser application, in order to provide a feedback mechanism for the treatment.
An SRT (treatment) and an OCT(imaging) system are aligned and guided to the retina throught an optics system. The OCT signal that is acquired comprises of a time series of retinal depth profiles of the treatment spot. By evaluating those sequences we aim to infer the treatment progress, and automatically stop the laser application according to the deposited energy.
For the experiments, ex-vivo enucleated porcine data are used. 968 scans are acquired from roughtly 100 different eyes. The eyes are shot with different energy levels and their corresponding OCT scans are analyzed. For the analysis, we isolate a small part of the long M-scan, that can be assumed to be motion-free, and therefore belonging to the same spot on the retinal surface. We train a convolutional neural network (CNN) to classify this short M-scan as corresponding to treated or not treated tissue.
The system showed promising results in this classification task. As it is able to evaluate the accumulated treatment effect by only seeing a short after-treatment or during-treatment M-scan, it can be used for online evaluation and automatic stopping of the SRT laser application when the desired effect is achieved.
Simultaneous Recognition and Pose Estimation of Instruments in Minimally Invasive Surgery
1ARTORG Center for Biomedical Engineering Research, Ophthalmic Technology Laboratory, University of Bern, Switzerland; 2cole Polytechnique Fédérale de Lausanne, Switzerland; 3University College London, United Kingdom; 4Department of Ophthalmology, Inselspital, Bern University Hospital, Switzerland
Detection of surgical instruments plays a key role in ensuring patient safety in minimally invasive surgery. In this paper, we present a novel method for 2D vision-based recognition and pose estimation of surgical instruments that generalizes to different surgical applications. At its core, we propose a novel scene model in order to simultaneously recognize multiple instruments as well as their parts. We use a Convolutional Neural Network architecture to embody our model and show that the cross-entropy loss is well suited to optimize its parameters which can be trained in an end-to-end fashion. An additional advantage of our approach is that instrument detection at test time is achieved while avoiding the need for scale-dependent sliding window evaluation. This allows our approach to be relatively parameter free at test time and shows good performance for both instrument detection and tracking. We show that our approach surpasses state-of-the-art results on in-vivo retinal microsurgery image data, as well as ex-vivo laparoscopic sequences.
Time-Series Learning Through Hidden Population Shaping by Somatic Nudging
Department of Physiology, University of Bern, Switzerland
Many behavioural patterns, like walking or memorization, are grounded in the production of neural time-series. In order to shed light on how biological circuits solve this problem, models of recurrent neural networks can be used. Many algorithms have been proposed for solving temporal learning tasks, like FORCE learning or Back-propagation through time. However, how these could be implemented in a biological neural network remains unclear.
It is the aim of this work to propose an approach to temporal sequence learning that is biologically realistic, by using a suitable neuron model and learning rule as proposed by [Urbanczik and Senn, 2014]. In this model the dendritic synapses adapt in such a way that the dendrite correctly predicts the somatic membrane potential. The somatic potential is nudged towards a desired value, through inputs proximal to the soma. After learning the dendrite can produce this desired value independent of somatic nudging. The network consists of two neuron populations. The visible population, reproducing the desired temporal sequence and the hidden population, providing a memory of past neural activity. In the visible population somatic nudging represents the desired time-series and in hidden neurons nudging is provided either by visible neurons or by other hidden neurons. Dendritically the network is all-to-all connected. In the case of successful learning, the visible neurons can reproduce the teacher without external input.
This setup is capable of learning non-Markovian patterns as well as time-series with long activity gaps. The memory extension of the visible population, necessary for these tasks, is due to the formation of delay lines and loops in the hidden population. Both structures can keep information alive over long periods without any visible activation. Hence, we conclude that our biological plausible approach is successful in shaping the hidden population for time-series prediction.
Prediction of Outcome by Imaging: An Automated Approach for Lesion Load Mapping
1Support Center for Advanced Neuroimaging (SCAN), Department of Diagnostic and Interventional Neuroradiology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Institute for Surgical Technology & Biomechanics, University of Bern, Switzerland
Stroke is a major neurological disease with respect to life impact and incidence. In Switzerland approximately 16’000 persons are subjected to acute ischemic stroke every year, with every fourth remaining disabled and unable to live independently. Hence, the search for appropriate surrogate biomarkers constitutes a paramount step towards accurate prediction of outcome. Recovery from ischemic stroke depends on a variety of individual factors, e.g. lesion location and extent. However, the impact of focal damage on eloquent areas is still a field of extensive research. We have thus aimed to develop tools for lesion load mapping to i) quantify lesion load and ii) to predict outcome using recently developed tools for quantification of the tissue at risk in case of successful vs. unsuccessful thrombectomy. Two steps are mandatory to perform this kind of analysis, i) region-wise mapping of the cortex and white matter and ii) searching for the optimal clinical parameters to correlate them with outcome. We use a histological atlas encompassing grey and white matter structures to create a normalized map of the lesion load in the areas under investigation. We further compared the impact on lesion load on 2 different clinical outcome scales, NIHSS and mRS after 3 months of recovery.
The correlation between the lesion loads onto the defined brain regions and the three month outcome scores indicated that the NIHSS provides a more accurate correlation with lesion loads in eloquent brain areas compared to the mRS. A stratification into two cohorts, based on revascularization success improved the correlation analysis substantially. Using our in-house developed fully automated prediction software FASTER (McKinley, 2016), we were able to demonstrate that lesion load to outcome correlation can be improved using prediction maps compared to simple diffusion based segmentations.
Does Cartilage Composition Change After Joint Preserving Hip Surgery for FAI—Preliminary Data of a Prospective Comparative Longitudinal Study Using Delayed Gadolinium Enhanced MRI of Cartilage (dGMERIC) at 3 T
1Department of Orthopedic Surgery and Traumatology, Inselspital, Bern University Hospital, Switzerland; 2Institute for Surgical Technology and Biomechanics, University Bern, Switzerland
Purpose: Delayed gadolinium enhanced MRI of cartilage (dGEMRIC) allows a 'monitoring' of glycosaminoglycan depletion of cartilage and can be used as biomarker for osteoarthritis. Longitudinal data on the change in hip cartilage composition in patients with femoroacetabular impingement (FAI) is lacking. We aimed to assess whether changes in T1 values occur 1 year after FAI surgery compared to patients with non-operative treatment and if changes correlate with clinical short-term outcome.
Methods: IRB-approved prospective, comparative, longitudinal study of two groups (61 hips, 55 patients). After exclusion 17, 18 hips were assigned to the `operative` (open/arthroscopic correction) respectively to the `non-operative` (conservative treatment) group. At baseline, groups were comparable regarding radiographic osteoarthritis (Tönnis score), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and T1 values. Patients had baseline and 1-year follow-up i.v. dGEMRIC scans with the same 3 T magnet (0.4 ml/kg, 0.2 mmol gadopentate dimenglumine /kg). Radial T1 maps were reformatted from a 3D dual flip-angle volume interpolated breathold examination (VIBE). Central, peripheral femoral (12 `hour´ positions) and acetabular (10 `hour` positions) T1 values were manually measured separately from each other with the help of 2D radial high-resolution PD-w images. Baseline and 1-year follow-up WOMAC scores were obtained. Statistical analysis included paired/unpaired Student's t-Tests (p<0.05).
Results: Acetabular T1 values decreased significantly (p<0.05) in 17/20 (85%) zones and in 21/24 (88%) of femoral zones in the operated group. In the non-operative group, no acetabular zone and 2/24 (8%) femoral zones presented with a significant drop. After 1 year the WOMAC significantly improved (58±42 to 33±42; p= 0.007) for the operative group, while there was no significant change (55±45 to 48±50; p= 0.825) for the non-operative group.
Conclusion: FAI surgery led to a decline in cartilage composition on MRI after 1 year despite significant clinical improvement. This decline may result from postoperative inflammation and/or may reflect altered biomechanic stress within the cartilage and requires longer follow-up.
What Is the Optimal Voxel Size for Single Voxel Magnetic Resonance Spectroscopy in Global Brain Pathology?
Department for BioMedical Research, Magnetic Resonance Spectroscopy and Methodology, University of Bern, Switzerland
Introduction: Achieving high quality spectra for quantitative evaluation of metabolite content is a challenge in clinical Magnetic Resonance Spectroscopy (MRS), where a high Signal-to-Noise-Ratio (SNR), but also good resolution is essential. For non-focal disease, the voxel size (VS) can be increased to achieve better SNR and hence smaller fitting inaccuracies, reflected in Cramer-Rao Lower Bounds (CRLB). For very large VS, spectral resolution deteriorates and outer-volume artifacts arise, explaining the need to compromise between increasing SNR and decreasing linewidth under the additional condition of minimal artifacts. Plus, multichannel receive arrays offer different spatial views on the overall signal, so optimizing combination of subspectra is of interest.
Methods: 6 volunteers at 3T; 8 different VS; accurate/misset shim simulating limited field homogeneity.
Results & Discussion: CRLB decrease with VS to a minimum and increase for larger VS, hinting at an optimal VS of ~60 cm3. Including lineshape information improves fitting accuracy.Spectral quality and peak shapes obtained from unsuppressed water can be vastly varying for different coil channels, indicating that simultaneous fitting of subspectra with inclusion of lineshape information may be superior to weighted data combination before fitting the summed spectrum. Alternatively, signals from coil elements with large artifacts can be omitted from the average, costing signal intensity and leading to increases in CRLB, which can be mitigated by use of averaged lineshapes.
Conclusions: CRLB minimum at VS of ~60 cm³. Combination of multi-channel data should also be based on artifact levels and lineshapes of subspectra from single coil elements. Elimination of subspectra with bad quality may increase overall spectral quality. Inclusion of lineshape information promises to extend the useful range of VS. Simultaneous fitting of subspectra enforcing lineshape and coil sensitivity is promising for future studies.
Sequentially Optimized Reconstruction Strategy: A Meta-Strategy for Perimetry Testing
ARTORG Center for Biomedical Engineering Research, Ophthalmic Technology Laboratory, University of Bern, Switzerland
Perimetry testing is an automated method to measure visual function and is heavily used for diagnosing ophthalmic and neurological conditions. Its working principle is to sequentially query a subject about perceived light using different brightness levels at different visual field locations. At a given location, this query-patient-feedback process is expected to converge at a perceived sensitivity, such that a shown stimulus intensity is observed and reported 50% of the time. Given this inherently time-intensive and noisy process, fast testing strategies are necessary in order to measure existing regions more effectively and reliably. In this work, we present a novel meta-strategy which relies on the correlative nature of visual field locations in order to strongly reduce the necessary number of locations that need to be examined. To do this, we sequentially determine locations that most effectively reduce visual field estimation errors in an initial training phase. We then exploit these locations at examination time and show that our approach can easily be combined with existing perceived sensitivity estimation schemes to speed up the examinations. Compared to state-of-the-art strategies, our approach shows marked performance gains with a better accuracy-speed trade-off regime for both mixed and sub-populations.
In vitro Biomimetic Models of Brain Microvasculature to Investigate Red Cells Phase Separation and Blood Rheology
ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland
In networks with complex topology, such as the cerebral ones, the partitioning of red blood cells (RBC) between daughter branches plays a key role in the local perfusion because it is the most important determinant of RBC heterogeneity inside the network. Moreover, RBC velocity and hematocrit (HT) affect the blood apparent intrinsic viscosity (KT) and pressure/flow field in the whole network.
The research project aims at developing an in vitro biomimetic model of the brain microvasculature to study the RBC phase separation. In this study we show how two methods, Particle Tracking Velocimetry (PTV) and Line Scan Method (LSM), can be applied to measure the motion of RBCs and derive micro-rheological properties of blood. PTV is based on the analysis of a RBC image sequence; it involves the following steps: detection of RBCs, establishment of connection between matching RBCs in different frames and extraction of motion parameters. In LSM, a line of pixels is scanned out from each frame of a video recording and then all the line scans are stacked in a temporal sequence. The number of RBCs is computed by counting the line traces in this stack and their velocity is the slope of these traces.The methods comparability was tested and we found that PTV and LSM are able to trace the motion of RBCs and they provided very similar results.
As a test case in which LSM has been applied, we reported here an experiment to study the apparent intrinsic viscosity of blood in function of RBC velocity and hematocrit. The PDMS microdevice had a single straight channel of 500μm length with square cross-section (8μm x 8μm) and we used RBCs suspensions with HT=2.5% and HT=5%. To drive the flow, two hydrostatic pressure differences were set between the reservoir and the outlet of the device (4.5 cmH20 and 7.5 cmH20). We found a significant decrease of KT for increasing hematocrit but we did not find a sufficient statistical evidence to prove that KT changes in function of the RBC velocities.
Fast Learning Without Forgetting by Synaptic Consolidation
Department of Physiology, University of Bern, Switzerland
For some synaptic plasticity protocols the induced change in the synaptic strength decays partially back with a time constant of minutes before it reaches the long-term value. This early phase of long-term plasticity can be observed both for long-term potentiation (LTP) and for long-term depression (LTD).
We hypothesize that the synaptic weight overshoot of early LTP is advantageous for learning in an environment where one context follows another. In different contexts, similar sensory stimuli may require different responses. Consolidating all synaptic weight changes in one context may therefore cause learning interferences in a next context and lead to the forgetting of previously learned stimulus-response associations.
Early LTP and its consolidation process is one of nature’s choices to deal with the plasticity-stability dilemma. To keep the benefit of strong synaptic plasticity while avoiding context interferences, the network should only consolidate a minimum number of synapses, but still be able to change enough synapses if required by novel learning. We show in the framework of reinforcement learning that fast synaptic plasticity which decays to zero when the postsynaptic neuron is not active, solves this conflict. The activity of the postsynaptic neuron sets a tag that protects synapses from forgetting. The plasticity rule is derived from gradient descent procedure of a cost function which is composed of the expected reward and a penalty term that punishes strong synaptic weights and large voltage deflections. The rule gives a functional explanation for the biological phenomena of early/late LTP and synaptic tagging.
Examination of True and False Predictions of a Quantitative EEG Method to Support Epilepsy Surgery
1Support Center for Advanced Neuroimaging (SCAN), Department of Diagnostic and Interventional Neuroradiology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department of Neurology, Inselspital, Bern University Hospital, University of Bern, Switzerland
Resection of the brain areas causing seizures is a treatment option for pharmacoresistant epilepsy patients. Quantitative analysis of intracranial EEG signals may support clinicians by suggesting brain regions for resection. Because only the effect of actually performed resections on post-surgical seizure-control can be known, the validation of such methods is hindered by the shortage of real data. Regarding a prospective application in the future it is essential to further evaluate their accuracies. Therefore, we identify in this study resections predicted as highly beneficial according to a recently developed quantitative method and check their medical plausibility and applicability.
Identification of a patient’s optimal resection is a combinatorial optimization problem and, due to the method’s inherent nonlinearity, cannot be done by an exact algorithm in feasible time. We thus implemented several nature-inspired metaheuristic optimization algorithms to efficiently obtain approximately optimal solutions. After appropriate adaptation of the algorithms to the specific problem, they all found solutions which can be assumed to be among the global best solutions. However, the contacts of these solutions are typically distributed over many of the implanted electrodes and do not coincide with known successful resections above chance level. Since epilepsy is considered a network disease, we cannot exclude these resections would be successful but due to their spatial dispersion they are medically hardly justifiable and consequently highly unlikely to get applied in clinics.
