Browsing by master's degree program "Translationaalisen lääketieteen maisteriohjelma (Translational Medicine)"

The contact site between the endoplasmic reticulum and mitochondria, also known as the mitochondria endoplasmic reticulum contact sites (MERCS), have a crucial role in maintaining the homeostasis within the cell. Across the MERCS multiple functions, such as regulation of calcium (Ca2+) homeostasis, lipid metabolism, ER stress, mitochondrial quality control (MQC) and regulation of unfolded protein response (UPR) take place. These processes have been shown to be implicated in numerous different neurodegenerative diseases, such as Parkinson’s disease. Parkinson’s disease is the second most common neurodegenerative disease that at the moment has no cure. The main obstacle in developing a neuroprotective treatment for the disease is the limited understanding of the key molecular events leading to neurodegeneration. One of the things in Parkinson’s disease that has eluded scientists for years is the selective death of the dopaminergic (DA) neurons in substantia nigra pars compacta. One hypothesis that could explain the selective death is the Ca2+ hypothesis, looking at the Ca2+ vulnerability of SNpc DA neurons as a plausible cause leading to the selective cell death. This project focused looking at the protein level and morphological changes of the ER and MERCS in stressed neurons, hypothesizing these as possible sites that contribute to the neuron vulnerability, as they are known to be the key modulators of the intracellular Ca2+ homeostasis. This study looked closer at two MERC proteins, GRP75 and BAP31, and one ER protein, SERCA2, to see how they are affected in stressed dopamine-like neurons. Firstly, the in vitro model was established by differentiating SH-SY5Y neuroblastoma cells to dopamine-like neurons expressing tyrosine hydroxylase. Three different molecular compounds were tested as possible stressors affecting the Ca2+ homeostasis within the neurons, and we concluded that thapsigargin, a commonly used stressor to model PD like pathology, leads to the highest measurable ER Ca2+ depletion. Lastly, we quantitatively and qualitatively analyzed the effect of 24-hour treatment with each stressor on the differentiated SH-SY5Y neurons. Thapsigargin treatment lead to an increased level of GRP75 and SERCA2. A slight increase in BAP31 was also detected after thapsigargin treatment, but no apparent changes of the ER morphology were detected. The results, together with previous research, show GRP75 to be a possible contributor to the pathology of the disease, but further research is needed to see if it could be a possible target for treatment.

Aims: Reading ability is a fundamental skill in the modern society, yet some individuals have difficulties in learn-ing to read and write. There is a lot of variability in reading skills, and one reason that can cause reading difficulty is a neurodevelopmental disorder called dyslexia. It is the most common learning disability, and the core deficit in dyslexia lies in word decoding, which is the process of connecting letter combinations into their corresponding auditory representations. Dyslexia is familial and is recognized to have strong genetic background. A dozen dyslexia susceptibility genes have been suggested, but DYX1C1, DCDC2 and KIAA0319 have been associated with dyslexia most commonly. The function of these genes is however not yet fully understood. In previous studies variation in these genes have been linked to struc-tural brain alterations in left hemispheric regions where language is mostly processed. The aim of this study was to examine the connection between dyslexia susceptibility genes DCDC2, DYX1C1 and KI-AA0319 and variation in brain activity during reading tasks in the left middle temporal gyrus (MTG), infe-rior Frontal Gyrus (IFG) and intraparietal sulcus (IPS), by combining functional magnetic resonance imag-ing data and genetic data in a neurotypical population. Previous studies have reported that weaker reading skills are associated with decreased brain activity in these regions, and reading incongruent sentences has been associated with increased brain activity in the left IFG and MTG. Methods: During fMRI, participants were presented with sentences with illogical and logical endings, and judged them as either congruent or incongruent, in distracted and undistracted conditions. Auditory speech stimuli were used as distractor. Regions of Interest analyses were conducted to examine brain activation in the aforementioned brain regions during distracted and non-distracted reading separately for different allelic groups in single nucleotide polymorphisms of the three genes. Results and Conclusions: DYX1C1 showed significant interaction with brain activation in the IPS. A significant interaction of DCDC2 with logic was found in the IFG and IPS showing that individuals carrying susceptibility alleles have reduced brain activation when reading incongruent sentences. Additionally, DCDC2 showed inter-action with distraction in the IFG, as individuals carrying susceptibility alleles had reduced brain activa-tion when a speech distractor was presented. In the MTG, there was a significant interaction of DCDC2 with logic and distractor showing that in different allelic groups, speech distractor modulated the activa-tion elicited by incongruent sentences in different ways. These results provide a link between variation in dyslexia susceptibility genes and brain activation during reading. Previous studies have mostly linked dyslexia susceptibility genes to structural brain alterations, and dyslexia and lower reading skills have been linked to variation in brain activity. The current study therefore expands the current understanding of genetic basis on reading and linguistic processing.

