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

The early life gut microbiota plays a major role in establishing neonatal immunity and child’s long-term health. However, relatively little is still known about the role of individual bacteria as most studies so far have focused on characterizing the diversity and the individual and temporal variations of the infant gut microbiome. The genus Bacteroides is of particular interest since its abundance is remarkably decreased in infants born via C-section, and relatively little is known about the genomic and phenotypic characteristics of early Bacteroides colonizers despite their anticipated role in the increased morbidity following C-section birth. This thesis aims to contribute to the isolation and characterization of Bacteroides strains from infant and mother stool samples from the Health and Early Life Microbiota (HELMi) cohort study using culture-based and metagenomic approaches. Gram-negative bacteria were isolated from stool samples of 9-week-old infants and identified by Sanger sequencing. In total, seven isolates identified as unique species of Bacteroides, isolated from infant samples or previously from mother samples in late pregnancy, were then characterized for their potential to activate innate immunity in vitro by using HEK-Blue™ hTLR2-hTLR6 reporter cells either as live cells or filtered culture media. Whole genome shotgun sequenced stool metagenomes obtained from 88 infants during the first year of life were leveraged as well. A computational pipeline able to scale to the large size of the dataset was developed to obtain metagenome assembled genomes (MAGs) from the metagenomes. MAGs obtained from Bacteroides species were further taxonomically and functionally annotated. Among the seven Bacteroides spp. isolated from HELMi mother and infant samples, the majority were able to activate the TLR2/6 receptor in vitro. The isolates varied in their potential to activate the receptor via their cell surface molecules and substances they excreted to the culture media. In addition, over 2500 MAGs could be retrieved from the infant metagenomes, of which 18 belonged to Bacteroides spp. Based on predicted open reading frames, majority of the identified proteins of these MAGs were involved in housekeeping functions. Most of predicted proteins involved in cellular metabolism were, however, related to carbohydrate metabolism, amino acid metabolism, and glycan metabolism, stressing the role of Bacteroides spp. in the gut as important and versatile carbohydrate consumers. The results indicate that the Bacteroides spp. colonizing infant gut have an immunologically and metabolically active role. Further work is needed to characterize the molecules responsible for the TLR2/6 activation as well as the nature of the downstream immune responses elicited by the isolated Bacteroides spp.

Large granular lymphocytic (LGL) leukemia is a rare form of chronic lymphocytic leukemia, that is characterized by clonal expansion of mature cytotoxic T- or natural killer (NK)- cells. As the white cell count in patients is predominantly not distinguishably altered, it often goes underdiagnosed or is diagnosed accidentally. T lymphocytic LGL leukemia (T-LGLL), that makes up 85% of all LGL leukemia (LGLL) cases is characterized by a prolonged expansion of peripheral blood T-lymphocytes, mostly CD8+ lymphocytes. 40 % of T-LGLL patients harbor mutations in the Signal Transducer and Activator of Transcription 3 (STAT3) gene. Y640F mutation of STAT3 (STAT3 Y640F) is the most commonly occurring alteration, present in approximately 17% of all T-LGLL patients, and 42% of patients that bear a mutation in STAT3. Furthermore, a higher prevalence of rheumatoid arthritis (RA) can be observed in patients with mutated STAT3 (26% vs 6%, p=0.02). As T-LGLL patients with the Y640F mutation have a higher incidence of co-occurring RA, we aimed to understand the possible role CD8+ T-cell clones carrying somatic mutation of STAT3 may play in the autoimmune process. We applied lentiviral vectors to express STAT3 wild type (wt) and STAT3 Y640F in murine and human CD8+ T cells. We were able to show their successful integration into the host genome using droplet digital PCR (ddPCR). ddPCR showed high selectivity in its ability to differentiate between the hosts’ gDNA and virally inserted cDNA. The custom-designed probes showed high specificity for either STAT3 wt or STAT3 Y640F, proving the functionality of the assay. Sensitivity studies provided us with accurate quantification even with the presence of STAT3 wt or STAT3 Y640F cDNA under 1%, displaying successful detection of rare variants in low concentration samples. In our expression studies, using Flow cytometry and Western Blotting (WB), we detected a modest rise in STAT3 expression in the virally transduced CD8+ cells. We hypothesized that the CD8+ cells were successfully transduced, but unable to accommodate sufficient STAT3 expression.To determine the role of Y640F mutation in the migration of CD8+ lymphocytes in different tissues in vivo, we injected lentivirally transduced cells, mixed in a 1:1 ratio (wt:mut), into the mice. Unfortunately, our ddPCR method was not sensitive enough to reliably quantitate the transduced cells in the diverse tissue samples. Consequently, we decided that any further mouse experiments cannot be justified. In conclusion, we present successful integration of lentivirally expressed wt and mutant STAT3 in both human and mouse primary CD8+ T lymphocytes and human peripheral blood mononuclear cells. The successfully constructed and optimized ddPCR assay was not, however sensitive enough for in vivo quantification of the transduced cells. As the lentivirally mediated expression of STAT3 variants was low, new approaches and tools are needed to study the role of STAT3 mutated T cells in the pathogenesis of RA.