In the literature, the assessment of such quantitative methods is confined to the very limited amount of data about performed surgeries. We investigated in this study in more depth the amount of false positive predictions (type I error) a method produces. We conclude its current version to be prone to type I errors. The same analysis remains to be done for similar methods to complement their assessments.
How Can Antibiotic Consumption and Mortality Be Reduced in the Swiss Veal Industry? A Novel Approach Reconsidering Transport and Housing
1Clinic for Ruminants, Vetsuiss Faculty, University of Bern, Switzerland; 2Veterinary Public Health Institute, Vetsuisse Faculty, University of Bern, Switzerland
Antibiotic consumption and mortality are important issues in calf-fattening facilities, causing bacterial resistance and economic loss. Reasons for this are prevailing working methods which impair calf health, including commingling, crowding and suboptimal barn climate. Pneumonia is reported to be the main cause for the consumption of antibiotics.
The present study proposes a three-part solution, which will be scrutinized on 20 farms. Calves are transported directly among neighbors from the farm of birth to the fattening facility instead of commingling calves in livestock dealer trucks. Outdoor hutches are used for the entire fattening period of four months. Consequently, all calves have unlimited access to fresh air and are less exposed to gas emission from the bedding, such as ammonia. Specific management measures are taken to support calf health. For instance, each calf completes a three-week quarantine in an individual hutch, is subsequently fattened in groups not exceeding 10 calves and is vaccinated against predominant lung pathogens.
The motives for this study are the growing threat of antibiotic-resistant bacterial infections in veterinary and human medicine as well as the declining reputation of veal with its economic consequences.
Evaluation of Management Factors and Personality Profiles of High-Level Sport Horses and Amateur Horses Regarding Their Association with the Response to an ACTH Stimulation Test
1Swiss Institute of Equine Medicine, Department of Clinical Veterinary Medicine, Vetsuisse Faculty, University of Bern, and Agroscope, Bern, Switzerland; 2Niederlenz, Switzerland; 3Animal Welfare Division, Veterinary Public Health Institute, Vetsuisse Faculty, University of Bern, Liebefeld, Switzerland; 4Division of Veterinary Physiology, Vetsuisse Faculty, University of Bern, Switzerland; 5Unit of Physiology, Pathophysiology and Experimental Endocrinology, Department of Biomedical Sciences, University of Veterinary Medicine, Vienna, Austria
We recently showed that moderate/severe Equine Glandular Gastric Disease (EGGD) is associated with a quicker, exaggerated cortisol release following an ACTH stimulation test (Sauer et al., submitted; Scheidegger et al., 2017). In another previous study, crib-biting horses also showed an increased response (Freymond et al., 2015).
The aim of the present study was to determine if temperament traits and management factors affect cortisol levels after ACTH stimulation test in sport and amateur horses that compete in different disciplines.
The field study lasted from January 2017 until August 2017, with a total of 149 horses included. Ninety-five high-level sport horses of the disciplines show jumping (n=17), dressage (n=14), eventing (n=20), endurance (n=20), para-equestrian (n=5), combined driving (n=14) and vaulting (n=5) and 54 amateur horses of the disciplines show jumping (n=18), dressage (n=16), eventing (n=9) and endurance (n=11) participated.
For the evaluation of the character of the horse, we adapted a validated questionnaire created by Momozawa et al. (2005). All questions about management factors, stereotypies and clinical signs were taken from other validated questionnaires used in our clinic. We added some further questions, concerning for example recovery times or sleeping behavior of the horse.
Each participating horse was visited at home in its normal environment and underwent an ACTH stimulation test. Saliva samples were collected before and 60 and 90 minutes after the intravenous injection of a dose of 1 µg/kg BW of synthetic ACTH1-24. In addition to the test, we collected fecal samples and riders of the horses completed the questionnaire.
The statistical evaluation process is still going on. Results and conclusions will be presented at the GCB Symposium.
Heart Rate Variability in Horses During Exercise
1Clinic for Equine Internal Medicine, Equine Department, Vetsuisse Faculty, University of Zurich, Switzerland; 2Department of Veterinary Clinical Sciences, Swiss Institute of Equine Medicine, University of Bern, Switzerland
The aim was to quantify beat-to-beat (B2B) heart rate variability (HRV) in horses during both standardised high-speed treadmill exercise and ridden exercise tests.
Study 1 (HSET) was performed in 30 healthy horses (Group H), 9 horses with cardiac disease (Group C) and 11 horses with non-cardiac causes of poor performance (Group O). Study 2 (RET) was performed in 9 horses used for pleasure riding (Group P) and 23 horses used for eventing competitions (Group E). Velocity, heart rate, lactate concentration and ECGs were recorded during each exercise test. The ECGs were corrected for artefacts and arrhythmias noted. HRV analysis was performed on each segment of the exercise test (trot, gallops, recovery) and the % B2B variability was calculated. Performance indices (velocity at lactate 4 mmol/L and velocity at HR 200 min-1) were calculated.
In Study 1 there was no difference in performance indices between the 3 groups. The % B2B variability was narrow (median % variation – 0% for all groups and time points). Only a small number of beats varied from each other (1st percentile RR shortening gallop, mean± SD %: Group H -4.9 ± 1.8, Group C -4.3 ± 1.8, Group O -3.8 ± 1.6; p=0.348) and this was similar between the 3 groups at each time point (p=0.073).
In study 2 Group P had lower performance indices than Group E. The % B2B variability was narrow (median % variation – 0% for all groups and time points). Only a small number of beats varied from each other (1st percentile RR shortening gallop 1, median % (range): Group P -3.8 (-5.5 to -2.4), Group E -3.6 (-15 to -1.9), p=0.574) and this similar between the gallops phases (p>0.05).
Overall, B2B variability during exercise was very low in both studies. Fitness level and the presence of cardiac disease did not influence HRV during exercise. Arrhythmias occurred infrequently in all groups. These arrhythmias did not affect B2B variability although the timing of arrhythmias were easily visualised in the HRV graphical analysis.
Creatures of Habit? Laying Hens Show Consistent Behavior Patterns in Aviary Systems
1Division of Animal Welfare, VPH Institute, Vetsuisse Faculty Bern, Universitty of Bern, Switzerland; 2VPH Institute, Vetsuisse Faculty Bern, Universitty of Bern, Switzerland
Aviaries are the most common housing system for laying hens in Switzerland. In this system, resources (food, water, perches, nest box, litter) are allocated to functional areas differing in height and structure. Since the resources provided don’t allow synchronized behavior, hens might be challenged to adapt their behavior in order to meet their demands. Therefore, the development of individual movement patterns is likely. In this study, an aviary system housing 225 laying hens was equipped with infrared emitters dividing the aviary into five different levels (litter, lower tier, nest boxes, upper tier, and outdoor area) sending out specific infrared signals for each zone. Infrared receivers were mounted on 15 focal birds, recording their location with a frequency of 1 Hz over seven days. The number of overall level changes per day was highly variable between birds, ranging from 27.3±5.9 to 129.3±21.2 (mean: 80.3±43.6), whereas individual coefficients of variation (CV) were rather low (ranging from 0.09 to 0.30, mean: 0.17±0.06). The number of level entries was highest (27.7±11.4) and mean CV was lowest (0.15±0.05) in the litter level. The upper tier was visited least often (3.9±4.4). Hens could be assigned to three groups depending on their nest box visiting behavior: daily visit every 24h; daily visit every 25.5h with a break of 1-2 days every 6-10 days; and no nest box visits longer than 15min. The time of the last level change before night time was consistent within most birds but varied considerably among individuals (range: 14:05h±00:28h to 16:59±00:13h). Also, most birds were consistent in their choice for a specific level for night time roosting. These data revealed individual movement patterns, consistency within and variation between laying hens. It offers the opportunity to study not only widely known welfare issues like keel bone fractures or feather damage, but also potential limitations of the housing system.
Mood as Cumulative Expectation Mismatch: A Test of Theory Based on Data from Nonverbal Cognitive Bias Tests
1Centre for Proper Housing of Ruminants and Pigs, Federal Food Safety and Veterinary Office, FSVO, Agroscope, Ettenhausen, Switzerland; 2Division of Animal Welfare, Veterinary Public Health Institute, Vetsuisse Faculty Bern, University of Bern, Switzerland
Affective states are known to influence behavior and cognitive processes. To assess mood (moderately long-term affective states), the cognitive judgement bias test was developed and then widely used in various animal species. However, little is known about how mood changes, can be experimentally manipulated, and feeds back into cognitive judgement. A recent theory argues that mood reflects the cumulative impact of differences between obtained outcomes and subject expectations (so called mismatches). We take advantage of the large number of studies published on nonverbal cognitive bias tests in recent years (95 studies with a total of 162 independent tests) to test whether cumulative mismatch could indeed have led to the observed mood changes. Based on a criteria list, we assessed whether a mismatch had occurred with the experimental procedure used to induce mood (mood induction mismatch), and/or whether a mismatch had occurred during the actual testing (testing mismatch). We then investigated whether these two types of mismatch can predict the actual outcome of the cognitive bias study. The present evaluation shows that mood induction mismatch cannot well predict the success of a cognitive bias test. Additionally, testing mismatch can modulate or even inverse the expected outcome. We think that cognitive bias studies should more specifically aim at creating expectation mismatch while inducing mood states to test the cumulative mismatch theory more properly. Besides, testing mismatch should be avoided as much as possible because it can reverse the affective state of animals as measured in a cognitive judgement bias paradigm.
Keywords: Affective states, Mood, Cognitive bias, Judgement bias, Animal Welfare
Impact of Cholesterol Deficiency on Metabolism and Milk Production in Lactating Dairy Cows
1Division of Veterinary Physiology, DCR-VPH, Vetsuisse Faculty, University of Bern, Switzerland; 2Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland; 3Swissgenetics, Switzerland; 4Swissherdbook Cooperative, Switzerland; 5Holstein Switzerland, Switzerland
The recently detected cholesterol deficiency (CD) in Holstein cattle is associated with reduced cholesterol in calves and bulls and caused by a loss of function mutation of the APOB gene. We set out to investigate if this mutation influences also the metabolism and performance of dairy cows. Twenty pairs of full sibling cows from same farms were investigated. Each pair contained a heterozygous carrier (CDC) and a non-carrier (CDF) of the disease associated APOB mutation. Blood samples were taken in early lactation (18 to 40 days in milk) and mid-lactation (141 to 186 days in milk) to measure cholesterol metabolites. Pearson’s correlation coefficients were calculated with the PROC GLIMMIX procedure of SAS to assess the differences between study groups. Total and free cholesterol, high density lipoproteins, low density lipoproteins, phospholipids and cholesterol esters, but not triacylglycerides and very low density lipoproteins were lower in CDC cows compared with matched CDF siblings at both time points (P < 0.05). Concentrations of glucose, free fatty acids, beta-hydroxybutyrate, and aspartat-aminotransferase and gamma-glutamyl-transferase activities did not differ between both groups (P > 0.05). Milk yield, milk protein and milk fat in the current lactation as well as days open did not differ between groups (P > 0.05). A parallel development of cholesterol concentrations from early to mid-lactation of CDC and CDF was observed. In conclusion, although total cholesterol and associated lipoprotein cholesterol concentration in CDC are reduced, neither energy metabolites, liver enzyme activity nor performance within matched sibling pairs differ over the course of lactation.
The Immunological Role of Reproductive Hormones in Rainbow Trout (Oncorhynchus mykiss)
Centre for Fish and Wildlife Health, Vetsuisse Faculty Bern, University of Bern, Switzerland
Sex differences in immune function and diseases resistance are common. It has been hypothesized that the reproductive strategy of females, which selects for prolonged egg production, favours high immunocompetence and disease resistance, whereas the reproductive strategy of males, which selects for high mating rates and secondary sexual characteristics, goes on cost of immunocompetence and disease resistance. In vertebrates, sex-steroids – androgens and oestrogens – are candidate physiological mediators of the influence of sex immune functions, what means that oestrogens should have immunosupportive and androgens immunosuppressive effects. However, immunological effects of oestrogens such as 17β-estradiol (E2) appear to be rather complex, including both stimulation and inhibition of immune functions.
In my thesis, I am investigating the immunological role of estrogens in teleost fish, using rainbow trout (Oncorhynchus mykiss) as a model. So far studies with fish focused on cellular responses with in-vitro tests, what gives an idea of how E2 can affect different types of cells, but a better understanding is needed in an animal perspective, implicating the need for in-vivo experiments. More specifically, I am studying how immune parameters of rainbow trout are changed as a consequence of changes in the endogenous estrogen and energy status of rainbow trout: during different life stages 1) juvenile age versus adult, non-reproductive stage), 2) in adult stage during the reproductive cycle, and 3) in juvenile fish during energy restriction.
Role of ACKR3 in Leukocyte Migration and the Microarchitecture of Secondary Lymphoid Organs
1Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 2Institute for Research in Biomedicine, Università della Svizzera Italiana, Bellinzona, Switzerland
Background: Effective trafficking of immune cells in secondary lymphoid organs (SLOs) is crucial for an adequate immune response. Leukocyte trafficking in these organs is mainly driven by the generation and maintenance of locally confined chemokine gradients. Here, atypical chemokine receptors (ACKRs) have recently emerged as key regulators of chemokine patterns. AKCRs act as “sinks” in the “source and sink” model proposed by F. Crick. The receptors scavenge the ligand and deliver them for degradation in lysosomes, hence contributing to the formation of chemokine gradients.
High levels of ACKR3 mRNA on splenic marginal zone B cells (MZB) have been reported, and destruction of the ACKR3/CXCL12 axis leads to the destruction of this microarchitecture. However, the role of this receptor in the marginal zone of the spleen remains elusive.
Hypothesis: We hypothesize that ACKR3 plays a key role in the MZ formation, maintenance and function.
Results: Using ACKR3GFP/WT reporter mice we investigated the microarchitecture of the spleen. As expected GFP-positive B-cells were observed in the MZ. The MZ could easily be visualized after injection of encapsulated attenuated and fluorescently labelled S. pneumoniae. A strong GFP expression was observed in a vascular structure located in the red pulp. The structures appear to be distinct from endothelial cells of blood vessels. ACKR3 expressed on these vessels appear to actively scavenge an ACKR3-selective chimeric chemokine suggesting depletion of CXCL12 in its near surroundings.
Linking Metabolism and Inflammation in Macrophage Foam-Cells
Division of Experimental Pathology, Institute of Pathology, University of Bern, Switzerland
Atherosclerosis is a chronic inflammatory disease affecting the major arteries and remains the primary cause of death worldwide. The development of atherosclerotic lesions is driven by a sterile inflammatory response to oxidized lipids in the sub-endothelial space, which trigger endothelial activation and leukocyte recruitment. Infiltrating monocytes differentiate into macrophages, phagocytose the lipid material and transform into lipid-loaded “foam cells” (FCs) that represent a hallmark of the plaque. These FCs remove potentially cytotoxic lipids but also promote atherogenesis by releasing pro-inflammatory mediators. Macrophages differentiate into specialized subsets with distinct effector functions and metabolic programs in response to cues from their microenvironment. A potential regulator of metabolic adaptations during FC-formation is the mitochondrial uncoupling protein 2 (UCP2) which has been shown to have a protective role in atherogenesis. We aim to investigate pathways that link lesion-related influences, such as inflammation, dyslipidemia and hypoxia, to FC-formation and to examine a potential role of UCP2 on metabolic polarization. Under normoxia, FCs displayed a pro-inflammatory metabolic M1-like phenotype with enhanced glycolysis and reduced OxPhos compared to macrophages. This, interestingly, was not observed in UCP2KO FCs, indicating a UCP2-dependend M1-polarization of FCs. In hypoxic conditions, both, WT and UCP2KO FCs showed the M1-phenotype, suggesting hypoxia-induced M1-polarization upstream of UCP2. Both, macrophages and FCs metabolized mostly pyruvate – fatty acid and glutamine restriction did not reduce metabolic rates. Taken together, our findings suggest that hypoxia and UCP2 contribute to the FC-phenotype developing in response to the dyslipidemic and inflammatory conditions present in the atherosclerotic plaque. We are currently further investigating potential mechanisms linking the inflammatory state and metabolic programs in FCs.