Multiple myeloma (MM) is a malignancy of antibody-secreting plasma cells in the bone marrow (BM). MM is the second most common hematological malignancy, accounting for about 14% of blood cancers. Despite the improvements in the treatment of MM, the disease remains incurable, and essentially all patients end up relapsing. Deletion of chromosome 16q, the location of tumor suppressor CYLD, occurs in 35% of MM patients. CYLD is a deubiquitinase, most recognized for its function as a negative regulator of the nuclear factor kappa B (NF-κB) pathway. The loss of CYLD is associated with disease progression and worse survival in MM, but its significance in drug response is unknown. As the loss of CYLD is common in MM, determining its effect on drug response is essential. Analysis of data gained from MM patient samples were used to study the effect of CYLD copy number status on drug response. The treatments were selected based on previous research performed by our group. The effect of homozygous deletion was most significant in inducing resistance to BMS-754807, an inhibitor of insulin-like growth factor 1 receptor. To investigate the role of loss of CYLD on drug response in vitro, cell lines with CYLD knockout (KO) were created using the clustered regularly interspaced short palindromic repeats (CRISPR) – CRISPR associated protein 9 (Cas9) technology. An established CYLD-KO cell line was treated with carfilzomib, bortezomib, dexamethasone and BMS-754807 to assess the effect of CYLD-KO. The CYLD-KO slightly increased the sensitivity to BMS-754807, but essentially no differences were detected in response to the drugs. The effect of CYLD-KO was additionally explored to NF-κB - and Wnt-pathway activation by Western blot analysis, but due to technical difficulties, the results were inconclusive. The loss of CYLD is a common genetic aberration in MM, giving a survival benefit for the malignant cells. Based on the results from patient data analysis, the loss of CYLD could promote drug resistance to BMS-754807, and the effect should be further studied with more cell lines with CYLD-KO. As the population ages, and as the median age of newly diagnosed patients is 70, the need for efficient MM therapies increases. Studying the mechanisms behind drug resistance and sensitivity is essential in the aim of improving the efficacy of MM therapy and, in the end, the overall survival of the patients.

Although approximately 80% of children with cancer survive five years after diagnosis, many childhood cancer survivors suffer from side effects and long-term health complications caused by the treatments. In addition, the diseases can often recur due to ineffective therapy. In pediatric sarcoma treatment, the recurrence of the disease is one of the major challenges. In rhabdomyosarcoma (RMS) 30% of cases end up relapsing and the cure rate is low with current therapies. Functional precision medicine approach is of interest especially with children since childhood cancers are likely diseases of upregulation or downregulation of protein expression with an overall low mutational burden and a lack of identifiable targets. In functional precision medicine the information from drug sensitivity and resistance testing (DSRT) of patient-derived cells (PDCs) is combined with other molecular tumor data, such as genomic and transcriptomic data, to better guide the selection of personalized treatments. In this study, we first established the PDC culture conditions for two longitudinal tumor samples of a young patient with RMS. DSRT was performed to age-matched healthy control fibroblasts and aortic smooth muscle cells, four RMS cell lines with different molecular backgrounds, and as a proof-of-concept study, to two longitudinal tumor samples of the RMS patient, to find compounds able to inhibit RMS cell viability. In addition, the clinically relevant genetic findings of the patient were highlighted. Specific drug responses relating to the molecular characteristics of the RMS cell lines were observed, such as MEK inhibitors in RAS mutated fusion-negative RMS cells. The findings of this work show that the DSRT assay can be utilized to discover RMS specific drug vulnerabilities, and it reveals the importance of proper control cells when using DSRT. This study provides evidence that combination of DSRT and NGS-based diagnostics may provide additional value for cancer treatment.