Breast cancer, which is the most common cancer among women, has a good overall survival rate with over 90% of the patients being alive after 5 years of the diagnosis, and 80% being alive after 10 years. Metastatic breast cancer with metastasis in the brain, however, is associated with a poor prognosis with less than 20% survival one year after the detection of a brain metastasis. Understanding the mechanisms behind brain metastasis formation is crucial for developing therapeutic drugs in the future. The Fatty Acid-Binding Protein 3 (FABP3) has been discovered to have both tumor suppressing and tumor promoting effects depending on the cancer type. This thesis elucidated the role of FABP3 in the formation of breast cancer brain metastases. The results of in vitro blood-brain barrier (BBB) studies show that breast cancer cells overexpressing FABP3 penetrate the in vitro BBB significantly better than the control breast cancer cells. Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) revealed further genes associated with the invasiveness of FABP3-overexpressing breast cancer cells. The results of the RT-qPCR combined with immunofluorescent staining suggest that a transcription factor plays a part in the formation of brain metastases. Further studies need to be conducted to fully understand the role of this transcription factor. The affinities of the FABP3-binding peptide CooP and its alanine scan-generated variants CooP-K and [Ala5]CooP to FABP3 has previously been studied through microscale thermophoresis. The abilities of the CooP peptide variants to pass through the in vitro BBB were studied in this thesis. The results may be of importance for the development of targeted therapies.

Ischemic stroke is a complex disease involving multiple pathophysiological mechanisms. To date, many therapeutic intervention strategies such as anti-inflammatory treatments have been tested, but none of them has been successful. Previous studies have shown that mesencephalic astrocyte-derived neurotrophic factor (MANF) improves stroke recovery and increases the expression of phagocytosis related genes. In this study, the phagocytic and inflammatory effect of monocyte chemoattractant protein 1 (MCP-1), macrophage colony-stimulating factor (M-CSF), complement component 3 (C3), adhesion G protein-coupled receptor E1 (ADGRE1), MER receptor tyrosine kinase (MerTK) and mesencephalic astrocyte-derived neurotrophic factor (MANF) on microglia were studied simultaneously for the first time. The phagocytosis related genes were transiently transfected into a microglial cell line and studied in vitro utilizing phagocytosis assay, fluorescence-activated cell sorting, Western blot and enzyme-linked immunosorbent assay. MCP-1, M-CSF and C3a were shown to enhance microglial phagocytosis without inducing a pro-inflammatory response. In addition, MerTK induces phagocytosis and the synthesis of pro-inflammatory cytokines. In conclusion, the real therapeutic potential of MCP-1, M-CSF, C3a and MerTK in stroke treatment should be further characterized and tested in vivo.