Diet-Induced Obesity Potentiates IgE-Mediated Food Allergic Reponses
1Institute of Pathology, University of Bern, Switzerland; 2Department of Fundamental Microbiology, University of Lausanne, Switzerland
Introduction: The past few decades have witnessed an increase in the prevalence of allergic diseases and number of obese individuals in tandem. Epidemiological studies suggest that obese children are more likely to have IgE mediated food allergies. However, the mechanisms by which diet-induced obesity may modulate susceptibility to IgE mediated food allergies remain unknown. Employing an experimental model of IgE mediated food allergy, we sought to test alterations in food allergen sensitization, immunological responses and gut barrier integrity in food allergic mice fed either a control diet or a diet rich in sugars and fat.
Method: Mice were fed a Western diet for 12 weeks prior to epicutaneous food allergen sensitisation on an atopic dermatitis like skin lesion followed by intra-gastric allergen challenge. Allergic manifestations in response to oral food allergen challenge were assessed by a clinical allergy score. Total- and allergen specific IgE levels were measured by ELISA. Accumulation of mast cells in the small intestine was quantified by flow cytometry or on tissue sections by means of chloroacetate esterase staining. In ongoing studies, we are currently analysing the impact of a Western diet on gut barrier integrity and allergen uptake.
Results: While no changes in food allergen sensitization through an atopic dermatitis-like skin lesion were observed between the two groups, mice fed a Western diet presented with increased serum IgE levels and mucosal mast cell infiltration, two hallmarks of intestinal food allergy, that manifested in an overall increased clinical allergy score compared to food allergic mice on a control diet.
Conclusion: Consumption of a Western diet promoting obesity potentiates the magnitude of IgE-mediated food allergic responses in an experimental model of food allergy.
The Leucine-Rich Repeats (LRR) Protein, Ribonuclease Inhibitor (RNH1) Limits the Inflammasome Activation
1Department of Hematology, Inselspital, Bern University Hospital, Switzerland; 2Department for BioMedical Research, Hematology, University of Bern, Switzerland
Background: RNH1 is a ubiquitously expressed protein, which binds to and inhibits pancreatic type ribonucleases. Interestingly, bioinformatic studies revealed that LRRs of RNH1 are very similar to those of NLRP family and belong to the same group. NLRP proteins form inflammasomes complexes. Inflammasomes are caspase-1 activating complexes containing an adaptor protein (ASC) and caspase-1. Therefore, we investigated the role of RNH1 in inflammasome activation.
Methods: We generated RNH1-deficient human monocyte THP1 cells and mouse immortalized macrophage cells by using CRISPR-Cas9 method and performed inflammasome experiments. Inflammasome activation leads to capase-1 cleavage, which precedes to cell death (pyroptosis) and release of active-IL-1β, which were measured by LDH and ELISA assay respectively. Western blot was performed to check caspase-1 and IL-1β cleavage and other inflammasome components, such as NLRP3 and ASC.
Results: Activation of RNH1-deficient, human THP1 and murine macrophage cells, with NLRP3 stimuli leads to increased IL-1β production, caspase-1 cleavage and cell death compared to wild type cells. However, we also found increased AIM2 and NLRC4 inflammasome activation in RNH1-deficient cells, suggesting that RNH1 might not specifically affect NLRP3 inflammasome activation but regulate general inflammasome activation. Indeed, we found increased Pro IL-1β and NLRP3 protein levels in RNH1-deficient cells with TLR activation, suggesting that RNH1 might be involved in priming signal.
Conclusions: We found that RNH1 negatively regulates inflammasome activation. Further, we are performing yeast two-hybrid (Y2H) screen and mass-spec experiments to identify several binding partners to understand the molecular mechanism. We also generated macrophage-specific Rnh1 knock out mice (by crossing floxed Rnh1 mice with LysM Cre+ mice) to challenge the physiological relevance of our observations.
Active and Passive Immunization Against Type 2 Diabetes Mellitus
1Department of Rheumatology, Immunology and Allergology (RIA), Inselspital, Bern University Hospital, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland
Amyloid aggregates composed of extracellular fibrils of islet amyloid polypeptide (IAPP, also called amylin) - a peptide synthetized in the pancreatic β-cells and co-secreted with insulin- are found in most type 2 diabetes mellitus (T2DM) patients and has been associated with the progression of the disease. As aggregates are considered to be a key factor in β cell death, we aim at developing a vaccine targeting these pathogenic aggregates to prevent and/or reverse accumulation and enhancing β cell survival. To study this, a transgenic mouse model expressing human IAPP (hIAPP) is used.
The vaccines were designed using amylin peptide sequences chemically cross-linked to virus like particles (VLPs). To test the induced antibodies response against each peptide, C57BL/6 mice were immunized and the serum antibodies were analyzed by ELISA assay. The peptides coupled to the VLPs inducing the highest IgG titers against IAPP where then tested in the mouse model. Monitoring analysis of the transgenic mice showed spontaneous development of T2DM around the 8th week of age only in the homozygous male group. For this reason, immunizations with the vaccine were performed only in these mice. Interestingly, the first analyzed group of immunized male transgenic mice showed no symptoms of T2DM up to 12 weeks. We are currently repeating these experiments and assess inflammatory state, hIAPP load and disease progression.
Splenic Red Pulp Provides a Secondary Stem Cell Niche for Chronic Myeloid Leukemia Stem Cells
1Tumorimmunology, Department for BioMedical Research, University of Bern; 2Department of Hematology, University Hospital and University of Zurich; 3Department of Medical Oncology, Inselspital, Bern University Hospital and University of Bern
Background: Leukemic stem cells (LSCs) in chronic myeloid leukemia (CML) are responsible for initiation, progression and relapse of the disease. Resistance of LSCs to treatment is frequent and mediated by cell-intrinsic characteristics but also by the interactions of LSCs with their microenvironment.
Splenomegaly is a hallmark of CML. Nevertheless, the role of the spleen on leukemia and LSCs is unknown. Therefore, this project aims to investigate the potential role of the spleen as an independent secondary niche for CML stem cells and its contribution to disease development and progression.
Methods: CML-like disease was induced in mice by retroviral transfection of FACS sorted Lineage- Sca-1+ c-Kit+ bone marrow cells (LSKs) with pMSCV-p210BCR/ABL-IRES-GFP. Transduced LSKs were then injected i.v. into recipient mice.
Results: Analysis of the leukemic compartment in BM and spleen revealed a significant accumulation of LSCs in the spleen. Splenectomy prior to CML induction significantly improved CML survival and decreased the amount of LSCs in the BM. Microscopy on optically cleared spleens further revealed that LSCs exclusively localized within the red pulp, which is partially populated by macrophages. Using clodronate liposomes, we depleted total macrophages in CML mice and analyzed the leukemic compartment. Spleen weight and numbers of splenic LSCs were significantly reduced in clodronate treated animals whereas LSC numbers in the BM were not affected. In addition, Imatinib treatment revealed that splenic LSCs are quiescent and more drug-resistant compared to LSCs from the BM.
Conclusions: These results suggest the presence of an independent splenic LSC niche which promotes resistance to conventional therapy and disease progression and is mainly composed of macrophages in the red pulp.
Transcriptional Regulation of Autophagy-Related Genes in Melanoma Development and Progression
1Institute of Pharmacology, University of Bern, Switzerland; 2Life Sciences Research Unit, University of Luxembourg, Luxembourg; 3Department of Dermatology, Inselspital, Bern University Hospital, Switzerland
Autophagy is a highly conserved cellular process in which intracellular proteins and organelles are sequestered and degraded in double-membrane vesicles known as autophagosomes. Depending on different stages and types of cancer it can act as a pro-survival or pro-death mechanism. Our first goal was to investigate the expression of autophagy-related proteins in melanoma. A collection of nevus, primary and metastatic patient samples were used to construct tissue microarrays (TMA). With immunohistochemistry we determined that ATG5, ATG7 and BECLIN 1 are significantly differently expressed between benign nevi and melanoma. In addition, analysis of p62 and LC3 indicated possible lower levels of autophagy in primary melanoma compared to benign nevi. The analysis of disease free survival (DFS) for primary and metastatic melanoma patients showed that patients with higher levels of ATG5 and ATG7 had an improved DFS than those with lower levels of ATG5 and ATG7 in their tumors. The second aim was to investigate the molecular mechanism behind the downregulation of ATG proteins in melanoma. With bisulfite sequencing we excluded the possibility of ATGs promoter methylation in melanoma patients. We hypothesize that altered expression or activity of transcription factors is responsible for downregulation of ATG5, ATG7 and BECLIN 1 in melanoma. With DNase-seq data from primary melanocytes we performed a genome-wide chromatin accessibility analysis to predict the binding of putative transcription factors (TF) in promoter regions of ATG5, ATG7 and BECLIN 1. Data of siRNA – mediated knockdown of selected TFs shows an impact on ATG5, ATG7 and BECLIN 1 mRNA and protein levels in melanoma cell lines. Furthermore, we validated the transcriptional regulation of NRF1, NFE2L2, TCF3, MZF1 and GMEB1 on ATG5, ATG7 and BECLIN 1 using the Luciferase Dual-Glo assay. In addition, knockdown of NRF1, TCF3 and GMEB1 leads to an increased migration potential of melanoma cells.
Development of Conformational Biosensors to Investigate CCR7 Signaling Pathways
1Biotechnology Institute Thurgau (BITg), University of Konstanz, Kreuzlingen, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
How is it possible that immune cells such as dendritic cells maintain CCR7 signaling permitting directed migration over hours, while the receptor is internalized within minutes upon ligand binding? As we have shown previously, internalized CCR7 resides in endosomes. Here, we propose that persistent CCR7 signaling from endosomes facilitates sustained directional migration. To verify this hypothesis, we aim to develop conformation-specific nanobodies recognizing ligand-bound and inactive CCR7. Such new nanobody-based biosensors would permit to investigate whether CCR7 signaling is only confined to the plasma membrane or whether sustained signaling derives from CCR7 residing in endosomes. In addition, we aim to address whether CCR7 signaling at endosomes exert biological actions distinct from transient plasma membrane signaling. Besides controlling cell migratrion, persistent CCR7 signaling might additionally influence regulation of gene expression and tumor progression. It is of particular interest to investigate if binding of CCL19 and CCL21 triggers distinct active states of CCR7 and accounts for biased signaling. Exploring nanobody-CCR7 interactions in different immune and tumor cells will finally highlight distinct features of CCR7 signaling and will thus contribute to reveal new prospects for pharmacological interventions affecting CCR7-mediated cell migration during autoimmune diseases and metastasis formation.
The Role of the p73-ATG5 Axis in Regulating Autophagy in Atopic Dermatitis and Psoriasis
1Institute of Pharmacology, University of Bern, Switzerland; 2Department of Dermatology, Inselspital, Bern University Hospital, University of Bern, Switzerland
Autophagy is a lysosomal degradation pathway that is essential for cellular survival, differentiation and homeostasis. Autophagy principally serves as an adaptive process to protect organisms against diverse pathologies, but its regulation can become critical for the whole organism, for example, in neurodegenerative diseases and cancer. We hypothesize that autophagy must be also important under pathological conditions such as inflammation. We propose to investigate how autophagy is regulated in epithelial cells in atopic dermatitis and psoriasis patients. p73 is a transcription factor, which has a role in the regulation of many cellular processes, such as apoptosis, cell cycle and senescence, especially following DNA damage. Based on the previous observations that p73 is responsible for inducing ATG5 expression and regulates autophagy in hepatocytes, we hypothesize that this mechanism may also play an important role in atopic dermatitis and psoriasis. By applying immunofluorescence techniques, we observed an increased expression of p73 in keratinocytes in skin samples of patients suffering from atopic dermatitis and psoriasis compared to normal controls. While under normal conditions, only cells of the basal layer expressed p73 in their nuclei, we observed additional strong expression of p73 in the stratum spinosum under inflammatory conditions. Until now the role of p73 and autophagy in inflammatory skin diseases has not been studied yet. We are investigating the expression and interaction of p73 and ATG5 in atopic dermatitis and psoriasis.The preliminary results indicate that reducing p73 expression by shRNA, reduces ATG5 expression in human skin keratinocyte cell line (HaCaT). Presently it is known that autophagy has a role in inflammation, investigating this process in atopic dermatitis and psoriasis may lead to new insights and a better understanding of the pathophysiology of these diseases, which may result in identifying new drug targets.
Involvement of the Adaptive Immune System in Metamizole-Induced Agranulocytosis
1Clinical Immunology/Allergology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Division of Clinical Pharmacology and Toxicology, University Hospital Basel, University of Basel, Switzerland; 3Department of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Switzerland
Introduction: Metamizole is a pro-drug with analgesic and antipyretic proprieties. Commonly prescribed in western countries because of its favorable efficacy/toxicity balance, metamizole can cause in rare cases agranulocytosis. So far, the mechanism leading to the disease has not been yet elucidated. Few case reports showing that patients re-exposed to metamizole develop more rapidely and more severely symptoms suggest a possible involvement of the adaptive immune system in metamizole-induced-agranulocytosis (MIA).
Methods: Patients with history of MIA, metamizole tolerant patients and unexposed donors were selected to build a clinical cohort. Peripheral blood mononuclear cell (PBMCs) were cultured with metamizole and its main metabolites. Activation induced by metamizole were analyzed by flow cytometry using the early activation marker CD69. Proliferation was monitored as well by CFSE staining and by 3H-thymidin incorporation.
Results: After three days of culture, expression of CD69 was significantly up-regulated in T cells of MIA-patients, but not in the control groups. Furthermore, this upregulation increased with metamizole concentration, suggesting a dose-dependent response, although statistical significance was not met. The expression of CD69 was lost after seven days of culture. Although cell activation could be observed in T cells, proliferation induced by the drug could not be observed.
Conclusion: These findings indicate that T cell response against metamizole is more prevalent in MIA-patients than in control groups. Since T cell activation is only transient, further investigations of metamizole activity on T cells are rendered difficult. Whereas the observed activation of MIA-patients T cells is relevant for the pathogenesis of MIA still needs to be demonstrated.
Mechanisms of Regulation of Human T Cell Activation and GM-CSF Production
1Institute for Research in Biomedicine, Università della Svizzera Italiana, Bellinzona, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Experimental Oncology, European Institute of Oncology (IEO), Milan, Italy
T helper (Th) lymphocytes have a central role in orchestrating immune responses against pathogens. Dysregulation of such responses can however lead to disease, including autoimmune disorders like Multiple Sclerosis (MS). Indeed, aberrant activation of auto-reactive T cells can induce responses against self-antigens like myelin components, and determine immune-mediated pathology.
We aim at identifying molecular determinants able to influence the pathogenicity of pro-inflammatory T cells in MS.
As readout of pro-inflammatory, potentially pathogenic T lymphocytes, we utilized the cytokine GM-CSF, which was shown to be critical for the induction of disease in EAE, a mouse model of MS, and to be increased in MS patients. We therefore investigated the molecular signature associated with GM-CSF production. We separated primary human Th lymphocytes in GM-CSF-producing and non-producing fractions by cytokine secretion assay, and assessed differential expression of mRNAs and miRNAs.