Uterine leiomyomas are common smooth muscle tumours, with a prevalence as high as 80%. Even though they are benign, they present severe symptoms such as heavy menstrual bleeding, pelvic pain and reproductive dysfunction. Uterine leiomyomas can be classified to conventional tumours and leiomyoma variants based on their histopathology. The tumours usually harbour one of the three driver alterations: MED12 mutations, HMGA2 overexpression or biallelic FH inactivation. Known risk factors for leiomyoma development are African ancestry, family history and age. Uterine leiomyomas are most typically treated by surgery, through either uterus preserving myomectomy or by definitive hysterectomy. This Master’s thesis is continuation of a study from Äyräväinen et al. 2020, a retrospective study of 234 patients undergoing myomectomy at Helsinki University Hospital during 2009-2014. The aim of this study was to analyse how many of these patients had developed recurrent leiomyomas and how often the tumours in subsequent operations were potentially clonally related. In addition, clinical characteristics associated with the operations were analysed. In total 18% of these patients had recurrent operations, leading to the screening of 77 individual uterine leiomyomas from 32 patients. The mutational statuses were studied systematically with molecular screening using Sanger sequencing and immunohistochemistry. Altogether 33 tumours from 21 patients were found to have identical mutational status with a tumour from the original study. Of these tumours, 14 had a MED12 mutation. All the MED12 mutations were found in exon two affecting either codons 44 or 36. Six tumours had HMGA2 overexpression, and eight tumours were FH deficient. Five tumours were triple negative for all studied alterations. Whereas 81% of the patients had had two removal operations, the rest of them had had three to five operations. The years between operations ranged from performing them on the same year to performing them ten years apart. Even though most of the recurrent tumours were sporadic, almost half (43%) of them had identical mutations, suggesting that though uterine leiomyomas usually arise independently, some might be clonally related. The mutational distribution was different in the recurrent tumours than in uterine leiomyomas in general, indicating that in addition to germline predisposition, the driver related characteristics might also contribute to the potential of recurrence and to the likelihood of developing clonal lesions. Tumours harbouring MED12 abnormalities were the least probable to be clonally related. The tumours showing identical HMGA2 overexpression were likely clonally related. The number of identical FH deficient ULs was high, but not unexpected, since all the patients harbouring the mutation in the recurrent tumours had HLRCC, and therefore having a predisposition. Most surprisingly, all patients with recurrent triple negative tumours had identical mutation statuses in the recurrent tumours, which points to previously unknown clonal development of these lesions. Most of the patients with more than two surgeries had recurrent mutations, suggesting that multiple surgeries might indicate the development of clonally related tumours. However, further research is required to confirm the clonal relationships and to investigate the pathological nature of the tumours with different driver alterations.

Recently, several novel post-translational modifications (PTMs) have been identified as important regulators in biology. Succinylation, the reversible addition of a succinyl group from a free succinyl-CoA into a protein lysine, is one such novel PTM. The last decade of research has unveiled succinylation as a powerful regulator of metabolism, prevalent in every organism it has been studied in and with functional effects on target proteins in several key metabolic pathways. A major contribution of this thesis is to catalogue the recent advances in succinylation research into the most comprehensive literary review currently available on succinylation. While the biological role of this PTM is being established, the relevance of succinylation in human disease has remained unclear. Meanwhile, mitochondrial DNA depletion syndrome caused by defective SUCLA2 (SUCLA2 disease) is a progressive hereditary mitochondrial disease with no available treatment. SUCLA2 disease is caused by defective mutations in the ß-subunit SUCLA2 of the TCA cycle enzyme succinyl-CoA synthetase. While the characteristic manifestations, including impairment of respiratory complexes, and the etiological mutations in this disease are well established, the pathogenic model for SUCLA2 disease has remained incomplete. As succinyl-CoA synthetase shares a substrate, succinyl-CoA, with succinylation, this thesis set out to probe SUCLA2 mutants for a potential succinylation phenotype. An extensive hypersuccinylation phenotype was characterized in fibroblasts and tissue samples from SUCLA2 mutant patients by immunochemical methods. The hypersuccinylation target identities in SUCLA2 mutants were revealed with proteomics by mass-spectrometry. Hypersuccinylation in SUCLA2 mutants was shown to be enriched in proteins participating in mitochondrial energy metabolism, including respiratory complex proteins. In addition, several novel metabolic phenotypes were characterized in SUCLA2 mutants with metabolomics by mass-spectrometry, most prominently a significant depletion of aspartate metabolism. While identification of extensive hypersuccinylation in SUCLA2 mutants establishes a novel concept of succinylation relevance in human metabolic disease, the prospect of altered regulation of the respiratory complexes due to hypersuccinylation lays the foundation for a novel pathogenic model for SUCLA2 disease. Meanwhile, the observed novel metabolic phenotypes significantly contribute to the current understanding on SUCLA2 mutant metabolism and inspire a hypothetical model on how the defective succinyl-CoA synthetase could be circumvented in the TCA cycle of SUCLA2 mutants.

