Department of Biological Sciences (BIO)

Master theses submitted in 2013

Main content

Mari Katrine Berg

Mari Katrine Berg (A. Goksøyr, R. Lille-Langøy, H. Routti)

Peroxisome proliferator-activated receptors(PPARs) in polar bear (Ursus maritimus) as targetreceptors for environmental pollutants

Environmental pollutants with metabolic disrupting abilities are present in the Arctic. Particularly high levels of pollutants have been found in the apex predator of the Arctic marine food web, the polar bear. This species is dependent on a highly regulated metabolic system as it undergoes a distinct feasting and fasting regime due to seasonal variation in availability of food. Key regulators in consumption and storage of energy are the peroxisome proliferator-activated receptors (PPARs, NR1C), members of the nuclear receptor superfamily. PPARα and PPARγ are particularly important due to their positions in regulation of lipid metabolism and adipose cell differentiation, respectively.

The aim of this study was to clone polar bear PPARα and PPARγ, study their activation by contaminants, and locally establish a method to investigate how environmental pollutants affect lipid metabolism in vitro. Purifying RNA from polar bear tissue and cloning of polar bear PPARα and PPAR γallowed for phylogenetic comparisons and construction of plasmids. Sequence comparison with the human orthologs revealed 94 % and 98 % amino acid identity for PPARα and PPARγ, respectively. The constructed plasmids were used to establish a luciferase reporter assay (LRA) to assess PPAR activation by environmental pollutants. Activation by known ligands showed that the established LRA system was functional. Out of eleven single compounds and three synthetic POP mixtures, only PPARγ was activated by one compound,namely tetrabromobisphenol A (TBBPA). The local establishment of the preadipogenic cellline3T3-L1 as a model for studying adipogenic capacity was completed. Preliminary exposures to mixtures of MeSO2- metabolites of POPs as identified in polar bears suggested that this mixture can increase lipid accumulation in the preadipogenic cell line both in the absence and presence of a hormonal cocktail (MDI). Cells exposed to OH-metabolites of POPs increased lipid accumulation in the absence of MDI, but decreased accumulation in its presence. This suggests that one or several compounds in the mixture has an inhibiting effect on differentiation.

Due to lack of replicates in this time-limited study, experiments should be repeated before final conclusions are drawn. The established LRA and 3T3-L1 cell-line as a method for studying a dipogenic potential of environmental pollutants can further be used to assess the metabolic effects of unlimited amounts of compounds, alone or in combination.


Ingvill Tolås

Ingvill Tolås (R. Male, L. Sandlund, P. Bhattachan)

Characterization of Ultraspiracle and Hormone Receptor 38 in the salmon louse Lepeophtheirus salmonis

The salmon-louse, Lepeophtheirus salmonis, is a ecto-parasitic crustacean attacking salmonids, harming wild populations, and resulting in huge financial losses in the salmon-farm industry. Salmon lice are fast developing resistance to most treatments currently in use, making the need for new treatment methods profound. Physiological processes regulated by nuclear receptors (NRs), like growth, reproduction, and molting could be possible targets for new pesticides, and manipulation of NRs in the molting-process is common in insecticides.

The NR superfamily are largely ligand-dependent transcription-factors that modulate down-stream gene expression. In arthropods, the NR-mediated ecdysteroid-signalling pathway is important. Ecdysteroids interact with a heterodimeric receptor consisting of ecdysteroid-receptor (EcR) and ultraspiracle (USP) to regulate essential physiological responses, such as molting. In insects, ecdysteroid-function is regulated by juvenile hormones (JH), hypothesized to function via USP-binding.