We found that GM-CSF expression was associated with a general pro-inflammatory gene signature. Mechanistically, this gene signature was linked to increased activation of the NF-kB pathway and to the preferential expression of the transcription factor BHLHE40. Importantly, in vitro treatment of T cells with vitamin D or IFNβ, a cytokine therapy approved for MS, dampened the pro-inflammatory gene signature as well as GM-CSF expression, although the two treatments appeared to affect T cell effector functions with different underlying mechanisms. Finally, analysis of miRNA expression in GM-CSF-producing and non-producing cells revealed selective expression of miRNAs in the GM-CSF-negative population.
We are investigating the functional role of selected transcription factors and miRNAs in cell activation, proliferation and cytokine production using primary Th lymphocytes from both healthy donors and MS patients, in order to finally evaluate relevance for disease.
Influence of Bacterial RNA on Memory Generation Against Virus-Like Particles
1Department of Rheumatology, Immunology and Allergology, Division of Immunology, Inselspital, Bern University Hospital, Switzerland; 2Department of Dermatology, Zurich University Hospital, Switzerland
Due to their highly repetitive nature and bacterial RNA content, virus-like particles (VLPs) derived from bacteriophage Qβ elicit strong and long lasting humoral immune responses. After immunization memory B cells (MBC) and plasma cells (PCs) are generated in germinal center (GC) responses. It has been reported that MBC differentiate to secondary plasma cells after challenge. Secondary PCs produce higher amounts of antibodies than PCs in primary responses early during the recall response1. To determine the influence of bacterial RNA, a ligand for TLR7/8 inside the VLPs, on MBC and secondary PC generation, adoptive transfers of MBCs generated in the presence or absence of TLR7/8 stimulation followed by challenge with Qβ VLPs were performed. The antibody response of secondary PCs, derived from Qβ RNA induced MBCs, is higher and starts earlier, than the primary response. In contrast, the response of secondary PCs, derived from MBCs induced by Qβ without RNA is more similar to the primary response. Thus, TLR7/8 signalling seems to drive differentiation of MBCs capable of generating secondary plasma cells, responsible for rapid IgG responses during secondary antibody responses.
Moreover, we are interested to determine differences in gene expression patterns between activated VLP specific GC cells versus circulating memory and naïve B cells of Qβ immunized mice. Using fluorescent activated cell sorting VLP specific and naïve B cells were sorted to extract RNA. The gene expression pattern was determined using RNA sequencing. Vast differences in gene expression between Qβ specific and naïve, as well as between GC and MBC are observed. GC cells express higher amounts of genes related to cell cycle whereas in MBCs genes of various immune system categories are upregulated.
1. Zabel, F. et al. Viral particles drive rapid differentiation of memory B cells into secondary plasma cells producing increased levels of antibodies. J. Immunol. 192, 5499–508 (2014).
Deciphering a Prototypical MAPK Signaling Network at the Single Cell Level Using an Optogenetic Circuit
1Institute of Cell Biology, University of Bern, Switzerland; 2Department of Biosystems Science and Engineering, ETH Zurich, Switzerland
Receptor tyrosine kinases (RTK) enable to convert extracellular INPUTs as growth factors into specific cellular OUTPUTs through the activation of dynamic signaling networks. Nowadays, we have a good idea about the components of such networks, but we still miss information about how they are wired in a coherent signaling pathway. Our lab has recently brought new insights into the feedback and feed-forward structures regulating the ERK-MAPK network using a microfluidic device to deliver dynamical growth factors inputs and an ERK biosensor to record single cell ERK activity outputs(1). Here, we propose to use a similar INPUT/OUTPUT approach combined with system perturbations to identify the molecular players involved in the network regulation. To increase the experimental throughput, we have built an INPUT/OUTPUT system enabling to activate the MAPK network with light pulses thanks to an optogenetic FGF receptor and to record dynamic ERK signaling with a spectrally compatible ERK biosensor. This new INPUT/OUTPUT system provides both the spatial and temporal resolution required to study the effect of targeted perturbations on the network behaviour at the single cell level.
(1)H. Ryu et al., Frequency modulation of ERK activation dynamics rewires cell fate, Molecular Systems Biology, 2015
Recycling of IgE Immune Complexes in Human B Cells Facilitates Transfer of Antigen to Dendritic Cells for Antigen Presentation
1Department of Rheumatology, Immunology and Allergology, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland; 2Nuffield Department of Medicine, The Jenner Institute, University of Oxford, Oxford, UK
IgE immune complexes (IgE-IC) have been shown to enhance antibody and T cell responses in mice by binding CD23, the low affinity IgE receptor. In humans, the mechanism by which CD23-expressing cells take up IgE-IC and process them is not well understood. Here, we aimed at studying CD23/IgE-dependent antigen processing and presentation in human B cells as well as monocyte-derived dendritic cells (moDCs). IgE-IC binding to CD23 caused CD23 aggregation followed by IgE-IC internalization in B cells and moDCs. However, while IgE-ICs were degraded in moDCs, B cells did not degrade the complexes but recycled them in native form protected from degradation. Moreover, IgE-IC pulsed B cells were only able to induce antigen-specific T cell proliferation in contact with moDCs that have taken up the antigen recycled from B cells. Our findings argue for a novel model in which human B cells promote specific T cell proliferation upon IgE-IC encounter by acting as carriers transferring antigen to specialized antigen presenters.
Inhibition of Hsp90 Overcomes Acquired Chemotherapeutic Resistance in KRAS-Mutant Lung Cancer by Abrogating eIF4E-Myc Signaling Axis
1Division of General Thoracic Surgery, Inselspital, Bern University Hospital, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland; 3Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
Introduction: KRAS mutation is one of the most common oncogenic drivers in non-small cell lung cancer (NSCLC). Currently, the standard therapy for KRAS-mutant NSCLCs remains cytotoxic chemotherapy due to lack of effective targeted therapies. Patients experience recurrence and ultimately death because of chemotherapeutic resistance, which is a significant clinical challenge. Therefore, strategies to overcome chemotherapeutic resistance are urgently needed.
Aim: To identify the molecular mechanism driving chemoresistance in KRAS-mutant lung cancer and develop novel therapeutic strategies to treat KRAS-mutant lung cancer.
Methods: KRAS-mutant NSCLC cell lines were chronically exposed to clinically relevant chemotherapeutic agents to generate chemotherapy resistant cells. Pharmacological screens were then conducted to identify the pathway whose inhibition specifically reverts chemotherapy resistance in KRAS-mutant NSCLC cells.
Results: Heat shock protein 90 (HSP90) outstood as one of the top hits on our screen. Further analyses showed that Hsp90 regulates a key survival pathway through eukaryotic translation initiation factor 4E (eIF4E)-Myc axis, which is essential for chemotherapy resistance in KRAS-mutant NSCLC cells.
Conclusions: Our studies identified a novel signaling axis involving eIF4E-Myc that confers acquired chemotherapy resistance, and validates a rational strategy by targeting Hsp90 in reverting chemotherapy resistance in KRAS-mutant lung cancer.
Glycan-Checkpoint Inhibitor Unleashing CD8+ T Cells Against Cancer
1Institute of Pharmacology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department for BioMedical Research, University of Bern, Switzerland; 4Department of Haematology and Central Haematological Laboratory, Inselspital, Bern University Hospital, Switzerland; 5Department of Fundamental Oncology, University of Lausanne, Switzerland; 6Department of Dermatology, Inselspital, Bern University Hospital, Switzerland; 7Department of Biomedecine, University of Basel, Swizerland
In the past few years, the role of cytotoxic T lymphocytes (CTL) in the fight against cancer has been put into light and is playing a growing importance into the development of check-point inhibitor therapies.
Siglecs are inhibitory receptors recognizing sialoglycan ligands and are able to trigger inhibitory functions. The ligands of Siglec receptors have been shown to be highly upregulated in various types of tumors.
We hypothesized that Siglec-9 expressed on CD8+ T cells might be used by tumor cells to hide from the immune system.Thus, we investigated the Siglec-9+ CD8+ T cell pool characteristics in healthy and in patients with melanoma.
CD8+ T cells were isolated from patients and healthy donors to be analyzed using FACSverse. Cells have been screened for various membrane receptors expression, cytokine production, proliferation capacity and fundamental tissue quantification for a complete overview of the Siglec-9+ CD8+ T cell pool characteristics, as well as fonctional assays.
Our study has put in evidence an increased Siglec-9+ CD8+ T cell pool presence in the Tumor Infiltrating Lymphocytes (TILs) isolated from patients with melanoma compared to healthy donors. The Siglec-9+ CD8+ T cell pool represent a more differentiated, more cytotoxic and more proliferative subset of CD8+ T cell; this activity can be inhibited by Siglec-9. Moreover, the RNA and protein levels of PD-1 and Siglec-9 appeared to behave similarly in patients with melanoma. Finally, we demonstrated that the ligands of Siglec-9 is highly upregulated on tumor surface of patients with melanoma, as are PD-1 Ligands. Finally, we show that the blocking of siglec-9 interaction with its Ligands on tumor increased the cytotoxic potential of the T cell towards the tumor cells.
Taken all together, our data suggest that Siglec-9 on CD8+ T cells may represent a novel potential therapeutic targets for immune check-point therapy of malignancies with high expression of sialoglycans, such as melanoma.
Accumulation of Thymic-Derived Regulatory T Cells in the Bone Marrow of Leukemia Mice Leads to Disease Progression and Immune Evasion
1Tumor Immunology, Department for BioMedical Research, University of Bern, Switzerland; 2Department of Medical Oncology, Inselspital, Bern University Hospital and University of Bern, Switzerland
Leukemia stem cells (LSCs) in the bone marrow (BM) are the origin of any leukemia and resistant against conventional therapies and immune control. This resistance is partially mediated by protective mechanisms of the hematopoietic stem cell niche in the BM. In leukemia, the BM microenvironment changes dramatically amongst others with regulatory T cells (Tregs) accumulating. However, little is known how Tregs affect LSCs and leukemia.
In a model of murine chronic myeloid leukemia (CML)-like disease, we could demonstrate that Treg frequency within CD4 T cells in the BM was significantly increased in CML mice and positively correlated with leukemia load. These accumulating Tregs were phenotypically identified as activated (CD62L- CD44+) and thymic-derived (Helios+). Depletion of Tregs 13 days after leukemia induction resulted in long-term control and elimination of the disease in the majority of the mice. Importantly, Treg-depleted CML mice showed a decrease of LSCs compared to controls by phenotypical and functional analysis. To determine whether an increased anti-leukemic CD8 T cell response is responsible for the reduction in LSCs after Treg ablation, Tregs were depleted in the presence and absence of CD8-depleting mAb and LSC numbers and function were assessed. Parallel depletion of Tregs and CD8 T cells restored LSC numbers in leukemia mice, suggesting that Tregs protect LSCs from CD8-mediated elimination.
Lastly, we analyzed the mechanism how Tregs in the BM of CML mice get activated. Therefore, we generated a CML derived from MHC-II-deficient and –proficient LSCs in a preliminary set of experiments. CML derived from MHC-II-deficient LSCs developed significantly slower than the control CML. Importantly, the frequency of activated BM Tregs in MHC-II-/- CML mice was significantly reduced compared to controls.
In summary, our data indicate that thymic-derived, MHC-activated Tregs protect LSCs from elimination by cytotoxic CD8 T cells and promote leukemia development.
Understanding Single Cell-Level MAPK Activation Dynamics for Manipulation of Neuronal Stem Cell Self-Renewal and Differentiation Fates
1Institute of Cell Biology, University of Bern, Switzerland; 2Department of Biomedicine, University of Basel, Switzerland
Dividing stem cells face the decision of either generating more uncommitted stem cells or lineage-committed differentiating cells. The proper balance between these cell fates is crucial for development and tissue homeostasis. The ERK/MAPK signalling pathway is a key regulator of both proliferation and differentiation fates, which poses the question of how the pathway instructs a decision between them. In the well-studied PC12 cell line dynamic temporal patterns of ERK activation are responsible for specifying cell fate. These cells proliferate in response to a transient activation of ERK, which results from stimulation with the growth factor EGF and differentiate into neuron-like quiescent cells in response to a more sustained ERK activation, which results from stimulation with NGF. Importantly, growth factor identity influences the likelihood that a cell will activate ERK in a transient or sustained way but ultimately the response is heterogeneous across the cell population, with a subset of cells proliferating in response to NGF or differentiating in response to EGF. In stem cell niches, this "noisy" response might provide a way to balance self-renewal and differentiation. The aims of this project are to investigate if and how ERK activity dynamics determine cell fate at the single cell-level in a primary neural stem cell model – adult dentate Gyrus stem cells which are responsive to EGF and FGF – and to find ways to manipulate this fate choice efficiently.
Signaling Mechanisms of Non-Genetic Cancer Drug Resistance in Melanoma
1Institute of Cell Biology, University of Bern, Switzerland; 2Department of Dermatology, University Hospital Zurich, Switzerland
Cutaneous melanomas are among the most heterogeneous and aggressive types of cancer. More than 50% of all melanomas harbors serine-threonine kinase B-Raf mutations (over 90% BRAF V600E). As a result, the mitogen-activated protein kinase (MAPK) ERK pathway switches to a constitutive activation. Activated ERK pathway leads to tumor progression. Recent target-oriented anti-cancer drugs greatly inhibit ERK pathway and patients with late-stage metastatic melanomas are now treated with drugs such as vemurafenib (inhibitor of the mutated BRAF V600E). Despite these new generation drugs have improved survival in late-stage melanoma patients, drug-induced resistance mechanisms often lead to lethal tumor relapses. The mechanisms that lead to the emergence of drug-induced resistance are not fully understood. Besides secondary genetic mutations in the ERK pathway, it is becoming more and more evident that non-genetic mechanisms that operate at timescales of hours allow drug tolerance after short exposure. Short term resistance mechanisms arise from the topology and robustness (negative and positive feedbacks) of the ERK network.
We want to ask whether a computational model that recapitulates basic properties of ERK network in the context of short term non-genetic melanoma drug resistance can provide a rationale for a better combinatorial drug dosing scheme to robustly shutdown ERK activity. To address this hypothesis, a quantitative analysis of heterogenous signaling states in hundreds of single melanoma cells will allow us to grasp basic properties on how ERK network is rewired after drug exposure. Single cell measurements of ERK dynamics, which require fluorescent biosensors encoded by the cells, feed and refine the computational model, which in the end should predict better therapies to test. Finally, coupling of ERK measurements with cell-cycle stage will allow us to gain insights on why the emergence of short term resistance involves a subpopulation of melanoma cells.
Studying Transient Cell Fate and Targets of IL-9-Producing T Helper Cells Using a GFP-Reporter Construct
1Department of Dermatology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department of Rheumatology, Immunology and Allergology, Inselspital, Bern University Hospital, University of Bern, Switzerland
T helper type 9 (TH9) cells mediate robust anti-tumor immunity in mouse models of melanoma and mediate acute allergic skin inflammation. Our previous experiments showed that Interleukin 9 (IL-9) is predominantly expressed by a subset of skin-tropic chemokine receptor 4+(CCR4+) and CCR8+ TH2 cells. Here, we aim to substantiate the cell fate of transiently IL-9-producing cells and investigate the genetic regulation of IL-9 expression in these cells.