STK11/LKB1 is a tumor suppressor gene and mutated in 18% of lung adenocarcinomas. Tumor suppressor liver kinase B1 (LKB1) is known to activate adenosine monophosphate-activated protein kinase (AMPK) and 12 AMPK-related kinases (ARKs) by phosphorylating a conserved threonine residue in their T-loop region. A number of studies focused on investigating the influence of LKB1-AMPK signaling on cancer cell proliferation. However, there is no systematic study for identifying the critical LKB1 kinase substrates in suppressing lung cancer cell growth. In this project, the LKB1-deficient lung adenocarcinoma cell line A549 cells were sequentially overexpressed with constitutively active mutants of AMPKα1, AMPKα2, MARK1, MARK2, MARK3, MARK4, NUAK1, NUAK2, SIK1, SIK2, SIK3. The overexpression status was confirmed at both genetic and protein levels by qPCR and Western blotting, correspondingly. In vitro growth assays demonstrated up to 33% reduced growth rate of A549 cells overexpressing AMPKα1, AMPKα2 and NUAK1. Furthermore, siRNA knockdown of the selected substrates in LKB1-overexpressing A549 cells significantly rescued the cell growth defect. These findings suggest, that AMPKα1, AMPKα2 and NUAK1 kinases are critical for LKB1-mediated cell growth defect in lung adenocarcinoma.

Transposable elements (TEs) are repetitive DNA elements that have an autonomous capability of replicating and inserting themselves into new loci within genomes. TEs have often been thought to be a part of the non-functional “junk DNA”, but advances in next-generation sequencing technology has revealed that TEs have rich biochemical functions: Specific TEs bind a large fraction of transcription factors and harbor marks of active chromatin in human genomes. However, there have been few genome-wide functional enhancer activity studies on the role of TEs in gene regulation, especially in the context of human cancers. In normal cellular homeostasis, TE activity is tightly controlled by epigenetic mechanisms. In contrast, the cell state in cancer genomes is often permissive for TE activation: genome instability and mutations, notably p53 inactivation, and nonmutational epigenetic reprogramming such as DNA hypomethylation are common characteristics of cancers. TE transcription, somatic retrotransposition and activation of cryptic promoter elements occur frequently in tumors. TE activation is highly heterogeneous between cancer types: for example gastrointestinal tract cancers such as colorectal cancer show high somatic insertion activity, whereas retrotransposition is rare in hematolymphoid malignancies. Due to the widespread TE activation in cancers, we posited that this may also be seen in the activity of cryptic TE enhancers, with specific active families of TEs characteristic to different cell types due to lineage-specific TF binding. We asked if TEs contribute to the enhancer landscape of cancers and to which extent, what are the differences between cancers in the activity and transcription factor binding and whether TE enhancers may have a role in tumorigenesis and the regulation of cancer-specific genes. To functionally study TEs, we utilized a high-resolution, unbiased, genome-wide massively parallel reporter assay (MPRA). We utilized colorectal and hepatocellular cancer cell lines to study the differences and similarities between TE activation and combined the MPRA data with orthogonal epigenetic data to study the in vivo signatures of the TEs. We found that both cell lines show common and highly enriched TE subfamilies that were mostly specific for p53, as well as TEs that were highly unique in both cell lines. By using in vitro methylated MPRA libraries, we found that CpG methylation has relatively minor a role in regulating the enhancer activity of some TE subfamilies in the reporter assay. By comparing the epigenetic context of the TEs, we found that especially colorectal cancer has specific highly active TE subfamilies with signatures of canonical active enhancers. We also used an in silico model to predict TE enhancer to gene contacts and found that these subfamilies regulated genes that were frequently overexpressed. Thus, we present the widespread functional activity of TE enhancers in cancers, providing evidence for further functional validation of TEs and their effects on transcriptional programs and especially dysregulation of gene expression in cancer.

