Logo GCB Symposium 2018
 

Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

 
Session Overview
Session
Posters B: Poster Session B
Time:
Thursday, 01/Feb/2018:
1:30pm - 3:00pm

Location: DCB, Foyer to Room U113 & Cafeteria, basement
Department of Chemistry and Biochemistry, basement, Freiestrasse 3, 3012 Bern

Posters B-092 to B-182


Show help for 'Increase or decrease the abstract text size'
Presentations

Tight Junction Proteins for Normal Liver Regeneration

Felix Alexander Baier, Giulio Loforese, Ramesh Kudira, Fadi Jebbawi, Daniel Candinas, Guido Beldi, Deborah Stroka

Visceral Surgery Research Laboratory, Department for BioMedical Research, University of Bern, Switzerland

Background & Aims: Failure to regenerate after surgical resection is a major cause of death in advanced liver diseases. To improve the clinical outcome after intervention, it is crucial to first understand the basic mechanisms which initiate and regulate liver regeneration. Within this project, we studied tight junction proteins as potential triggers of the regenerative process.

Methods: Expression of junctional genes was measured by RNAseq, RT-PCR and GEO database analysis. Liver regeneration was induced using a mouse 70% - partial hepatectomy model. Hepatic proliferation was analyzed using immunofluorescent staining in regenerating livers of C57BL/6 mice and CLDN3 -/- or IL-22 -/- mice. Human primary isolated hepatocyte cultures were incubated with recombinant IL-22 protein.

Results: Murine and human liver express a distinct set of tight junction associated genes, in particular Cldn1, Cldn3, Cldn12, Tjp1, Tjp2, Jam-a, Ocln and Tric. 6h after 70% - partial hepatectomy, we observed a significant downregulation of the mRNA expression of Cldn1, Cldn3, Tjp, Ocln followed by a restoration or upregulation above the initial expression levels. CLDN3 Protein levels coincided with the observed mRNA expression. Immunofluorescent histology revealed a hepatocyte nuclear accumulation of CLDN3 at 48h post partial hepatectomy. Aged CLDN3-/- mice had significantly decreased hepatocyte proliferation and decreased cell cycle gene expression when compared to aged CLDN3 +/+ control mice. IL-22-/- mice showed downregulated Cldn1, Cldn3, Cldn12, Tjp1, Ocln and Jam- levels at 24h and 48h after partial hepatectomy.

Conclusions: Tight junction gene expression is regulated during liver regeneration. CLDN3 -/- mice had decreased hepatocyte proliferation, indicating importance of tight junctions for normal liver regeneration. Mice lacking IL-22 had inhibited tight junction upregulation after partial hepatectomy, suggesting an involvement of cytokine for tight junction regulation in the liver.


Directional Effects of Accurately Controlled Uniaxial Strain on Action Potential Propagation in Cultured Cardiomyocyte Strands

Andrea Buccarello1, Michela Azzarito1, Frédéric Michoud2, Stéphanie P. Lacour2, Jan P. Kucera Kucera1

1Department of Physiology, University of Bern, Switzerland; 2Bertarelli Foundation Chair in Neuroprosthetic Technology, Laboratory for Soft Bioelectronic Interfaces, Institute of Microengineering, Institute of Bioengineering, Centre for Neuroprosthetics, École Polytechnique Fédérale de Lausanne (EPFL), Geneva, Switzerland

Background: Slow cardiac conduction is a well-known mechanism of arrhythmias. Deformation of cardiac tissue can modulate tissue resistance, membrane capacitance and ion currents (e.g., stretch-activated channels), and hence it can affect conduction velocity. In this study we investigated whether uniaxial strain influences conduction velocity (θ) in different manners when applied parallel vs. perpendicular to the direction of action potential propagation.

Methods: Custom stretchable microelectrode arrays were used to determine θ during steady-state pacing of patterned cardiomyocyte strands. Uniaxial strain (5%), accurately controlled by imaging a grid of markers, was applied for 1 min parallel to propagation (orthodromic strain) and perpendicular (paradromic strain). Results were interpreted in terms of cable theory.

Results: Uniaxial strain induced immediate changes of θ upon application and release. In material coordinates, orthodromic strain decreased θ significantly more (p<0.001) than paradromic strain (2.2±0.5% vs 1.0±0.2% in mouse cardiomyocyte cultures, n=8; 2.3±0.4% vs 0.9±0.5% in rat cardiomyocyte cultures, n=4, respectively). According to our theoretical analysis, the larger effect of orthodromic strain can be explained by the increase of axial myoplasmic resistance, which is not altered by paradromic strain. Thus, changes in tissue resistance accounted for a substantial part of the changes of θ during strain, in addition to other influences (e.g., stretch-activated channels). Besides these immediate effects, strain also consistently led to a slow decrease of θ when the preparations were kept stretched and to a slow recovery of θ after release.

Conclusion: Our findings suggest that slowing of cardiac conduction is not only determined by the extent of acute strain but also on the orientation of strain itself relative to impulse propagation. Thus, arrhythmogenic effects of strain depend on the spatiotemporal relationship between strain and propagation.


Adult Sox10+ Cells Contribute to Myocardial Regeneration in the Zebrafish

Marcos Sande-Melon1,2, Ines Marques1,2, Maria Galardi2, Juan Manuel Gonzalez-Rosa3,4, Fernando Rodriguez-Pascual5, Nadia Mercader1,2

1Institute of Anatomy, Developmental Biology and Regeneration Research Group, University of Bern, Switzerland; 2Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain; 3Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA, USA; 4Harvard Medical School, Harvard University, Cambridge, MA, USA; 5Hospital La Paz, Comunidad de Madrid, Spain

During heart regeneration in the zebrafish, fibrotic tissue is replaced by newly formed cardiomyocytes that derive from pre-existing ones. It is unclear whether all cardiomyocytes have an equal capacity to replace lost myocardium. In opposition to mammals, sox10+ neural crest cells were proposed to contribute to the embryonic zebrafish myocardium.

Here we examined the contribution of sox10-derived cells to the adult heart. Embryonic sox10-derived cardiomyocytes persisted in adult hearts but did not participate during regeneration. Surprisingly, a resident sox10+ cell pool in the adult heart expanded massively after cryoinjury. Inhibition of collagen maturation impaired this cell expansion and blocked heart regeneration. Finally, genetic ablation of sox10+ cells severely impaired cardiac regeneration.

Thus, our results show that rather than being detrimental, transient fibrosis is necessary for regeneration and that sox10 marks a subset of cardiomyocyte with high regenerative potential.


In vitro Assessment of Xenogeneic Complement and Endothelial Cell Activation Using a Microvascular Flow Model

Riccardo Sfriso1, Oliver Steck1, Andrea Bähr2, Nikolai Klymiuk2, Eckhard Wolf2, Robert Rieben1

1Department for BioMedical Research, University of Bern, Switzerland; 2Molecular Animal Breeding and Biotechnology, Ludwig Maximilian University Munich, Germany

In vitro models aiming to mimic the in vivo situation present in small vessels are lacking. Traditional systems present a major gap of not providing a sufficient endothelial surface to exploit the natural anticoagulant properties of the quiescent endothelium. In xenotransplantation, such advanced vascular model would be helpful to test genetically modified porcine aortic endothelial cells (PAEC) in a more physiological system. Here we introduce a 3D microfluidic system in which PAEC were cultured within micro-channels and subjected to peristaltic flow.

Microfluidic chips were fabricated in PDMS using needles of 100-550 µm in diameter as molds. PAEC cells were seeded and a physiological flow of 10 dyn/cm2 was applied once a monolayer of cells had formed. Immunofluorescence staining of F-actin and VE-cadherin was performed to assess the coverage of the entire microchannel and to highlight morphological changes of the cells under physiological flow. Visualization of the heparan sulfate proteoglycans was also performed. Different PAEC cells were perfused with normal human serum (NHS) to mimic a xenotransplantation setting and complement C3b and C4b depositions were assessed by confocal microscopy. Perfusates were analyzed to detect pro-inflammatory cytokines and soluble complement components.

A nice monolayer of endothelial cells elongated in the direction of the flow was observed. Deposition of C3b and C4b was strong on WT PAEC while it was markedly reduced on transgenic PAEC. A strong increase of pro-inflammatory cytokines and soluble complement activation markers was observed after perfusion with NHS.

The study revealed the potential of a new 3Dmicrofluidic model in reproducing key findings of antibody- and complement-mediated hyperacute rejection. The system might be useful to screen novel, genetically modified PAEC for activation of complement and coagulation when perfused with human serum, plasma, or eventually whole blood.


Lesion-Dependent Plasticity of Spinal Cord Circuits in vitro With and Without Stem Cell Insertion

Samuel Buntschu, Jürg Streit

Department of Physiology, University of Bern, Switzerland

Unlike peripheral nerves, the central axons fail to regenerate after injury and at present there exists no cure for spinal cord injury (SCI) with only a limited spontaneous recovery observed depending on the extent of the lesion. Transplantation of embryonic stem cells to improve SCI repair has been a promising therapeutic strategy investigated in several studies over the last decades, but the exact cellular mechanisms involved are still unclear. A further hypothesis is that recovery may be promoted after SCI by inducing propriospinal plasticity in an activity-dependent manner. In a previous project our group developed a new in vitro model for the study of functional regeneration after SCI consisting of organotypic slice co-cultures of E14 embryonic rat spinal cord. Using this model and optogenetic tools we aim to better characterize the specific plasticity of endogenous neuronal circuits as well as the effect of the transplantation of pre-differentiated embryonic stem cells into SC lesions.

Synchronous optogenetic stimulation of both co-cultures after lesion over two to three weeks in vitro to promote circuit plasticity failed to improve functional regeneration. In contrast, transplantation of embryonic stem cells pre-differentiated into embryoid bodies (EB) induced a re-establishment of neuronal connections between the two transected slices, allowing propagation of activity over the lesion site and thus, improving functional recovery.

A greater impact on recovery occurred when the transplantation was performed at the time of the lesion, while significant increase in recovery was observed only under disinhibition condition when the EBs were transplanted one week after the lesion.

Together these findings suggest that pre-differentiated embryonic grafts are able to induce the formation of functional connections between two transected spinal slices in vitro to induce functional regeneration with a critical role of the time of application.


Characterization of Novel Subtypes of Macrophages During Zebrafish Fin and Heart Regeneration

Andres Sanz Morejon1,2, Ana Garcia-Redondo2, Ines Marques2, Nadia Mercader1,2

1Institute of Anatomy, University of Bern, Switzerland; 2Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain

Organ regeneration is preceded by an inflammatory response and recruited macrophages play an active role during repair and regrowth. Macrophage subtypes are well characterized in mammals, but not in the zebrafish, an animal model with high regenerative capacity. Using a wilms tumor 1 b (wt1b) reporter line, we identified a subpopulation of macrophage-like cells, which accumulate at the site of injury in a model of cardiac cryoinjury and fin amputation. This population revealed a different gene expression profile and migratory behaviour compared to the rest of mpeg1+ macrophages. After arrival at the amputation plane, mpeg1+ cells start to express wt1b, and mpeg1+ /wt1b+ macrophages are retained at the injury site. Functional inhibition of wt1b in macrophages promoted their motility, suggesting that wt1b is involved in the retention of macrophages at the site of injury thus allowing them to promote regeneration of the missing body part. This is the first description of a pro-regenerative macrophage subtype in the zebrafish and expands our understanding on the molecular mechanisms through which macrophages contribute to organ regeneration.


The Role of Brain-Barrier Tight Junctions in Neuroinflammation

Mariana Dias1, Caroline Coisne1, Pascale Baden1, Ivana Lazarevic1, Michael Vanlandewijk2,3, Liqun He2,3, Masaki Hata4, Noriko Iwamoto4, David Francisco5, Adolfo Odriozola6, Isabelle Gruber1, Ruth Lyck1, Gaby Enzmann1, Urban Deutsch1, Remy Bruggmann5, Christer Betsholtz2,3, Schoichiro Tsukita4, Benoit Zuber6, Mikio Furuse4, Britta Engelhardt1

1Theodor Kocher Institute, University of Bern, Switzerland; 2Karolinska Institutet, Huddinge, Sweden; 3Uppsala University, Uppsala, Sweden; 4Division of Cerebral Structure, National Institute for Physiological Sciences, Okazaki, Japan; 5Interfaculty Bioinformatics Unit, University of Bern, Switzerland; 6Institute of Anatomy, University of Bern, Switzerland

During neurological disorders, such as multiple sclerosis (MS) or its animal model experimental autoimmune encephalomyelitis (EAE), focal loss of brain barriers’ integrity and tight junctions (TJ) alterations are described. Claudin-3 is thought to be localized at the blood-brain barrier (BBB) and blood-cerebrospinal fluid barrier (BCSFB) TJ. Claudin-3 specific contribution to BBB integrity is suggested by its selective loss in inflamed microvessels in EAE and by its identification as a transcriptional target of the Wnt/β-catenin signalling pathway involved in BBB maturation in development. This prompted us to have a first focus on the role of claudin-3 in barriers integrity in C57BL/6 mice. Unexpectedly, claudin-3-/- mice did not have aggravated EAE in comparison to the WT condition. In vitro assays on the barrier properties of our BBB and BCSFB models revealed no differences between WT or claudin-3-/- mice. Interestingly, its absence increased BCSFB expression of claudin-1 and claudin-2 in vivo. However, we concluded that absence of claudin-3 does not impair barrier properties nor does it impact EAE pathogenesis in C57BL/6 mice.

Besides TJ alterations in MS context, autoaggressive T cells cross the BBB and enter the CNS parenchyma. T-cell diapedesis across the BBB can be paracellular, through the endothelial TJ, or transcellular, through the endothelial cell body. In our in vitro BBB model, different endothelial ICAM-1 levels direct T cells to paracellular or transcellular pathways. By using serial block face scanning electron scanning microscopy (SBF-SEM), we now analyze, in a 3D ultrastructural level, T-cell interactions with the ICAM-1hi or ICAM-1lo BBB under physiological flow. Our preliminary data suggest that paracellular T-cell diapedesis occurs preferentially at the unique tricellular junctions. We will therefore assess the molecular composition of these BBB tricellular junctions to understand their role in T-cell diapedesis in neuroinflammation.


Depletion of Paneth Cells Is Associated with Decreased Portal Hypertension and Angiogenesis After Partial Portal Vein Ligation in Mice

Mohsin Hassan1, Sheida Moghadamrad1, Philipp Kellmann1, Coralie Trentesaux2, Marie Fraudeau2, Béatrice Romagnolo2, Andrea De Gottardi1

1Department for BioMedical Research, Hepatology, University of Bern, and Inselspital, Bern University Hospital, Switzerland; 2Institut Cochin, INSERM, Biomedical Research Institute, Department of Development, Reproduction and Cancer, Paris, France

Background and Aims: Paneth cells may contribute to the regulation of splanchnic hemodynamics in response to microbial stimuli from intestinal flora. However, the detailed mechanistic links have not been elucidated yet. We hypothesized that conditional deletion of intestinal Paneth cells might have regulatory effects on portal hypertension and angiogenesis after partial portal vein ligation (PPVL) in mice.

Method: Math-1 Lox/LoxVilcreERT2 (a tamoxifen-dependent conditional model of Paneth cell depletion) or wild type mice were injected three consecutive doses of tamoxifen and subsequently underwent PPVL or sham surgery. Portal pressure (PP) and the development of portosystemic shunts (PSS) were assessed 14 days after PPVL. Intestinal and mesenteric angiogenesis was assessed by immunohistochemistry (IHC) using anti-CD-31 and anti-Lyve-1 antibodies. Expression of genes involved in the regulation of angiogenesis was evaluated by RT2 profiler PCR array in intestinal tissue.

Results: PP was significantly attenuated in Paneth cell depleted mice compared to control mice after PPVL (n=7/group, 8.78±1.23 cmH2O vs 11.75±1.53 cmH2O, respectively, p<0.01). This was associated with a decrease in PSS (n=3/group, 42.8% ± 4.6 vs 65.3% ± 3.0, p=0.13). Depletion of Paneth cells also resulted in a significantly decreased density of blood and lymphatic vessels as assessed by IHC (n=5, pixel ratio, 2.7% ±1.9 vs 13.9%±2.3 p=0.05 and 10.5%±2.9 vs 29.0%±7.2 p=0.03, respectively). Quantitative gene expression measurements showed a differential expression in 69 angiogenic genes, whereby 47 proangiogenic genes including Ang-1, Angpt-1, VEGFR1&2, VEGFb, VEGFc, and Nrp1 were downregulated in Paneth cell depleted mice.

Conclusion: Portal hypertension was significantly decreased in mice lacking Paneth cells. The decreased density of blood and lymphatic vessels suggest that Paneth cell-derived factors may act on portal hypertension and angiogenesis.


Expression and Localization of the Canine Luteal and Hypophyseal Relaxin (RLN) System: Functional Implications for Luteal Function

Marta Nowak, Alois Boos, Mariusz P. Kowalewski

Institute of Veterinary Anatomy, Vetsuisse Faculty, University of Zurich, Switzerland

Relaxin (RLN) plays several roles during pregnancy in mammals. By acting through its receptors, RXFP1 and RXFP2, it leads to remodeling of the pelvic girdle in order to prepare the birth canal for parturition. Additionally, indirect luteotropic effects have been suggested for primates and pigs, through stimulation of prolactin (PRL) release from the hypophysis. In the domestic dog, RLN of placental origin is the only known endocrine marker of pregnancy. Since the canine placenta lacks steroidogenic activity, circulating steroids originate from the corpus luteum (CL). Progesterone profiles are similar in pregnant and non-pregnant dogs, however numerically higher concentrations were measured during pregnancy. Starting around the second third of diestrus, luteal maintenance becomes dependent predominantly on PRL. Serum PRL increases shortly after placental RLN appears in the circulation. By applying immunohistochemistry (IHC) and RT-qPCR, we investigated the expression and localization of RLN, RXFP-1 and -2 in canine CL during pregnancy: pre-implantation (n=5), post-implantation (n=5), mid-gestation (n=5) and at normal (n=3) and 24/72h after antigestagen-induced luteolysis (Alizine®, 10 mg/kg bw; 2x/24 h apart, n=10). Moreover, co-localization of the RLN-system and PRL was assessed in canine hypophysis by IHC. Luteal expression of RLN was time-dependent, increasing following implantation and decreasing at prepartum. Antigestagen-treatment led to suppression of luteal RLN and RXFP2, but not of RXFP1. RLN was localized in luteal cells, whereas receptors were additionally strongly represented in luteal macrophages. All RLN-system members were found in canine hypophysis. They were co-localized in pituitary cells expressing PRL. In conclusion, RLN seems to be involved in auto/paracrine regulation of CL function in dogs in a time-dependent manner. Circulating, but also locally produced RLN, appears to act in adenohypophyseal cells providing PRL. SNSF grant nr: 31003A_160251.


Evaluation of a New Drug for Its Intraocular Fibrinolytic Efficacy in a Rabbit Model

Katrin Voelter1, Karina Klein2, Nicole Borel3, Deborah Brütsch1, Simon Pot1

1Equine Department, Division of Ophthalmology, Vetsuisse Faculty, University of Zurich, Switzerland; 2Center for Applied Biotechnology and Molecular Medicine CABMM, University of Zurich, Switzerland; 3Insitute of Veterinary Pathology, University of Zurich, Switzerland

Purpose: A new substance was evaluated and compared to tissue plasminogen activator (t-PA) for its intraocular fibrinolytic effect in a rabbit uveitis model. D-dimer concentrations were evaluated for their use in assessing fibrinolytic activity in the eye.

Methods: In an ex vivo pilot study corneal and scleral permeability of the drug were evaluated. In an in vivo trial fibrin clots were induced in anterior chambers of 44 rabbits and drug efficacy was measured by clot size reduction over 24 hours (h). Eyedrops containing the new agent were compared to carrier solution in a multiple drop regime in 8 animals per group. Intracameral new substance (n=14) and 25ug t-PA (n=14) were assessed for their fibrinolytic capacity. Biocompatibility was determined performing clinical examinations and histopathology. Animals were euthanized 24h after drug application. D-dimer concentrations in aqueous humor were measured at baseline, before and after drug application with an enzyme-linked immunosorbent assay.

Results: The drug permeated corneoscleral tissue in different species ex vivo. No significant differences were seen between topically treated animals in different groups. Intracameral t-PA had higher fibrinolytic efficacy at early timepoints. Twenty-four hours post application, no significant difference was seen between intracameral groups. Animals with t-PA treatment showed significantly more side effects compared to animals treated with the new agent. No ocular toxicity was noted in histologic examinations of all globes. D-dimer concentration differed neither between treatment groups nor over time.

Conclusions: The new agent is a promising fibrinolytic drug with similar 24h fibrinolytic efficacy and less side effects than t-PA when injected into the anterior chamber of rabbits. D-dimer measurements in aqueous humor provided no additional information concerning fibrinolysis in a rabbit model.


Intra-Graft Tacrolimus-Loaded Hydrogel—A Local Alternative to Systemic Immunosuppression in Composite Allografts?

Dzhuliya Vihrenova Dzhonova1,2, Radu Olariu3, Jonathan Leckenby3, Yara Banz4, Jean-Christophe Prost5, Ashish Dhayani6, Praveen Vemula6, Esther Voegelin3, Adriano Taddeo1,3, Robert Rieben1

1Department for BioMedical Research, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Plastic and Hand Surgery, Inselspital, Bern University Hospital, Switzerland; 4Institute of Pathology, University of Bern, Switzerland; 5Center of Laboratory Medicine, Department of Clinical Chemistry, Inselspital, Bern University Hospital, Switzerland; 6Institute for Stem Cell Biology and Regenerative Medicine, Bangalore, India

Localized immunosuppression is an alternative to systemic immunosuppression in VCA. However, modalities of application, effects and pharmacokinetics of local immunosuppression must be carefully evaluated. We performed Brown Norway-to-Lewis rat hind-limb allotransplantations using intra-graft injections of a Tacrolimus-loaded hydrogel (TGMS-TAC) and compared this to systemic immunosuppression by Tacrolimus (TAC).

Recipient rats were randomly divided in two groups: TGMS-TAC (intra-graft injection of 1 mL hydrogel containing 7 mg TAC, repeated every 70 days, n=6) and systemic-treatment (s.c. TAC, 1 mg/kg daily, n=6). Graft survival was monitored for 280 days or until grade III rejection. TAC levels, toxicity markers, hematopoietic chimerism and Tregs in blood and skin were examined.