To study cell fate of alive IL-9-expressing cells, we generated a reporter construct in order to express green fluorescent protein (GFP) under the IL-9 promoter. The IL-9 promoter:eGFP construct was transduced with a lentivirus into different TH cell subsets and an IL-9 producing human lymphoid cell line HDLM-2. GFP- and cytokine expression levels were measured by FACS before and after stimulation with αCD3/2/28 beads.
A subset of IL-9 promoter:eGFP transduced T cells and HDLM-2 cells expressed GFP whereas cells transduced with control vector did not. As expected, IL-9-expressing CCR4+/CCR8+ Th2 cells expressed GFP, but IFN-y producing CXCR3+ TH1 cells did not. Also, the transduced IL-9-producing cell line HDLM-2 expressed GFP. However, co-expression of GFP and IL-9 protein on the single cell level was low as measured by i.c. FACS.
Our results indicate a possible discrepancy between promoter activity of the endogenous IL9-locus as compared to the transduced locus which is randomly inserted into the genome by the lentivirus. The difference between IL-9 protein and GFP expression thus indicates differential regulation at these two loci either on the epigenetic, the transcriptional, or the translational level.
The tools developed here will allow us to shed light on the transcriptional and post-transcriptional regulation of IL-9 expression. Knowledge of these unique mechanisms are key to understand the transient expression of IL-9 in activated TH2 cells and their contribution to pathogenesis in acute inflammatory skin diseases.
ATG5 Regulates Eosinophil Differentiation
1Institute of Pharmacology, University of Bern, Switzerland; 2Theodor Kocher Institute, University of Bern, Switzerland
Background: Autophagy is a highly regulated catabolic process in which cell constituents are targeted for lysosomal degradation. The autophagic pathway is incorporated in various biological mechanisms including multiple innate and adaptive immune pathways. Our research group demonstrated that autophagy negatively affects neutrophil generation in the bone marrow and is not essential for neutrophil granulopoiesis.
Methods: To study the role of autophagy in eosinophils, we generated an eosinophil-specific Atg5 knockout mice (designated Atg5eoΔ). Because eosinophils in mice comprise only 1-3% of circulating leukocytes, we provoke eosinophilia by crossbreeding our genetically modified mice with IL-5 transgenic mice.
Results: ATG5 was efficiently and specifically deleted within the eosinophilic lineage and these cells were autophagy deficient. We measured eosinophil numbers and discovered decreased absolute and relative numbers in the peripheral blood of IL-5tg_Atg5eoΔ mice as compared with control IL-5tg_Atg5flox/flox mice. We found that IL-5tg_Atg5eoΔ mice have greater numbers of immature eosinophils in bone marrow and spleen than counterpart controls. After assessing the proliferative status of eosinophils we observed that eosinophil precursors in the bone marrow of IL-5tg_Atg5eoΔ mice exhibit a significantly lower proliferation rate both under in vitro and in vivo conditions, indicating that ATG5 supports eosinophil maturation. Lack of Atg5 does not compromise the viability for in vitro culture of eosinophils isolated from bone marrow.
Conclusions: Decreased counts of eosinophils in the peripheral blood of IL-5tg_Atg5eoΔ mice, higher numbers of immature Atg5-deficient eosinophils at sites of eosinophilopoesis and their lower proliferation rate suggest that autophagy is required for normal eosinophil production.
Towards Unidirectional Reconstitution of Membrane Proteins into Liposomes
Department of Chemistry and Biochemistry, University of Bern, Switzerland
Functional studies of membrane proteins (MP) in their native environment (the cellular membrane) is challenging due to the complexity of the native membranes and measurement interferences from other membrane constituents or reactions. Therefore, MPs are often purified and integrated into systems which mimic their natural environment in a membrane as e.g. liposomes or nanodiscs, a process called reconstitution, allowing to investigate its function and structural aspects without any disturbing background.
The most important problem during MP-reconstitution is the often random orientation of the MP in the liposomal membrane after reconstitution. For functional studies of the MP of interest and quantitative analysis of its properties, unidirectional orientation in the liposomal membrane is required. Previous work of other groups, where MPs have been reconstituted and studied, did not include a final and universal approach (Rigaud JL, Levy D. (2003) Methods Enzymol. 372:65-86.; Tunuguntla R et al. (2013) Biophys. J. 105(6):1388-96.) and procedures have to be individually optimized for an enrichment of enzyme orientation. In most cases, however, orientation cannot be influenced and is thought to depend on the three-dimensional structure of the protein (Tunuguntla R et al. (2013) Biophys. J. 105(6):1388-96; Nordlund G et al. (2014) Nat. Commun. 5:4303.).
We are currently developing and establishing a universal method to force unidirectional reconstitution of MPs by the aid of a molecular unit that can be attached to every protein. Our method is based on the observation that the F1F0 ATP synthase with its large hydrophilic head piece (F1 part) orients highly uniformly when embedded in preformed vesicles. We try to mimic this situation with a bioengineering approach that should be applicable to a wide variety of proteins. Once such a method is fully established, the interplay of two or more membrane proteins can be investigated more quantitatively.
Genetic Analysis of Genodermatoses: A de novo Variant in the ASPRV1 Gene in a Dog with Ichthyosis
1Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland; 2DermFocus Laboratory, University of Bern, Switzerland; 3Department for BioMedical Research, Molecular Dermatology and Stem Cell Research, University of Bern, Switzerland; 4Dermavet, Switzerland; 5Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Switzerland; 6Hubrecht Institute, The Netherlands; 7Department of Dermatology, Inselspital, Bern University Hospital, Switzerland; 8Division of Clinical Dermatology, Vetsuisse Faculty, University of Bern, Switzerland
The skin is the largest organ of the body and serves as a barrier between the organism and the environment. In humans, hundreds of genetic disorders affecting the skin or its appendages are described, many of them assumed to be monogenic. However, the causative gene or the gene function remains unclear for many of them. The skin undergoes a constant process of self-renewal and keratinocytes migrate from the basal layer of the epidermis to the uppermost, cornified layer, as they differentiate. A defect in the differentiation of keratinocytes can lead to cornification disorders such as ichthyosis (scaling disease). Ichthyosis vulgaris is the most common form of ichthyosis in humans and caused by genetic variants in the FLG gene encoding filaggrin.
We studied a German Shepherd dog with a novel form of ichthyosis. Comparing the genome sequence of the affected dog to 288 control dogs from various different breeds revealed a private heterozygous variant in the ASPRV1 gene encoding “aspartic peptidase, retroviral-like 1”, also known as skin aspartic protease (SASPase). The identified candidate variant was a missense variant (c.1052T>C) affecting a conserved residue close to an autoprocessing site (p.L351P). Since the variant was not present in the non-affected parents, it was most likely due to a de novo mutation event in the germline of one of the parents or during the early embryonal development of the case. The retroviral-like protease encoded by ASPRV1 is known to be involved in profilaggrin-to-filaggrin processing. By immunofluorescence staining, we showed that the filaggrin expression pattern was altered in the affected dog.
Thus, our findings provide evidence that the identified de novo variant is causative for the ichthyosis in the affected dog and that ASPRV1 plays an essential role in skin barrier formation. ASPRV1 is thus a novel candidate gene for unexplained human forms of ichthyoses (Bauer et al. 2017, PLoS Genetics).
Mms19: A Mitotic Gene and a Novel Microtubule Associated Protein
1Institute of Cell Biology, University of Bern, Switzerland; 2Department of Biology, University of Fribourg, Switzerland
Mms19 is a multifunctional protein which was first identified as a regulator of DNA repair in a Saccharomyces cerevisiae mutagenesis screen. Further studies identified Mms19 as a part of the cytoplasmic iron-sulfur cluster (Fe-S) assembly (CIA) complex, which transfers an Fe-S cluster to target enzymes to regulate their activity. The cell cycle and transcription regulator Xpd is one such enzyme requiring an Fe-S cluster. Xpd binds to Mms19 to form a complex termed ‘MMXD complex’. Downregulation of Mms19 in human cells leads to mitotic defects and Mms19 loss-of-function mutant Drosophila larvae lack imaginal discs. These findings indicate that Mms19 regulates mitosis and this regulation may be linked to its interaction with Xpd. The objective of this study is to decipher a mechanism by which Mms19 regulates mitosis and whether its interaction with Xpd affects cell cycle progression. We observed that in Mms19 mutants, the larval brain size is reduced alongwith a high proportion of cells arrested in mitosis. Spindle associated defects such as multipolar spindles, short spindles and mis-oriented spindles were seen. A mass spectrometry screen further revealed that Mms19 interacts with microtubules (MTs) and microtubule associated proteins. These interactions were validated by co-IPs. Furthermore, Mms19 may regulate the assembly and structure of the mitotic spindle as it enhanced the polymerization of microtubules in vitro and in vivo. These findings point towards a novel role of Mms19 as an MT regulator.
Ribosome Profiling in Mammalian Cells to Reveal the Role of NMD Factors in Translation Termination
1Department of Chemistry and Biochemistry, University of Bern, Switzerland; 2Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland
Eukaryotes have acquired numerous ways to control the integrity and quality of mRNA. Nonsense-mediated mRNA decay (NMD) is one of these conserved pathways that contribute to mRNA surveillance. It is thought NMD is activated when a ribosome encounters a termination codon (TC) in a context in which it fails to properly terminate translation (He and Jacobson, 2015; Karousis et al., 2016).
In the past, NMD was believed to degrade exclusively aberrant mRNAs with premature TCs and hence avoid production of potentially deleterious truncated proteins. However, evidence from several transcriptome-wide profilings pointed out a role of NMD in gene expression regulation (He and Jacobson, 2015; Karousis et al., 2016). Indeed, depletion of NMD factors affects the level of about 10% of all mRNAs in mammalian cells as we have recently confirmed by the identification of a bona fide set of NMD-targeted mRNAs (Colombo et al., 2017).
In order to gain insights into the role of the NMD factor UPF1 in translation termination, we performed ribosome profiling in mammalian cells depleted of endogenous UPF1 and supplemented with exogenous wild-type or mutant UPF1. This yielded information about the effect of UPF1 on ribosome occupancy along the coding sequence, at TCs and in 3’ UTRs. For a comparison, we also investigated the effect of ABCE1 depletion on ribosome occupancy along mRNAs. Latest results of this analysis will be presented.
Decolonization Experiments with a Commercial Bacteriohpage Cocktail Lytic Against a Multidrug-Resistant Escherichia coli in a Chemostat Culture System
1Institute for Infectious Diseases, University of Bern, Switzerland; 2Institute of Veterinary Bacteriology, Vetsuisse Faculty, University of Bern, Switzerland
Background: Multidrug-resistant Escherichia coli (MDR-Ec) are spreading worldwide due to hyperepidemic high-risk clones (e.g., ST131), whereby the production of ESBLs (e.g., CTX-M-15) is one of the main resistance mechanisms. The use of antibiotic treatment to decolonize this at-risk population demonstrated limitations. In this context, bacteriophage treatment could represent a potential and effective alternative strategy.
Material/Methods: A glass fermentation vessel operated under the control of a New BrunswickTM BioFlo®/CelliGen® 115 Unit was used as a continuous flow culture system for 48 hrs. Two different pools of faeces (i.e., Study I and II) each obtained from three different healthy volunteers were added to the system. Faeces were spiked with 107 CFU/L of CTX-M-15-producing ST131- E. coli strain 4901.28 and three doses of INTESTI Bacteriophage (Eliava BioPreparations) were added at four hours intervals. Bacterial and phage l populations were monitored at different time points using selective plates as well as performing viral titrations on the host strain.
Results: In Study I the three doses of INTESTI gave an immediate decrease in the population size of MDR-Ec from 105 to 101 CFU/mL with no further recovery. In contrast, in Study II the MDR-Ec dropped below the limit of detection, yet starting to re-grow and further re-stabilizing itself at 103 CFU/mL till conclusion of the experiment. In both studies, the total E. coli microbial community as well as the bacteriophage population increased for the first hours and then maintained themselves at 108 and 105 CFU/mL, respectively.
Conclusions: Our data seems to demonstrate that the individual microbiota composition has an impact on in vitro bacteriophage activity. Further studies including next-generation sequencing will help shed some light on the potential factors contributing to the emergence of bacteriophage resistance.
Investigating the Role of NMD as an Innate Cellular Antiviral Defence Mechanism
Department of Chemistry and Biochemistry, University of Bern, Switzerland
Due to their multicistronic nature, viral genomes encode transcripts with features that could render them susceptible to nonsense-mediated mRNA decay (NMD). Such features include uORFS that result in stop codons that could trigger NMD, and long 3’UTRs. A number of studies done in plants, insects and mammalian cells have demonstrated the degradation of viral transcripts by NMD, suggesting that NMD could make up part of the host cell innate immune response. The NMD factors, UPF1, SMG5 and SMG7 were shown to restrict viral replication of the alphaviruses, Semliki Forest Virus (SFV) and Sindbis Virus (SINV) in mammalian cells. Surprisingly, shortening of the 3’UTR of SFV still rendered the viral genome sensitive to NMD (Balistreri et al., 2014). The mechanism by which NMD recognises and degrades the SFV genome remains unclear. This project therefore aims to provide further insight into this mechanism and identify the features that render the SFV genome susceptible to NMD. UPF1-RNA immunoprecipitation (UPF1-RIP) experiments in cells infected with SFV in the absence and presence of the viral entry inhibitor, Bafilomycin (Baf), revealed a substantial enrichment of SFV viral RNA bound to UPF1 in samples without Baf compared to Baf-treated samples. In addition, UPF1 RNA proximity ligation assays (RNA-PLA) performed in cells infected with SFV compared to uninfected controls, revealed a clear interaction between UPF1 and the SFV genome. Together, these results provide convincing evidence of a direct interaction between UPF1 and the viral SFV genome. These techniques will allow us to investigate potential interactions between other NMD factors and SFV. Since we have been able to successfully optimise the production and purification of virus like particles (VLPs) of SFV lacking the 3’UTR (SFV 3- ), these interactions will also be explored in the context of the SFV 3- genome. Together, we aim to provide insight into the mechanism/role of NMD in antiviral defence.
Regulation of CDX2 in Colorectal Cancer via Methylation and Histone Deacetylation
Institute of Pathology, University of Bern, Switzerland
In stage II and III colorectal cancer (CRC) CDX2 is a diagnostic, prognostic and potentially predictive biomarker. Aberrant expression of the CDX2 gene, detected in approximately 20% of CRCs, is associated with tumorigenesis. The aim of the study was to restore CDX2 by targeting epigenetic modifications.
Two CDX2-negative CRC cell lines, Colo-205 and SW-620 were treated with increasing concentrations of DNA methyltransferase inhibitor (DNMTi), Decitabine alone and combined with the Histone deacetylase inhibitor (HDACi) Trichostatin A (TSA), as well as HDAC4 and HDAC5 inhibitor LMK-235 and HDAC1 and HDAC2 inhibitor Romidepsin. CDX2 restoration was assessed by western blot, qPCR and immunohistochemistry (IHC).
After treatment with Decitabine a significant dose-dependent 15-fold increase of CDX2 mRNA and protein were detected. Additionally additive effects of Decitabine and TSA treatment were observed with a 25-fold increase of CDX2 mRNA upregulation and protein level. Further evaluations revealed synergistic effects of Decitabine in combination with LMK-235 and Romidepsin showing an up to 35-fold and 30-fold increase of CDX2 mRNA, respectively.
Methylation and histone deacetylation may play a major role in CDX2 regulation. Targeting specific HDAC’s or DNMT’s involved in the epigenetic regulation of CDX2 could provide an alternative treatment possibility for patients with CRC.