A third nuclear receptor, and hormone receptor 38 (HR38), modulates ecdysteroid-response by competing with EcR for binding to USP. Deduced amino-acid sequences of LsUSP and LsHR38 revealed a high degree of conservation, but variations in the ligand binding domain. Variations could affect drug-affinity and could therefore contribute to L.salmonis specific pesticides. LsHR38 was found to encode two distinct A/B-domain specific isoforms, and variant 2 was dominantly expressed in the adult female. Each isoform contained two internal sub-variants. Relative expression of LsEcR, LsHR38 and LsUSP from several developmental stages was quantified using real-time quantitative PCR. Mature females showed elevated LsEcR and LsHR38-expression, indicating that they participate in female-specific processes, like vitellogenesis. USP-expression increased at the adult stage, but did not vary between genders. To characterize the interaction of these NRs and possible ligands in L.salmonis, a mammalian two-hybrid system was used. Ponasterone A was shown to trigger heterodimerization between EcR and USP in a dose dependent relationship. 20-E was found to effect the response to Pon A, but not to trigger dimerization alone. JHIII did not appear to affect dimerization, but its crustacean counterpart, methyl farnesoate, had an effect when added in low concentration. HR38 appeared to interact with both USP and EcR, and its role could vary depending on its partner. Ultimately, the results obtained in this study contribute to the understanding of the ecdysteroid signalling-pathway in L.salmonis.


Helene Stuart

Helene Stuart (B. E. Grøsvik, K. E. Fladmark)

Mechanistic studies of Atlantic cod responses to alkyl-substituted naphtalene and phenanthrene
Current and potential future biomarkers of exposure in atrial and hepatic tissues

Polycyclic aromatic hydrocarbons (PAH) from crude oil are a constant threat to the local ecosystem causing acute and chronic stress. Despite representing the dominant toxins in fossil fuels, low molecular weight (MW) PAH have received limited attention in toxicity studies and environmental risk assessment. Recently, exposure of teleost fish to the three-ring benzenoid phenanthrene during early life stages (ELS) was found to cause substantial malformations in cardiac, spinal and cranial structures in addition to widespread oedema.

Disputes about opening the Barents Sea to offshore oil and gas industry has lead to concern about the unique properties of areas around Lofoten, Vesterålen and Senja in terms of recruitment of an estimated 70% of commercial fish species. One of these species is the already threatened, commercially important species Northeast Atlantic cod.

Liver and hepatic tissues are commonly utilised in toxicity testing and biomarker analyses. However, the endocardial endothelium of Atlantic cod, and other bony fish have shown to be equivalent in function to liver sinusoidal endothelial cells (SEC) of higher vertebrates in being general scavengers of a broad range of solutes and particular matter. In addition, endothelial cells are thought to be significantly more sensitive to exposure to ecotoxins.

Current biomarkers of organic xenobiotics are either general in reflecting physical or chemical stress (e.g. Hsp70), or specific to large PAHs and dioxin-like compounds (e.g. CYP1A). However, their ability to reflect exposure to naphtalene and phenanthrene is disputed.

This master thesis looks at the use of primary endcardial endothelial cell (EEC) cultures from Atlantic cod to study the mechanism of toxicity of low molecular weight PAHs. We looked at some of the most commonly applied biomarkers of PAH toxicity to assess the use in detecting exposure to these substances. We also used liquid chromatography tandem mass spectrometry (LC-MS/MS) to screen for proteomic changes indicative of proteins and pathways affected.Efforts at establishing primary EEC cultures were unsuccessful, and it was decided to attempt culturing whole atrial tissue and exposing to 2-methylphenanthrene (2-MP) in vitro. Atrial tissue exposed to benzo(a)pyrene (B(a)P), was used as a positive control for CYP1A-induction. A dosage-dependent expression pattern was observed in tissues exposed to benzo(a)pyrene, and at low dosages of 2-methylphenanthrene. However, higher dosages had an inhibiting effect.

Atlantic cod yearlings were exposed to 2-methylnaphtalene (2-MN) and 2-methylphenanthrene by intraperitonial (ip) injections of various dosages. Eight days later, ivorgans were harvested and hepatic tissues were analysed for CYP1A protein expression and activity using enzyme-linked immunosorbant assay (ELISA), western blot and ethoxyresorufin-O-deethylation (EROD) assay. Expression of Hsp70 was assessed by ELISA and western blot. Both toxins inhibited CYP1A protein expression in an inverse dosage-related fashion. However, 2-methylphenanthrene induced EROD activity. Hsp70 expression showed a general inhibition of protein synthesis.

Atrial and hepatic tissues were sent for proteomic analyses by label-free MS/MS. Atrial tissues showed an induction of actins, troponins and myosins in response to both toxins, and proteins involved in intracellular calcium regulation in response to 2-MP only. Hepatic tissues had a significant alterations in expression of sorting nexin 12 (SNX12) and allograft inflammatory factor 1 (AIF1) when exposed to 2-methylphenanthrene.