In both groups 83.3% of grafts survived for 280 days. TAC concentration in VCA skin was 1.1±1.49 ng/mg in TGMS-TAC treated and 0.46±0.39 ng/mg in systemically TAC-treated animals (p>0.05). In skin of contralateral limbs TAC concentrations were 0.25±0.19 and 0.29±0.16 ng/mg, respectively, in the TGMS-TAC and TAC groups (p>0.05). Blood levels of TAC were 9.29±5.89 ng/mL in the TGMS-TAC and 13.44±4.44 ng/mL in the TAC group (p=0.03). Blood urea nitrogen was 12.62 ± 1.6 mmol/l in TGMS-TAC treated animlas and 20.86 ± 4.86 mmol/l in systemic TAC treated (p<0.0001). Creatinine was 27.2 ± 4.21μmol/l in TGMS-TAC treated animlas and 49.8 ± 17.08 μmol/l in systemic TAC treated (p=0.0117). Agressive diffuse large B-cell lymphoma and pseudocyst infected with Staphilococcus aureus and Proteus mirabilis were found in systemic TAC treated rats. Hematopoietic chimerism was detectable but declining until the endpoint, with TGMS-TAC animals having significantly higher levels compared to systemic TAC. Treg counts were comparable between the two groups.

Our findings suggest that localized immunosuppression with hydrogel could reduce the burden of systemic immunosuppression for composite allotransplant recipients.


ProteinY* as a Treatment Option for Hearing Loss Caused by Experimental Pneumococcal Meningitis

Silvia Erni1,2,3,4, Michael Perny1,2,3, Michelle Buri1,2,3, Gabriella Fernandes1,2,3, Alessio Balmelli1,2,3, Rolf Jan Rutten5, Denis Grandgirard1,2, Pascal Senn3,6, Marta Roccio2,3,7, Stephen L. Leib1,2

1Neuroinfection Laboratory, Institute for Infectious Diseases, University of Bern, Switzerland; 2Cluster for Regenerative Neuroscience, Department for BioMedical Research, University of Bern, Switzerland; 3Laboratory of Inner Ear Research, Department for BioMedical Research, University of Bern, Switzerland; 4Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 5Audion Therapeutics, Amsterdam, The Netherlands; 6Department of Otorhinolaryngology, Head & Neck Surgery, Hôpitaux Universitaires de Genève, Switzerland; 7Department of Otorhinolaryngology, Head & Neck Surgery, Inselspital, Bern University Hospital, Switzerland

Over 5% of the world’s population suffer from hearing loss. Treatment options are currently limited and novel pharmaceutical treatments are needed. Sensorineural hearing loss is the most common long-term deficit after pneumococcal meningitis, occurring in up to 30% of surviving patients.

Here, we tested whether the recombinant ProteinY*, with previously reported anti-inflammatory, anti-apoptotic and neuronal protective functions, improves hearing loss in experimental pneumococcal meningitis.

Pneumococcal meningitis was induced in infant rats randomized for treatment with systemically applied ProteinY (1-3x 50 µg) (n=30) or vehicle (n=31). Histological assessments and hearing thresholds determined by auditory brainstem responses (ABR) were performed at different times after infection.

Neutrophil extracellular trap (NET) formation was observed in the scala tympani of the cochlea during the acute phase of the infection and increased hearing thresholds were measured 1 week later, persisting 3 weeks post infection.

ProteinY treatment significantly lowered ABR at 1 and 3 weeks post infection and reduced the percentage of animals with high hearing thresholds (80-100 dB). No significant differences between groups were observed in NET formation in the early stages of the disease.

Three weeks after infection, spiral ganglion neuron density, inner and outer hair cell counts did not differ significantly between groups. When animals were stratified according to the severity of hearing loss, we observed significantly more surviving inner and outer hair cells in the ProteinY-treated group in animals with severe hearing loss (threshold above 80 dB). In contrast, no significant differences were detected at the level of presynaptic ribbons between the two groups.

Further studies will be needed to unravel the mechanisms behind the protective role of ProteinY and validate its use as therapeutic option in case of severe pneumococcal meningitis.

*ProteinY (Patent currently being filed).


Epithelial Growth Factor Receptor Expression Influences 5-ALA Induced Glioblastoma Fluorescence

Deborah Barbara Angela Sarah Maria Piffaretti1,2, Andrea Orlando Fontana2, Francesco Marchi2, Floriana Burgio2,3, Ana Bela Faia-Torres3, Paolo Paganetti2, Sandra Pinton2, Uwe Pieles3, Michael Reinert2,4

1Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 2Laboratory for Biomedical Neurosciences (LBN), Department of Neurosurgery, Neurocenter of Southern Switzerland, Ente Ospedaliero Cantonale, Torricella-Taverne, Ticino, Switzerland; 3Fachhochschule Nordwestschweiz (FHNW), Campus Muttenz, Switzerland; 4Neurosurgery, Neurocenter of Southern Switzerland, Lugano, Switzerland

The extent of 5-aminolevulinic acid (5-ALA) guided tumor resection has determining impact in highgrade glioma and glioblastoma (GBM) surgery. Intensity of the 5-ALA induced fluorescence is variable in biomolecularly and histologically confirmed GBM. Furthermore the detection method is currently subjective and non-quantitative. Epithelial Growth Factor Receptor (EGFR) expression is variable in GBM. We therefore aimed to correlate 5-ALA induced fluorescence with the quantitative expression EGFR in vitro.

To elucidate the role of EGFR in the metabolism of 5-ALA in glioblastoma cell lines (U87MG, BS153 and LN229EGFR) with variable EGFR status, we analyzed the activation of EGFR by its primary ligand EGF and its downstream effect on Heme oxygenase-1 (HO-1), a key enzyme regulating the metabolism of Protoporphyrin IX (PpIX). Effects of pharmacological inhibition by Tin(IV)-Protoporphyrin (SnPP) or gene knockdown (KD) by HO-1-specific siRNA were analyzed in respect to 5-ALA induced fluorescence.

A significant difference in 5-ALA induced fluorescence was observed in U87MG (low EGFR expression) and LN229EGFR cells (EGFR overexpression) compared to BS153 (EGFR overexpression/ EGFRvIII+). Treatment of U87MG and LN229EGFR cells with EGF lead to reduced cellular fluorescence, by promoting HO-1 transcription and expression in a concentration-dependent manner. Inhibition of HO-1 activity by SnPP or gene KD were able to restore the fluorescence in all the cell lines, independently of EGFR expression.

In GBM cell lines, 5-ALA induced fluorescence is variable and influenced by EGF-induced downstream activation of HO-1. HO-1 protein expression was identified as a negative regulator of 5-ALA induced fluorescence. We further propose that co-expression of EGFRvIII but not quantitative EGFR expression influence HO-1 activity and therefore cellular fluorescence. These findings justify the clinical observation of different intensities of 5-ALA induced fluorescence during glioblastoma surgery.


Estrogen Regulation of Thymus and T Cell Differentiation in Teleost Fish: An Eco-Toxicological Perspective

Larissa Kernen, Helmut Segner

Center for Fish and Wildlife Health, Vetsuisse Faculty, University of Bern, Switzerland

In mammals, an important target organ of the immunomodulatory actions of estrogens is the thymus and T cell differentiation. Teleost fish are evolutionary the first vertebrates to possess a thymus. The role of estrogens in thymus development and function of teleost fish is unknown to date. In my thesis, I test two hypotheses: the first hypothesis is that the thymus and T cell differentiation of teleosts fish are responsive to estrogens, as it is the case in mammals. The second hypothesis is that environmental estrogen-active compounds disrupt thymus development and function of teleost fish. The relevance of this question comes from the fact the aquatic environment is a sink for estrogen-active compounds, and they are able to disrupt sexual differentiation and reproductive performance of fish. In mammals it is well known that thymus changes with life cycle (age, sex and reproductive stage), however in fish it is not well known, thus at a first step, I am studying the development of thymus size and morphology as well as the expression of estrogen receptors throughout the entire life cycle of zebrafish. The preliminary result is that according to the age, the thymus seems to increase from 5 to 10 weeks post fertilization and then slowly decrease. If we look at the sex difference in thymus development, the females and the males both seems to have the same thymus maximum size at 10wpf. But of course 5 individuals per sex and age are not enough to have a robust conclusion and I need to finish the study and sampling. Also the reproductive stage can be related to the thymus size in both males and females.


Tolerance of Coronary Endothelium to Ischemia and Reperfusion: Studies in an Isolated Rat Heart Model of Donation After Circulatory Death

Natalia Méndez Carmona, Rahel K. Wyss, Maria Arnold, Thierry P. Carrel, Hendrik T. Tevaearai Stahel, Sarah Longnus

Department of Cardiovascular Surgery, Inselspital, Bern University Hospital, University of Bern, Switzerland

Background: Donation after circulatory death (DCD) could significantly improve cardiac graft availability, thereby allowing a greater number of transplantations for patients in need of this life saving therapy. However, DCD hearts undergo potentially deleterious warm ischemia and reperfusion (I/R).

As endothelial damage is a key factor in cardiac I/R injury, we aimed to characterize hemodynamic and endothelial function following various durations of warm ischemia to improve the timing and choice of cardioprotective therapies.

Methods: Isolated, working, rat hearts were perfused for 20’ aerobically, then underwent various periods of warm global ischemia (I), followed by reperfusion (R) of 30’ or 60’. Hemodynamic parameters were monitored and endothelial function was assessed by comparing vasodilatory responses between endothelium-dependent (bradykinin; 10-9 and 10-8 M) and endothelium-independent (sodium nitroprusside; 3x10-5 M) vasodilators.

Results: At 60’ R, recovery of left ventricular work (heart rate–developed pressure product) was significantly lower with 27’ I (76% ± 12), 30’ I (67% ± 15) and 33’ I (31% ± 20) compared with non-ischemic hearts (0’ I; 100% ± 11; p <0.05 for all), but was unchanged with 21’ I (91% ± 5) and 24’ I (88% ± 13; n=7-8/ group).

After 30’ R, endothelial function was impaired after 24’ I compared with 0’ I hearts, while smooth muscle function was impaired only with ≥27’ I (p <0.05 for all; n=6/group).

At 60’ R, the proportion of phosphorylated endothelial nitric oxide synthase significantly increased with ≥21’ I compared with 0’ I hearts (p <0.05), but was not changed with 33’ I.

Conclusion: Endothelial dysfunction occurs with shorter periods of ischemia than those required to induce hemodynamic and smooth muscle dysfunction. A window of opportunity therefore exists for application of endothelial-based therapies aimed at optimizing both endothelial and myocardial recovery.


Genetic Dissection of Notch Pathway's Role in Mice Papillary Thyroid Carcinoma

Florian Traversi1, Alexandre Sarre2, Amandine Stooss1, Matthias Dettmer3, Roch-Philippe Charles1

1Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland; 2Cardiovascular Assessment Facility, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland; 3Institute of Pathology, University of Bern, Switzerland

Thyrocyte-derived cancers are categorized into different sub-types based on histology. Papillary thyroid cancer (PTC) is the most prevalent with up to 80% of the cases. Among the various mutations described in thyroid cancer, the most common one affects BRAF, encoding for BRAFV600E. This mutation constitutively activates the MAP-kinase pathway leading to the aberrant behavior of the cancer cells by promoting proliferation and reducing apoptosis. Several mechanisms could be co-operating with MAP-kinase pathway to promote tumor progression. One putative pathway is NOTCH signaling which is implicated in diverse functions. In thyroid cancer, current data on NOTCH pathway’s involvement are insufficient, contradictory and therefore inconclusive.

To investigate NOTCH pathway’s role in tumor progression, a BRAFV600E-PTC mouse model has been combined with an ectopic expression of NOTCHIC, the active form of NOTCH1. Inversely, this BRAFV600E-PTC model has been associated with an obliteration of NOTCH pathway by deleting RBPj, an essential transcription-activator of the NOTCH pathway. All of these alterations are induced specifically in thyrocytes, in a timely controlled manner.

Interestingly, while mice with a BRAFV600E mutation and NOTCH pathway inactivation (RBPj knockouts) present a classical PTC, our PTC model coupled with an over-activation of NOTCH signaling (NOTCHIC) led to significantly bigger tumors, with the expression of more aggressive variants of PTC associated with a dramatic decrease in overall survival.

Recent studies propose a possible interaction between the MAP-kinase pathway and NOTCH. Our findings suggest that NOTCH could modulate MAP-kinase pathway resulting in worsening symptoms of PTC in mice. Further in vitro analysis are required to investigate in more depth, the possible interaction between the MAP-kinase and the NOTCH pathway.


Increased Sensitivity to Apoptosis upon ER Stress-Induced Activation of the UPR in Chemotherapy-Resistant Malignant Pleural Mesothelioma

Duo Xu1,3, Shun-Qing Liang1,3, Hai-Tang Yang1,3, Ren-Wang Peng1,2, Ralph Alexander Schmid1,2

1Division of General Thoracic Surgery, Inselspital, Bern University Hospital, Bern, Switzerland; 2Department for BioMedical Research,Thoracic Surgery Group, University of Bern, Switzerland; 3Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Introduction: Malignant pleural mesothelioma (MPM) is a rare but aggressive cancer, notoriously known for lack of therapeutic options and for extremely poor prognosis. The present standard of care for patients with advanced MPM is the dual chemotherapeutic regimen that combines cisplatin and pemetrexed(MTA), which, however, is frequently confronted by intrinsic and/or acquired drug resistance, leading inevitably to tumor relapse. Identifying molecular mechanisms underlying MPM resistance to standard chemotherapy holds promise to unveil new path towards development of effective therapeutic strategies.

Methods: In vitro three-dimensional (3D) cultivation of MPM cell lines and ex vivo organotypic culture of patient-derived xenograft (PDX) tumors were used to enrich chemotherapy-resistant MPM cells and to mimic in vivo MPM environment, respectively. Quantitative real-time polymerase chain reaction (qRT-PCR), Western blots, immunohistochemistry (IHC) and flow cytometry (FACS) were used to probe endoplasmic reticulum (ER) stress and apoptotic cell death. Cell viability was determined by a colorimetric approach based on the activity of acid phosphatase (APH assay). Clonogenic assay was used to assess growth inhibition.

Results: MPM cells resistant to standard chemotherapy (cisplatin plus MTA) displayed deregulated UPR, which renders the cells hypersensitive to agents that induce ER stress. Bortezomib, a clinically approved drug that targets proteasome activity, preferentially impaired chemotherapy-resistant MPM cells. Mechanistically, hyperactivation of the PERK/ATF4/CHOP pathway is important for bortezomib-induced UPR and apoptosis, as genetic perturbation of the pathway by CHOP depletion attenuated the effectiveness of bortezomib.

Conclusions: Our study provides a mechanistic link between UPR signaling and chemoresistance in MPM, and shows for the first time that altering ER stress might be a rationale to treat patients with relapsed chemoresistant MPM.


Gestational Diabetes Mellitus Affects Placental Iron Homeostasis

Jonas Zaugg1,2, Malgorzata Wegner1,2, Xiao Huang1,2, Marc Baumann3, Daniel Surbek3, Meike Körner4, Christiane Albrecht1,2

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; 4Pathologie Länggasse, Bern, Switzerland

During pregnancy iron is transferred from the mother to the fetus across the placental barrier involving the transferrin receptor (uptake), DMT1 (release from endosomes) and FPN1 (fetal transfer). In a clinical context, gestational diabetes mellitus (GDM) has been associated with elevated maternal iron plasma levels, but an impact of GDM on placental iron transport has never been investigated. Thus, we measured 24 genes involved in iron homeostasis in placentas of GDM patients and controls. Results showed that FPN1 and DMT1 were significantly downregulated in GDM, while levels of the oxidoreductase zyklopen were increased. Using immunohistochemistry in placental tissue, we found DMT1 apically accentuated, while FPN1 was predominantly basolaterally oriented. These results support the notion that FPN1 is involved in iron transfer at the fetal side, whereas DMT1 plays a yet poorly defined role in intracellular iron uptake. To reveal the underlying mechanisms associating hyperglycemia with altered placental iron homeostasis, we generated novel BeWo trophoblast cell lines adapted to normo-, hyperglycemic and hyperglycemic-obese conditions. Under hyperglycemic conditions, iron uptake was significantly reduced confirming an impact of elevated glucose levels on iron homeostasis and involved transport systems. The underlying mechanisms are currently investigated focusing on oxidative stress, ER stress, apoptosis and autophagy. Moreover, to avoid antibody specificity problems detecting transmembrane proteins in immunoblot assays, we developed a mass spectrometry based assay for DMT1 and FPN1 protein quantification.

In conclusion, this is the first time the effect of diabetes and iron homeostasis is reported in placental cells. The placenta seems to reduce uptake of highly oxidative iron under hyperglycemic condition to protect both the fetus and itself from oxidative damage. Alternatively, this could also explain the higher risk to develop GDM due to iron supplementation.


Prostaglandin-Mediated Effects on Vascularization and Immunomodulation in Early Canine Corpus Luteum

Miguel Augusto Tavares Pereira1, Aykut Gram1, Bernd Hoffmann2, Tomasz Janowski3, Alois Boos1, Mariusz P. Kowalewski1

1Institute of Veterinary Anatomy, Vetsuisse Faculty, University of Zurich, Switzerland; 2Clinic for Obstetrics, Gynaecology and Andrology, Faculty of Veterinary Medicine, Justus-Liebig-University, Giessen, Germany; 3Department of Animal Reproduction, University of Warmia and Mazury, Olsztyn, Poland

The canine corpus luteum (CL) has some unique features when compared with those of other domestic species. In fact, this organ is the only producer of progesterone in the dog. Additionally, in non-pregnant animals, no active luteolysis is observed. Also, the canine CL shows a transitional gonadotropic independency in the first two to four weeks after ovulation. At this time, local PGE2 is considered to be among the main regulators of luteal function. This is supported by the high luteal expression of PTGS2/COX2, PGE2-synthase and cAMP-dependent PGE2 receptors (EP2 and EP4). Moreover, in vitro stimulation of canine luteal cells with PGE2 increases their steroidogenic function and modulates the expression of vascular factors. This was also substantiated by in vivo observations where the suppression of prostaglandin (PG) synthesis in the early luteal phase, by a specific COX2-blocker (firocoxib), suppressed luteal production of PGE2, lowered circulating progesterone and negatively affected luteal development. Using samples from this previous in vivo study, this follow-up project aimed to understand PG-mediated effects by assessing possible vascular and immunomodulatory effects. The most important findings following blocking of PGE2 synthesis are: increased expression of endothelin-1 (END-1) and diminished expression of angiopoietin (ANG) family members. Concerning the immune system, expression of CD4 and pro-inflammatory interleukins (IL) 1β, -6 and -12 was increased. These data show a modulatory role of PGE2 over angiogenesis, blood vessel stabilization and local immunomodulation in the canine CL. (Supported by SNSF grant nr: 31003A_160251)


Mitochondrial Integrity at Early Reperfusion Predicts Post-Ischemic Cardiac Graft Recovery

Rahel Kathrin Wyss1,3, Natalia Mendéz Carmona1,3, Maria Nieves Sanz1,3, Maria Arnold1,3, Thierry Carrel1,3, Siamak Djafarzadeh2,3, Hendrik Tevaearai1,3, Sarah Longnus1,3

1Department of Cardiovascular Surgery, Inselspital, Bern University Hospital, Switzerland; 2Department of Intensive Care Medicine, Inselspital, Bern University Hospital, Switzerland; 3Department for BioMedical Research, University of Bern, Switzerland

Introduction: Cardioprotective therapies and reliable means for graft evaluation are essential in facilitating heart transplantation with donation after circulatory death (DCD). Given the key role of mitochondria in cardioprotection, we investigated the tolerance of cardiac mitochondria to ischemia (IS) and reperfusion (RP), and determined the predictive value of early RP mitochondrial parameters for cardiac recovery.

Methods: Isolated, working rat hearts underwent 0, 21, 24, 27, 30, or 33 min warm, global IS followed by 60 min RP. Left ventricular work (LVW; heart rate * developed pressure), cardiac O2 efficiency (O2 consumption/ LVW), and cytochrome c release were monitored.

Additional hearts were stopped after 10 min RP. ATP and PCr levels, mitochondrial respiration, Ca2+ retention capacity, and ROS production were measured.

Results: LVW recovery at 60 min RP significantly decreased with IS ≥27 min compared with no IS (p<0.01; n=4-8/grp for all results). Cytochrome c release at 10 min RP, also increased with ≥27 min IS vs. no IS (p<0.05).

IS of ≥21 min decreased cardiac O2 efficiency, ATP levels, mitochondrial coupling, and Ca2+ retention capacity versus no IS (p<0.01 for all).

ROS production through reverse electron transfer increased with 27 min IS vs. no IS (p<0.001), but was unchanged with 33 min IS. Direct ROS production by complex I increased with 33 min IS only (p<0.05).

Cardiac O2 efficiency (ρ=0.77), ATP (ρ=0.80), complex I/II mitochondrial coupling (ρ=0.82, ρ=0.73) and cytochrome c (ρ=0.76) at 10 min RP significantly correlated with LVW at 10 min RP (representative for recovery; p<0.01 for all).

Conclusions: Disruption of mitochondrial integrity occurs with shorter periods of IS than hemodynamic dysfunction. Mitochondrial function is particularly sensitive to early reperfusion damage, suggesting potential targets for cardioprotection. Early reperfusion indicators of mitochondrial integrity appear to be promising predictors for post-ischemic cardiac recovery.


Automatic Localization of Lumbar Vertebral Landmarks in CT Images Using Context Features

Dimitrios Damopoulos1, Ben Glocker2, Guoyan Zheng1

1Institute for Surgical Technology and Biomechanics, University of Bern, Switzerland; 2Biomedical Image Analysis Group, Imperial College London, United Kingdom

The goal of this study is the development of a method for the automatic localization of the lumbar vertebral bodies (VBs) in CT images and the localization of key landmarks of the vertebral processes. Such a method can facilitate subsequent automated procedures, such as the segmentation of the vertebrae, the automatic assessment of vertebral pathologies and the analysis of the spinal shape.

The developed method consists of two modules. The first one localizes the VBs and estimates their pose. It performs this task via the detection of the vertebral endplates on an unwrapping of the input image along the curve of the spine. The second module identifies the vertebrae based on the output of the first module and localizes key landmarks on each of them. For this purpose, two layers of random forest regressors are used. The second layer employs context features, extracted from the probability maps generated by the first layer.

The method is evaluated on a dataset of 28 lumbar-focused CT images. 5 anatomical landmarks are defined on every lumbar vertebra: The center of the VBs and the tips of the 4 inferior articular processes. 20 images are randomly selected to be used for training and the held-out 8 images are the testing set. During evaluation, the method achieves a mean localization error of 3.0 mm, with a 1.6 mm standard deviation and a median value of 2.7 mm. 95.4% of the detections have a localization error of below 6 mm.