Induced pluripotent stem cells (iPSCs) can be derived from somatic cells by transgenically expressing the four transcription factors OCT4, SOX2, KLF4, and C-MYC. This technology has revolutionised the stem cell field, yet cellular reprogramming is still inefficient and slow. To become fully applicable in regenerative medicine, the robust generation of safe and high-quality iPSCs from patient samples is essential. Various methods and potent reprogramming factors have been described to date. Yet, none have been able to circumvent these limitations markedly. The recently published activator-mediated approach (CRISPRa) is considered to be more physiological compared to the forced transgenic expression as the cell’s own genes are activated. Here, guide RNAs (gRNAs) mediate sequence-specific recruitment of non-cutting Cas9 (dCas9) activator proteins to the promoter region. Unlike other methods, it holds great multiplexing capacity and can also target enhancer and non-coding sequences. CRISPRa reprogramming still needs to be optimised since its efficiency is low. Thus, we aimed at enhancing this aspect and the temporal kinetics by targeting the micro RNA (miRNA) clusters 302/367 and miR-371-373, which both have been described as powerful cell fate regulators. We demonstrate successful reprogramming by targeting the miR-302/367 promoter alongside OCT4, SOX2, KLF4, C-MYC, LIN28A, REX1, NANOG, and EEA-motifs with CRISPRa. Activating the miRNA cluster results in a 2.5 fold efficiency increase in human foreskin fibroblast (HFF) reprogramming compared to the published basal CRISPRa system, quantified by staining for alkaline phosphatase. In HFFs, the CRISPRa efficiency is now comparable to the commonly used transgenic approach. Aiming to clarify the molecular mechanisms of these results, we characterised the expression of direct and downstream targets of miR-302/367 at different time points throughout the reprogramming process. Furthermore, validated with immunocytochemical stainings, the generated bona fide iPSCs express pluripotency markers and spontaneously differentiate into the three germ-layers, both signs of high-quality iPSCs. Beyond that, we report that miR-302/367 activation appears to result in earlier iPSC colony formation resulting in faster proliferating stem cell colonies shown with live-cell imaging. Employing a conditionally stabilised activator construct, we further show that with miR-302/367 targeting, the dCas9 activator expression seems to be required for only a short time period, sufficient to induce pluripotency. At the end of the project, the miR-302/367 cluster targeting was optimised and the best-working gRNAs were selected for further studies, which when combined further increase the CRISPRa-induced expression of the miR-302/367 cluster markedly. All in all, this study demonstrates that non-coding genetic elements like the miR-302/367 cluster can be targeted with CRISPRa, and its targeting significantly improves the reprogramming efficiency. Implications of the study for regenerative medicine and future steps are discussed.

Autophagy is an essential pathway that evolved to sustain cellular integrity by removing damaged and aged organelles. During this process, our cells sense, encapsulate and deliver defective cellular components to the lysosome for destruction. Over the past decade, many laboratories have demonstrated that damaged mitochondria can be selectively eliminated, during a process known as "mitophagy". Mitophagy senses, targets, and engulfs defective mitochondria for elimination via lysosomal hydrolysis. The identification of factors that promote or prevent mitophagy has high therapeutic relevance, particularly those that alter PINK1/Parkin-independent mitophagy. Recent research in the McWilliams lab uncovered a novel role for lipid metabolism in the regulation of PINK1/Parkin-independent mitophagy. Briefly, the team discovered that DGAT1-dependent lipid droplet (LD) biosynthesis occurred several hours upstream of mitochondrial clearance, with LDs accumulation upon iron chelation. LDs accumulate in a DGAT1-dependent fashion as mitochondria are eliminated. Pharmacological or genetic inhibition of DGAT1, restricts mitophagy levels in vitro and in vivo. However, the mechanism that linked defective lipid metabolism to reduced mitophagy remained mysterious. We hypothesized that defective lipid signalling may compromise lysosomal activity leading to reduced levels of mitophagy. Accordingly, my project examined the functional contribution of DGAT-dependent LD biogenesis to lysosomal homeostasis in the context of PINK1/Parkin-independent mitophagy. After first verifying the DGAT1-dependent nature of LD accumulation in human cells, I established assays to investigate lysosomal homeostasis in the context of iron chelation-induced mitophagy. Using a variety of labelling approaches, live cell imaging experiments revealed a significant displacement of endolysosomes upon DGAT1/2 inhibition, in addition to possible alterations in lysosomal dynamics. My data suggest that loss of DGAT1 activity impairs lysosomal homeostasis when iron levels are low. This likely explains the mitophagy impairments and might account for additional phenotypes of impaired cell viability upon DGAT1 inhibition. Changes in lysosomal acidity were inconclusive, indicating further timepoints may need to be analysed to detect transient impairments in hydrolysis. My results emphasize the importance of organelle crosstalk in mitophagy and the emerging role of LDs in cellular integrity. These data further highlight that targeting lipid metabolism may provide a means to sustain efficient mitochondrial turnover.