Although this was just a preliminary study into the mechanisms of low MW PAH-induced toxicity in Atlantic cod, it suggests that using atrial tissues and EEC cultures may bring significant information in toxic mechanism. Cardiac muscle proteins, SNX12 and AIF1 may be potential candidates for biomarkers of crude oil-exposure.


Helene H. Sandnes

Helene Heitmann Sandnes (Ø. Halskau, A. Lewis)

Creating bamlet using ALPs and investigating their effect on differentiated and non-differentiated human breast cancer cells 

HAMLET (Human Alpha-lactalbumin Made Lethal to Tumour Cells) was discovered coincidentally when a group of researchers at Lunds University in Sweden were studying properties of human breast milk. Something in the milk appeared to be killing the cancerous cell in their experiment, while the non-cancerous cell-lines were unaffected. It was later found that the HAMLET complex was responsible for this interesting occurrence. HAMLET consists of human alpha-lactalbumin (HLA) and oleic acid (OA), and the phenomenon has grown to include several functionally similar compounds, such as the bovine equivalent BAMLET (Bovine Alpha-lactalbumin Made Lethal to Tumour Cells). The complex has a promising potential as a cancer drug, and several promising in vivo assays have further demonstrated the clinical potential of this drug, including on bladder cancer, glioblastomas,and very recently colon cancer.

The molecular mechanism of how HAMLET carries out its cytotoxicity, the specific action against cancer cells, and the roles of the individual components are all currently under scrutiny and debate. To investigate these matters, this study looked into the effect of BLAOA (Bovine Alpha-Lactalbumin-Oleic Acid, a complex closely resembling BAMLET) on differentiated and non-differentiated cell-lines from breast tissue. Presto Blue viability assay and Trypan Blue exclusion assay were performed to investigate the toxic effects on the celllines. To explore the cellular response to the treatments, DNA fragmentation and PARP cleavage assays were carried out.

Cell viability assays established a marked cytotoxicity for BLAOA, and the complex portrayed specific action against the cancerous cells. We also attained indications that OA could be the main cytotoxic component of the complex. The OA component in the complex can be replaced with anticancer Alkylphospholipids (ALPs), but the resulting complexes (BLA-M and BLA-E) behave differently than BLAOA. While OA in complex greatly increases OA-toxicity, ALPs in complex seem to have lower toxicity than ALPs only. This suggest that the BLA-ALP complexes represent a different phenomenon altogether. Results from DNA fragmentation and PARP-cleavage assays indicated that both BLAOA and ALP derived substances induce an apoptotic response in the cancerous cells subjected to treatment, indicating that the difference, if any, lies elsewhere.


Christopher Holte

Christopher Holte (K. E. Fladmark)

Azaspiracid-1 effects on the developing zebrafish

Azaaspiracid-1 (AZA-1), a marine polyether marine phycotoxin discovered in 1996 after an outbrake of shellfish intoxication in the Netherlands. The azaspiracids are noted for DSP like and neurotoxic-like symptoms in vivo. AZA-1 and AZA-2 are the predominant forms of azaspiracid, with all other isoforms considered metabolic bi-products. AZA-1 caused multiple organ damages when orally administered to mice, with pronounced intestinal necrosis. AZA-1 has also shown neurological symptoms in mice. In in vitro experiments AZA-1 was found to reduce cellular adhesion and disorganize the actin cytoskeleton. A large number of proteins have been found differentially regulated in cultured cells exposed to AZA-1. AZA-1 is further known to be a finfish teratogen, with adverse effects such as edemas, reduced heart rates, decreased development rates and reduced hatching success, observed in Japanese medaka.

The purpose of this study was to determine the adverse effects of AZA-1 in developing zebrafish embryos, and to examine for differential expression in the proteome of zebrafish exposed to AZA-1.