Our evaluation suggests that the developed method can detect reliably the vertebral landmarks on the lumbar spine. Although we focused on the articular processes, we expect that the method can be applied to more vertebral landmarks, such as key endplate landmarks for the measurement of spondylolisthesis. In the future, we would like to explore such a direction. Future research also includes more extensive evaluation on larger datasets.


How Do Critically Ill Patients in the Intensive Care Unit Perceive and Process Visuo-Acoustic Virtual Reality (VR) Stimulation: Feasibility and Proof of Concept of a VR Setup

Stephan M. Gerber1, Marie-Madlen Jeitziner2, Patric Wyss1, Alvin Chesham1, Prabitha Urwyler1,3, Urs P. Mosimann1, René Müri1,4, Stephan M. Jakob2, Tobias Nef1

1ARTORG Center for Biomedical Engineering Research, Gerontechnology & Rehabilitation Group, University of Bern, Switzerland; 2Department of Intensive Care Medicine, Inselspital, Bern University Hospital, Switzerland; 3Department of Geriatric Psychiatry, University Hospital of Psychiatry and Psychotherapy, Bern, Switzerland; 4Department of Neurology, Division of Neurorehabilitation, Inselspital, Bern University Hospital, University of Bern, Switzerland

Background: Around 70% of patients in the intensive care unit (ICU) suffer long-term functional deficits and reduction of quality of life after prolonged stay in the ICU. It is assumed that the noisy and stressful ICU environment exposes patients to both stimulus overload and deprivation and puts them at risk of alterations in sensory perception that can result in cognitive dysfunction. The visual exposure to natural environments (e.g. landscape) restores physiological, emotional and attentional functions and there is some evidence, that the early onset of an intervention (Stimulation) has the biggest impact on later outcome.

Aim: The aim of this observational study was to measure if critically ill patients after a complex heart surgery can perceive and process visuo-acoustic virtual reality (VR) stimulation.

Method: The VR stimulation presented inside a head-mounted display isolated patients from disturbing environmental audio- and visual input. The stimulation consisted of an immersive nature video (five minutes in length), which was presented prior to ICU admission, during their stay and after discharge. In addition, a built-in eye tracker to measure the level of attention, sensors to assess physiological parameters and neurological testing were used.

Results: Oculomotor data in the prior to ICU admission measurement matches findings in healthy volunteers, whereas after surgery oculomotor data was reduced. Fixation/saccade ratio was decreased when no visual target was presented, reflecting enhanced visual search and reduced visual processing.

Conclusion: Overall stimulation did not evoke any negative reactions and oculomotor data indicates post-surgery fatigue of patients. Importantly, critically ill patients were able to perceive and process the VR Stimulation even in the ICU. From these findings, we conclude that VR stimulation in ICU settings is feasible for patients as soon as sedative medication has been stopped.


An Optimal Multi-Path Tracking Approach for Gaze-Based Ground Truth Annotation of Medical Sequences

Laurent Lejeune, Jan Grossrieder, Raphael Sznitman

ARTORG Center for Biomedical Engineering Research, Ophthalmic Technology Laboratory, University of Bern, Switzerland

Modern machine learning strategies, especially when applied to medical imaging,
require large image datasets with associated ground-truth examples to optimize
models with increasing number of parameters. The present work aims at replacing
the traditional mouse annotation interface with an eye-gaze tracker. In this
setting, the user simply observes an object of interest visible in a video or
volumetric sequence. The eye-gaze coordinates allow us to generate pixel-wise
segmentations in volumetric and video image data. Our approach tracks observed
regions over time at superpixel level solving efficiently a maximum a posteriori problem,
involving appearance, location and motions priors.
Our approach not only provides superior performance to existing methods but is applicable to a variety of applications, even in cases where the object is non-compact with branching parts.


Feasibility and Accuracy of Robotic Multi-Port Cochlear Implantation—Evaluation on a Phantom

Daniel Schneider1, Igor Stenin2, Juan Ansó1, Markus Huth3, Lukas Anschütz3, Jan Hermann1, Wilhelm Wimmer1, Jörg Schipper2, Marco Caversaccio3, Stefan Weber1, Thomas Klenzner2

1ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland; 2Department of Otorhinolaryngology, University Hospital Düsseldorf, Germany; 3Department of Ear, Nose, Throat (ENT), Head and Neck Surgery, Inselspital, Bern University Hospital, Switzerland

Robotic cochlear implantation through a multi-port approach (Stenin et al., 2017) comes with advantages relating to surgical manipulation and visualization and contact with the ossicular chain can be avoided. This study investigated the feasibility of a minimally invasive cochlear implantation (CI) via a multi-port approach.

Based on medical images (CT, 0.15 x 0.15 x 0.2mm) of a temporal bone (Phacon), four trajectories (∅ = 1.8mm, l = 21.5 to 31.8mm) from the surface of the temporal bone to the round window were planned collision-free to anatomical structures: retro-facial (RF), suprameatal (SM), sub-facial (SF) and through the facial recess (FR), and drilled with a robotic (Weber et al., 2017). Six attempts for electrode insertion were undertaken (visualization through EndoGnost, ∅ = 0.55mm, 0° angle, PolyDiagnost and manipulation with a straight needle, Medicon). Every approach consisted of a trajectory for electrode insertion (always FR), visualization and surgical manipulation (combination of RF, SM and SF). For each combination the insertion depth and the time used for insertion was determined. On the basis of postoperative CT images, the trajectories were geometrically gauged.

The mean drilling error was 0.15mm (SM 0.12mm, FR 0.2mm, RF 0.07mm, SF 0.2mm). For a full electrode-array insertion, an approach consisting of the trajectories FR, RF and SF was most suitable (12 of 12 electrodes inserted within 4:30 minutes). Visualization and manipulation via the trajectories RF and SF respectively allowed the visualization and reachability of the round window and additional manual assistance during the insertion of the electrode-array. The remaining approaches were inappropriate due to the lack of visualization or reachability of the round window.

Direct view of the round window and reachability with the tool axis being tilted away from the round window membrane are essential for CI. Choosing the correct combination of trajectories, a multi-port CI is feasible.


Semi-Automatic MRI-Based 3D Models of the Hip Joint Can Replace CT-Based 3D Models for Virtual Impingement Detection and Range of Motion Analysis for Patients with Femoroacetabular Impingement

Till Lerch1, Simon Steppacher1, Florian Schmaranzer1, Klaus Siebenrock1, Guoyan Zheng2, Moritz Tannast1

1Department of Orthopaedic Surgery and Traumatology, Inselspital, Bern University Hospital, Switzerland; 2Institute for Surgical Technology and Biomechanics, University of Bern, Switzerland

Introduction: Femoroacetabular impingement (FAI) and hip dysplasia are three-dimensional deformities of the hip joint that can cause hip pain and osteoarthritis in young and active patients in the child-bearing age. CT-based 3D reconstruction for virtual impingement detection is the actual gold standard for diagnosis FAI. But CT scans are limited in clincal routine due to radiation exposure. To replace CT scans, we performed semi-automatic 3D reconstruction based on radiation-free routine MRI of the hip joint of symptomatic patients with FAI.

Therefore, we compared CT-based 3D models of the hip joint with MRI- based 3D models of the same patients and asked the following questions:

(1) What is the mean surface distance?

(2) Do diagnostic parameters correlate?

(3) Do range of motion and impingement zones correlate between CT-based 3D models and MRI-based 3D models?

Methods: We performed a retrospective comparative study including a total of 31 hips of 26 symptomatic patients with hip pain due to FAI or hip dysplasia. Out of 63 hips we selected 31 hips with a CT scan and MRI of the entire pelvis and distal femoral condyles of the same patient that were obtained routinely for evaluation for hip-preserving surgery. Threshold-based manual CT segmentation was performed using commercial software (AMIRA) and served as ground truth.

Results: (1) Median surface distance was 0.4 ± 0.1 mm (0 – 1) for the proximal femur and 0.5 ± 0.2 mm (0 – 1) for the acetabulum comparing CT-based and MRI-based 3D models.

(2) Correlation for six diagnostic parameters was excellent (r=0.993, p<0.001) between MRI-based and CT-based 3D models. Mean absolute difference for inclination was 2.1 ± 1.8 (0 – 7), for anteversion was 1.3 ± 1.1 (0 – 5).

Conclusion: Based on these results, we change our clinical practice and we use the novel technique for MR-based reconstruction of 3D models. This can potentially reduce the need for CT scans in young and active patients in the childbearing age with hip pain due to FAI.


Personalized Adaptive Basal-Bolus Insulin Algorithm Using Data from Glucose Meters and Continuous Glucose Monitoring Systems

Qingnan Sun1,2, Marko V. Jankovic1,3, Christoph Stettler4, Stavroula Mougiakakou1,4

1ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland; 2Graduate School of Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Emergency Medicine, Inselsptial, Bern University Hospital, Switzerland; 4Department of Endocrinology, Diabetes and Clinical Nutrition, Inselspital, Bern University Hospital, Switzerland

The majority of individuals with Type 1 diabetes (T1D) are using either Self-Monitoring of Blood Glucose (SMBG) devices or Continuous Glucose Monitors (CGMs) to measure the glucose concentration. Scope of this work is to introduce an algorithmic approach for estimating the basal and bolus insulin to be delivered by a pump, in a personalized and adaptive manner, independent of the used glucose measuring devices.

The proposed adaptive basal-bolus algorithm (ABBA) is based on reinforcement learning (RL), a type of Artificial Intelligence (AI) algorithm, for optimizing the insulin to be delivered to individuals with T1D independent of the used glucose measuring devices. The ABBA uses glucose information of the current day to adjust the basal rate and three Carbohydrate-to-Insulin ratios (CIRs), for breakfast, lunch and dinner respectively, for the next day.

The algorithm was in silico evaluated with FDA approved UVa/Padova T1DM Simulator v3.2 with 33 virtual subjects. The total simulation duration was 98 days with the last 7 days for evaluation. The used scenario involved three main meals and one bedtime snack per day, along with different variabilities for insulin sensitivity, meal time and carbohydrate amount. Furthermore, uncertainty was introduced to simulate the error when patient estimates the meal’s carbohydrate content. Both variabilities and uncertainty follow uniform distribution.

The proposed system achieves comparable performances for both CGM and SMBG input signals without affecting the total daily insulin dose. The implementation of the system will offer diabetic patients the possibility of personalized adaptive insulin administration and glucose control independent of the used glucose measuring devices.


Conduction Speed on the Left Atrial Posterior Wall—Towards Non-Invasive Functional Substrate Assessment

Romy Sweda1,2,3, Andreas Moser4, Reto Wildhaber4, Thomas Niederhauser4, Marcel Jacomet4, Josef Goette4, Hildegard Tanner2, Andreas Haeberlin1,2

1ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland; 2Department of Cardiology, Inselspital, Bern University Hospital, Switzerland; 3Institute of Cell Biology, University of Bern, Switzerland; 4Institute for Human Centered Engineering, Bern University of Applied Sciences, Biel, Switzerland

Slow conduction speed (CS) is well known to increase the risk for cardiac arrhythmias, including atrial fibrillation. It is frequently observed in diseased hearts, where tissue fibrosis, pathological membrane excitability and abnormal cell coupling lead to an unphysiological propagation of depolarization. Currently, contrast enhanced MRI can be used to morphologically assess the degree of substrate anomaly, while methods to functionally characterize atrial tissue without the need for expensive imaging or invasive testing are missing. One possibility to fill the gap might be esophageal electrocardiography (eECG), as the close anatomical relationship provides it with an excellent view of the left atrium (LA).

We hypothesized, that by using eECG, it would a) be possible to estimate the CS on the LA posterior wall and that b) this could be used as a marker for the functional integrity of the underlying tissue.

As currently available esophageal ECG catheters only allow calculation of projected velocities, a novel ECG catheter with 3-dimensional electrode arrangement was designed. Subsequently, an algorithm to compute the 3D CS from catheter recordings was developed. The algorithm’s performance was tested in an in vitro setup. For this, a prototype of the catheter was placed in a pot filled with water. A printed circuit board (PCB) plate with adjustable azimuth and inclination angle that was connected to digital outputs was used for the generation of stimuli. The digital outputs were controlled by a MATLAB script in which the movement of a dipole along a straight line was simulated. The speed of the dipole as well as the angle of the PCB were set to different values and the resulting signals acquired. After adequate filtering, the algorithm correctly estimated the dipole speed in the chosen range of 0.1–2m/sec.

This shows that estimation of CS is possible from eECG recordings. In a next step, the methodology will be tested under real-life conditions in a clinical study.


Development of a Lung Alveoli Array-on-Chip with a Collagen-Elastin Membrane

Pauline G. V. Zamprogno1, Sven Achenbach1, Janick D. Stucki1, Nina Hobi1, Nicole Schneider-Daum2, Claus-Michael Lehr2, Hanno Huwer3, Ralph A. Schmid4, Olivier T. Guenat1,4,5

1ARTORG Center for Biomedical Engineering Research, Organs-on-Chip Technologies, University of Bern, Switzerland; 2Drug Delivery (DDEL), Helmholtz-Institute for Pharmaceutical Research Saarland (HIPS), Saarland University, Saarbrücken, Germany; 3SHG Clinics, Department of Cardiothoracic Surgery, Völklingen Heart Center, Völklingen, Germany; 4Division of General Thoracic Surgery, Inselspital, Bern University Hospital, Switzerland; 5Department of Pulmonary Medicine, Inselspital, Bern University Hospital, Switzerland

Introduction: Standard in vitro lung alveoli models poorly reproduce the microenvironment of the distal airways. Recently reported, advanced in vitro models, called lung-on-chips, reproduce the air-blood barrier including the cyclic stress of the breathing movements. However, they are made of a polydimethylsiloxane (PDMS) membrane, whose chemical and physical properties differ by far from those of the alveolar basal membrane.

Objective: The goal of the project is to develop an array of tiny lung alveoli with physiological dimensions, equipped with a stretchable biological membrane made of extracellular matrix (ECM) proteins instead of PDMS.

Material & Methods: We develop a new membrane made of collagen and elastin, two main components of the basal membrane, which provide the elasticity to the tissue. The membrane is supported by a hexagonal gold mesh, which mimics the physiological dimension of the lung alveoli.

For the development of the alveoli model, we used human alveolar type II cells from patient provided by the N. Daum group. Additionally, we used primary lung endothelial cells from Angiocrine Bioscience (Veravec).

Results: The newly developed biological membrane is porous and stretchable. It allows the development of an in vivo-like lung alveoli model, with a monolayer of type II (ATII) and type I (ATI) like cells in co-culture with endothelial cells. The primary cells form a tight barrier, and they express markers of tight junction even at ALI condition. Preliminary results demonstrate that the lung alveoli model can be cultured for several days under cyclic stress.

Discussion: This new advanced model mimics the composition, dimensions, and mechanical stretch of the lung alveoli. The long-term stability of the membrane and its entirely biological nature makes this model a promising tool for drug discovery.


The Effect of Cardiac Risk Factors on Coronary Atherosclerosis Burden Among Patients Undergoing Percutaneous Coronary Intervention: A Multicenter Three Vessel Optical Coherence Tomography Study

Thomas Leonardo Gian-Luca Zanchin1,2, Tomoyo Sugiyama2, Erika Yamamoto2, Lei Xing2, Krzysztof Bryniarski2, Hang Lee2,3, Ik-Kyung Jang2

1Department of Cardiology, Inselspital, Bern University Hospital, Switzerland; 2Massachusetts General Hospital, Harvard University, Boston, USA; 3Biostatistics Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA

Background: Risk of recurrent cardiovascular events among patients undergoing percutaneous coronary intervention (PCI) has been shown to vary largely between individuals depending on their cardiac risk profile.

Purpose: The aim of the current study was (I) to assess atherosclerosis severity and distribution by means of optical coherence tomography (OCT) in the entire coronary artery tree among patients undergoing clinically indicated PCI, and (II) to elaborate the impact of cardiac risk factors on coronary atherosclerosis severity.

Method: A total of 101 patients with de novo coronary artery disease underwent three-vessel frequency-domain optical coherence tomography within the MGH OCT Registry. Fibrous plaques (FP), fibroatheromas (FA) and fibrocaclific plaques (FCP) were assessed according to current consensus standards.

Results: A total of 421 plaques were identified (30% FP, 39% Thick cap FA, 17% Thin-cap FA and 15% FCP). Number of plaques per patient varied with 21% of patients having one or two plaques, 38% having three or four plaques and 39% having five or more plaques. The distribution of FP, FA and FC within each coronary vessel is shown in figure 1. Independent predictors for total plaque length were age (1.13mm per year, 95% CI 0.37-1.9, p=<0.01), BMI (3.29mm per unit, 95% CI 0.63 – 5.95, p=0.02), diabetes mellitus (19.3mm, 95% CI 1.16-37.5, p=0.04), male sex (29.2mm, 95% CI 12.19-46.1, p=<0.01) and there was a trend for smoking (18.6mm, 95% CI -0.3–37.4, p=0.05). Total fibroatheroma length was independently predicted by male sex (29.4mm, 95% CI 11.2-47.6, p=<0.01) and there was a trend for diabetes mellitus (19.0mm, 95% CI -0.7-38.8, p=0.06) and smoking (19.3mm, 95% CI -0.8–39.5, p=0.06).

Conclusions: Coronary atherosclerosis severity as assessed by three-vessel OCT largely differs among patients undergoing coronary revascularization. Gender, diabetes and smoking represent independent predictors of total plaque length and presence of fibroatheroma.


Human Lung Microvasculature-on-Chip for Drug Testing: Anti-Angiogenic Efficacy of Nintedanib

Soheila Zeinali1, Colette A. Bichsel2, Nina Hobi1, Manuela Funke3,4, Olivier T. Guenat1,3,5, Thomas Geiser3,4

1Organs-on-Chip Technologies Laboratory, ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland; 2Vascular Biology Program, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA; 3Department of Pulmonary Medicine, Inselspital, Bern University Hospital, Switzerland; 4Department for BioMedical Research, University of Bern, Switzerland; 5Division of General Thoracic Surgery, Inselspital, Bern University Hospital, Switzerland

Differences in cellular responses observed in standard in vitro studies, animal models, and human clinical studies, decrease efficiency of developing new therapeutic strategies. Recent studies have shown the feasibility of in vitro self-assembled perfusable microvessel networks. This microfluidic co-culture system enabled the formation of the microvessel network that resembled human lung microvasculature in terms of morphology, vascular marker expression, permeability, and vasoactive response. By reproducing specific physiological functions, in vitro tissue models own capability to investigate the efficacy, toxicity and mode of action of therapeutic agents. The aim of this study is to explore the mode of action of nintedanib, the drug agent used for the treatment of idiopathic pulmonary fibrosis (IPF), on in vitro human lung microvasculature architecture.

We studied the efficacy and action of nintedanib on in vitro human lung microvasculature by investigating vasculogenesis within a 3D fibrin scaffold by co-culturing primary ECs and lung fibroblasts (Fb). To observe the effect of the drug, different concentrations of nintedanib in cell culture medium were investigated.

The microvasculature-on-chip model provides the possibility to study the effect of nintedanib on permeability, perfusablility and vascularized area of a human lung microvasculature model. The anti-angiogenic impact of nintedanib is significant for concentrations starting at 10nM, with an increase in vessel permeability coefficient and decrease in vessel density. As a replacement to animal models, advanced in vitro microvasculature-on-chip could open new prospects to study the mode of action of therapeutic compounds, such as nintedanib. Experimental platforms, such as this one, can address fundamental questions of drug effects on angiogenesis and vasculogenesis, and can be used to optimize drug treatments on personalized vasculature models.


High Performance Computational Fluid Dynamics for Improving the Hemodynamics of Prosthetic Heart Valves

Hadi Zolfaghari, Barna Becsek, Dominik Obrist

ARTORG Center for Biomedical Engineering Research, University of Bern, Switzerland

We use theoretical and high-performance computational fluid dynamics to investigate mechanisms of blood flow transition to turbulence through and past a prosthetic heart valve in the aortic position. Despite nearly four decades of theoretical and computational work on cardiac flows, an understanding of such mechanisms still does not exist. Such an understanding will enable us to maintain a favourable blood flow through a prosthesis, which is less separated or turbulent. Of particular interest, are the bileaflet mechanical heart valves (BMHV). BMHVs are reputed for high durability, however they do not offer an excellent hemodynamic performance, for which, patients need life-long anticoagulation therapies. Therefore, controlling the blood flow in those valve types can provide longevity and hemo-compatibility in the same picture, which will positively impact patients’ post-operative life. To that end, we perform massively parallel computations (~10K core-hours) on ultra-high grid resolutions (~10M grid points in plane) to model the transient nature of systolic blood flow. Results uncover an active flow separation close to leaflet’s leading edge. Kelvin-Helmholtz instability waves also appear in the form of von Karman or Burger vortex streets past the valve, depending on the Reynolds number. The non-linear interaction of those modes with boundary layer separation TS waves seems to trigger the transition to turbulence, and hence, one needs to push the separation point down to the trailing edge, which is the subject of our current investigation.


Factors Associated with Antimicrobial Resistant Gonorrhoea Infections in Men Who Have Sex with Men in Switzerland: Case-Control Study

Million Abraha1, Barbara Bertisch2, Christoph Hauser3, Michael Kluschke2, Dianne Egli-Gany1, Joost Smid1, Magnus Unemo4, Andrea Endimiani5, Valentina Donà5, Hansjakob Furrer3, Sara Kasraian5, Nicola Low1

1Institute of Social and Preventive Medicine, University of Bern, Switzerland; 2Checkpoint Zürich, Switzerland; 3Department of Infectious Diseases, Inselspital, Bern University Hospital, Switzerland; 4Department of Laboratory Medicine, Microbiology, Örebro University, Sweden; 5Institute for Infectious Diseases, University of Bern, Switzerland

Introduction Antimicrobial resistant (AMR) Neisseria gonorrhoeae (NG) is a global public health challenge. In Switzerland, gonorrhoea reports increased from 404 in 2000 to 2270 in 2016 with about 50% in men who have sex with men (MSM). To our knowledge, no studies have examined factors associated with antibiotic resistant NG in Switzerland.

Methods We enrolled MSM at clinics in Zürich and Bern from May 2015 to June 2016. MSM completed an online questionnaire and health care providers completed a clinical case report form. Specimens were tested with a nucleic acid amplification test (NAAT) and culture. For culture positive specimens, the minimum inhibitory concentration (MIC) was determined using European Union standard 2017 resistance breakpoints for ciprofloxacin, spectinomycin and cephalosporins and ≥2mg/L for azithromycin. Cases were MSM with NG and MIC above the breakpoint, controls were MSM with NG and MIC less than or equal to the breakpoint. We used logistic regression to estimate odds ratios (OR) with 95% confidence intervals (CI).