Characterization of the Hypothetical kDNA Kinetochore in Trypanosoma brucei
1Institute of Cell Biology, University of Bern, Switzerland; 2Department of Chemistry and Biochemistry, University of Bern, Switzerland
Kinetoplastid protozoa, such as T. brucei, possess a single large organelle with a single mitochondrial genome named kinetoplast DNA (kDNA). The kDNA consists of ~5000 plasmid like small circles (minicircles) and ~ 25 larger circles (maxicircles), which start to replicate prior to the initiation nuclear replication. Segregation of the replicated kDNA depends on the physical connection of the mitochondrial genome to the basal bodies and their movement during cell division. The physical connection has been described as the tripartite attachment complex (TAC). Several proteins including TAC40, TAC102, p166 and p197 have been described to be part of the TAC, however many components remain to be discovered. To identify new potential TAC proteins, we performed co-immunoprecipitation (Co-IP) with N-terminal PTP-tagged TAC102. Here I describe the results of one candidate (7590) identified in the PTP-TAC102 Co-IP mass spectrometry analysis.
7590 is a 110 kDa protein. It shows sequence similarities to the HMG box containing micronuclear linker histone polyprotein in Tetrahymena thermophila. 7590s isoelectric point is 8 and it has an N-terminal signal peptide for mitochondrial import. We showed that 7590 co-localizes with the TAC component TAC102 in T. brucei blood stream form (BSF) cells and that TAC102 is lost upon depletion of 7590. Interestingly we did not observe the typical kDNA loss and kDNA missegregation phenotype of a TAC component when 7590 is depleted.
Due to the fact that 7590 has a putative HMG box domain, we hypothesize that it could be involved in the connection of the TAC to the kDNA, and thus be part of a hypothetical kDNA kinetochore. It is of great interest to discover the function of 7590. For this we will knock-out 7590 in T. brucei BSF cell and perform blue native analysis to determine whether 7590 and TAC102 reside in the same protein complex. Further we will perform Co-IP to identify more candidates of the hypothetical kDNA kinetochore.
Novel Insights into the Pathways Regulating the Canine Hair Cycle and Their Deregulation in Alopecia X
1Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Switzerland; 2DermFocus Laboratory, University of Bern, Switzerland; 3Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland; 4Division of Clinical Dermatology, Vetsuisse Faculty, University of Bern, Switzerland; 5AniCura Tierärztliche Spezialisten, Hamburg, Germany; 6Clinics of Small Animals and Horses, University of Veterinary Medicine, Vienna, Austria
Alopecia X is a hair cycle arrest disorder in Pomeranians. Histologically, kenogen and telogen hair follicles predominate, whereas anagen follicles are sparse. The induction of anagen relies on the activation of hair follicle stem cells and their subsequent proliferation and differentiation. Stem cell function depends on finely tuned interactions of signaling molecules and transcription factors, which are not well defined in dogs. We performed transcriptome profiling on skin biopsies to analyze altered molecular pathways in alopecia X. Biopsies from five affected and four non-affected Pomeranians were investigated. Differential gene expression revealed a downregulation of key regulator genes of the Wnt (CTNNB1, LEF1, TCF3, WNT10B) and Shh (SHH, GLI1, SMO, PTCH2) pathways. In mice it has been shown that Wnt and Shh signaling results in stem cell activation and differentiation Thus our findings are in line with the lack of anagen hair follicles in dogs with Alopecia X. We also observed a significant downregulation of the stem cell markers SOX9, LHX2, LGR5, TCF7L1 and GLI1 whereas NFATc1, a quiescence marker, was upregulated in alopecia X. Moreover, genes coding for enzymes directly involved in the sex hormone metabolism (CYP1A1, CYP1B1, HSD17B14) were differentially regulated in alopecia X. These findings are in agreement with the so far proposed but not yet proven deregulation of the sex hormone metabolism in this disease.
Rare Forms of Canine Neurological Disorders in Leonberger Breed
1Institute of Genetics, Vetsuisse Faculty Bern, University of Bern, Switzerland; 2Department of Veterinary and Biomedical Sciences, University of Minnesota, USA
Clinical signs of canine neurodegenerative disorders have variable rates of progression and ages of onset, and are present in different breeds. We focus on Leonbergers affected by inherited polyneuropathy (LPN) and leukoencephalomyelopathy (LEMP).
Many neuropathies affecting peripheral nerves in dogs show similarities to Charcot-Marie-Tooth (CMT) neuropathies in humans. LPN is caused by variants in several genes with similar clinical and histopathological features. Previously, a frameshift variant associated with juvenile-onset form (LPN1) was identified in ARHGEF10 gene. We mapped a second form (LPN2) associated with a later age of onset by GWAS on chromosome 15. Whole genome re-sequencing (WGS) of a LPN-affected individual revealed a frameshift variant in the GJA9 gene. While mutation of this gene has not been reported in CMT patients before; GJA9 does belongs to a group of CMT-associated genes. Population studies for LPN2 support a dominant mode of inheritance and incomplete penetrance. Presently, ~30% of LPN-affected Leonbergers can be explained by the ARHGEF10 or GJA9 variants, and we assume additional genetic heterogeneity likely exists in the breed.
Canine LEMP is a juvenile-onset disorder of the central nervous system. Affected dogs show characteristic lesions in the cervical spinal cord white matter. GWAS mapping allowed us to associate LEMP with chromosome 18. Subsequent WGS of a LEMP-affected individual identified a missense variant in NAPEPLD gene. This gene encodes an enzyme of the endocannabinoid system, which is assumed to have neuroprotective function. Population studies support a recessive mode of inheritance with reduced penetrance. We then investigated NAPEPLD in LEMP-affected Rottweilers and identified a novel recessive frameshift variant. Our results provide evidence for allelic heterogeneity in canine LEMP, and a novel description of defects in the endocannabinoid system associated with severe neurodegenerative disease.
Mechanistic Insight into Function and Processing of a Ribosome-Associated ncRNA (rancRNA)
Department of Chemistry and Biochemistry, University of Bern, Switzerland
Translation of mRNAs into proteins is an essential process in every cell. Therefore it is important to understand the mechanism of translation as well as the regulation of this process. Nowadays several classes of non-coding RNAs (ncRNAs) have been identified that are involved in translation regulation. The molecular mechanisms behind are still heavily studied in the field and most known ncRNA target the mRNA. In our laboratory we found a new class of ncRNAs, which directly bind to the ribosome, and regulate translation. Ribosome-associated ncRNAs (rancRNAs) have been shown to be involved in translation regulation in various organisms . Besides their direct availability and ribosome association little is known about their molecular mechanism and processing. We could already show that an mRNA exon-derived 18 residue long rancRNA in Saccharomyces cerevisiae has a significant influence on protein biosynthesis in vivo and in vitro and is crucial for growth upon high salt stress condition .
The goal is to gain detailed mechanistic insights into the function and the processing of the rancRNA_18. We have evidence that rancRNA_18 competes with yeast bulk tRNA for binding. Further investigation revealed the ribosomal A-site as the binding site for rancRNA_18. We could also show that the rnt1 enzyme is involved in the processing of rancRNA_18 and additional mutant clones strengthen these findings. Furthermore we know that there is a minimal length of the ORF needed for proper processing. To get more detailed information about a possible mechanism, we started a collaboration with the group of B. Zuber for a CryoEM structure of 80S ribosomes with bound rancRNA_18.
 Pircher A., Gebetsberger J., and Polacek N., Ribosome-associated ncRNAs: An emerging class of translation regulators. RNA Biology, 2014. 1335-1339
 Pircher, A., Bakowska-Zywicka, K.,Schneider, L., Zywicki, M., and Polacek, N., An mRNA-derived non-coding RNA targets and regulates the ribosome. Mol. Cell, 2014
Investigation of the Multifaceted Alba-Domain Proteins in Trypanosoma brucei
1Institute of Cell Biology, University of Bern; 2Graduate School for Cellular and Biomedical Sciences, University of Bern
RNA binding proteins (RBPs) play essential roles in mRNA stability, translation, repression and degradation, thereby acting as key modulators of gene expression. This is of prime importance for trypanosomes, whose gene regulation relies largely on post-transcriptional processes. Expression of GPEET, a stage-specific surface protein, is regulated post-transcriptionally by elements in the 3’ untranslated region of its mRNA. Our laboratory originally identified Alba-domain proteins in T.brucei as RBPs interacting with one of these elements. Further characterization of these proteins revealed their interactions with translation initiation factors and poly(A)-binding proteins, indicating that they might be involved in translational control. Recent studies have shown increasing evidence of functional plasticity among Alba family of proteins in different organisms. This stimulated us to explore the function of Alba 1-4 in the four life cycle stages of T. brucei accessible to culture: early and late procyclic forms, long slender and stumpy bloodstream forms.
Here, we show that upon knockdown of Alba 3 and Alba 4 by RNAi, Alba 1 and Alba 2 proteins are also depleted, resulting in a slow growth phenotype. This was more pronounced in late procyclic forms than early procyclic forms, indicating a probable stage-specific function of Alba proteins. For the first time, we were successful in establishing knockdown of Albas in bloodstream forms of a pleomorphic strain of T. brucei (i.e. a strain capable of generating stumpy forms). To have a better understanding of the function of Alba proteins, we have performed in-situ tagging of individual Alba proteins in both procyclic and bloodstream forms. These are being used to identify their mRNA and protein interacting partners and should provide insights into the role of Alba proteins in the course of the trypanosome life cycle.
Non-Canonical Function of β-PheRS B5 Domain
1Institute of Cell Biology, University of Bern, Switzerland; 2Max Planck Institute for Biology of Ageing, Cologne, Germany
Aminoacyl-tRNA-Synthetases (aaRSs) are essential enzymes for loading the appropriate amino acid onto their cognate tRNAs. Recent publications revealed fascinating non-canonical functions for an increasing fraction of these proteins. Several diseases and tumorigeneses were shown to be associated with malfunctioning aaRSs. Among these aaRSs Phenylalanyl-tRNA-Synthetase (PheRS) is overexpressed in multiple cancers. Interestingly, β-PheRS, the non-catalytic subunit, contains the domain B5, which is not involved in aminoacylation and has been reported to be able to bind DNA and possibly RNA. We mutated potentially important sites in the B5 domain and found that two of these mutants were not able to rescue a null mutant, while overexpressing these two B5 mutants in wild type animals caused lethality. These overexpressing larvae showed reduced growth in larval stage L3 and do not pupate. This demonstrates that the domain B5 is essential for viability and it points to the possibility that the domain B5 could perform a growth regulating function. We are presently testing this hypothesis.
Bovine Epithelial in vitro Models for Mycoplasma bovis Infection
1Institute of Veterinary Bacteriology, Vetsuisse Faculty Bern, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Institute of Veterinary Physiology, Vetsuisse Faculty Bern, University of Bern, Switzerland
Mycoplasma bovis causes bovine mycoplasmosis. The major clinical manifestations range from pneumonia to mastitis, arthritis and otitis. In Western Central Europe, an increase in the number of severe mastitis cases was reported in the mid-2000s. Subsequent studies demonstrated a parallel switch in the genotype of circulating strains that might be associated with the current mammary gland pathology. At the molecular level, there is limited understanding of the mechanisms of pathogenicity of M. bovis. Host-pathogen interactions were primarily studied using primary bovine blood cells. To date little is known about the impact of M. bovis on other cell types present in infected tissues. Furthermore, it is still not known if the severity of mastitis is related to genetically distinct M. bovis strains. For this reason, we are developing bovine in vitro systems to investigate epithelial cells-M. bovis interaction using a cell line (MDBK: bovine epithelial kidney) and primary cells (PECT: bovine embryonic turbinate cells and bMEC: bovine epithelial mammary gland cells). Two strains representative of the circulating genotypes were selected to perform the experiments.
Differences in co-culture of bovine cells with the two M. bovis strains were detected more particularly concerning bMEC cells. Infection with M. bovis resulted in variable induction of apoptosis and cytotoxic effects depending on the cell type and M. bovis strain. In summary, the findings showed a potential increased virulence of the newly circulating M. bovis genotype to bovine mammary gland cells. In conclusion, thorough analysis of the host cell response after M. bovis infection in appropriate cell models is an important step to understand the mechanisms of pathogenicity of different M. bovis strains in different bovine tissues.
Astrovirus Associated with Encephalitis in a Sheep—Indication of Cross-Species Transmission with Cattle
1Division of Experimental Clinical Research, DCR-VPH, Vetsuisse Faculty, University of Bern, Switzerland; 2Interfaculty Bioinformatics Unit, University of Bern and Swiss Institute of Bioinformatics, Bern, Switzerland
Introduction: Astroviruses are best known to cause gastroenteritis in different mammal species. Lately, some strains have been associated with encephalitis in humans, minks, cattle and sheep. We report the identification of such a neurotropic astrovirus in a sheep, named ovine astrovirus CH16 (OvAstV-CH16); interestingly, this virus is genetically almost identical to an astrovirus recently described in neurologically diseased cattle, bovine astrovirus CH15 (BoAstV-CH15).
Material and Methods: We submitted three ovine encephalitis samples from our archives to next-generation sequencing (NGS) and a bioinformatics pipeline for virus discovery. We Sanger sequenced the whole genome length of the virus that we obtained in one of these samples, and performed a phylogenetic comparison of the recovered sequence with other known astroviruses. We also tested brain samples of the affected sheep by immunochemistry (IHC) with antibodies aimed at BoAstV-CH15.
Results: One ovine sample that went through next-generation sequencing presented sequences with high similarity to BoAstV-CH15, a virus found recently in cattle with encephalitis. Once the viral sequence was confirmed by Sanger sequencing, a phylogenetic analysis confirmed its close relationship to BoAstV-CH15, as well as to other ruminant neurotropic astroviruses. The samples tested also reacted positively in the IHC.
Discussion: Our study indicates that astroviruses of the same genotype species may cause encephalitis in different species.
Conclusion: Our results confirm a very close relationship of OvAstV-CH16 to BoAstV-CH15, as well as to other astroviruses that were found in association with encephalitis in cattle and sheep.
Acknowledgements: This work was funded in part by the Federal Food Safety and Veterinary Office (grant MON-108), by the Swiss National Science Foundation (grant 31003A_163438) and by the Bangerter-Rhyner-Foundation.
Investigating the Function of Spo11, MSH4, and MSH5 in the Genome Reorganization Process in Paramecium tetraurelia
1Institute of Cell Biology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland
Paramecium tetraurelia, although unicellular, performs a number of complex molecular processes. One such phenomenon is the genome reorganization that occurs during development through a trans-nuclear comparison of two distinct types of nuclei: the germline micronucleus (MIC) and the somatic macronucleus (MAC). During sexual development which starts with meiosis of the germline nuclei, MIC-limited sequences such as Internally Eliminated Sequences (IESs) and transposons are eliminated from the developing MAC in a process guided by noncoding RNAs (scnRNAs and iesRNAs). Our current understanding of this process is still very limited. To understand this phenomenon in more detail my work focuses on investigation the function of proteins expressed during early development. Here, I describe the importance of meiotic proteins, such as Spo11, MSH4 and MSH5 in genome reorganization. Knockdowns of these genes in Paramecium is lethal and leads to the retention of IESs and transposons. However, sequencing data suggests that the levels of scnRNAs and iesRNAs are not affected upon these knockdowns. Also, although these proteins localize to the germline nuclei early during the development, both meiosis and mitosis seem not to be affected upon the knockdown of the genes. Moreover, mRNA-seq data from MSH4 silenced cells have shown that many of the genes involved in genome reorganization process as well as cell cycle were affected. So far, our data indicate that Spo11, MSH4 and MSH5 are necessary for proper genome rearrangement and progeny production in Paramecium. Nonetheless, it still remains puzzling what is the exact mechanism behind these proteins that influence the subsequent genome reorganization processes.