AZA-1 was found to cause increased mortality in zebrafish embryos exposed to doses of 100 pg AZA-1 per egg and above.AZA-1 exposure did not have an associated phenotype for the dosages in this study (0.02pg/egg to 400 pg/egg). AZA-1 exposure was not found to decrease measured embryo or developmental rates of zebrafish embryos at all concentrations used in this study. There was observed a significant heart rate reduction in embryos exposed to dosages of 50 pg AZA-1 per egg and above, with a significant decrease in heart rate when increasing dosages. Two dimensional gel electrophoresis of whole embryo proteome showed differences in spot intensities indicative of differentially expressed proteins.


Alex K. Datsomor

Alex Kojo Datsomor (J. Sagen, G. Mellgren)

The impact of SRC-2 and Lipin-1 co-activator functions on lipid metabolism

Regulation of lipid metabolism is crucial in maintaining energy homeostasis. Peroxisome proliferator-activated receptor α (PPARα) is a transcription factor that controls lipid metabolism via interaction with ligands and transcriptional co-regulators.

Lipin-1 is a bifunctional protein that serves as a transcriptional co-activator and an enzyme in triglyceride synthesis. Studies have shown that lipin-1 co-activates PPARα together with PPARγ co-activator 1α (PGC-1α). Additionally, lipin-1 co-operates with PGC-1α to regulate the expression of PPARα.

Recent studies have identified steroid receptor co-activator 2 (SRC-2), a member of SRC family, as a potential binding partner of lipin-1. This thesis sought to investigate the hypothesis that SRC-2 and lipin-1 interact to regulate the expression of PPARα and its target genes involved in fatty acid oxidation. In vitro transcriptional activation assays showed that SRC-2 and lipin-1 co-activate PPARα gene promoter in co-operation with PGC-1α. RT-qPCR analyses, following siRNA-mediated knockdown of lipin-1 demonstrated downregulation of PPARα mRNA.

Conversely, SRC-2 knockdown had no effect on the expression of PPARα. Knockdown of lipin-1 downregulated selected PPARα-target genes involved in fatty acid oxidation, whereas SRC-2 knockdown had no effect on most PPARα-target genes, with exception of carnitine palmitoyltransferase 1A (CPT1A). In conclusion, SRC-2 and lipin-1 do not seem to act together to co-activate PPARα gene expression and transcriptional activity in relation to fatty acid oxidation.


Marianne Goris

Marianne Goris (A. Lewis, T. Karlsson, V. Ardawatia, H. When)

Subcellular PI3K Signaling in Adipocyte Differentiation

Phosphoinositides (PIs) are a group of amphiphatic phospholipids involved in many intracellular signaling events due to their function as direct or second messengers, both in the cytoplasm and the nucleus. Special focus has been given PtdIns(3,4)P2 and PIP3 in the recent years due to their pivotal role in insulin--‐induced PI3K/AKT signaling.

This pathway is implicated in a myriad of cellular processes. One of these is adipogenesis, which is the event of adipocyte differentiation. Exactly how PtdIns(3,4)P2 and PIP3 exert their effect on nuclear PI3K signaling separately from their effect in the cytoplasm during adipogenesis is still highly unclear.

The principal aim of this study was to investigate the respons of some nuclear PI3K/AKT signaling constituents, namely PI3K, AKT, PtdIns(3,4)P2 and PIP3, during adipogensis. The present study demonstrated an almost instant activation of cytoplasmic AKT during insulin-induced adipogenesis of 3T3-L1 cells. AKT was observed in the nucleu of non-stimulated cells, followed by the detection of active AKT in the nucleus after 20 minutes of stimulation. Indicating a possible time and spatially restricted activation of AKT in the nucleus, compared with the cytoplasm.

Furthermore, increase in cytoplasmic and nuclear PtdIns(3,4)P2 and PIP3 levels were also observed post-stimulation. The Class I PI3K isoforms p110α and p110β were observed in the cytoplasm at constant levels during adipogenesis, while they were undetectable in the nucleus under equal conditions. The PIP3 deregulating phosphatases PTEN and SHIP2 were found to locate in the cytoplasm, but not in the nucleus. Moreover, PI3K C2α and Type IIα PIPK were observed as highly nuclear.

Taken together, these findings could indicate alternative nuclear PtdIns(3,4)P2 and PIP3 metabolizing routes, differing from the well known cytoplasmic ones, upon insulin induced adipocyte differentiation of 3T3-1 cells.