Results In total, 164/230 (71%) MSM were positive for NG by NAAT; culture was positive in 118/164 (72%). We compared 45 (38%) cases with ciprofloxacin resistance with 73 (62%) controls. Cases were more likely than controls to have received oral sex (OR 5.3, 0.6-44.2), to have had sex abroad in the last 3 months (OR 2.3, 1.04-4.9), to have concurrent sexual partners (OR 2.2, 95% Cl 0.8-6.1) and for the most recent partner to be casual (OR 2.6, 0.8-8.5). In a multivariable model, sex abroad in the last 3 months remained associated with AMR (adjusted OR, 2.34, 1.01-5.44). We found azithromycin resistance in one MSM but no cephalosporin or spectinomycin resistance.

Conclusions Surveillance of antimicrobial susceptibility is important for the early recognition of emerging AMR in NG in Switzerland. Research about pharyngeal AMR NG and transmission via oral sex is needed. Interventions for MSM should promote safer sex, especially when abroad.


Immunodeficiency at the Start of Combination Antiretroviral Therapy in Low-, Middle- and High-Income Countries

Nanina Anderegg1, Ole Kirk2, Matthias Egger1,3, IeDEA Collaboration4, Cohere Collaboration5

1Institute of Social and Preventive Medicine, University of Bern, Switzerland; 2University of Copenhagen, CHIP, Centre for Health & Infectious Diseases Research, Department of Infectious Diseases, Rigshospitalet, Copenhagen, Denmark; 3Centre for Infectious Disease Epidemiology and Research (CIDER), University of Cape Town, Cape Town, South Africa; 4International epidemiology Databases to Evaluate AIDS; 5Collaboration of Observational HIV Epidemiological Research in Europe

Background: Early initiation of combination antiretroviral therapy (cART), at higher CD4 cell counts, prevents disease progression and reduces sexual transmission of HIV. Changes in guidelines on when to start cART are expected to result in increased CD4 cell counts at cART initiation. We describe temporal trends in the median CD4 cell count at cART start in adult men and women.

Methods: We used data from the International epidemiology Databases to Evaluate AIDS (IeDEA) sub-Saharan Africa, Latin America, Asia-Pacific and North America regions and from the Collaboration of Observational HIV Epidemiological Research in Europe (COHERE). We included all HIV-positive adults initiating cART 2002-2015. We imputed missing CD4 cell counts at cART start through multiple imputation. We calculated median CD4 cell counts by calendar year, country and sex and used weighted additive mixed models to smooth temporal trends in median CD4 cell counts.

Results: We included 951,855 adults from 16 low-, 11 lower middle-, 9 upper middle-, and 19 high-income countries. Overall, the modeled median CD4 cell count (95%-confidence interval) at the start of cART increased 2002 to 2015: from 78 (58-104) to 287 (250-328) cells/µL in low-income countries (+268%), from 99 (71-140) to 234 (192-285) cells/µL in lower middle-income countries (+136%), from 71 (49-104) to 311 (255-379) cells/µL in upper middle-income countries (+338%), and from 161 (143-181) to 327 (286-372) cells/µL in high-income countries (+103%). In low and middle-income countries the increase was more pronounced in women; in high-income countries the opposite was observed.

Conclusions: Median CD4 cell count at cART start increased in all income groups, but generally remained below 350 cells/μL. Substantial additional efforts and resources are required to increase testing coverage with the aim of achieving earlier diagnosis, linkage to care, and initiation of cART.


Detection of Cardiac Dysfunction by Echocardiography in Swiss Childhood Cancer Survivors—A Pilot Study

Christina Schindera1,4,5, Mladen Pavlovic2, Sebastiano Lava2, Iso Hutter2, Kurt Leibundgut3, Nicolas von der Weid4, Thomas Suter2, Claudia Kuehni1,5

1Institute of Social and Preventive Medicine, University of Bern, Switzerland; 2Department of Cardiology, Inselspital, Bern University Hospital, Switzerland; 3Division of Pediatric Hematology/Oncology, Inselsptial, University Children`s Hospital Bern, Switzerland; 4Division of Pediatric Hematology/Oncology, University Children`s Hospital Basel, Switzerland; 5Department of Pediatrics, Inselspital, University Children`s Hospital Bern, Switzerland

Background: Cardiovascular disease (CVD) is the leading non-malignant cause of late deaths in childhood cancer survivors (CCS). Early detection and medical intervention might be important to improve outcome, as CVD and cardiac dysfunction remain asymptomatic for many years.

Aims: 1) To evaluate the prevalence of cardiac dysfunction in CCS using conventional and novel echocardiographic techniques; 2) To investigate the association between cancer treatment and cardiac dysfunction.

Methods: We invited CCS registered in the Swiss Childhood Cancer Registry, diagnosed with cancer between 1976-2012, aged 0-16 years at diagnosis, treated at the University Children`s Hospital Bern and survived ≥ 5 years after cancer diagnosis, to a cardiac follow-up visit to the University Children`s Hospital Bern. We performed a standardized medical history, physical examination and an echocardiography including conventional (left ventricular ejection fraction/LVEF) and novel (2-dimensional strain/2d strain) techniques. Data on cardiotoxic cancer treatment including cardiac radiation and chemotherapy with anthracyclines were collected from medical records. Recruitment is ongoing and we aim to include 711 CCS.

Results: We assessed 49 CCS with a median age of 35.2 years (interquartile range/IQR 30.8-39.6 years; 59% male) and a median time since diagnosis of 26.0 years (IQR 20.1-31.7 years). Median exposure to cardiac radiation was 24.6 Gray (IQR 18.0-35.2 Gray) and to anthracyclines 254 mg/m² (IQR 180-310 mg/m²). At examination, only 3 CCS (6%) had abnormal LVEF (<55%), whereas 17 CCS (34%) had abnormal 2d strain (> -18%). Median LVEF was 63.4% (IQR 59.8 – 67.9%) and median 2d strain was -20.7% (IQR -23.0 to -17.7%).

Conclusions: Prevalence of cardiac dysfunction in CCS seems to be higher when using novel echocardiographic techniques (2d strain) compared to conventional echocardiographic techniques (LVEF). However, data are preliminary as this study is ongoing.


Patterns of Chronic Hand Eczema: A Semantic Map Analysis of the CARPE Registry Data

Simone Cazzaniga1,2, Christian Apfelbacher3, Thomas Diepgen4, Robert Ofenloch4, Elke Weisshaar4, Sonja Molin5, Andrea Bauer6, Vera Mahler7, Peter Elsner8, Jochen Schmitt9, Barbara K. Ballmer-Weber10, Philippe Spring11, Luigi Naldi2,12, Luca Borradori1, Dagmar Simon1

1Department of Dermatology, Inselspital, Bern University Hospital, Switzerland; 2Centro Studi GISED, Bergamo, Italy; 3Institute of Epidemiology and Preventive Medicine, University of Regensburg, Germany; 4Department of Clinical Social Medicine, Occupational and Environmental Dermatology, University Hospital, Ruprecht Karls University, Heidelberg, Germany; 5Department of Dermatology and Allergy, Ludwig Maximilian University, Munich, Germany; 6Department of Dermatology, University Hospital Carl Gustav Carus, Technical University Dresden, Germany; 7Department of Dermatology, University Hospital of Erlangen, Friedrich Alexander University Erlangen-Nuremberg, Germany; 8Department of Dermatology and Dermatological Allergy, Friedrich Schiller University Jena, Germany; 9Centre for Evidence-based Healthcare, Medical Faculty Carl Gustav Carus, Technical University Dresden, Germany; 10Allergy Unit, Department of Dermatology, University Hospital Zürich, Switzerland; 11Department of Dermatology, University Hospital of Lausanne CHUV, Switzerland; 12Department of Dermatology, San Bortolo Hospital, Vicenza, Italy

Background: Hand eczema (HE) has a high incidence and prevalence, and is associated with disturbances of both physical and psychological well-being, and risk of persistence as chronic HE (CHE). Epidemiological studies on HE provided mainly descriptive and risk analyses, but pattern analyses of variables associated with HE, in particular CHE have not been explored so far.

Objectives: To investigate and display the semantics of associations among variables of HE obtained by the Swiss and German registries of CHE (CARPE) to dissect patterns and novel links.

Methods: This was a cross-sectional study on selected variables of the CARPE registries. Associations among variables were analyzed by means of an auto-associative system. A semantic connectivity map was generated by using the maximum spanning tree algorithm.

Results: Baseline data sets of 1466 CHE patients (Switzerland: 199; Germany: 1267) were analyzed. Occupational exposure had the highest impact on CHE in the total and country cohorts. We identified two areas of exposure linked to corresponding occupations that clearly demarcated genders.

Conclusions: This study using a semantic connectivity as novel method of data analysis reveals the complexity of features characterizing CHE as well as novel association patterns that deserve further investigations.


Primary Human Intervertebral Disc Cells Effect Bone Formation in Primary Human Osteoblasts

Rahel Deborah May1, Daniela Angelika Frauchiger1, Christoph E. Albers2, Lorin Michael Benneker2, Sandro Kohl2, Benjamin Gantenbein1

1Tissue and Organ Mechanobiology, Institute for Surgical Technology & Biomechanics, University of Bern, Switzerland; 2Department of Orthopaedic Surgery and Traumatology, Inselspital, Bern University Hospital, University of Bern, Switzerland

Introduction: For discogenic back pain the currently most popular treatment strategy is lumbar interbody fusion where the disc space is cleared and replaced with a cage (± bone substitute) to allow an intersomatic spondylodesis. However, clinical observation showed that partial intervertebral disc (IVD) tissue removal could lead to a failure of spinal ossification. It was previously shown that human IVD cells inhibit ossification of human mesenchymal stem cells (MSC). Nonetheless, whether this is a direct effect of IVD cells on osteoblasts (OB) is still unknown. We hypothesized that human primary OB co-cultured with IVD cells show similar inhibitory effects in ossification as previously demonstrated for MSC.

Methods: IVD cells (nucleus pulposus cells (NPC), annulus fibrosus cells (AFC) and cartilaginous endplate cells (CEPC)), isolated from patients, were encapsulated in 1.2% alginate beads and co-cultured via inserts with OB (N=7) in monolayer. Six, nine and twelve beads, respectively, were investigated. The experimental groups were stimulated with osteogenic medium. To quantify ossification, matrix mineralization of OB was visualized after 21 days by alizarin red (ALZR) staining and alkaline phosphatase (ALP) activity.

Results: After 21 days, the OB culture in osteogenic medium (positive control) (0.804 ± 0.189, mean ± SEM) and empty beads control (0.500 ± 0.053) showed a statistically significant higher mineralization compared to negative control (0.051 ± 0.010) (p-value < 0.03). OB cultured with nine beads showed a significant lower calcium deposition compared to the positive control (nine beads: NPC: 0.374 ± 0.085, AFC: 0.383 ± 0.070, CEPC: 0.402 ± 0.062, p-value < 0.03). The ALP activity showed a trend to be lower with increasing bead number in every cell type.

Discussion: We could show a tendency that mineralization of primary OB could be inhibited by direct exposure to IVD cells. IVD cells directly influence OB, by apparent inhibition of bone formation.


Regenerative Therapeutic Approaches for Ischemic Brain Damage After Cardiac Arrest: Development of an ex vivo Model to Evaluate Grafting of Neuronal Progenitor Cells

Patricia Meyer1,2, Denis Grandgirard1, Matthias Haenggi3, Stephen L. Leib1

1Institute for Infectious Diseases, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Intensive Care Medicine, Inselspital, Bern University Hospital, University of Bern, Switzerland

Sudden cardiac arrest (CA) is the most important cause of global brain ischemia, which affects the hippocampal cornu ammonis segment (CA1) in particular. Due to a lack of effective therapies, most patients are left with incomplete neurological recovery.

In this project we aim to develop an ex vivo model for this brain injury, using organotypic hippocampal cultures (OHCs) for the evaluation of grafting of neuronal progenitor cells (NPCs) as possible therapy.

Hippocampi from rats pups were cut into slices, cultivated for one week, and subjected to oxygen-glucose deprivation (OGD) to reproduce in vitro the hypoxic injury observed after CA. Neuronal damage was quantified by Fluoro-Jade (FJ) staining, specific for degenerating neurons. For grafting experiments, NPCs were isolated from hippocampi of rat embryos, expanded as neurospheres and grafted into injured cultures. Survival and differentiation of NPCs were assessed by immunofluorescent stainings (IF).

OHCs submitted to 33 minutes of OGD developed similar damage as observed in vivo. A significantly higher amount of FJ-positive cells was found after OGD in the hippocampal CA1 segment compared to the normoxic control. The neurosphere composition, tested by IF, showed the presence of numerous nestin-, doublecortin- and Ki67-positive cells, confirming the presence of NPCs. Differentiation of these cells towards neurons yielded a significant proportion of βIII-tubulin-positive cells. After NPCs transplantations into OGD-injured OHCs, viable cells were found at different time points after grafting, confirming the overall feasibility of the procedure.

We successfully reproduced damage to hippocampal CA1 neurons ex vivo after OGD. Neurospheres contained proliferating neuronal progenitors with the potential to differentiate into mature neurons. Their fate after grafting is currently assessed further. This project represents a first step towards a cell-based regenerative therapy for ischemic brain damaged consecutive to CA.


Improved Outcome by Combining Adjuvant Therapies in Experimental Pneumococcal Meningitis

Lukas Muri1,2,3, Denis Grandgirard1,3, Michael Perny1,3, Michelle Buri1,3, Stephen L. Leib1,3

1Institute for Infectious Diseases, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Cluster for Regenerative Neuroscience, Department for BioMedical Research, University of Bern, Switzerland

Background: Pneumococcal meningitis (PM) is associated with high mortality and morbidity rates. Up to 50% of survivors show neurologic sequelae, including severe hearing loss, cerebral palsy, cognitive impairments and learning disabilities.

Methods: Eleven day old Wistar rats were infected intracisternally with S. pneumoniae and randomized for treatment with either combination of 2 adjuvant therapies (2AT) consisting of daptomycin plus Trocade with ceftriaxone, single adjuvant therapies or standard ceftriaxone mono-therapy. Cortical damage, hippocampal apoptosis and inflammatory cytokine expression were assessed during acute PM. Three weeks after infection, Morris Water Maze was used to evaluate the effect of 2AT on learning and memory formation. Evoked auditory potentials were performed to assess hearing capacity.

Results: After acute PM, hippocampal apoptosis was significantly reduced in infant rats receiving adjuvant Trocade and 2AT. Adjuvant daptomycin and 2AT significantly reduced cortical necrosis. Six hours after treatment initiation, cerebrospinal fluid levels of TNF-α, IL-1β, IL-6 and IL-10 were significantly reduced in animals treated with 2AT. In long-term experiments 2AT significantly reduced neurologic sequelae by improving learning and memory formation compared to standard ceftriaxone mono-therapy. In addition, 2AT significantly improved average hearing capacity. Multivariate linear regression modelling revealed that 2AT during acute PM preserves 21.7 dB hearing capacity compared to animals treated with the standard ceftriaxone therapy.

Conclusion: Combination adjuvant therapy consisting of non-bacteriolytic daptomycin and Trocade—a potent matrix metallo-proteinase (MMP) inhibitor—is neuroprotective during acute bacterial meningitis by reducing inflammatory cytokines, and reduces multiple long-term neurofunctional deficits. These findings identify 2AT as a very promising therapeutic option to improve the outcome of paediatric pneumococcal meningitis.


Diagnostic Accuracy of a Portable Colposcope, the Gynocular, for Detecting Pre-Cancerous Lesions of the Cervix: Systematic Review and Meta-Analysis

Katayoun Taghavi, Eliane Rohner, Julia Bohlius

Institute of Social and Preventive Medicine, University of Bern, Switzerland

Background: Colposcopic examination is the gold standard for detecting pre-cancerous lesions of the cervix. Recently, studies examine the potential of a portable colposcopic device, called the Gynocular. We aimed to conduct a systematic review and meta-analysis of studies assessing its test accuracy in detecting histologically proven pre-cancerous lesions of the cervix.

Methods: We included randomized and non-randomized interventional studies comparing the Gynocular for cervical cancer screening with a histopathological reference standard. A systematic search in Embase, Medline and the Cochrane Library was performed and additional studies found through reference searches. We assessed trial quality using the Quadas-2 tool and performed descriptive analyses and a random-effects meta-analysis to derive the combined overall estimates of test accuracy. We explored causes of heterogeneity using meta-regression.

Results: We included four trials, involving 3280 women. We found a moderate risk of bias for test accuracy against histopathological reference. The pooled sensitivity of the portable colposcope to detect histology confirmed CIN2+ was 84% [95% CI 78 – 88%]. The pooled specificity was 57% [95% CI 13 – 92%], however, we found strong evidence for heterogeneity (I₂ 94%). Differences in the preceding screening tests used and colposcopist experience were considered possible causes of this. When introducing these covariates, the specificity decreased to 6% [95% CI 0-64%, I₂ 42%]. With only four studies included in the analysis this has to be considered as exploratory only.

Discussion: Mobile colposcopy increases the sensitivity of detecting precancerous lesions of the cervix. A large variation in specificities was observed between the studies. Additionally we found differences in study settings and methods that could not be meaningfully explored due to the small number of studies. We conclude that more research is needed before recommendations on mobile colposcopy can be made.


An Engineered IgE-Fc Variant Inhibits Basophil Degranulation ex vivo

Pascal Gasser1,2, Luke Pennington3, Daniel Brigger1,2, Noemi Zbären1,2, Theodore Jardetzky3, Alexander Eggel1,2

1Department for BioMedical Research, University of Bern, Switzerland; 2Department of Rheumatology, Immunology and Allergology, Inselspital, Bern University Hospital, Switzerland; 3Department of Structural Biology, Stanford University School of Medicine, Stanford, California 94305, USA

Background: Allergen-specific IgE plays a major role in the development of allergic reactions. It binds with high-affinity to the primary IgE receptor FcεRI on basophils and mast cells. Upon exposure to the cognate allergen IgE-loaded cells immediately degranulate and release soluble mediators causing allergic symptoms. The therapeutic anti-IgE antibody Omalizumab is known to neutralize free IgE and to prevent binding of IgE to basophils and mast cells. Recently, we have reported that Omalizumab actively desensitizes basophils at high concentrations. Furthermore, we have provided evidence that a mutated IgE-Fc variant, which is resistant to Omalizumab binding, may be used to actively replace the IgE-repertoire on the surface of primary human basophils when co-applied with Omalizumab. This combination treatment significantly increased inhibition of antigen-mediated basophil activation ex vivo. Here, we aim to further investigate the exact mechanism of basophil inhibition for the mutated IgE-Fc variant.

Methods: Human primary basophils were isolated from whole blood donations of grass-pollen allergic individuals and treated with wiltype IgE-Fc or mutated IgE-Fc variants alone or in combination with Omalizumab. Subsequently, cells were challenged with a grass-pollen allergen mix. Basophil activation was measured by flow cytometry.

Results: Interestingly, the mutated IgE-Fc variant alone diminished basophil activation in a competition-independent manner, while the wildtype IgE-Fc variant showed no effect. Furthermore, the IgE-Fc variant showed synergistic and dose-dependent inhibition already at low concentrations when used in combination with Omalizumab.

Conclusions: Our data indicate that the mutated IgE-Fc variant might engage an inhibitory receptor on the surface of basophils. However, further studies are required to confirm this hypothesis. The IgE-Fc variant could potentially be used as an efficient add-on treatment to the current Omalizumab therapy.


Death Regulation of Neutrophil Subsets by Intravenous Immunoglobulin (IVIg)

Stefanie Graeter1,2, Ulf Kessler3, Joerg C. Schefold4, Jane Burns5, Stephan von Gunten1

1Institute of Pharmacology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Surgery, HFR Fribourg – Cantonal Hospital, Fribourg, Switzerland; 4Department of Intensive Medicine, Inselspital, Bern University Hospital, Switzerland; 5Kawasaki Disease Research Center, UCSD School of Medicine, San Diego CA

Band neutrophils are bone marrow-resident precursors of mature, polysegmented neutrophils only found in small numbers in the circulation of healthy individuals. In severe inflammation, more immature neutrophil precursors as well as other myeloid precursors are circulating, resulting in a so-called left shift. Mature neutrophils circulating in the periphery usually undergo spontaneuous cell death to maintain homeostasis. If activated, neutrophils prolong their half-life, migrate into inflamed tissues and die there after performing their effector function. A disequilibrium in neutrophil cell death due to chronic inflammation or autoimmune diseases results in severe tissue damage. To date, not much is known about cell death regulation in immature neutrophils. In our project, we are investigating the role of immature neutrophils in a Kawasaki disease (KD) model. This self-limited systemic vasculitis of infants and children is the most common cause of acquired heart disease in Asia and Western countries. The inflammation responds to treatment with high-dose intravenous immunoglobulin (IVIg) and aspirin. However, approximately 15% of patients are resistant to IVIg therapy and have persistent fever following treatment associated with an elevated band neutrophil count in the peripheral blood. Our previous data demonstrated that IVIg triggers cell death in mature neutrophils. We postulate that band neutrophils are more resistant to IVIg-mediated cell death and persistence of these cells in the periphery contributes to systemic inflammation and IVIg-resistance. It was recently reported that CD10 expression on neutrophil surfaces is a marker to separate band and other premature neutrophils from polysegmented neutrophils. This marker allows us to to perform both descriptive and functional assays by discriminating the different neutrophil subsets.


Generation of Cross-Presenting DC for Immunotherapy of Melanoma

Thomas Gruber, Hassan Sadozai, Mirjam Schenk

Institute of Pathology, University of Bern, Switzerland

A healthy immune system has the ability to recognize and eradicate tumor cells via a process that is termed the “Cancer-Immunity Cycle”. Upon release of tumor-associated antigens (TAA) from cancer cells, they are captured by antigen presenting cells (APCs) and cross-presented to naive CD8+ T cells inducing their maturation into antigen-specific effector T cells. Primed CD8+ cytotoxic T cells (CTL) play a crucial role in antitumor immunity due to their ability to kill tumor cells and thereby release more TAA. Despite the recent advances in cancer immunotherapy, especially anti-CTLA-4 and anti-PD-1 / -PD-L1 antibodies, there are limitations to the treatment due to the mechanisms of tumor immune evasion. A promising target cell population for improving the efficacies of modern immunotherapies are the antigen presenting cells known as dendritic cells (DC). DC in cancer patients are reported to be insufficiently activated and thus cannot efficiently cross-present melanoma antigens to generate an effective CTL response. Our data shows that IL-32 induces maturation of mouse bone marrow derived DC, which correlated with an increased potential to induce antigen-specific T cell proliferation compared to immature DC in vitro. Administration of IL-32 into murine B16 tumors resulted in reduced growth of the directly treated as well as contralateral, untreated tumors thereby indicating the induction of systemic immunity. IL-32 generated a more prominent immune infiltration with DC and T cells compared to untreated tumors. Depletion of CD8+ T cells as well as absence of a distinct cross-presenting DC subset (BATF3+ DC) abrogated the anti-tumor effect by IL-32. These data suggest that IL-32 may induce tumor immunity through activation of DC and subsequent priming of tumor specific CD8+ CTLs. Interestingly, the combination with anti-PD-L1 further increased the efficacy of intra-tumoral IL-32, suggesting a beneficial effect of this combinatorial treatment for cancer patients.