Multidrug Resistance Plasmids in Enterovirulent Escherichia coli Strains from Diarrheic Pigs in Switzerland
Institute of Veterinary Bacteriology, Vetsuisse Faculty, University of Bern, Switzerland
Background: Antibiotics represent the first therapeutic choice for porcine enteric colibacillosis caused by enterotoxigenic and shiga toxin-producing E. coli (ETEC and STEC, respectively), and their use has selected for resistant strains (Brand et al. 2017, Schweiz Arch Tierheilkd, 159: 373-380). Among them, two strains identified as ETEC (strain 14OD0056) and STEC (strain 15OD0495) exhibited a multidrug-resistance (MDR) profile and were further characterized by whole genome sequencing (WGS).
Methods: Minimum inhibitory concentration of antibiotics was measured according to the Clinical and Laboratory Standards Institute guidelines. WGS was performed by PacBio technology and the assembled scaffolds were corrected by using read-mapping paired-end MiSeq Illumina reads. Resistance genes and plasmid typing were obtained using ResFinder and PlasmidFinder (Center for Genomic Epidemiology), respectively. Conjugation was performed at room temperature and 37°C.
Results: Both strains were non-susceptible to ampicillin, gentamicin, sulfamethoxazole, tetracycline, trimethoprim and tobramycin. WGS revealed a MDR plasmid in each strain. 14OD0056 (ETEC) carried plasmid p14ODMR which harbored the resistance genes aac(3)-IIa, aadA1, aph(6)-Id, aph(3")-Ib, aph(3’)-Ia , blaTEM-1, dfrA1, sul1, sul2, tet(B)and tet(C); while 15OD0495 (STEC) carried plasmid p15ODMR which harbored aph(6)-Id, blaTEM-1, dfrA14, aph(3')-Ia (partial), sul2 and tet(A). Both plasmids were transferable in the conjugation assays at both temperatures.
Conclusions: WGS of MDR ETEC and STEC revealed that almost all resistance genes were located on a single conjugative plasmid. Thus, this plasmidic-mediated co-resistance poses the risk of simultaneous selection of several resistance traits, even if only one antibiotic is used during pig husbandry. Targeted therapy following antibiotic susceptibility testing are necessary for a prudent use of antibiotics in pigs.
Supported by FSVO grant Nº1.15.07 (ERA-Net Project PRAHAD).
Translation Regulation by Ribosome-Associated Non-Coding RNA in Haloferax volcanii
Department of Chemistry and Biochemistry, University of Bern, Switzerland
As translation is the final step in gene expression it is particularly important to understand the processes involved in translation regulation. It was shown in the last years that a class of RNA, the nonprotein-coding RNAs (ncRNAs), is involved in regulation of gene expression via various mechanisms (e.g. gene silencing by microRNAs). Most of these ncRNA discovered so far target the mRNA in order to modulate protein biosynthesis. This is rather unexpected considering the crucial role of the ribosome during gene expression. However, recent data from our laboratory showed that there is a new class of ncRNAs, which target the ribosome itself [(1),(2)]. These so called ribosome-associated ncRNAs (rancRNAs) have an impact on translation regulation, mainly by interfering / modulating the rate of protein biosynthesis.
The goal of this project is to identify and describe novel potential regulatory rancRNAs in the halophilic archaea H. volcanii. Northern blot analyses already revealed interactions of rancRNA candidates with the ribosome and showed differential expression of rancRNAs. To investigate the biological relevance of these rancRNAs, knock-outs were generated in H. volcanii which were used for phenotypic characterization studies. The rancRNA s194 showed association with the 50S ribosomal subunit in vitro and in vivo. This RNA was capable of inhibiting peptide bond formation and showed inhibition in in vitro and in vivo translation. These preliminary data for the rancRNA s194 make it an interesting candidate for further functional studies to identify the molecular mechanisms by which rancRNAs can modulate protein biosynthesis.
1. A. Pircher et al., An mRNA-derived noncoding RNA targets and regulates the ribosome. Molecular Cell 54, 147 (Apr 10, 2014).
2. J. Gebetsberger, L. Wyss et al., A tRNA-derived fragment competes with mRNA for ribosome binding and regulates translation during stress. RNA biology, 1 (Nov 28, 2016).
A New Resource: The Cancer Long Non-Coding RNA Census
1Department for BioMedical Research, University of Bern, Switzerland; 2Department of Medical Oncology, Inselspital, Bern University Hospital and University of Bern, Switzerland; 3Centre for Genomic Regulation (CRG), The Barcelona Institute of Science and Technology, Barcelona, Spain; 4Universitat Pompeu Fabra (UPF), Barcelona, Spain
Most of our attention in cancer research has focused on finding protein-coding genes responsible for dysregulation of normal cell growth. These cancer causing genes have been collected in the Cancer Gene Census (CGC) database; a valuable resource for studying cancer gene biology and testing new methods for identifying novel cancer genes.
However, increasing evidence now points non-coding regions as key regulatory elements, and non-coding genes have raised cancer-researchers interest. A promising group for cancer study is the long non-coding RNAs (lncRNAs). LncRNAs have been linked to tumorigenic process, however, little is known about their general implication: how deep their contribution is, which cancer type profile they show, and what their particular features are. Hence, there is the need for a public gene collection resembling CGC, which we can use for the study of lncRNAs in cancer.
Therefore, we created the Cancer LncRNA Census (CLC), a collection of all lncRNAs with functional or genetic evidence for causative roles in tumorigenesis based on the literature. For each entry, we recorded evidence and cancer type, and biological activity (oncogene or tumour suppressor). We also collected lncRNA genomic and expression data and studied what are the relevant features that distinguish CLC genes from other lncRNAs.
CLC comprises 122 lncRNAs implicated in 29 distinct cancers. They are distinguished by a series of features consistent with biological function, including gene length, high expression and sequence conservation. We identified a trend for CLC genes to be co-localised with CGC genes along the human genome. Finally, we show that some mouse orthologues of CLC genes are also cancer genes.
In conclusion, we report CLC, a manual-curated collection of lncRNAs experimentally validated to be involved in tumorigenic process. We aim CLC to become a reference resource to study lncRNA in cancer, as well as a positive set to test new methods for their discovery.
Genetics of Metamizole-Induced Agranulocytosis: Development of a Resequencing Method for HLA Genes
1Department of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department of Clinical Pharmacology and Toxicology, University Hospital Basel, University of Basel, Switzerland; 3Department of Rheumatology, Immunology and Allergology, Inselspital, Bern University Hospital, University of Bern, Switzerland; 4Department of Clinical Pharmacology and Toxicology, Inselspital, Bern University Hospital, University of Bern, Switzerland
Background: Metamizole-induced agranulocytosis (MIA) is a rare and potentially fatal adverse drug reaction that leads to a loss of mature myeloid cells in the blood. While, the mechanism underlying MIA is currently not understood, an increasing number of studies have recently reported associations between specific human leukocyte antigen (HLA) variants and agranulocytosis caused by other drugs. To investigate genetic susceptibility factors for MIA, we thus aimed to develop a method for the comprehensive genetic analysis of HLA genes.
Methods: For the resequencing of eight HLA genes, a high-throughput sequencing (HTS) method was optimized and tested on 10 samples from HLA-typed individuals. Target enrichment by long-range PCR used primers to amplify each gene (4-13kb). For HTS library preparation, genes were grouped by size (pool 1: HLA-A, -B,-C and –DQA1; pool 2: –DRB1, -DQB1, -DPA1 and –DPB). Pooled PCR products were subjected to enzymatic fragmentation and Illumina adapter ligation. After equimolar pooling of barcoded libraries, the sequencing was performed on a Miseq instrument using a Nano flowcell with 2X250 bp reads. Sequencing data was analyzed using the CLC Biomedical Workbench.
Results: All eight HLA genes were successfully amplified and used to construct libraries with a consistent size range of 300 to 3000bp. The final pool of all library was characterized by an average fragment size of 850bp. The sequencing run yielded a total of 3.12 million reads (85,688-166,736 reads per library) and 88.9% of basecalls had a quality score of ≥ 30. A more homogenous distribution of sequence reads across the target genes was observed compared to a previous fragmentation method.
Outlook: HLA alleles determined from HTS data will be compared to HLA typing data to optimize alignment parameters and rule out allelic drop out. This method will subsequently be applied to 20 MIA cases and 20 drug-tolerant controls to identify candidate variants that are associated with MIA.
Epigenetic Changes in DAXX and/or ATRX Negative Pancreatic Neuro-Endocrine Tumors
1Institute of Pathology, University of Bern, Switzerland; 2University College London, London, United Kingdom; 3Interfaculty Bioinformatics Unit, University of Bern, Switzerland
MEN1, DAXX and ATRX are the most commonly mutated genes in pancreatic neuroendocrine tumors (PanNETs). MEN1 mutations are described as an early event in PanNET progression, while DAXX/ATRX mutations are a rather late event. DAXX/ATRX mutations often co-exist with MEN1 mutations. MEN1, DAXX and ATRX are involved in chromatin remodelling and epigenetic regulation.
Our group previously shown that DAXX/ATRXmutated PanNETs have a more aggressive phenotype and global hypo-methylation. Tumor pathways associated with DAXX/ATRX mutation are still unclear. We hypothesize that DAXX/ATRX and MEN1 mutations mediate PanNET progression via epigenetic dysregulation. We aim to describe epigenetic profiles of MEN1, DAXX/ATRX mutated PanNETs, identify pathways epigenetically dysregulated and assess their contribution in tumor development.
We performed methylation profiling on 55 primary PanNETs for which DAXX and ATRX status was assessed by IHC and MEN1 was sequenced. The highest differences in DNA methylation were found when we compared MEN1mut and DAXX/ATRX negative tumors with MEN1wt and DAXX/ATRXpositive tumors (3’078 differentially methylated regions corresponding to 594 genes). MEN1mut yet DAXX/ATRX positive samples represented an intermediate group, suggesting that methylation changes are evolving during tumor progression. Cell development, cell cycle and transcription regulation are few of the main pathways we found to be dysregulated from the DNA methylation analysis.
We assessed the expression of 800 miRNAs in 24 primary tumors and 31 miRNAs were differentially expressed in DAXX/ATRX negative vs DAXX/ATRX positive tumors. Notably Mir548N was down-regulated and simultaneously hypo-methylated in the gene body of DAXX/ATRX negative samples.
Our preliminary results strongly support the hypothesis that DAXX/ATRX loss and MEN1 mutation in PanNETs result in DNA methylation dysregulation which in turn affect RNA expression and that epigenetic changes are involved in PanNET progression.
BBBHub: An Expression Datahub for Brain Barriers
Interfaculty Bioinformatics Unit, Department of Biology, University of Bern, Switzerland
Due to the highly specialized nature of the brain barriers tissues, as well as their development and maintenance process, transcriptome profling is a vital tool to better understand processes in the blood-brain Barrier (BBB). In particular, RNA-seq techniques have provided the opportunity of more thorough studies in many felds of life sciences. Interestingly, the number of studies taking advantage of this technique to study the brain barriers is still very low when compared to the total number of publications concerning the brain barriers. However, we expect that (i) the number of studies using RNA-seq will increase dramatically and (ii) these methodologies will help to bring the knowledge of the BBB to a new level.
We are currently developing «BBBHub»—a web based data hub platform (i) to collect existing and new data (ii) to integrate and organize the omics-data (mainly RNA-seq) (iii) to implement data-mining tools and (iv) to provide the data to both researchers and general public. For the implementation of the BBBHub, we are using the frame work «openBIS», an open source platform for managing scientifc information, as its backbone.
Furthermore, we are developing an easy-to-use frontend system to allow all researchers to use it independently of their computational expertise. In the context of the BBBHub, we are setting up standardized data analysis workfows to allow for a more efcient comparison between datasets and are currently working on quality control determination features. These features will be very helpful to understand the utility and limitations of each dataset.
ATOM19, a Somewhat Special Subunit of the Protein Translocase of the Outer Mitochondrial Membrane in Trypanosoma brucei
Department of Chemistry and Biochemistry, University of Bern, Switzerland
Mitochondria are essential organelles in eukaryotes, best known for their role in ATP synthesis. Because of their endosymbiontic origin, mitochondria are surrounded by two membranes. Thus, the delivery of nucelar encoded mitochondrial proteins that are synthesized in the cytosol is a major and crucial challenge for the eukaryotic cell. Mitochondrial protein import machineries have been extensively studied in Opisthokonts. However, during the last years the archaic translocase of the outer mitochondrial membrane (ATOM) of the early branching Excavate Trypanosoma brucei has also been described. Interestingly, the ATOM complex contains only two subunits that are remotely homologous its yeast counterparts: The beta-barrel protein and actual translocon ATOM40, and ATOM14, which forms together with ATOM40 the core complex. The other four subunits, the two receptors (ATOM69 and ATOM46) as well as the two regulatory subunits (ATOM12 and ATOM11) are trypanosomal specific.
We have shown that the mitochondrial morphology factor TbLOK1 is actually linked to mitochondrial protein import and in fact is a stably integrated subunit of the ATOM complex. Thus we proposed to rename it to ATOM19 in line with the ATOM complex nomenclature. Analysis of the functional interactions between TbLOK1/ATOM19 and the other known ATOM subunits by steady state protein levels and blue native PAGE showed a complex network of interdependencies that revealed a central role of ATOM19 in the ATOM complex organization and stability.
Interestingly, ATOM19 has two predicted transmembrane helixes. This is a unique feature for a subunit of a mitochondrial outer membrane translocase, setting the ATOM complex even further apart from the known Opisthokont translocases. Presently, we are trying to confirm this prediction by biochemical assays. Moreover, truncations of the N- or the C-termini will hopefully give us insight in the essential domains of the protein as well as in their function.
Characterization of Kinase Activity by Phosphoproteomics in Myeloid Cell Lines for Identification of Driving Oncogenic Pathways
1Department of Haematology and Central Haematology Laboratory, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland; 3Interfaculty Bioinformatics Unit and Swiss Institute of Bioinformatics, University of Bern, Switzerland
Introduction: Myelodysplastic syndromes (MDS) are heterogeneous disorders caused by sequential accumulation of genetic lesions in haematopoietic stem cells. MDS are characterized by dysplasia, ineffective haematopoiesis and a propensity to evolve towards acute myeloid leukemia (AML).
The aim is to build a pipeline to integrate phosphoproteomics data with genetic information to characterize targettable kinase-activities of involved oncogenic pathways. Here, we present data on the ongoing phosphoproteome characterization and the inferred kinase-activity from 5 myeloid cell lines.
Methods: Five myeloid cell lines K562, NB4, THP1, OCI-AML3 and MOLM13 were used. Cytogenetic analysis and ionTorrent panel sequencing were performed. Phosphoproteomes were enriched by Titanium-dioxide (TiO2) and samples were analyzed in triplicates by reversed-phase nano liquid chromatography coupled to tandem-mass spectrometry (nanoLC-MS2). Raw data was analyzed using MaxQuant software and processed with R. For kinase-activity enrichment analysis (KAEA), SetRank R-package was used while integrating substrate-kinase datasets from 5 databases.