Marianne K. Holmedal

Marianne Kristvik Holmedal (R. Aasland, C. Issalene)

The Molecular Basis for Ligand Binding Specificity of the CW Domain

The CW domain is a histone recognition module found in chromatin proteins and epigenetic regulators. This domain has specificity towards H3 histone tails with lysine 4 methylations. Different CW domains have different specificity for histone H3 with of different CW domain-containing proteins have specificities for mono-, di-, or trimethylations on lysine 4. The molecular basis for these differences is currently unknown, but previous reports have suggested that the C-terminal region of these domains have a role in the determination of their specificities.

The specificities of the CW domains of the plant histone methyl transferase ASHH2 and the human ZCWPW1 protein have been studied. Previous studies of both proteins have shown that although the two proteins share sequence and structure for most of the protein domain, they have specificities towards different modifications of histone tail modifications (H3K4me1 and H3K4me3, respectively) (He et al., 2010; Hoppmann et al., 2011). While the two proteins share the conserved CW core domain, their C-terminal regions are non-conserved (Hoppmann et al., 2011).

Ligand specificity has been tested for the CW domains of a number of proteins (Hoppmann et al., 2011), a nd all domains tested bind to histone H3 peptides methylated on lysine 4. They do, however, have different specificities, and the different subfamilies of CW domains have different, non-conserved C-terminal regions. The fact that binding specificities seem to follow differences in C-terminal region rather than differences in the conserved region, makes it reasonable to think that the C-terminal region, in addition to being crucial for binding, is also involved in ligand recognition.

To put to test this hypothesis, chimeric proteins were engineered, using the CW domains of ASHH2 and ZCWPW1. The C-terminal region of both proteins were replaced with the C-terminal region of the other, resulting in two chimeric proteins; one with the N-terminal conserved region of ZCWPW1 CW domain connected to the C-terminal non-conserved region of the ASHH2 CW domain. The other contained the ASHH2 N’-terminal conserved region connected with the ZCWPW1 C-terminal non-conserved region.

These chimeras were then tested in pull-down assays with differently modified H3 histone peptides, in addition to isothermal titration calorimetry measurements with H3K4 (me1/me2/me3) modifications. The data presented in this thesis supports the hypothesis that the C-terminal region indeed has high influence on ligand recognition for the CW protein domain. If confirmed, this is a novel mechanism for determination of ligand specificity for histone recognition modules.


Marie H. Solheim

Marie Holm Solheim (L. Bj. Gundersen, I. Aukrust, P. R. Njølstad)

GCK-MODY due to protein instability; does compound GKA50 function as a stabilizing agent for glucokinase?

GK plays a central role in blood glucose homeostasis, and functions as a glucose sensor in pancreatic β-cells as well as a regulator of hepatic glycolysis, glycogen synthesis and gluconeogenesis. More than 700 naturally occurring mutations in the glucokinase encoding gene (GCK) have been found, most of which result in GCK-MODY, dominantly inherited mild fasting hyperglycemia. Due to the central role of glucokinase in insulin secretion and the various glycemic disorders found associated with mutations in GCK, the enzyme has been an obvious candidate for diabetes research, and a promising drug target for type 2 diabetes (T2D). The discovery of glucokinase activators (GKAs) has further encouraged its potential for therapeutic targeting.

V62M and G72R are naturally occurring GCK mutations found to cause GCK-MODY in patients. Functional characterization of the kinetic characteristics of the mutant V62M and G72R enzymes has so far given inconsistent results. This study demonstrates that both mutant enzymes show a reduced rate of glucose turnover (kcat) and reduced glucose affinity ([S]0.5), indicating a lower enzyme activity compared to the wild-type (WT) enzyme, using both the tag-free and GST-tagged proteins at 30 and 37 oC. Investigating protein stability by limited proteolysis the V62M and G72R hGK mutants were more susceptible to trypsin digest than the WT enzyme, and demonstrated also slightly increased rates of degradation, determined by pulse chase analyses in stably transfected mouse insulinoma β-cells (MIN6). In silico analyses predicted an instability of both mutant proteins in the open and closed hGK conformation (G72R>V62M in open conformation).