Aspects of Cellular Tropism and Innate Immune Response During Zika Virus Infection

Nathalie Jane Vielle1,2,4, Beatrice Zumkehr1,4, Obdulio García-Nicolás1,4, Miloš Stojanov3, David Baud3, Artur Summerfield1,4, Marco Alves1,4

1Federal Department of Home Affairs, Institute of Virology and Immunology, Mittelhäusern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Materno-fetal and Obstetrics Research Unit, Lausanne University Hospital, Switzerland; 4Department of Infectious Diseases and Pathobiology, Vetsuisse Faculty, University of Bern, Switzerland

While Zika virus (ZIKV) circulated for decades without report of outbreaks and severe complications, its emergence was associated with description of severe neurological defects in newborns/neonates and adults. ZIKV is primarily vectored by mosquitoes of the Aedes genus but evidence of transmission through sexual contact and blood transfusion has been reported. So far, Flavivirus infections have not been defined as sexually transmissible. With the aim to identify susceptibility to ZIKV and mechanisms of innate immune response at mucosal surfaces, we investigated viral infection profiles in human monocyte-derived dendritic cells (MoDCs), in relevant cell lines and primary cultures of the respiratory and the genital tract. Infection of MoDCs by ZIKV was characterized by low percentages of infected cells and weak interferon responses, although the later was dependent on the virus strain used. Bronchial and nasal primary epithelial cells cultured at the air liquid interface (ALI) were susceptible to ZIKV infection and showed live virus shedding. Preliminary data using donor-derived vaginal epithelial cells showed no signs of infection in contrast to isolated vaginal mesenchymal stem cells, which showed to be susceptible to the virus. Our data highlight the need for further investigation of non-classical route of Flavivirus infection for ZIKV, the importance of variability among donors and show that great precaution shall be taken when extrapolating findings in cell lines to primary cells.


Effective HER2 Targeted Therapy Blocks Autophagy in Esophageal Adenocarcinoma Cells, While Treatment Resistance Is Associated with Autophagy Induction

{Ariane} Félice Janser1,2, Olivia Adams1,2, Vanessa Bütler1, Bastian Dislich1, Christian A. Seiler3, Dino Kroell3, Mario P. Tschan1,2, Rupert Langer1

1Institute of Pathology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Department of Visceral Surgery and Medicine, Inselspital, Bern University Hospital, Switzerland

Esophageal adenocarcinomas (EAC) are highly lethal (survival rate 30%) and exhibit in 20-30% a human epidermal growth factor receptor 2 (HER2) amplification. HER2 is a transmembrane receptor tyrosine kinase (RTK) promoting cell growth and proliferation. HER2 targeting drugs such as Herceptin (antibody), or RTK small molecule inhibitor Lapatinib are approved for the treatment of breast and gastric cancer. Therapy resistance is well documented, whereat underlying mechanisms remain unclear. Autophagy is a lysosome dependent catabolic process implicated in cancer. In the present study the effect of HER2 targeted therapy on autophagy in the context of treatment responsiveness was investigated in EAC.

Autophagy modulation upon HER2 targeted therapy was assessed in sensitive OE19 and resistant OE33 EAC cells. In OE19 cells a block in autophagy upon treatment was observed, while OE33 cells showed autophagy induction. These data suggest that effective HER2 targeted therapy correlates with a block of autophagy.

Furthermore, the role of autophagy in acquired resistance to HER2 targeted therapy was investigated. To this end a Lapatinib resistant OE19 subline was generated. These cells showed higher autophagic flux compared to the parental cells. From this observation we conclude that autophagy is involved in acquired resistance to HER2 targeted therapy in EAC.

Additionally, the expression of Her2 and levels of autophagy markers (LC3B; p62, previously published) were compared in a treatment naïve EAC patient cohort (104 cases of all stages) by immunohistochemistry. In this ex vivo part, no significant correlations were found.

Even though the ex vivo data indicates that basal HER2 overexpression does not affect autophagy levels, our in vitro data shows that responsiveness to HER2 directed therapy correlates with a block in autophagy. Therefore, we propose that autophagy induction may contribute to intrinsic and acquired resistance to HER2 targeted therapy in HER2 positive EAC.


The IL-33/ST2 Pathway Contributes to Intestinal Cancer by Promoting Intratumoral Regulatory T Cell Accumulation

Marie-Hélène Wasmer1, Eva Pastille2, Alexandra Adamczyk2, Tilman Rau1, Inti Zlobec1, Robert Geffers2, Astrid Westendorf2, Philippe Krebs1

1Institute of Pathology, University of Bern, Switzerland; 2Institute of Medical Microbiology, University Hospital Essen, University Duisburg-Essen, Germany

Colorectal cancer (CRC) develops through a multistep process and is modulated by inflammation. However, the inflammatory pathways that support intestinal tumors remain incompletely understood. We previously established a role for IL-33/IL-33 receptor (ST2) signaling in promoting CRC in humans and mice (Mertz et al. 2016).

Here we investigated the identity and the function of ST2-expressing immune cells in tumor lesions. We identified regulatory T cells (Tregs) as the main expressers of ST2 in the colons of CRC mice. ST2 expression on CRC Tregs determined their activation and migratory phenotype, and St2-deficiency was associated by a reduction in tumor-infiltrating Treg frequencies. Moreover, studies with bone marrow chimeric mice suggested a cell-intrinsic role of ST2 for the preferential accumulation of Tregs in CRC lesions. Lastly, we found that frequencies of ST2 Tregs in CRC lesions negatively correlated with the proportions of IFNγ- or GzmB- producing cytotoxic T cells and with the tumor score.

Taken together, our findings indicate that the IL-33/ST2 pathway promotes Treg accumulation in CRC lesions. This, in turn, triggers intestinal tumorigenesis by impairing the function of tumor-infiltrating cytotoxic T cells.


Digital Pathology Meets Mass Spectrometry: A Novel Approach to Biomarker Discovery in Formalin-Fixed Paraffin-Embedded Colorectal Cancer Tissue

Stefan Zahnd1, Sophie Braga-Lagache2, Manfred Heller2, Inti Zlobec1

1Institute of Pathology, University of Bern, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland

The TNM staging system for colorectal cancer (CRC) suboptimally predicts patient survival, particularly for patients classified into subgroups of TNM stage II and stage III. We aimed to identify novel prognostic protein biomarkers in these patients by selectively isolating tumor epithelial cells from formalin-fixed paraffin-embedded tissue. This was performed with a digital pathology-based approach followed by shotgun-proteomics based mass spectrometric analysis.

Epithelial cell populations were isolated by annotating regions of interest on a digital copy of CRC tissue from 25 patients with stage II and 25 patients with stage III cancer. Complete clinical pathological data was available for all patients. Annotated regions of interest were isolated using next generation Tissue Microarray (ngTMA®) technology, embedded in a new paraffin block, cut at 15 µm, and used for protein extraction and shotgun proteomic mass spectrometry analysis. Label-Free Quantification scores (LFQ) as obtained by the MaxQuant software were used as the basis for statistical evaluation.

Our findings reveal 1948 proteins reliably identified from FFPE tissue. Several identified proteins show significant dysregulation and thus serve as potential biomarkers. The most promising candidate among identified biomarkers is CTSB, also known as Cathepsin B.

Our data suggest that we successfully isolated and analyzed epithelial cell populations from FFPE tissue on the protein level using mass spectrometry, and identified several previously unreported biomarkers in a semiquantitative fashion. Our findings are further reinforced by the fact that Cathepsin B, which is a lysosomal cysteine protease with both endo-and exopeptidase activity, has repeatedly been associated with various types of cancer in the literature. Next steps in this project include validation of the exact role of Cathepsin B on a much larger cohort in order to elucidate the exact role and significance of this biomarker in colorectal cancer.


Membrane Dynamics During the Liver Stage of Plasmodium berghei

Livia Niklaus1,2, Volker Heussler1

1Institute of Cell Biology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Plasmodium parasites are the causative agent of malaria and are transmitted by Anopheles mosquitoes. Before the clinically relevant blood stage, the parasites infect hepatocytes where they reside in a special compartment called parasitophorous vacuole (PV). A host cell autophagy-related process plays an important role in the control of liver stage Plasmodium parasites. Previously, we have shown that autophagy markers, like LC3, localise to the PV membrane (PVM) of early liver stage parasites. However, long-term live cell imaging suggested that viable parasites can exclude LC3 and other membrane proteins from the PVM at later liver stages by membrane shedding.

Removal of autophagy marker via membrane shedding depends on a constant supply of phospholipids. We are currently investigating whether the host cell ER/Golgi or even the parasite ER contributes to this supply. Putative contact site proteins that could be involved in PVM and ER phospholipid exchange have been identified and now we investigate their localisation and their function in P. berghei-infected host cells.

The understanding of autophagy dynamics in infected host cells might help to generate measures against the Plasmodium liver stage and prevent the manifestation of the life-threatening blood stage.


FcεRI Cross-Linking Prolongs Survival of Human Basophils at Steady State and Under Intrinsic Apoptotic Stress

Lionel Rohner1,2, Ramona Reinhart2,3, Björn Hagmann1,2, Andrea Odermatt1, Annet Babirye1, Thomas Kaufmann3, Michaela Fux1

1Department of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences Bern, University of Bern, Switzerland; 3Institute of Pharmacology, University of Bern, Switzerland

Background: Basophils play a non-redundant role during the late- and chronic-phase of allergic inflammation by exerting various effector and immunoregulatory functions. Evidence from a growing number of sources, including animal models and clinical studies, support the concept that negative regulation of basophil survival might be an interesting approach to modulate the outcome of allergic inflammation.

Objective: This study is aimed at understanding the underlying mechanisms regulating intrinsic apoptosis in primary human basophils at steady state and under allergic conditions. We studied the impact of the anti-apoptotic Bcl-2 family members, Bcl-2, Bcl-xL and Mcl-1 on basophil survival using pharmacological inhibition with BH3-mimetics. Allergic conditions were mimicked by treating basophils with IL-3 and cross-linking anti-FcԑRIα antibody. Cell death and proteins associated with apoptosis were assessed by flow cytometry and western blot analysis.

Results: We show that spontaneous survival of basophils was largely dictated by constitutive expression of Bcl-2. In response to IL-3, basophils upregulated Bcl-xL and Mcl-1, resulting in efficient protection from BH3-mimetic-driven apoptosis. The protective effect of IL-3 was overcome by co-inhibition of Bcl-2, Bcl-xL and Mcl-1. In addition, we found that cross-linking of FcԑRI reduced the extent to which basophil undergo apoptosis over time and in response to BH3-mimetics. Using an IL-3-receptor antagonist, we show that the pro-survival effect was dependent on autocrine IL-3-signalling, which resulted in a marked increase of Mcl-1 and to a lesser extent Bcl-xL.

Conclusion: Our data indicate that FcԑRI-activated basophils have a survival advantage compared to resting basophils, enabling basophils to outlast the stressful process of degranulation, to resist potential pro-apoptotic stimuli and to potentially perpetrate and maintain allergic inflammations.


ACSL3 Is Required for the Maintenance of Pancreatic Ductal Adenocarcinoma

Matteo Rossi Sebastiano1, Mirco Galiè2, Deis Haxolli1, Maria Saliakoura1, Georgia Konstantinidou1

1Institute of Pharmacology, University of Bern, Switzerland; 2Department of Neuroscience, Biomedicine and Movement, University of Verona, Italy

Pancreatic cancer is the leading cause of cancer-related death worldwide. Tumor-associated mutations of KRAS occur in about 97% of pancreatic ductal adenocarcinomas (PDAC), the most common type of pancreatic cancer. The presence of these mutations is associated with aggressive, therapy resistant tumors both in humans and in mouse models. Indeed, over the past decades, the 5-year survival rates of pancreatic cancer have remained the same. The metabolism of PDAC, has not been extensively studied yet, therefore its better understanding and the identification of new therapeutic targets is urgently needed.

With the use of computational tools we compared the gene expression profiling of healthy pancreas with PDACs and we found significant alterations in lipid metabolic genes. Among the genes which expression is most significantly upregulated in PDAC is the Acyl-CoA transferase 3 and 4 (ACSL3, 4), a set of genes controlling lipid synthesis and catabolism.

Through selective RNAi-mediated ACSL3 knockdown, we found that ACSL function is required for PDAC cell proliferation and survival and that the majority of this effect is mediated by ACSL3. Our results thus identify ACSL3 as crucial liability in pancreatic cancer.

Still a lot of work is required in order to further understand the regulation of lipid methabolism in PDAC. We are currenly enstblishing mice colonies in order to study the effect in vivo of acsl3 knockout on a kras mutated, p53 deleted, PDAC model.

The evidence that ACSL3 is required for PDAC proliferation and survival sets the basis for this in vivo further validation as a new potential therapeutic target. This will hopefully play a role in the development of efficacious and resolutive pharmacological treatments against PDAC.


Echinococcus multilocularis: From Drug Screenings to the Energy Metabolism

Reto Rufener, Luca Dick, Laura D'Ascoli, Dominic Ritler, Andrew Hemphill, Britta Lundström-Stadelmann

Institute of Parasitology, Vetsuisse Faculty, University of Bern, Switzerland

Alveolar Echinococcosis (AE) is among the most lethal parasitic diseases in humans. Its etiological agent is the fox tapeworm Echinococcus multilocularis, which can be found in the Northern hemisphere all across the globe. The adult tapeworms live in the intestines of their final hosts (canids like foxes and dogs) and release eggs into the environment upon defecation of the canid host. Humans and other accidental hosts get infected by ingesting these eggs, which later give rise to the larva (metacestode). Metacestodes grow infiltrative in the liver, but can also grow and metastasize into other organs. Thus, AE has many pathological similarities with a malignant hepatic tumor.

Current chemotherapeutic treatment against AE consists of the benzimidazoles albendazole and mebendazole. However, they do not kill the parasite but only arrest its growth, and they can have serious side effects. Therefore, new treatment options against AE are urgently needed.

Our aim is to find and explore new compounds that are active against E. multilocularis. The Medicines for Malaria Venture (MMV) pathogen box is a library of 400 diverse molecules active against a wide range of pathogens. We screened the MMV pathogen box against in vitro cultivated metacestodes of E. multilocularis using different methods to assess the viability of the metacestodes as well as the drug-induced damage. Within this screen, four compounds were highly active, and the most promising compound (Buparvaquone) was further investigated to elucidate its mode of action within the energy generating pathways of Echinococcus. Furthermore, Buparvaquone was also tested in vivo in experimentally infected mice, but thus far Buparvaquone failed to be active.


Radiosensitization Strategy Based on DNA-PK Inhibition in Head and Neck Cancer: Emphasis on HPV Status

Selina Moara Roth, Daniel Aebersold, Yitzhak Zimmer, Michaela Medova

Department of Radiation Oncology, Inselspital, Bern University Hospital, and Department for BioMedical Research, University of Bern, Switzerland

Head and Neck cancer is the sixth most common non-skin cancer worldwide, leading to more than 350’000 deaths per year. The important physiological role of the affected anatomical regions can be corrupted, as well as complicate the complete removal of the tumor by surgery. Despite the use of improved organ-sparing treatments consisting of chemoradiation, loco-regional or distant relapses after completion of definitive therapy occur in up to 50% of all patients. Therefore, identification of new radiosensitization strategies to improve outcome of these patients are needed. Though head and neck cancers caused by classical risk factors such as tobacco and alcohol are declining, the incidence of human papillomarvirus (HPV)-infected tumors is increasing. Tumorigenic HPV strains target the tumor suppressors p53 and retinoblastoma protein (Rb), leading to their degradation and hence inducing aberrant proliferation and tumor formation. DNA-PK, a DNA-dependent nuclear serine/threonine kinase, is required for non-homologous end-joining (NHEJ) repair, where it re-ligates blunt DNA ends after induction of DNA double strand breaks (DSB). Functional DNA-PK repair pathway is essential for resolution of DSBs and its inhibition leads to cell cycle arrest and genomic instability. In this study, we aim to investigate the impact of DNA-PK inhibition in sensitizing HPV positive and HPV negative head and neck squamous cell carcinoma cell lines (HNSCC) to radiation therapy. Specifically, we would like to examine the impact of irradiation, DNA-PK inhibitor M3814 (EMD Serono) and the combination of these two treatments on a panel of HNSCC cell lines with respect to their HPV status. Our data suggest that inhibition of the NHEJ pathway alone does not affect proliferation and viability of cancer cells. However, when combined with ionizing irradiation induced DSBs, survival is reduced and apoptosis is increased in all cell lines, with enhanced sensitization effect on HPV positive cell lines.


Alterations in Gut Vascular Barrier in Experimental Portal Hypertension

Marcel Sorribas Olivera1, Ilaria Spadoni2, Maria Rescigno2, Reiner Wiest3

1Department for BioMedical Research, University of Bern, Switzerland; 2Department of Experimental Oncology, European Institute of Oncology, Milan, Italy; 3Department of Gastroenterology, University Clinic for Visceral Medicine, Inselspital, Bern University Hospital, Switzerland

Pathological bacterial translocation (PBT) in liver cirrhosis (LC) is the pathophysiological hallmark for spontaneous bacterial infections increasing mortality several-fold. Increased intestinal permeability in LC is a known contributor to PBT. We hypothesize that the recently described gut vascular barrier (GVB) is impaired in experimental portal hypertension (PHT) leading to an increased accessibility to the vascular compartment for translocating bacteria.

Different models of experimental PHT, namely Partial Portal Vein Ligation (PPVL), Bile Duct Ligation (BDL) or Carbon Tetrachloride (CCl4)-induced LC were used. A novel in vivo Confocal Laser Endomicroscopy (CLE) technique was established to visualize the intestinal vasculature. 70kDa or 150kDa FITC-dextran was injected intravenously and confocal probe was placed in the ileal lumen. Leakage from villi capillaries to lamina propria (LP) was measured. Immunofluorescence (IF) staining of the fenestral diaphragms (FD) marker plasmalemma vesicle-associated protein-1 (PV-1) was performed for GVB analysis.

In control animals CLE depicted an earlier permeation of 70kDa Dextran-FITC compared to 150kDa which was not extravasating after 30 minutes. In BDL, CCl4 and PPVL 70kDa FITC-dextran leaked earlier compared to sham. Interestingly 150kDa FITC-Dextran only leaked in the LC models (BDL and CCl4) but not in PHT mice (PPVL). Translocation from ileum to liver of 4kDa FITC-Dextran was only observed in the LC models but not in PPVL or Sham. Interestingly GVB staining showed increased signal for PV1 in intestinal vessels (CD34+) of BDL but not PPVL.

PHT per se has an impact on the GVB increasing FITC-70kDa-dextran leakage from intestinal capillaries to LP in CCl4, BDL and PPVL. However the IF showed only in BDL an increased PV-1 expression indicative of a wider opening of the FD than in PPVL. Therefore, different mechanisms appear to be involved in alterations of the gut-vascular barrier in pre-hepatic PHT and LC.


Bumped Kinase Inhibitor 1294 Induced Formation of Multi-Nucleated Complexes in Neospora caninum and Transformation of Protein Expression

Pablo Winzer1, Adriana Aguado-Martinez2, Joachim Müller1, Nicoleta Anghel1, Kayode K. Ojo2, Wesley C. Van Voorhis2, Andrew Hemphill1

1Institute of Parasitology, Vetsuisse Faculty, University of Bern, Switzerland; 2Center for Emerging and Reemerging Infectious Diseases (CERID), Division of Allergy and Infectious Diseases, Department of Medicine, University of Washington, Seattle, Washington, USA

The bumped kinase inhibitor BKI-1294, targets N. caninum calcium-dependent protein kinase 1 (NcCDPK1), interferes in tachyzoitesproliferation in vitro, and blocks its transplacental transmission and fetal infection in pregnant mice, as well as cerebral parasite load in dams. We have studied the localization of NcCDPK1 in N. caninum tachyzoites, by immunofluorescence and immunogold-electron microscopy. In intracellular tachyzoites, NcCDPK1 is localized within the cytoplasm. Localization shifts towards the apical part once parasites are maintained extracellularly, and NcCDPK1 is associated with the supellicular membrane, few micronemes, and the conoid. However, even at high concentrations, BKI-1294 does not exert parasiticidal effects in vitro.

We have shown that BKI-1294 treatment leads to the build-up of large schizont-like multinucleated complexes (MNCs). We now show that these MNCs remain viable for extended periods of time. While parasites are blocked during cytokinesis, newly formed tachyzoites emerge from these MNCs 6-8 days after the drug is removed. During treatment, MNCs exhibit a deregulated mRNA and antigen expression pattern, and we hypothesize that this could cause the exposure of new, potentially protective epitopes that might not be otherwise accessible during infection with N. caninum tachyzoites. We hypothesize the increased antigen expression during BKI-1294 therapy leads to protective immunity that leads to clearance of the parasites.


An Independent View at Vaccine Immunity in Feline Calicivirus Infection

Andrea Monika Spiri1,2, Barbara Riond1, Marilisa Novacco1,2, Marina Meli1,2, Felicitas Boretti3, Regina Hofmann-Lehmann1,2

1Clinical Laboratory, Vetsuisse Faculty, University of Zurich, Switzerland; 2Center for Clinical Studies, Vetsuisse Faculty, University of Zurich, Switzerland; 3Clinic for Small Animal Internal Medicine, Vetsuisse Faculty, University of Zurich, Switzerland

Feline calicivirus (FCV) is a common viral pathogen in domestic cats worldwide. The virus causes oral ulcerations, gingivitis/stomatitis and occasionally virulent systemic disease. Despite vaccines available for many years, FCV still poses a severe problem especially in multi-cat areas. This study compares the immune reaction towards the vaccine and towards a later FCV infection under experimental conditions. In a blind study, five specified pathogen-free (SPF) cats were twice vaccinated with a modified-live FCV vaccine and five SPF cats were twice placebo injected. All ten cats were housed together. Six months after the second vaccination all animals were experimentally infected with a FCV field strain. During the entire experiment clinical signs were recorded with a clinical score sheet and, viral shedding was tested by cell culture and direct real-time (RT-) qPCR and the humoral and cellular immune reaction was characterized by measuring antibody response, immune cell markers, cytokines expression and acute phase protein reaction. Environmental FCV contamination was determined in the facility. After vaccination no clinical signs were observed and a shedding of FCV vaccine virus from the oropharyngeal region was not detected. The five vaccinated cats developed an antibody response towards FCV, all unvaccinated cats stayed negative. After the experimental infection all cats were FCV positive by RT-qPCR and cell culture and showed typical signs of FCV infection with fever, depression, oral ulceration and faucitis at various degrees. All cats developed antibodies against FCV. The vaccinated cats developed a much stronger response than the unvaccinated cats. Most cats ceased viral shedding between day 29 and 50 post challenge. FCV nucleic acids were found on several items in the cat facility, however none of the environmental samples contained replication-competent virus as determined in cell culture.