Results: We identified 15’698, 14’087, 13’969, 13’993 and 14’201 unique phosphopeptides corresponding to 3’536, 3’363, 3’411, 3’410 and 3’403 unique phosphoproteins, respectively. KAEA identified 77 different kinases across the 5 cell lines. Phenotypically related cell lines clustered together with unique kinase-activity pattern for everyone. Two cell lines are driven by oncogenic kinases; K562 by BCR-ABL1 and MOLM13 by FLT3-ITD. An ABL1-kinase signal was detectable in K562 as well as additional downstream kinases contrary to the FLT3-kinase in MOLM-13.
Conclusion: Our pipeline detected unique phosphoproteins and enriched for kinase-activities in distinct cell lines. It reproduced an expected pattern of kinase-activities for two relevant driving kinases. Further improvement using heavy labeled cell lines (SILAC) and kinase-motif analysis is expected.
The Role of CRALBP in the Chemistry of Vision
1Departement of Chemistry and Biochemistry, University of Bern, Switzerland; 2Ruprecht-Karls-Universität Heidelberg, Deutschland; 3Department for BioMedical Research, Clinical Radiopharmacy, University of Bern, and Inselspital, Bern University Hospital, Switzerland
Continuous vision depends on the provision of 11-cis-retinal for assembly with membranous opsin to form light-sensitive rhodopsin. The initial absorption of a photon by rhodopsin effects a change of the shape of bound retinal inside the protein from the cis-form to the trans-form. In bacteria, all-trans-retinal is re-isomerized instantly back to 11-cis-retinal by light driven processes. However, the necessity for higher animals to be able to see at low light conditions (in the night) has led to the development of two highly conserved biochemical pathways for independently driving the trans- to the cis-state in the absence of light. Both pathways afford the concerted action of independent enzymatic activities and are located in adjacent retinal pigment epithelium cells (RPE) and Müller cells respectively. In both cell types, the cellular retinaldehyde binding protein (CRALBP) is essential for facilitating retinoid flux.
We aim to elucidate the structure-function relation of naturally occurring mutations of CRALBP for providing molecular clues to associated blinding diseases. We compare photoisomerization- and transfer-rates of wild-type CRALBP to the ones of disease mutants by time based UV/VIS measurements. We also aim to resolve the structures of the diseased mutants by X-ray crystallography. Our data indicate that distinct disease phenotypes of CRALBP do not rely on a “loss of function” phenotype of the protein but rather on stronger or weaker binding of the visual chromophore. We observe a strong correlation between loss of photo-protection and premature release of bound ligand in two disease phenotypes. On the other hand, at least one disease phenotype effects too strong binding and increased photo-protection eventually leading to impaired release of the chromophore. Both situations may cause in vivo a massive depletion of 11-cis-retinoid flux into the photoreceptors.
DNA-PK Regulates the Radiosensitivity and Oncogenicity of MET-Addicted Cancer Cell Lines via a Novel MET Phosphosite
1Department of Radiation Oncology, Inselspital, Bern University Hospital and University of Bern, Department for BioMedical Research, University of Bern, Switzerland; 2Department of Visceral Surgery, Inselspital, Bern University Hospital and University of Bern, Department for BioMedical Research, University of Bern, Switzerland
MET, the receptor tyrosine kinase for hepatocyte growth factor, is a proto-oncogene mainly expressed in epithelial cells and involved in development, homeostasis and tissue regeneration. Deregulation of MET signaling has been reported in numerous cancers, leading to a broad interest in MET targeting for therapy. Additionally, there is emerging evidence that MET signaling extends to the DNA damage response (DDR) machinery and protects tumors from DNA-damaging agents. A study of post-translational changes in a MET-addicted cancer cell line upon MET inhibition and ionizing radiation revealed a yet unreported phosphorylation site on MET that is part of a consensus motif recognized by master DDR kinases. Within this study we aimed to investigate the function of this novel MET phosphosite in the context of radioresistance and oncogenesis. Our results show that this site can be phosphorylated by DNA-PK and that its phosphorylation fluctuates in response to MET inhibition. Interestingly, MET-addicted transformed cells expressing the phosphodeficient (Ser to Ala mutation) form of active MET are more radiosensitive than their nonmutated counterparts, both in vitro and in a mouse xenograft model. Additionally, we show that preventing this phosphorylation has an impact on the oncogenicity of MET by affecting the migration and anchorage-independent growth capacity of cells. Altogether, our results describe a novel MET phosphosite that bears an important function for radiosensitivity and oncogenicity of MET-addicted cancer cell lines, providing new insights into the crosstalk linking MET and the DDR.
Dosing Variability in Fluoropyrimidine-Treated Cancer Patients: Preliminary Data of a Multicenter Cohort Study
1Department of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Medical Oncology and Hematology, Cantonal Hospital St. Gallen, Switzerland; 4Division of Medical Oncology, Cantonal Hospital Lucerne, Switzerland; 5Department of Medical Oncology, Inselspital, Bern University Hospital, Switzerland
Approximately 10-25% of cancer patients treated with the fluoropyrimidine (FP) 5-Fluorouracil (5-FU) or with the oral prodrug Capecitabine (Cp) develop severe and even life-threating adverse drug reactions (ADRs). Common genetic variants in FP-metabolizing genes explain about 30% of these adverse drug events. FPs have a narrow therapeutic window and show a strong dose-dependent relationship with drug efficacy and toxicity. The high drug exposure variability observed in FP-treated patients, despite individual dose adjustments based on body surface area (BSA), raises some questions regarding current FP-dosing strategies. Indeed some studies reported that standard BSA based dosing reduces individual FP-exposure variability only by about 15-20% and that around 50% of FP-treated patients are actually underdosed, while an unknown fraction of the patients are overdosed. Furthermore, this standard dosing-practice prevents the accurate assessment of genetic and non-genetic factors associated with FP-related ADRs in clinical studies. Data from studies including genetic and phenotypic data on drug exposure are greatly lacking. One of the major objectives of our current multicenter cohort study is to close this knowledge-gap. In detail, our study aims at recruiting up to 500 cancer patients in four years. The drug exposure in blood plasma is monitored in all participants for dosing of 5-FU in the first two chemotherapy cycles using a commercially available immunoassay. Furthermore, patients are genotyped for all currently known FP-toxicity risk variants. This study does not only aim at validating the associations of genetic risk variants, but also to quantitatively assess their effect on the observed drug exposure variability by taking into account clinical and demographic factors, such as age and liver function. Here, we present the preliminary results encompassing data obtained in the first year of recruitment.
Targeting MET Modulates Global Metabolic Pathways and Induces dNTPs Depletion in MET-Addicted Cellular Systems
1Department of Radiation Oncology, Inselspital, Bern University Hospital, Switzerland; 2Department for BioMedical Research, Radiation Oncology, University of Bern, Switzerland; 3Institute of Molecular Systems Biology, ETH Zurich, Switzerland
Overexpression of the MET receptor tyrosine kinase observed in various tumors is often associated with poor prognosis and furthermore, HGF/MET signaling confers resistance to cancer treatments by protecting tumors from DNA-damaging agents. In order to gain an insight into the link between MET and the DNA damage response, we carried out MS-based metabolomics analysis to study METi-induced changes in metabolites and metabolic pathways in various cell lines overexpressing the MET receptor. In the current work, MET-addicted and non-MET-addicted gastric carcinoma and non-small cell lung cancer cell lines have been used. MET inhibition (METi) was accomplished using a specific small molecule tyrosine kinase inhibitor tepotinib, which is currently under clinical evaluation.
METi caused significant alterations in energy metabolism, tricarboxylic acid (TCA) cycle and amino acid metabolism in MET-addicted models. Interestingly, 5'-Phosphoribosyl-N-formylglycinamide, a metabolite involved in the de novo purine synthesis pathway, was consistently altered in all MET-addicted cellular systems upon METi. Complementary transcriptomics analysis showed that critical purine synthesis enzymes such as GART and PPAT are downregulated upon METi, a finding confirmed also by quantitative Real-Time PCR. Moreover, reduced levels of critical purine enzymes resulted in depletion of important DNA building blocks, the dNTPs. Importantly, as inferred from a decrease in yH2AX levels, dNTPs supplementation seems to prevent the induction of DNA damage that occurs in MET-addicted cells after administration of the MET inhibitor. Currently, we are focusing on the role of transcription factors known to be involved in the purine synthesis pathway (e.g., E2F1 that is being downregulated in the transcriptomics dataset upon METi) as the data obtained so far indicate that E2F1 overexpression could potentially lead to a partial rescue of MET-addicted cancer cells after METi.
Mapping Spatio-Temporal Rho GTPase Signaling Networks Through Acute Perturbation of Signaling Fluxes
Institute of Cell Biology, University of Bern, Switzerland
Cell migration is a necessary phenomenon during development and a fundamental process characteristic of several pathologies. It is required, for example, for leukocyte transmigration during inflammatory response and it occurs during cancer during cells spreading.
Rho GTPases are involved in the regulation of several aspects of cell migration. Their function is necessary for cell polarity establishment, cell-substrate adhesion regulation and cytoskeleton remodeling.
Rho protein family is composed of 20 distinct proteins of which 3 members have been intensively investigated, for their biological relevance: RhoA, Rac1 and Cdc42.
Rho GTPase signaling is activated by GEFs (Guanine nucleotide Exchange Factors) and deactivated by GAPs (GTPase Activating Proteins).
Up to now 80 GEFs, and 70 GAPs have been identified as involved in Rho GTPases regulation, thus revealing one layer of complexity, given by the high number of possible signaling modules that can play distinct roles in the cell.
Recently, the introduction of FRET-based biosensors for Rho GTPases allowed to investigate their spatio-temporal patterns of activation, in single living cells. This gave new insights on their regulation and unveiled a much more complex scenario.
Rho GTPases are spatially regulated, by GEFs and GAPs, to generate micrometer-sized sub-compartments that temporally fluctuates on time scales of tens of seconds.
New questions arise from the new insights on Rho GTPases signaling complexity:
– Why there are so many GEFs and GAPs compared to GTPases?
– In which aspect of cell migration are they specifically involved?
I will address these open questions investigating single cell migration through a semi-automated imaging and quantification pipeline.
With this aim I will exploit wide field, TIRF and FRET microscopy, for live cell imaging. Furthermore, I will adopt acute perturbation systems for protein manipulation at the right spatial and temporal resolution.
Bioprospecting for Anti-Chagas Medicinal Plants in Bolivia and Biological Evaluation
1Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland; 2Universidad Mayor de San Andrés, La Paz, Bolivia
In Latin America, an estimated 8 million people are infected with Trypanosoma cruzi, the etiologic agent of Chagas disease. The current chemotherapy regimen is outdated and produces frequent side effects. Bolivia has the world’s highest infection rates and primarily the lowest socioeconomic population is affected, which relies on traditional medicine for the treatment of many infectious diseases. We conducted an ethnobotanical study in selected parts of Bolivia with the aim to study the consensus on the efficacy of anti-Chagas plants and to collect plant material in situ for subsequent extraction and in vitro testing. Altogether, a total of 95 plant species have been deposited and identified at the National Herbarium of Bolivia.
135 plant extracts of the collected material were screened for general cytotoxicity in different cell lines and for anti-trypanosomal activity in T. cruzi epimastigote formsand T. brucei procyclic forms. Different plant species showed apparent toxic selectivity for trypanosomes over the screened mammalian cells and will be followed up in the T. cruzi amastigote stage. Moreover, 660 medicinal plant extracts described in Dioscorides’ De Materia Medica were screened with the aim to compare results of our ethno-directed approach with a random screening. Our findings indicate that the ethno-directed approach did not lead to the identification of a significant larger percentage of species with anti-trypanosomal activity but that plant natural products nonetheless constitute an interesting starting point for anti-Chagas drug discovery.
Mechanistically, we focus on plant natural products that inhibit the T. cruzi fatty acid amide hydrolase (FAAH) which is crucial for the generation of ethanolamine in the parasite. The parasite appears to take the FAAH substrates (N-acylethanolamines) from the host cells. To that aim, we have established the biochemical assays to screen for this target using our active extracts and focused libraries.
Fusion-Mediated Membrane Protein Incorporation into Giant Unilamellar Vesicles
Department of Chemistry and Biochemistry, University of Bern, Switzerland
One of the current challenges in synthetic biology is the production of stable membrane mimetic systems and the insertion of various components in these systems, such as membrane proteins (MPs). We employ fusion of oppositely charged liposomes to deliver separately reconstituted MPs into a common lipid bilayer of small and giant vesicles.
After a systematic evaluation of different lipid compositions by lipid mixing and size distribution analysis, suitable conditions were further investigated for proteoliposome fusion of small vesicles. With this technique, we functionally co-reconstituted different terminal oxidases and ATP synthase from Escherichia coli into unilamellar vesicles. Successful fusion was confirmed by measuring ATP synthesis upon redox driven proton pumping.
Furthermore, reconstitution of proteins into giant unilamellar vesicles (GUVs) is still a challenging task and is an active field of research. We have inserted different protein populations into small positively charged liposomes and fused these with negatively charged GUVs. The insertion of different proteins into the same GUV was confirmed by fluorescence microscopy and functional analysis. Observing giant vesicles directly under the microscope gives us the opportunity to detect protein activity and to follow kinetic process on the basis of a single vesicle.
By using our fusion-based approach as a tool for membrane protein reconstitution into GUVs, we are able to study a wide range of proteins in a cell mimicking system.
Characterization of a Novel Mutation in Human Growth Hormone: Insight into the Cause of Growth Hormone Deficiency
1Department of Pediatric Endocrinology, Inselspital, Bern University Children’s Hospital, and Department for BioMedical Research, University of Bern, Switzerland; 2Department of Endocrinology, Great Ormond Street Hospital for Children, London, UK, and Section of Genetics and Epigenetics in Health and Disease, Genetics and Genomic Medicine Program, UCL Institute of Child Health, London, UK; 3Pediatric Endocrinology, University of Minnesota Masonic Children’s Hospital, MN, USA
Human growth hormone (GH) has various physiological and metabolic effects in human including its pivotal role in postnatal somatic growth especially during childhood and adolescence. It is mainly expressed in the pituitary gland as a 191-amino acid polypeptide and acts via binding to growth hormone receptor (GHR). Mutations in GH1 gene cause isolated growth hormone deficiency (IGHD) by affecting the production, secretion and stability of GH as well as its binding to GHR. Several GH1 mutations had been shown to be associated with IGHD with reduced GH secretion combined with low concentrations of IGF-1 leading to short stature. Most studies reporting IGHD have focused on defects in transcription or secretion of GH protein due to mutations in GH1 gene. Here we aim to study the structural and functional effect of a novel missense mutation in the GH1 gene converting codon 61 from proline (P) to serine (S) yielding a mutant GH-P61S polypeptide found in a patient with GHD. GHD was identified in a patient and sequencing of GH1 gene revealed missense mutation in GH1 gene resulting in a single amino acid change of the GH protein (pP61S). Computational analysis showed that P61 residue is highly conserved in mammals and predicted it to have a functional role. In silico mutagenesis and molecular dynamics simulations indicated altered folding and structural stability in GH-P61S as compared to wt-GH. Biochemical procedures were further conducted to examine the prediction made from in silico study. GH-P61S showed lower stability towards thermolysin degradation during fast parallel proteolysis assay at different temperature as compared to wt-GH, confirming the bioinformatics prediction. However, intrisic tryptophan fluorescence assay revealed that the unfolding pattern of GH-P61S seemed similar to wt-GH. We proved that advanced protein chemistry and computational analysis methods could be used to characterize the molecular basis of defects in GH1 gene causing IGHD.