Further, the V62M and G72R hGK enzymes were both subjected to activation when exposed to the glucokinase activating compound GKA50. Moreover, in the presence of GKA50, the WT protein and both mutants also showed reduced susceptibility to trypsin digest in vitro, which could indicate that GKA50 acts as a stabilizing agent. This study shows that V62M and G72R most likely cause GCK-MODY by a combined mechanism of kinetic inactivation and protein instability. The stabilizing effect of GKA50 on the GCK-MODY mutants advocates a possible new approach in the development of GKAs.


Rasmus M. Ree

Rasmus Moen Ree (T. Arnesen, L. Myklebust, K. E. Fladmark)

In vitro and in vivo study of zebrafish NAT enzymes

N-terminal acetylation, a co- and post-translational modification seen in over 80% of human proteins, is little understood in terms of functional importance. N-terminal acetylation is catalyzed by N-terminal acetyltransferases (NATs), of which six types exist in humans. NatA-NatF each process a distinct population of proteins based on the N-terminal sequence of the substrates.

This thesis is an attempt to establish the zebrafish Danio rerio as a model system in which to study the NAT enzymes. MBP fusion proteins of zebrafish Naa10 and Naa60 were cloned, expressed, purified and assayed in order to establish their function as NATs and characterize their substrate specificity. The knockdown phenotype of the major eukaryotic NAT, Naa10, in zebrafish embryos has also been studied.

Based on substrate specificity and bioinformatic conservation analyses, zNaa10 was deemed a true homolog of the human Naa10. Due to difficulties in purifying the recombinant zNaa60 protein as well as low in vitro acetyltransferase activity, the same conclusion cannot be reached for zNaa60.

Morpholino-mediated knockdown of zNaa10 has a severe phenotype in zebrafish embryos, including increased mortality, decreased growth and morphological abnormalities. These effects were partially mitigated by co-injection of human NAA10-V5 mRNA supporting the specificity of these effects. The results presented here indicate that Naa10 is an essential NAT in normal zebrafish embryogenesis.  


Sepideh Mostafavi

Sepideh Mostafavi (K. E. Fladmark, A. Wolf, B. E. Oftedal)

Characterization of the expression pattern of the autoimmune regulator, Aire, in Zebrafish

Autoimmune polyendocrine syndrome type I (APS-I) is a monogenic autoimmune disease in which the Autoimmune regulator gene (Aire) is mutated. Due to unfunctional Aire transcription factor, the immune system of these patients loses one of its main properties, self-tolerance, which means the ability to discriminate between self and non-self components. In this disease the self-reactive immune system compounds turn and attack endocrine organs instead of reacting foreign antigens. Patients with APS-I usually exhibit ectodermal defects such as enamel dysplasia and nail pitting in addition to the main symptoms.

The etiology behind these manifestations is obscure. Recently some publications have indicated that these ectodermal abnormalities may be due to developmental errors and that Aire may play a role during development beyond its essential function in thymic education of the immune cells.

In present study, we have recruited zebrafish as a developmental model to investigate the expression pattern of Aire during zebrafish development. We also studied the expression of the developmental genes; GATA-1, Ikaros and Rag-1, as temporal check points for Aire expression. Whole mount in situ hybridization, qualitative and quantitative polymerase chain reaction (PCR) and Western blotting were employed to study the spatial and temporal expression pattern of the Aire at different stages during development.

In addition, an antisense morpholino oligonucleotide approach was used to knock down the expression of Aire in order to elucidate its probable developmental role. The in situ analysis showed that Aire expression initiates in a wide region in brain at 24 hours post fertilization (hpf), i.e. prior to thymi formation and immune system development. Over time, Aire expression becomes more restricted. PCR results indicated variation in Aire expression during development, which can refer to the probable developmental function of this protein. Appearance of Aire expression at early stages, 24 and 48 hpf, is followed by reduction of Aire transcript at 72 hpf. A sharp rise in expression level was observed at 96 hpf and continued until 7 days post fertilization (dpf). Aire transcript level dropped at 10 dpf.

Our observation, identifying Aire expression at early stages of developments, even before complementation of the immune system and in tissues with different origins, reinforced our hypothesis that Aire transcription factor may play role in development.