Validation of Novel Fingolimod Derivatives for EAE Therapy

Bisera Stepanovska1, Holger Stark2, Britta Engelhardt3, Andrea Huwiler1

1Institute of Pharmacology, University of Bern, Switzerland; 2Institute of Pharmaceutical and Medicinal Chemistry, Heinrich-Heine-University Düsseldorf, Germany; 3Theodor Kocher Institute, University of Bern, Switzerland

Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disorder of the CNS associated with systemic immune dysregulation. A useful animal model of MS is the experimental autoimmune encephalomyelitis (EAE) model in mice. Since 2010 among the FDA approved treatments for RR-MS is the sphingosine 1-phosphate (S1P) analog fingolimod. Fingolimod is phosphorylated in vivo and as such it targets sphingosine 1-phosphate (S1P) receptors subtypes 1, 2, 3 and 5. It acts as a functional antagonist of S1P1 receptor by triggering internalization and degradation of the S1P1 in T cells which has a consequence on their trafficking. By introducing changes in the fingolimod structure we have developed two novel derivatives of fingolimod, denoted ST-1893 and ST-1894. S1PR activation profiles revealed that they have a potent and selective S1P1 receptor activation effect. Both exerted functional antagonism towards S1P1 receptor in a receptor internalization assay. In vivo, they induced a profound lymphopenia in mice after 24h of injection. In a preliminary mouse EAE experiment, prophylactic application of ST-1894 strongly reduced the motor deficits, lowered the incidence, and prevented from developing EAE with highest clinical score. This was accompanied with reduced immune cell infiltration in the CNS, reduced expression of the pro-inflammatory marker IL1β, and of ICAM-1 and VCAM-1, that contribute to immune cell transmigration. Moreover, we found a decrease of GFAP as a robust marker of astrogliosis, but no difference in the expression of BDNF. In summary, these data show that ST-1893 and ST-1894 are novel derivatives of fingolimod that exert selective and potent S1P1 receptor modulation. As a result, both compounds reduce the number of blood lymphocytes in mice. Moreover, ST-1894 protects from developing EAE when administered as prophylaxis. ST-1893 and ST-1894 hold promise for treatment of EAE and MS, however, more experiments are needed for further validation.


Investigation of Transient Receptor Potential Melastatin Channel 4 (TRPM4) in Colorectal Cancer Cell Line HCT116

Paulina Stoklosa1,2, Sven Kappel1, Christine Peinelt1

1Institute of Biochemistry and Molecular Medicine, National Center of Competence in Research NCCR TransCure, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Store operated calcium entry (SOCE) is the main source of calcium in non-excitable cells. Calcium is a universal second messenger that regulates multiple cellular processes. Imbalances in SOCE contribute to several cancer hallmarks, including increased proliferation, higher cell motility, and reduced ability to go through apoptosis. Transient receptor potential melastatin channel 4 (TRPM4) is widely expressed monovalent non-selective cation channel, that conducts mainly sodium and potassium ions and is activated by the increase of intracellular calcium ions concentration. Upon activation TRPM4 conducts sodium ions into the cell, thereby depolarizing the membrane, which results in the decrease of driving force for further calcium ions entry and serves as a negative feedback mechanism of intracellular calcium signaling.

TRPM4 is mainly present in the plasma membrane, however it is also found in intracellular vesicles, which can undergo exocytosis. Additionally, it was previously shown that elevations of intracellular calcium can activate fusion of TRPM4 positive vesicles with the plasma membrane.

TRPM4 was shown to be expressed in various tissues, with the most pronounced expression in prostate and colon. We here show that human colorectal cancer cell line HCT116 expresses TRPM4, and in addition, with the use of whole cell patch clamp technique we were able to measure TRPM4 currents, after stimulation with 10 µM Ca2+ in the patch pipette. Simultaneously, we were able to detect changes in cell capacitance, which correspond to changes in cell membrane size. Stimulation of HCT116 cells with calcium resulted in the increase of cell capacitance, as a result of vesicle fusion to the plasma membrane.


Developing Virus Like Particles as a Platform for Allergy Therapy

Federico Storni, Lisha Zha, Elisa Roesti, Paul Engeroff, Monique Vogel, Martin F. Bachmann

Division of Immunology, Department of Rheumatology, Immunology and Allergology (RIA), Inselspital, Bern University Hospital, University of Bern, Switzerland

Food allergies are common diseases in Western countries. Especially peanut allergy is a frequent disease affecting all age groups (1% of the population in US) but develops often already early in life. In most cases peanut allergy does not resolve with age and is a serious health-treat as small amount of peanut can induce strong allergic reaction. The high reactogenicity of peanuts (symptoms are caused in affected people with less than 1 mg ingested peanut) may be also due to an elevated amount of allergen protein in the fruit: one peanut contains about 200 mg of protein; indeed, the dominant protein species in peanuts are allergens, which is vastly different for most other sources of allergens as e.g. pollen. The major allergens Ara h 1 and Ara h 2 are recognized by IgE from greater than 95% of peanut-sensitive patients.

At present most therapies block mast cell effector molecules (antihistamines) or nonspecifically suppress immune response (steroids); for food allergies the widely-used treatment is the elimination of the allergenic food from the diet. Although oral immunotherapy has shown successes, the procedures remain time consuming and bear a risk for undesired allergic reactions.

A VLP based vaccination has been shown to treat cat allergy in a mouse model of Feld1-sensitized animals. In this project we intend to develop the VLP-technology as a new specific therapy against food allergies using peanuts as model allergens.


mRNA Splicing and Epithelial Integrity

Lester Thoo1, Lukas F. Mager1, Kathy D. McCoy2, Philippe Krebs1

1Institute of Pathology, University of Bern, Switzerland; 2Department of Physiology & Pharmacology, University of Calgary, Canada

Intestinal epithelial cells (IECs) have co-evolved with its neighbouring microbiota and the mucosal immune system to form a tightly regulated network. Dysregulation of this tripartite network leads to chronic inflammation and intestinal disorders such as inflammatory bowel disease (IBD). The role of alternative mRNA splicing (AS) for intestinal homeostasis and pathology is poorly understood. Epithelial splicing regulatory protein 1 (ESRP1) is an epithelial cell-specific critical regulator of mRNA splicing.

Here we used Esrp1 mutant mice (called Triaka), characterized by reduced ESRP1 function, to assess the role of ESRP1-specific AS in IECs’ epithelial barrier integrity, the composition of the adjacent microbiota and the function of the mucosal immune system. Contrarily to neonatal lethal Esrp1-/-mice, Triaka mice develop normally, permitting investigation of the physiological role of Esrp1 during adulthood.

We previously found that while Triaka mice have no overt intestinal pathology at steady-state, they show slightly reduced intestinal barrier integrity. This translates into impaired colonic wound-healing and increased susceptibility to dextran sodium sulfate (DSS)-induced colitis. In this study, we characterized the mucosal immune cell populations to address the mechanisms promoting intestinal immunopathology in Triaka mice. At steady-state, we detected a significant increase in CD8αβ+ TCRαβ+ intestinal epithelial lymphocytes (CD8 IELs) in the Triaka versus wild-type (WT) mucosa. This increase in CD8 IELs is dependent on commensals as germ-free Triaka and WT mice have similarly low frequencies. Furthermore, DSS treatment of WT mice augmented CD8 IELs to frequencies measured in homeostatic Triaka mice.

Taken together, our results indicate that alteration of mRNA splicing in IECs modifies the mucosal immune compartment. Future work will include elucidating how this mechanistically influences intestinal immunopathology and which bacteria species are involved.


Tn-Sequencing of Generated Mycoplasma hyopneumoniae and Mycoplasma hyorhinis Mutant Libraries Reveals Non-Essential Genes of Porcine Mycoplasmas Differing in Their Pathogenicity

Bettina Trueeb1, Simona Gerber1, Walid Gharib2, Peter Kuhnert1

1Institute of Veterinary Bacteriology, Vetsuisse Faculty, University of Bern, Switzerland; 2Interfaculty Bioinformatics Unit, University of Bern and Swiss Institute of Bioinformatics, Switzerland

Mycoplasma hyopneumoniae and Mycoplasma hyorhinis are two genetically closely related species inhabiting the respiratory tract of pigs. M. hyopneumoniae is the etiological agent of enzootic pneumonia, a disease with high economic impact. M. hyorhinis is considered a commensal of the respiratory tract of pigs potentially acting as a pathogen in polyserositis and arthritis. Moreover, it has been claimed that M. hyorhinis may play a role in human gastric cancer. In the framework of generating attenuated M. hyopneumoniae mutants to be used as improved vaccines, we refined a transposon (Tn) mutagenesis approach applicable to both species. Furthermore, a Tn-sequencing technique for mycoplasma has been established. Using this approach will help to identify essential genes as well as potential attenuated mutants. Moreover, comparing sets of essential and non-essential genes between the pathogen and the commensal may elucidate their differences in pathogenicity. Mutant libraries for both species have been generated, each library containing about 4000 transposon mutants. Based on Tn-sequencing data of these mutant libraries, the essentiality index of each gene was calculated using Tn-explorer software. Preliminary results show that the newly established Tn-sequencing method for mycoplasmas can be applied to sequence pools of mutants in parallel. The essentiality of genes of the pathogenic as well as the commensal porcine mycoplasma species of the respiratory tract can thereby be investigated under various growth conditions. This may lead to the identification of potential virulence attributes and pathogenicity mechanisms. Furthermore, this will eventually define new target genes for an attenuated live vaccine.


Tissue Resident T Cells in the Intestinal Mucosa in Homeostasis and Inflammation

Diego von Werdt, Nadia Corrazza, Christoph Müller

Institute of Pathology, University of Bern, Switzerland

Tissue resident memory T cells (TRM) are a recently described memory T cell subset, that are mainly localized in non-lymphoid tissues such as the skin, lung and the intestinal mucosa. Re-challenge of previously infected mice results in significantly improved protection provided by TRM at the site of pathogen entry. Currently however, little is known about how TRM differentiate and are maintained in situ.

We aim to study cell intrinsic and extrinsic factors required for TRM generation and maintenance. We also seek to further define the role of distinct resident T cell subsets in the intestinal mucosa in homeostasis and inflammation.

Recently we determined that the gene Rgs1 (Regulator of G-Protein Signaling 1) is expressed at significantly increased levels in resident intestinal T cell subsets versus those in peripheral circulation. Moreover, Rgs1 was shown by others to modulate lymphocyte chemotactic responses.

Here, we posit that increased Rgs1 expression is a prerequisite for the generation and long-term retention of non-recirculating TRM in the intestinal mucosa. By comparing frequencies of wild type and Rgs1-/- T cell subsets (circulating and tissue resident) in mixed bone marrow chimeric mice as well as following TRM generation post oral Listeria monocytogenes infection, we wish to define the functional relevance of increased Rgs1 levels for TRM formation and maintenance in the intestinal mucosa. Together, the planned studies will yield mechanistic insight into TRM biology in general and hopefully reveal the specific conditions for optimal TRM maintenance. This would expand the current understanding of the relevance of localized memory T cells in the intestinal mucosa.


Unravel the Cellular and Molecular Role of the Methyltransferase SUV39H2 in Health and Skin Disease

Pierre Balmer1,2, Eliane J. Müller2,3,4, Petra Roosje1,2

1Division of Clinical Dermatology, Department of Clinical Veterinary Science, 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; 4Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Switzerland

Hereditary nasal parakeratosis (HNPK) is a canine genodermatosis affecting the nasal planum, characterized by crusts, fissures and abnormal presence of nucleated cells in the stratum corneum. Our laboratory has reported that HNPK is associated with a missense variant in SUV39H2. SUV39H2 isa histone-3 lysine-9 (H3K9) methyltransferase involved in the epigenetic process, but with scant knowledge on its function in the skin. My aim is to investigate the cellular and molecular role of SUV39H2 in epidermal differentiation and in particular its causality.

Preliminary investigations on dog nose biopsies have revealed an aberrant differentiation. RNAseq and differential gene expression analyses, performed on three affected and control dog nose biopsies, revealed misregulated expression of gene products involved in keratinocyte late differentiation, such as loricrine, involucrine and cornuline, supporting that the missense variant of SUV39H2 leads to a keratinocyte differentiation defect.

Pathway analysis showed deregulated Wnt and Notch signaling in addition to cell adhesion anomalies, which are key pathways in onset of keratinocyte differentiation.

RNA and protein, from cultured keratinocytes, were collected during proliferation, early and late differentiation to investigate the differentiation phenotype and validate previous RNAseq and pathway analysis results. Preliminary observations on HNPK keratinocyte 2D cultures confirmed the aberrant onset of differentiation observed in affected dogs. Furthermore, we obtained evidence of premature cell cycle exit and senescence in HNPK keratinocytes.

In conclusion, we corroborated that keratinocyte differentiation is impaired in HNPK dog nose keratinocytes and identified the Wnt and Notch signaling pathways as major targets. Target genes will be now analyzed and epigenetic pattern defined. To determine the causative role of SUV39H2, a knock-down assay on control keratinocytes is currently ongoing to reproduce the HNPK phenotype.


Role of tRNA Processing Enzymes in Mitochondrial Biogenesis of T. brucei

Shikha Shikha, Andre Schneider

Department of Chemistry and Biochemistry, University of Bern, Switzerland

tRNAs are essential for translation, which in the cells is carried out by a sophisticated RNA-protein machinery called ribosome. In eukaryotes, ribosomes are present in compartments- the cytosol and mitochondria. Mitochondria have a bacterial origin and hence their translation machinery is also similar to that of the bacterial type. Therefore, most of the core RNA components of mitochondrial translation machinery are encoded by the mitochondrial genome including the tRNAs.

tRNAs are transcribed as precursors and then processed through series of enzymatic reactions to generate a mature tRNA which can then be used in translation. tRNA processing enzymes in eukaryotes are therefore expected to be present in both the cytosol as well as the mitochondrion where they process the mitochondrially encoded tRNAs.

T. brucei which is a single cell eukaryote, presents a unique situation where all the tRNAs have been thrown out of the mitochondrial genome and instead are imported from the nuclear encoded mature cytosolic pool of tRNAs. Therefore, tRNA processing enzymes are not expected to be required in its mitochondria. However, recently it was found that these enzymes are still present in the mitochondria of trypanosomes. Since, tRNAs are imported in their mature form, the question here is what are these enzymes doing in mitochondria? Are they vestigial in trypanosomes or do they have other essential functions in the mitochondria?

My project is concerned with two of these tRNA processing enzymes, namely RNase P and CCA Adding Enzyme. In trypanosomes, two nuclear genes encode for the RNase P enzymes and one of these is exclusively localised in mitochondria. Similarly, a mitochondrial CCA adding enzyme also exists and interestingly derives from the same gene which gives rise to the cytosolic CCA adding enzyme. Through biochemical & cellular analysis along with genetic manipulation methods, I aim to understand the role of these enzymes in mitochondrial biogenesis.


Optimization of CRISPR/Cas9 System for Highly Efficient Genomic Deletions

Nuria Bosch-Guiteras1,2, Rory Johnson1,2

1Department for BioMedical Research, University of Bern, Switzerland; 2Department of Medical Oncology, Inselspital, Bern University Hospital and University of Bern, Switzerland

Despite only the 2% of the human genome corresponds to protein-coding genes, a larger proportion is actively transcribed, and includes long non-coding RNA genes (lncRNAs). Although the number of described lncRNA is increasing, there is an important lack of information regarding their role in cellular functions and disease. Thus, the improvement of experimental methodology is fundamental for the functional characterization of lncRNAs.

Nevertheless, loss of function of lncRNAs has typically been a challenge. The development of Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) technology enables an efficient cut and repair of any targeted genomic region. However, given the lack of a targetable open reading frame, a small alteration in lncRNA sequence will probably not result in a loss of function. This is solved by techniques such as DECKO system (Double Excision CRISPR KnockOut) which enables the deletion of larger DNA fragments by the expression of dual guide RNAs. Although effective, there are several areas where DECKO may be improved, particularly in terms of knockout efficiency. Thus, the aim of this project is to study which factors could increase the efficiency of CRISPR deletions.

To approach this question, we are setting up a cellular model with a straightforward measurement for deletion. Concretely, targeting the non-coding sequence of a surface protein (i.e. the transcriptional start site) will allow FACS detection of protein levels, and therefore estimate deletion efficiency. The first modification of the current DECKO system is the design and implementation of an inducible Cas9 expression vector. Once established, the system will become a platform for further modifications regarding alterations of non-homology based DNA repair pathways and cell cycle regulation.

All in all, the optimization of CRISPR deletion systems is a powerful tool not only to understand the function of lncRNAs but also to target any genomic region of interest.


Understanding the Role of Ribonuclease Inhibitor (RNH1) in Erythropoiesis

Martina Stilinovic, Nicola Andina, Aubry Tardivel, Ramanjaneyulu Allam

Department of Hematology, Inselspital, Bern University Hospital, and Department for BioMedical Research, University of Bern, Switzerland

Background: Ribonuclease Inhibitor (RNH1) is a ubiquitously expressed leucine-rich repeat protein, which binds to and inhibits pancreatic-type ribonucleases (RNase). Recently, we found that RNH1 is a ribosomal associated protein and regulates mouse embryonic erythropoiesis by controlling GATA1 mRNA specific translation. Whether RNH1 regulates erythroid differentiation in human cells and the molecular mechanisms are largely unknown. In this study, we report that RNH1 also regulate erythroid differentiation in human erythroleukemia K562 cells.

Methods: To check RNH1 involvement in erythroid differentiation of human cells, RNH1-deficient K562 cells (which expresses globin genes) were generated by using CRISPR-Cas9 method and analyzed for erythroid differentiation. Western blot performed for GATA1 and RNH1. Polysome profiling was performed to check translation efficiencies. qRT-PCR performed to check mRNA expression.

Results: GATA1 protein levels were decreased in RNH1-KO K562 cells compared to wildtype cells. However, there is no difference in total GATA1 mRNA levels in WT and RNH1-KO cells. To check whether decreased GATA1 protein levels reflected impaired translation, we profiled polysomes from WT and RNH1-KO cells. Polysomes were decreased in RNH1-KO cells and, when normalized to 18S rRNA, polysome and monosome fractions contained lower levels of GATA1 mRNA than those of WT cells, while mRNA levels of another erythroid gene, HOXB4, remained comparable in WT and RNH1-KO polysomes. These results suggest that even though the overall translation rate is affected in RNH1-KO cells, GATA1 mRNA translation is further specifically decreased. Further, benzidine-positive cells were decreased in RNH1-KO cells compared to WT cells, suggesting that decrease erythroid differentiation in RNH1-KO cells.

Conclusions: We found that RNH1 deficiency decreased GATA1 translation and leads to defective erythroid differentiation. Our results suggest that RNH1 regulates erythropoiesis.


Whole Blood miRNAs as Prognostic, Non-Invasive Biomarkers for Equine Sarcoid Disease

Lucia Unger1, Vinzenz Gerber1, Alicja Pacholewska2, Tosso Leeb2, Vidhya Jagannathan2

1Swiss Institute of Equine Medicine, Department of Clinical Veterinary Medicine, Vetsuisse Faculty, University of Bern, and Agroscope, Switzerland; 2Department of Clinical Research and Veterinary Public Health, Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland

Objectives: Sarcoids are benign, but locally invasive fibroblastic tumours and represent the most common equine skin neoplasm worldwide Recent studies have shown the diagnostic value of miRNA fingerprints in whole blood of human patients with cancer. In this study we investigated the use of whole blood miRNAs as prognostic biomarkers for monitoring of disease dynamics in equine sarcoid (ES) affected horses.

Methods: 5 horses with spontaneous regression, 5 horses with progression of ES disease and 5 control horses were included in this longitudinal study. Whole blood samples taken at the initial presentation were used as input material for high throughput sequencing (HiSeq3000, Illumina). Clinical outcome was reevaluated after 5 – 7 years.

Results: 14 differentially expressed miRNAs which are accounted for control condition were identified with an adjusted p-value <0.05 and a fold change of ~ >2 and up to 6 in either direction in the regression vs. progression comparison. 4 miRNAs (28.6%) were upregulated and 10 miRNAs (71.4%) were downregulated in whole blood of horses with regression of ES disease.

Discussion: miRNA deregulation seems to be associated with phenotype aggressiveness of ES disease. The direct targets of downregulated genes are top sarcoma biomarker- disease associations according to GeneGo disease Ontology (MetaCore™ software suite). The direct targets of upregulated genes show enrichment in network processes like Cell cycle_G0-G1, DNA damage_Core and DNA damage_MMR repair.

Conclusion: Whole blood miRNA fingerprints may allow to predict prognosis in ES-affected horses.

Significance: Non-invasive prognostic biomarkers for ES disease might be promising for prepurchase exams of ES-affected horses and decision-making whether or not treatment should be initiated.


Designing a COX-2 Selective «Clickable» Fluorescent Probe

Serafina Calarco, Jürg Gertsch

Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland

The endocannabinoid system (ECS) is a complex endogenous lipid signaling network that modulates the communication between cells and exerts protective roles. It consists of G-protein coupled cannabinoid receptors (CB1 and CB2), the endogenous lipid ligands, better known as endocannabinoids (2-arachidonoylglycerol, N-arachidonoyl ethanolamine) and proteins related to their synthesis, transport and degradation.

The cyclooxygenase 2 (COX-2), is one of this enzymes involved in arachidonic acid and endocannabinoid metabolism.

For this project, we are designing a selective “clickable” fluorescent probe to investigate the dynamic regulation of COX-2 enzyme in native biological systems by microscopy, FACS or in gel protein profiling. This new chemical tool would allow a kinetic assessment of COX-2 protein expression in life cells, such as macrophages, and in more complex biological system.


Peptide Endocannabinoids and Their Emerging Role as CB Receptor Modulators

Sandra Glasmacher, Vanessa Petrucci, Andrea Chicca, Jürg Gertsch

Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland

Recent studies show that cannabinoid receptors (CB) can be modulated by endogenous peptides named peptide endocannabinoids (pepcans).1 The whole family of pepcans consists of N-terminal extended versions of pepcan-12.

Pepcan-12 is known to function as a negative allosteric modulator at the CB1 receptor.1 We show that pepcan-12 also acts as a positive allosteric modulator at the CB2 receptor.2 The double allosteric effects at CB1 and CB2 may represent powerful physiological modulation of these receptors.

ELISA and LS-MS/MS were performed in the CNS and other tissues, showing that pepcans were specifically localized in noradrenergic neurons in the locus coeruleus of the CNS as well as in chromaffin cells of the adrenal gland medulla.3 The analytical quantification experiments indicated high amounts of pepcan-12 in mouse adrenal glands.

Thus, the adrenal glands are assumed to be a site of production or storage of pepcan-12. To prove this hypothesis, quantification studies with LC‑MS/MS on adrenalectomized mice were performed underlining this hypothesis.

In future studies, the enzymes involved in pepcan-12 generation in adrenal glands should be identified. In doing so, clickable activity-based probes for ABPP will be synthesized and applied on adrenal gland homogenate and combined with analytical chemistry.

1. Bauer, M. et al. Identification and quantification of a new family of peptide endocannabinoids (Pepcans) showing negative allosteric modulation at CB1 receptors. J. Biol. Chem. 287, 36944–36967 (2012).

2. Petrucci, V., Chicca, A., Glasmacher, S., Paloczi, J. & Cao, Z. Pepcan-12 (RVD-hemopressin) is a CB2 receptor positive allosteric modulator constitutively secreted by adrenals and in liver upon tissue damage. Sci. Rep. 1–14 (2017).

3. Hofer, S. C. et al. Localization and production of peptide endocannabinoids in the rodent CNS and adrenal medulla. Neuropharmacology 98, 78–89 (2015).


Engineering a Chemical Switch into a Light-Driven Proton Pump

Stephan Hirschi, Daniel Harder, Zöhre Ucurum, Roland Goers, Wolfgang Meier, Daniel Müller, Dimitrios Fotiadis

Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland

The assembly of biomolecular nanofactories requires modules with controllable and cooperating functions. The fundamental building blocks of self-sustaining systems are energizing modules, which are able to establish an electrochemical gradient across a vesicular membrane to power secondary, energy-dependent modules. Light-driven proton pumps like bacteriorhodopsin (BR) and proteorhodopsin (PR) are excellent candidates for the conversion of light into chemical energy. The green-light absorbing proteorhodopsin was selected over BR due to the ease of genetic manipulation and overexpression in Escherichia coli. We extended its versatility by implementing an on-off switch based on reversible chemical modification of an introduced cysteine residue. The position of this residue in proteorhodopsin (N220C) was identified by assessing the proton-pumping activity in E. coli cells overexpressing proteorhodopsin and proteorhodopsin proteoliposomes, and two rounds of structure-based cysteine scanning mutagenesis. The N220C proteorhodopsin mutant could efficiently be deactivated by chemical modification with the membrane impermeable modifying agent sodium 2-sulfonatoethyl methanethiosulfonate (MTSES) and reactivated with the reducing agent b-mercaptoethanol (β-ME). Proton pumps with this extended functionality are of particular interest when the reconstitution generates mixed orientations of pumps in membranes, resulting in a functional short-circuit. Deactivation of the unfavorably oriented population creates vectoriality that can be imposed on secondary modules also present in mixed orientations. The result is a system that is able to generate a proton gradient across a membrane for energy-dependent, co-reconstituted modules (e.g., proton-driven transporters).


Function of the Human Heteromeric Amino Acid Transporter 4F2hc-LAT2

Satish Kantipudi, Patrick Bosshart, Jean-Marc Jeckelmann, Zöhre Ucurum, Dimitrios Fotiadis

Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland

Human heteromeric amino acid transporters (HATs) play key roles in renal and intestinal reabsorption, cell redox balance and tumor growth. These transporters are composed of a heavy (SLC3 family) and a light subunit (SLC7 family), which are connected by a disulfide bridge. Heavy subunits are type II membrane N-glycoproteins, while L-type amino acid transporters (LATs) are the light and catalytic subunits of HATs. These transporters show a broad specificity for aromatic and neutral amino acids that are exchanged in a 1:1 stoichiometry. Mutations in this class of transporters have been shown to be associated with primary inherited aminoacidurias. In addition, the HAT 4F2hc-LAT1 is overexpressed in certain types of tumor cells. The human HAT 4F2hc-LAT2 and the light subunits LAT2 alone were successfully overexpressed in the methylotrophic yeast Pichia pastoris. Uptake experiments with these stable cell lines showed L- leucine transport via 4F2hc-LAT2 and LAT2. In contrast to mammalian cells, Pichia does not require the presence of 4F2hc for the correct trafficking of LAT2 to the plasma membrane. This offers the possibility of identifying new functional roles of 4F2hc on light subunits.


RHS Proteins: A Novel Multiprotein Family Involved in Transcriptional Regulation of Trypanosoma brucei

Francesca Florini1,2, Arunasalam Naguleswaran1, Isabel Roditi1

1Institute of Cell Biology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Unusually for a eukaryote, most genes in Trypanosoma brucei are organized in long polycistronic units and the parasite’s RNA Polymerase I transcribes certain mRNAs as well as ribosomal RNAs. Moreover, it lacks many transcription factors that are conserved in other eukaryotes. It is clear that transcription is orchestrated in a fundamentally different way compared to other eukaryotes, so we asked if regulation at the level of transcription could be carried out by trypanosome-specific factors that mimic the roles of other proteins in other organisms.

Retrotransposon hotspot (RHS) proteins are unique to African and South American trypanosomes, and some of them were demonstrated to be associated with RNA polymerase II (1) and with TbRRM1, a nucleoprotein that modulates chromatin structure in T. brucei (2).

We could show by chromatin-immunoprecipitation (ChIP) that RNA Polymerase II and members of the RHS2, 4 and 6 bind to the same regions of the genome and that they are depleted in silent regions or in regions where transcription by RNA Polymerase I occurs. Through analyses of incorporation of the nucleoside analogue 5-ethynyl-uridine, we could observe a direct effect of RHS on the production of nascent RNAs. We are currently sequencing these nascent RNAs to have a closer insight into the role of RHS proteins, and understand at which level of the transcriptional process they are involved.

  1. Das et al, Biochemical characterization of Trypanosoma brucei RNA polymerase II, 2006, Molecular and Biochemical Parasitology, 150(2): 201-210
  2. Naguleswaran et al, Trypanosoma brucei RRM1 Is a Nuclear RNA-Binding Protein and Modulator of Chromatin Structure, 2015, mBio, 6(2):1-11


Proton Pumping bo3 Oxidase from Vitreoscilla

Simone Graf1, Christoph von Ballmoos1, Peter Brzezinski2

1Department of Chemistry and Biochemistry, Universtity of Bern, Switzerland; 2Department of Biochemistry and Biophysics, Universtity of Stockholm, Sweden

In aerobic bacteria and mitochondria, membrane bound heme-copper oxidases reduce oxygen to water, using the released energy for vectorial proton transport across the membrane. Both reactions contribute to the generation of the proton motive force by electrogenic proton and electron transfer reactions in the membrane. These reactions are tightly coupled and precisely orchestrated to ensure energy conservation and suppression of ROS production. Experiementally, these reactions are investigated using spectroscopic techniques to observe single turnover measurements.
However, the discrimination between chemical protons required for oxygen reduction and pumped protons is typically not possible. Therefore, a terminal oxidase pumping a different ion than protons would provide an elegant system to study mechanistic aspects of these intricate pumps. The group of Webster describeda sodium dependent quinol oxidase in the bacterium Vitreoscilla in th early nineties, but no further studies have been described with this enzyme.

Our aim was thus to isolate, purify and analyse the properties of the Vitreoscilla quinol oxidase to see, if it could be used a model system to elucidate the pumping mechanism using single turnover measurements with the flow-flash technique. We successfully cloned, heterologously expressed and purified the enzyme in E. coli. To our disappointment, we could neither observe sodium dependency of the enzymatic activity nor sodium pumping of the protein in inverted membrane vesicles, purified protein in detergent, or the enzyme reconstituted into proteoliposomes. Instead, we conducted several experiments that directly show proton pumping by the Vitreoscilla cytochrome bo, as compared with the related enzyme of E. coli.


Effect of tRNAPro Half on Translation in Chinese Hamster Ovary Cells

Yulia Gonskikh1, Johannes Grillari2, Norbert Polacek1

1Department of Chemistry and Biochemistry, University of Bern, Switzerland; 2Department of Biotechnology, University of Natural Resources and Life Sciences, Vienna, Austria

The Chinese Hamster Ovary (CHO) cell line is the most widely used mammalian cell system for industrial production of therapeutic proteins [1]. For this reason, the investigation of mechanisms controlling gene expression in CHO cells could allow improving the productivity of recombinant CHO cells. Recent data revealed tRNA-derived RNA as new class of ncRNAs that are able to regulate gene expression. Northern Blot analysis revealed expression of several 5’tRNA halves in CHO cells including the tRNAPro half. The tRNAPro half was detected in the polysomal fractions in polysome profiling, suggesting its association with ribosomes. Considering that 5’tRNA halves are known to be involved in translation regulation, the effect of tRNAPro half on translation was tested in vitro. Addition of tRNAPro inhibits global translation and causes upregulation of the specific translational product migrating as a 17 KDa protein on SDS PAGE. The effect of tRNAPro half was observed in in vitro translation systems of several other species including yeast, HEK and HeLa cells suggesting its conservation. The detected translational product was shown to be protease and RNase sensitive demonstrating that it consists of both RNA and amino acids. Based on the migration of the product on acidic gels, insensitivity to copper sulphate treatment, which selectively deacylates aminoacyl-tRNA but not peptidyl-tRNAs, and association with monosomes, we speculate that the accumulated product is peptidyl-tRNA. The identification of the RNA body of the upregulated translational product would confirm or falsify our theory and would give more insights into the mechanism of the effect of tRNAPro on translation.


Identification of Novel Enterotropic Astroviruses in Cattle

Michel Christoph Koch1,2, Céline Louise Boujon1,2, Simea Werder1, Torsten Seuberlich1

1NeuroCenter, Division of Neurological Sciences, Vetsuisse Faculty, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Astroviruses are single-stranded positive-sense RNA viruses with a genome size of 6.2-7kb. The genome includes at least three open reading frames (ORF): ORF1a encodes the non-structural protein nsp1a, and, together with ORF1b and via a translational ribosomal frameshift mechanism, the non-structural protein nsp1ab; ORF2 encodes the capsid protein precursor. In humans and mammalian animals, astroviruses are mainly enterotropic. Recently, our lab has identified a divergent bovine astrovirus (BoAstV-CH13) in brain tissues of cattle with encephalitis and neurological disease. The source of BoAstV-CH13 infection in these animals remains unknown, in particular whether this virus is shed by subclinical infected cattle via the feces.

We collected 148 bovine stool samples and analyzed them with a pan-astrovirus RT-PCR. Nine RT-PCR positive samples were subjected to Next-Generation Sequencing (NGS) for more detailed genotyping.

Amplicon sequences of all 11 RT-PCR positive samples revealed highest similarities to known enterotropic BoAstV and not to BoAstV-CH13. Scaffolds displaying similarities to Mamastrovirus sequences were found in all nine NGS sequenced samples, 14 of them longer than 5’500 nucleotides and with three identified ORF.

The best hits for nsp1ab amino acid sequences in all scaffolds were to known Astroviruses with an identity of >90%. A very similar situation was found for the capsid precursor protein sequence in five scaffolds. However, in nine scaffolds, the capsid precursor protein sequence identity to those of known astroviruses was only between 74,1% and 88.5%.

The neurotropic BoAstV-CH13 was not found in our study, suggesting that it does not occur at a high frequency in feces of cattle in Switzerland. However, we identified astrovirus sequences with high similarity to known enterotropic BoAstVs. Nine of the putative astrovirus genomes encode divergent capsid proteins, which may indicate recombination events with so far unknown astrovirus genotype species.


Immunological Alterations in the Mesenteric Lymph Node in Response to Naturally Acquired Feline Coronavirus Infection in Cats With and Without Feline Infectious Peritonitis

Alexandra Malbon1,2, Marina L. Meli2,3, Emi Barker4, Anja Kipar1

1Institute of Veterinary Pathology, Vetsuisse Faculty, University of Zurich; 2Center for Clinical Studies, Vetsuisse Faculty, University of Zurich; 3Clinical Laboratory, Vetsuisse Faculty, University of Zurich; 4School of Veterinary Sciences, University of Bristol, UK

Feline infectious peritonitis (FIP) is a fatal disease of felids resulting from poorly defined mutations of a largely avirulent enteric feline coronavirus (FCoV) within infected individuals. In FIP, an inappropriate immune response mediates a widespread granulomatous inflammatory disease. The present study screens the innate immune system for functional changes, focussing on the mesenteric lymph node (MLN) as the presumed first site of viral spread from the intestine.

MLN samples were collected from cats euthanased without FIP (G1, n=33) and with (G2, n=28). RT-qPCR was performed for FCoV and feline TLR1-9, STAT1-3, IFNα, β, γ, IL-1β, -6, -10, -15, -17, MMP2, 9, 13, TIMP1, 3, TNFα, CXCL10, CCL8, G- and M-CSF with reference gene GAPDH. Relative cytokine expression was calculated using the comparative CT method, calibrated to the G1 mean. A 2-tailed Mann-Whitney test was used to compare results between groups whilst correlation between FCoV levels in G2 and panel molecules was analysed using a Spearman’s rank test.

In FIP (G2), TLR2, 4, 5 and 8, STAT1, 2, all IFNs, IL-1β and 6, TNFα, CXCL10, CCL8, TIMP1&3 and G-CSF mRNA levels were significantly (p≤0.05) higher, and MMP13 lower than in G1. Of these, the majority also showed correlation between viral load and mRNA level.

Results in the FIP cats suggest an excessive immune response with selective activation of TLR signalling pathways, including TLR8 which detects ssRNA, TLR2 which has been shown to detect the SARS S protein, and downstream inflammatory mediators e.g. pyrogenic cytokines. Of particular interest were cats without FIP but with evidence of systemic FCoV infection; six MLNs from G1 were FCoV positive. Compared to FCoV negative MLNs these showed significantly elevated TLR9, STAT2 and IFNγ levels, whilst compared to cats with FIP TLR9 was also higher and viral load significantly lower. This suggests more selective immune stimulation by FCoV than in cats with FIP and may possibly represent a protective response.


Study of the in vitro and in vivo Metabolism of the Tryptamine 5-MeO-Mipt Using Human Liver Microsomes and Real Case Samples

Katharina Elisabeth Grafinger1,2, Marianne Hädener1, Stefan König1, Wolfgang Weinmann1

1Institute of Forensic Medicine, Forensic Toxicology and Chemistry, University of Bern, Switzerland; 2Graduate School of Cellular and Biomedical Sciences, University of Bern, Switzerland

Metabolism studies are of special importance in forensic science. When it comes to New Psychoactive Substances (NPS), often no reference standards are available, hence identifying unique fingerprints of drug metabolites is crucial. This knowledge helps developing new analytical methods and further proof consumption of a drug of abuse.

This study presents a case of intoxication with the synthetic tryptamine 5‐methoxy‐N‐methyl‐N‐isopropyltryptamine (5‐MeO‐MiPT) and the structural elucidation of metabolites in pooled human liver microsomes (pHLM), blood, and urine. In August 2016, the police encountered a naked man, agitated and with aggressive behaviour on the street. His premises were searched and blood and urine samples were taken at the hospital. Using liquid chromatography–high resolution tandem mass spectrometry (LC–HRMS/MS), the obtained blood and urine samples were analysed for in vivo metabolites of 5‐MeO‐MiPT. The confiscated pills and powder samples were qualitatively analysed using Fourier transform infrared (FTIR), gas chromatography–mass spectrometry (GC–MS), LC‐HRMS/MS, and nuclear magnetic resonance (NMR). 5‐MeO‐MiPT was identified in 2 of the seized powder samples. General unknown screening detected cocaine, cocaethylene, methylphenidate, ritalinic acid, and 5‐MeO‐MiPT in urine. Seven different in vitro phase I metabolites of 5‐MeO‐MiPT were identified. In the forensic case samples, 4 phase I metabolites could be identified in blood and 7 in urine. The 5 most abundant metabolites were formed by demethylation and hydroxylation of the parent compound. 5‐MeO‐MiPT concentrations in the blood and urine sample were found to be 160 ng/mL and 3380 ng/mL, respectively. Based on the results of this study we recommend metabolites 5‐MeO‐NiPT, 5‐OH‐MiPT, 5‐MeO‐MiPT‐N‐oxide, and OH‐5‐MeO‐MiPT as biomarkers for the development of new methods for the detection of 5‐MeO‐MiPT consumption, as they have been present in both blood and urine samples.


A Comparative Approach to Unravel Pathogenicity Factors of the Brain-Eating Amoeba N. fowleri

Nicole Liechti1,2,3, Rémy Bruggmann1,2, Matthias Wittwer3

1Interfaculty Bioinformatics Unit, University of Bern, Switzerland; 2Swiss Institute of Bioinformatics, Bern, Switzerland; 3Spiez Laboratory, Federal Office for Civil Protection, Spiez, Switzerland

Naegleria fowleri, commonly known as the brain-eating amoeba, is a free-living eukaryote found in soil and fresh warm water sources all over the world. Once entered the nose, N. fowleri follows the olfactory nerves to the brain and causes primary amoebic meningoencephalitis (PAM), a fast progressing and mostly fatal disease of the central nervous system. The mechanisms involved in the pathogenesis are still poorly understood. To investigate pathogenicity factors, a comparative approach of N. fowleri and its closest non-pathogenic relative N. lovaniensis is used.

To establish high quality genome references, long read sequencing technology was applied on one side to improve the fragmented genome assembly of N. fowleri and on the other side to preform de novo assembly of the N. lovaniensis genome. Keeping the focus on pathogenicity, predicted proteins specific for N. fowleri were defined by clustering of orthologous gene families between different Naegleria species and their function was characterized by functional annotation and GO enrichment analysis. Further, the transcriptome of N. lovaniensis was compared to the transcriptome of N. fowleri grown under different conditions influencing the pathogenicity. Proteins which were ether specific for N. fowleri or upregulated under growth conditions elevating the pathogenicity are associated with components of the membrane and vesicular transport.

Based on the genomic data and preliminary RNAseq data of different media types influencing the pathogenicity, membrane proteins seem to play an important role in the pathogenesis of N. fowleri. Further investigations will therefore focus on the identification of membrane proteins as well as secreted proteins of the different Naegleria species using Mass Spectrometry.


Floral Morphology as a Pollination Syndrome Trait in Petunia Species

Tural Yarahmadov, Sarah Jane Robinson, Hagen Peter Reinhardt, Mathieu Julien Hanemian, Cris Kuhlemeier

Institute of Plant Sciences, University of Bern, Switzerland

Solanaceae family member Petunia is an important horticultural crop valued mostly for its floral trait variety. P. exserta is a rare hummingbird - pollinated species, one of trademark features of which is the exserting from the floral tube reproductive organs. A closely related to P. exserta species P. parodii, is remarkable for a particularly long flower tube with a matching in terms of length style. A well - studied P. axillaris, which compared to aforementioned species has an average in terms of morphology flower while being phylogenetically close to both P. exserta and P. parodii, is a good "control" plant to which the comparisons can be made.

Phenotypical data regarding the styles and tubes of these 3 Petunia species and their F1 hybrids, the high-throughoutput RNA sequencing data from the different stages of floral development and the biomechanical data from the stress tests performed on the tissue allow for a large-scale multiple approach study aimed for determining the genes and/or their functional groups responsible for the significant differences in floral morphology of the studied species. By application of the differential expression, allele-specific expression, SNP effect evaluation, phenotype correlation-based and the quantitative trait locus analyses on the produced data, groups of genes of interest were isolated and are being prepared for the evaluation of their impact on the flower phenotype by performing the gene knockout/silencing experiments.


The AAT7 Gene Family Facilitates Uptake of Small Neutral Amino Acids in Trypanosoma brucei

Alexander Christoph Haindrich, Anaëlle Dubois, Corina Wirdnam, Doris Rentsch

Institute of Plant Sciences, University of Bern, Switzerland

Trypanosoma brucei, a protozoan parasite that causes sleeping sickness, is depend on the import of a variety of amino acids that are not only used for protein biosynthesis but also for energy production, polyamine synthesis and osmoregulation.

In the procyclic form (PCF) of T. brucei, which is present during the insect stage of the parasite, energy is obtained from the oxidation of proline, while the bloodstream form lives primarily on glucose. In both cases pyruvate is one of the final products of energy production, and is either directly excreted or first converted to alanine and then exported from the cell.

The transporters of the amino acid transporter AAT7 family form two phylogenetic separate clusters. When expressed in Saccharomyces cerevisiae, all transporters of the AAT7 family are able to import small neutral amino acids, including alanine and proline, suggesting an important role in PCF T. brucei. We further showed that at least one of the studied transporters may also export its substrates, a feature required for the export of alanine in both PCF and BSF T. brucei. Additionally, members of one of the AAT7 gene clusters show high affinity for threonine and their RNAi depletion in PCF T. brucei was lethal under standard culture conditions.


H2AX Depletion Promotes PARPi Resistance in BRCA2-Deficient Mouse Mammary Tumors

Martin Liptay1, Ewa Gogola2, Alexandra Duarte2, Jos Jonkers2, Sven Rottenberg1,2

1Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Switzerland; 2Division of Molecular Pathology and Genomics Centre, The Netherlands Cancer Institute, Amsterdam, The Netherlands

Poly(ADP-ribose) polymerase (PARP) inhibitors are highly effective in patients suffering from BRCA1/2-deficient breast or ovarian cancer, which lack a functional homologous recombination repair pathway. Despite this success, resistance to PARP inhibitors frequently emerges in tumors that initially respond to treatment. Using genetically engineered mouse models (GEMMs) highly resembling BRCA-mutated breast cancer in humans, we have previously shown that restoration of homology-directed DNA repair explains PARPi resistance in BRCA1-deficient cancers. Intriguingly, BRCA2-deficient tumors that acquired PARPi resistance lack restoration of homologous recombination. In part, resistance can be explained by the depletion of poly(ADP-ribose) glycohydrolase (PARG) (Gogola et al., submitted for publication). Moreover, we found that replication fork stability confers drug resistance (Ray Chaudhuri et al., 2016). To identify the underlying mechanisms that may explain fork protection, we carried out genome-wide CRISPR-Cas9 screens in cell lines derived from BRCA2-deficient mouse mammary tumors. As a result, we found a significant enrichment of gRNAs targeting H2afx after drug selection and we could validate that loss of H2AX results in PARPi resistance. This finding is corroborated by H2AX depletion in a subset of PARPi resistant mammary tumors. We are currently investigating whether the loss of H2AX affects replication forks under treatment-induced replication stress conditions. We hope that by studying the role of H2AX in the context of PARPi treatment, we will gain new insights into basic mechanisms of the DNA damage response and its link to replication fork biology.


The Role of Eukaryotic Specific rRNA Expansion Segments in Protein Synthesis

Vaishnavi Shankar, Miriam Koch, Norbert Polacek

Department of Chemistry and Biochemistry, University of Bern, Switzerland

The ribosome translates the information encoded by mRNA into sequence of amino acids, thereby synthesizing all proteins required by the cell. The main functional core of the ribosome comprising the subunit interface sides, the peptidyl transferase centre (PTC) and the decoding site, are conserved across all three domains of life. However, compared to the prokaryotic ribosome (70S), the eukaryotic one (80S) is about 40% larger and contains additional ribosomal proteins as well as additional ribosomal RNA (rRNA) elements, the expansion segments (ES) (1,2). Eukaryotic ES are mainly located at the periphery of the ribosome though the precise functions remain uncertain. However there is accumulating evidence that ES play a role in translation possibly by recruiting regulatory factors to the ribosome (3). To unravel the function of ES in translation specifically, ES7S and ES27L, these two ES were deleted from the ribosomes in the model organism S. cerevisiae, and were studied for growth phenotypes and functionality in translation and subsets thereof. Preliminary results show a serious growth phenotype upon ES7S deletion, lethality for ES27L deletion and no growth phenotype for strains carrying ribosomes with a section of ES27L deleted (ES27Lb deletion). Although the translation rate is greatly slowed down in ES7S mutants, there appears to be no biogenesis defects.

(1) Armache, J.P., Jarasch, A., Anger, A.M., Villa, E., Becker, T., Bhushan, S., Jossinet, F., Habeck, M., Dindar, G., Franckenberg, S., et al. (2010a). Cryo-EM structure and rRNA model of a translating eukaryotic 80S ribosome at 5.5-A resolution. PNAS 107, 19748-19753.

(2) Yokoyama, T., and Suzuki, T. (2008). Ribosomal RNAs are tolerant toward genetic insertions: evolutionary origin of the expansion segments. Nucleic Acids Research 36, 3539-3551.

(3) Becker, T., et al. (2009). Structure of the monomeric yeast and mammalian Sec61 complexes interacting with the translating ribosome. Science 326, 1369-1373.


Noncanonical Functions of Phenylalanyl tRNA Synthetase

Tin Manh Ho1, Dominique Brunssen1, Jiongming Lu2, Beat Suter1

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 tRNA. Additionally, recent publications revealed non-canonical functions for an increasing fraction of these proteins. Several diseases and tumorigeneses are shown to be associated with malfunctioning aaRSs. Phenylalanyl tRNA synthetase (PheRS) is overexpressed in multiple cancers and we found that overexpression of PheRS increased growth and cell proliferation in different fly tissues and that downregulation of PheRS by RNAi had the opposite effect. To test if the function of PheRS in regulating cell growth and proliferation is carried out by a noncanonical function of PheRS, we made a mutant α-PheRS gene, which codes for a protein that is not able to carry out aminoacylation. We found that the mutated PheRS protein also increased the cell proliferation in the follicle cell twin-spot experiments, which indicates the effect is caused by the noncanonical function of PheRS.

We are also following up on an independent observation and are testing whether this connects to a growth regulating function. The α-PheRS and β-PheRS subunits need each other to be stabile. However, we found that some cell types can accumulate higher levels of β-PheRS even if only this subunit is overexpressed in larvae. Besides, upon overexpression in larvae, α-PheRS, the catalytic subunit that performs the main enzymatic activity in aminoacylation is cleaved into stable isoforms that, based on their size, are predicted to lack most of the catalytic domains. We are presently investigating whether we can connect these observations with a non-canonical function of PheRS.


CRISPR/Cas9-Based Knockout Screening for Long Non-Coding RNAs in Lung Cancer

Taisia Polidori1,2, Rory Johnson1,2

1Department for BioMedical Research, University of Bern, Switzerland; 2Department of Medical Oncology, Inselspital, Bern University Hospital and University of Bern, Switzerland

The human genome contains many thousands of long non-coding RNAs (lncRNAs), the majority of which are still functionally uncharacterised and may play roles in human disease. A small number of lncRNAs have been implicated in disease. For example, oncogenes HOTAIR or MALAT1 that promote the cellular proliferation and are upregulated in tumors. A compelling question is how many of the remaining 99% of lncRNAs may play similar disease roles.

Our hypothesis is first, that lncRNAs mediate and contribute to a cancer phenotype, and second, that such lncRNAs may be discovered using CRISPR-based pooled functional screening using cancer-related phenotypes. Thus, the aim of the project is to identify and characterise lncRNAs involved firstly in basic cancer phenotypes (proliferation) and progression, that could have a relevant significance and impact upon diseases, and may be targets for therapy.

In order to knock out lncRNAs, we will use the validated approach of promoter deletion. We will take advantage of two tools that we have developed for knockout of noncoding elements: a vector system called DECKO (Double Excision CRISPR Knockout) and CRISPETa, a bioinformatic pipeline for paired sgRNA design.

Using a high-throughput method that relies on CRISPR/Cas9, we propose to do a Loss of Function (LOF) screening on lung cancer cell lines in order to find candidates that, when lost, lead to reduced proliferation.

In conclusion, the increasing understanding of the mechanisms by which lncRNAs function will allow the identification of the best targets for diagnostics and therapies.


Self-Assembly Properties of α-TTP Are Associated with Transport of Vitamin E Across the BBB

Stephan Martin Kammer, Walter Aeschimann, Achim Stocker

Department of Chemistry and Biochemistry, University of Bern, Switzerland

Vitamin E represents a micronutrient of the human diet. It acts as general antioxidant of cellular membrane bilayers by protecting polyunsaturated lipid side-chains from autoxidation. Among the eight naturally occurring congeners of vitamin E exclusively RRR-α-tocopherol is retained in significant amounts in mammals. The retention of RRR-α-tocopherol occurs at the late endosomal compartment of liver parenchymal cells and is mediated by the α-tocopherol transfer protein (α-TTP). Subsequent to α-tocopherol binding it also facilitates re-secretion of the later into the cardiovascular system. It has been proposed that α-TTP associates with the inner leaflet of the plasma membrane where the ATP-binding cassette transporter (ABCA1) facilitates RRR-α-tocopherol transfer into the blood stream. Altogether, these data indicate that α-TTP is essential for Vitamin E homeostasis in man. However, the transport pathways of RRR-α-tocopherol into extrahepatic tissues, especially the transport through the blood brain barrier into the central nervous system (CNS) have not been studied in detail yet.

Recent findings in our group have revealed oligomerization of α-TTP into 24-meric spherical nanoparticles upon binding of α-tocopherol. In vitro experiments have evidenced significant flux across primary endothelial cell layers (HUVEC) of such particles. Current studies aim to elucidate the mechanisms of α-TTP oligomerization. For this amino acid residues located within the interfaces of the particle are mutated one by one through site directed mutagenesis. We aim to isolate putative intermediates, to characterize the mutant particles by X-ray crystallography as well as by stability measurements and to gain insight into the assembly process.


Exploring Social Motility of Trypanosoma brucei in vivo

Sebastian Millius1,2, Isabel Roditi1

1Institute of Cell Biology, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Trypanosoma brucei causes sleeping sickness in humans and nagana in cattle in sub-Saharan Africa. It shuttles between the mammalian host and its vector, the tsetse fly. In the fly, the parasites need to cross the peritrophic matrix to reach the ectoperitrophic space. From there, they migrate into the proventriculus (PV), an organ at the junction between the mid- and foregut, finally reaching the salivary glands. To enter the PV, trypanosomes have to cross the peritrophic matrix a second time. We hypothesise that parasites cooperate in order to cross barriers such as the peritrophic matrix.

When procyclic (midgut form) trypanosomes are inoculated onto agarose plates, they form multicellular communities that exhibit coordinated group movement in the form of radial projections. This is termed social motility (SoMo). Cyclic AMP-specific phosphodiesterase B1 (PDEB1) is one of 4 genes known to influence SoMo. It was shown previously that knockdown of PDEB1 by RNAi caused a profound defect in SoMo: the parasites stayed at the inoculation site with no formation of radial projections or migration. It was also shown that the PDEB1-depleted cells could be rescued by mixing them with wild-type (WT) parasites. In collaboration with Kent Hill’s laboratory (UCLA) we generated a PDEB1 knockout (KO) and investigated its behavior in tsetse. We found that the KO was extremely poor at colonizing the PV and that this phenotype could be rescued by expressing PDEB1 ectopically. Interestingly, infection of flies with a mixture of WT cells and KO cells did not show complementation in vivo and only limited complementation in vitro. The different reactions in SoMo may be due to the 10% residual PDEB1 present in the RNAi line.

So far, studies of trypanosome transmission have focused on the midgut and salivary glands. This study shows that the PV plays a central role, with successful colonization depending on appropriate cAMP signaling.


Transcriptional Basis of Natural Transdifferentiation

Jaime Osuna Luque1,2, Sophie Jarriault2, Peter Meister1

1Institute of Cell Biology, University of Bern, Switzerland; 2Institut de Génétique et de Biologie Cellulaire et Moléculaire, University of Strasbourg, France

Traditionally differentiated identity could not be reversed. However, landmark studies in the last decade have shown that not only can a differentiated cell identity be experimentally erased, but remarkably, that it can be naturally converted into a different cell type in vivo, a process called transdifferentiation.

I study a natural transdifferentiation event naturally occurring in vivo in a single cell in 100% of the animals and 100% efficiency. This cell transdifferentiates from a hypodermal fate into a moto-neuron identity. This system has contributed key insights on the transition and cellular steps involved in transdifferentiation and the identification of conserved nuclear factors crucial to the initiation of the process, or the relative importance and roles of transcription factors versus histone modifying factors for the dynamics and robustness of the conversion. This suggests that many genes are switched on/off. The exact transcriptional dynamics of these genes remains unknown.

If the transdifferentiation process is highly efficient, it cannot be modelled in vitro. I therefore need to use entire animals in which I would like to understand the transcriptional dynamics in a single cell. I am therefore establishing an RNA polymerase II footprinting technique. This is based on DamID, a technique which uses a fusion protein between RNA polymerase subunits and a bacterial adenine methyltransferases (Dam). Binding of the RNA polymerase to transcribed genes leads to DNA methylation, which can be subsequently identified using molecular techniques. To restrict expression of the Dam fusion to the transdifferentiating cell, I setup a cascade of recombinations. To temporally control when the footprint is set, I additionally use a degron system, in which an applied plant hormone leads to protein degradation. Once has been determined, I will validate these results using sm-FISH experiments. The role during transdifferentiation of genes turned on or off will be examined.


In silico Study of Human SRD5A1 and SRD5A2 Suggests that Underinvestigated SNPs Might Be Disease-Causing and Alter Androgen Biosynthesis

Efstathios Katharopoulos1,2,3, Amit Pandey1,2, Christa Flück1,2

1Department of Pediatrics, Division of Endocrinology, Diabetology & Metabolism, Inselspital, Bern University Hospital, Switzerland; 2Department for BioMedical Research, University of Bern, Switzerland; 3Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland

Background: In the biosynthetic pathway of androgens steroid 5α reductase (SRD5α) catalyses the conversion of testosterone (T) to its more potent androgen receptor-activator, dihydrotestosterone (DHT). In principal, there are two enzymes with similar catalytic activity: type 2 enzyme is mainly expressed in reproductive tissues (e.g. testis), while type 1 in peripheral tissues (e.g. skin). Mutations in the SRD5A2 cause disordered sex development. By contrast, for the SRD5A1 no phenotype is associated with variants. In addition, both proteins are poorly characterized even though they are essential for sex development and reproduction.

Aim: We sought to explore in silico how coding SNPs in human SRD5A2, SRD5A1 may affect the function of the enzymes.

Methods: A structural model of the proteins was built as no 3D model currently exists. All human SNPs were collected from different databases.

Results: Evolutionary conservation analysis shows that certain areas of both enzymes are highly conserved among species. Using our novel 3D model we found that these areas are affiliated with substrate and cofactor binding domains. In our study we found coding SNPs within or close to those highly conserved areas. In total we acquired 7 SNPs in SRD5A1 and 12 SNPs in SRD5A2, which have not been related to a human phenotype or functionally studied. Our model predicts them to cause protein instability and in addition to affect binding of substrate T or cofactor NADPH to the enzymes.

Conclusion/Significance: We found underinvestigated SNPs in SRD5A1 / SRD5A2, which might be of clinical importance according to our novel protein models and bioinformatic studies. It is therefore possible that individuals carrying particular SNPs show a minor phenotype that is not yet described in the medical literature. Ongoing enzyme kinetic studies of these SRD5A variants should be able to solve this question, and might trigger further detailed clinical investigations in individual carriers.


Streptococcus pneumoniae Binds and Responds to Peptides Found in Ribosomal Proteins of Other Bacterial Species

Fauzy Negiub Ali Nasher1,2, Manfred Heller3, Lucy Hathaway1

1Institute for Infectious Diseases, University of Bern, Switzerland; 2Graduate School for Cellular and Biomedical Sciences, University of Bern, Switzerland; 3Proteomics and Mass Spectrometry Core Facility, Department for BioMedical Research, University of Bern, Switzerland

Background and Aims: Pneumococcus colonizes the nasopharynx alongside other bacteria species. We predict that AmiA, AliA and AliB, substrate-binding proteins of a pneumococcal ABC transporter, play a role in interspecies communication. We have previouslyshown that peptides matching other bacterial species are recognized by homologues of AliB (AliB-like ORF1 and ORF2) in nonencapsulated S. pneumoniae. AmiA, AliA and AliB, unlike AliB-like ORFs, are universally present in virulent, encapsulated pneumococci. Here we aimed to determine: (i) whether AmiA, AliA and AliB also bind foreign peptides of other bacterial species and (ii) the effect of binding on pneumococcal phenotype.

Methods: We expressed recombinant AmiA, AliA and AliB proteins and incubated them with human nasopharyngeal swabs to capture specifically bound ligands. Manual de novo peptide sequencing was performed to confirm the sequences of the peptides from their MS/MS spectra. Tryptophan fluorescence binding assay confirmed binding of the proteins to their ligands. Growth assays were performed in defined peptide-free medium (CDM).

Results: We found a total of 11 possible ligands. Among these, we confirmed binding for one peptide for each protein: AmiA, AliA and AliB bind three different peptides matching ribosomal proteins of Gammaproteobacteria, including common colonizers of the nostrils and nasopharynx. Phenotypic assays showed binding of the peptides results in altered pneumococcal growth.

Conclusions: We propose a novel route of interspecies bacterial communication in the nasopharyngeal microbiota via ribosomal protein-derived peptides which bind AmiA, AliA and AliB proteins, resulting in changes to pneumococcal phenotype. Such interspecies communication is a potential target for intervention.


Use of Nuclear Magnetic Resonance (NMR) Spectroscopy for Analysis of Pneumococcal Serotype and Biosynthetic Mechanism of Capsular Polysaccharide (CPS)

Lukas J. Troxler1, Julien Furrer2, Markus Hilty1,3

1Institute for Infectious Diseases, University of Bern, Switzerland; 2Department of Chemistry and Biochemistry, University of Bern, Switzerland; 3Department of Infectious Diseases, Inselspital, Bern University Hospital, University of Bern, Switzerland

The composition of the capsular polysaccharide (CPS) of Streptococcus pneumoniae, a major human pathogen, is a main virulence factor and defines the pneumococcal serotype. Due to its direct and detailed depiction of molecular structures, Nuclear Magnetic Resonance (NMR) spectroscopy has great potential for elucidation of capsule structures and detection of potentially new serotypes. We will show examples of this from our earlier experiments using strains of serotypes 19A and 23B where new subtypes were suspected based on genetic information.

We furthermore hypothesized that NMR could also be used to detect incorporation of carbon source molecules from growth media into the capsule structure. We tested this by culturing S. pneumoniae in chemically defined medium (CDM) and Lacks medium containing sugars labelled with deuterium (2H) or 13C as the sole carbon source and extracting the capsule polysaccharide. This was then analyzed by 1H and 13C NMR spectroscopy and compared to unlabelled samples.

Our experiments showed that in S. pneumoniae serotype 6B, labelled glucose and galactose from the growth medium are directly incorporated into the capsular polysaccharide structure without change to their basic structure. These insights could help increase our understanding of the mechanism of CPS biosynthesis. NMR is thus useful for research into both the structure and metabolism of S. pneumoniae CPS.


In silico Analysis of Genetic Variants of the Equine CYP3A94, CYP3A95, and CYP3A97 Genes

Sara Vimercati1, Vidhya Jagannathan2, Amit Pandey3, Jana Zielinski1, Tosso Leeb2, Meike Mevissen1

1Division of Veterinary Pharmacology and Toxicology, Vetsuisse Faculty, University of Bern, Switzerland; 2Institute of Genetics, Vetsuisse Faculty, University of Bern, Switzerland; 3Department of Pediatrics, University Children's Hospital, Bern, Switzerland

CYP450 proteins contain a large number of genetic variants that might result in intra- and inter-species differences in drug metabolism and substrate interactions. Although CYPs share a high sequence identity, the enzymatic activity was shown to differ between species.

Analysis of genetic variants from equine CYP3A94, CYP3A95 and CYP3A97 obtained from whole genome sequences from 81 horses of different breeds showed the presence of various genetic variants. We found 3 protein-changing variants in CYP3A94. Remarkably, one of them was a splice site variant (c.798+1G>A), which probably leads to complete loss of function. The mutant allele had a frequency of 30% and 7 horses were homozygous for this variant. These findings indicate that CYP3A94 is not under strong selection in horses. In CYP3A95, we identified 9 missense variants. For 6 of them, horses with all possible genotypes were found and2 of these were very common (p.His214Asp, pThr392Ser, mutant allele frequency 93% and 30%, respectively). In CYP3A97, 7 missense variants were found and all 3 genotypes were observed for them in the 81 horses. The most common variants were p.Val500Met and p.Thre119Ile with a mutant allele frequency of 67% and 60%, respectively.

An in silico CYP3A94 model was created based on the CYP3A4 crystal structure. The sequence alignment showed that the wild type residues of CYP3A94 variants are conserved between species but their function has not been analyzed. The model demonstrated that the 3 wild-type residues are localized on the CYP membrane domain (Ser6), on the typical CYP F-helix (Asp217) and on a loop between the J and K helices (Ala343). p.Asp217Asn is closest to the heme group (16.5Å), while Asp217 showed a shorter distance (15.9Å) possibly leading to differences in substrate oxidation.

Mutations in the CYP3A4 F helix were reported to modify interactions with raloxifene. Modeling CYP3A isoenzymes will allow studying structural differences important for drug metabolism.


Engineering of a Proton-Driven Antibiotic Translocation Module

Mirko Stauffer, Hüseyin Ilgü, Zöhre Ucurum, Dimitrios Fotiadis

Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland

The contamination of natural water sources with antibiotics poses a growing threat as it can lead to antibiotic resistances in environmental microorganisms. One approach to the bioremediation of such water sources could be the use of liposome- or polymerosome-based bionanoreactors, which are able to actively accumulate and degrade antibiotics.

Peptide transporters were shown to be involved in the transport of various peptidomimetic drugs, including β-lactam antibiotics. This makes them an interesting target for the engineering of an antibiotic transport module for subsequent incorporation into a light/solar-driven molecular factory for bioremediation.

The apo crystal structure of the prokaryotic peptide/antibiotic transporter YePEPT was previously solved in our laboratory. The moderate affinity of YePEPT for β-lactam antibiotics was significantly improved by structure-based mutagenesis in the putative substrate binding site in combination with an established functional uptake assay. Two mutations were identified, which significantly increased the affinity of YePEPT for certain antibiotics.

Furthermore, we could successfully grow 3D crystals of this YePEPT-double mutant in the presence of a selected antibiotic, which will soon be analyzed by X-ray diffraction. An X-ray crystal structure of YePEPT-double mutant in complex with an antibiotic will reveal protein-antibiotic interactions at the molecular level and allow efficient structure-based mutagenesis for improving the affinity of YePEPT for antibiotics into the nanomolar range.


Abolition of the Dominant Negative Effect of Brugada Syndrome Variants in SCN5A

Zizun Wang1, Valentin Sottas2, Jean-Sébastien Rougier1, Hugues Abriel1

1Institute of Biochemistry and Molecular Medicine, University of Bern, Switzerland; 2Institute of Physiology and Pathophysiology, Medical Faculty, Heidelberg University, Germany

Introduction: The voltage-gated sodium channel Nav1.5, encoded by the SCN5A gene, is responsible for the inward depolarizing cardiac Na+ current. Loss-of-function mutations in SCN5A result in alterations in cardiac action potentials with diverse arrhythmia phenotypes, e.g. Brugada Syndrome. Nonetheless, as Nav1.5 is a monomeric channel, the mechanisms of the dominant-negative (DN) effect are still unclear. The calmodulin model serves as an adaptor between different Na+ channels. We present findings of a new Nav1.5 variant in a patient with Brugada syndrome, p.Y87C, which leads to a DN effect when co-expressed with wild-type Nav1.5 channel.

Aim: Investigation of molecular mechanism of the DN effect.

Methods: We use TsA-201 cells and patch-clamp technique to study the sodium current, co-immunoprecipitation (Co-IP) to study the interaction of the Nav1.5 alpha subunits. We use pull-down experiments and calmodulin-beads to show the interaction of Nter of Nav1.5 and calmodulin.

Results: By using transient transfected TsA-201 cells, we observed a strong reduction in current density in the presence of the p.Y87C variant. Interestingly, co-transfection of the N-terminal domain (Nter) of Nav1.5 with wild-type Nav1.5 induced a significant increase in sodium current. In addition, the decrease observed with the variant p.Y87C was abolished by co-transfecting Nter, suggesting a role of the Nter in the DN effect. Moreover, the DN effect was abolished after the deletion of the 25-Nterminal residue calmodulin binding site.

Conclusions: The N-terminal fragment of Nav1.5 increases the trafficking of the full channel. Whereas a variant with the N-terminal fragments, in which the variant can induce a dominant-negative effect, are not able to induce the same effect. There is an interaction between the full channel and the N-terminal part. Besides, these findings are consistent with the calmodulin model, which suggests that the multimerization of Nav1.5 channels depends on calmodulin.



 
Contact and Legal Notice · Contact Address:
Conference: GCB Symposium 2018
Conference Software - ConfTool Pro 2.6.116
© 2001 - 2018 by H. Weinreich, Hamburg, Germany