High-grade serous ovarian cancer (HGSC) is the most aggressive subtype of ovarian cancer. Most patients are diagnosed at an advanced stage, resulting in a poor prognosis. Although some targeted therapies have made good progress, immune checkpoint inhibitors (ICIs) still show limited efficacy in HGSC treatment. This is closely associated with the high heterogeneity and unique immunosuppressive tumor microenvironment (TME) of HGSC. As local immune centers in the TME, tertiary lymphoid structures (TLSs) contain diverse types of immune cells and play a crucial role in the anti-tumor immune process. Nonetheless, the spatial immune landscape of TLSs and key gene expression signatures affecting prognosis have not been fully elucidated. In recent years, spatially resolved gene expression profiling of clinical samples has provided insights into the transcriptome at high resolution. However, despite revealing a series of response mechanisms of anti-tumor immunity, these methods still have many limitations in non-single cell resolution, specific tissue sample types, and high cost. Here in this thesis, I established Nano-Pick, a novel method based on precise microtissue extraction and the nCounter analysis system. Nano-Pick enables cost-effective and rapid spatial gene expression profiling using formalin-fixed paraffin-embedded (FFPE) samples. All TLSs-enriched FFPE slides were obtained from patient-derived HGSC samples at Helsinki University Hospital (HUH). With the help of tissue cyclic immunofluorescence (t-CycIF), a highly multiplexed imaging method, selected regions of interests (ROIs) were annotated in different zones of TLSs. These ROIs were accurately transferred to adjacent tissue sections, guiding the target region selection for microtissue extraction. An improved immunofluorescence staining method was used to successfully label cancer cells, macrophages, and immune cells without antigen retrieval. With the optimized combination of parameters, efficient and consistent precise picking for different cell types was achieved. mRNA extracted from the collected microtissues was validated by qPCR for the range of pre-amplification cycles, confirming its suitability for nCounter gene panel analysis. Therefore, Nano-Pick offers the opportunity for higher-resolution spatial analysis and interactions of immune cell populations in the TME, especially in the TLSs. This low-cost, reliable and sensitive spatial gene expression profiling method has broad application prospects in many fields, with the potential to advance the development of precision medicine.

Ex vivo drug sensitivity testing is used widely in studies aiming at personalizing medicine for acute myeloid leukemia (AML) patients. However, different conditions, such as cytokines used in media and cryopreservation of cells, as well as varying readout methods can affect primary cell viability, cell composition and sensitivity results. Such affects have been previously studied in some AML treatments, however, not with flow cytometry or with venetoclax. In this thesis, we studied the responses of AML patients to venetoclax using ex vivo drug sensitivity testing with various settings. We first tested three media and two sensitivity readout methods on 29 fresh primary AML samples to determine the optimal media and method for determining ex vivo drug sensitivity. We then tested these same variables on 16 cryopreserved samples and compared these results to their fresh counterparts. Finally, we applied our platform to clinical use and tested its capability to predict in vivo responses to venetoclax in ten AML patients. Our platform was able to predict venetoclax responses in nine out of ten patients using condition media coupled with a flow cytometry-based method, determined as optimal in the first phase. Sensitivity results as well as cell composition obtained after cryopreservation differed from their fresh counterparts and, therefore, we conclude that cryopreserved samples should not be used in guiding treatment ex vivo. Our results give valuable information about sources of error associated with ex vivo drug sensitivity testing. Consideration of these results when designing preclinical studies will enhance their reliability and relevance. Ex vivo testing could be in the future implemented into clinical practice in guiding treatment, saving society and patients from costs and unnecessary adverse effects.

Colorectal cancer (CRC) is the second deadliest cancer in the world. Given the heterogeneity of the disease, a substantial percentage of patients do not benefit from the standard-of-care. The ability to identify patients that most or least likely to respond to the generic therapy prior to its implementation would improve the safety and efficacy of the anti-cancer regimen in CRC.The goal of this study was to assess the suitability of patient-derived organoids (PDOs) as in vitro models to evaluate the efficacy of chemotherapy in CRC. PDOs were generated from surgical tumor specimens of rectal cancer patients. Next, PDOs were treated with standard of care chemotherapeutics for rectal cancer, with or without neoadjuvant radiotherapy, and a commercial cell viability assay was used to assess drug response. PDOs were classified according to the consensus molecular subtype (CMS) system, based on the whole transcriptome sequencing of PDO-derived RNA before treatment. Clinical information was retrieved from the patient records. In vitro drug responses of PDOs revealed heterogeneous drug sensitivity profiles and highlighted patients who would benefit from standard of care. With respect to the consensus molecular subtype (CMS) classification, CMS2 organoids showed highest response to standard of care, while CMS1 PDOs exhibited a less responsive phenotype. For the majority of the PDOs, neoadjuvant radiotherapy prior to drug treatment had no effect on drug response. On the other hand, in certain cases, neoadjuvant therapy sensitized or desensitized PDOs to standard of care chemotherapeutics. This study adds to the literature demonstrating the feasibility of PDOs as platforms for modelling cancer treatment and highlighting their potential to facilitate progress in personalized medicine. More studies involving complex, co-culture PDO models and designed to better reflect the relevant interplay between tumor microenvironment and the anti-cancer regimen are needed to confirm the predictive qualities of the PDOs and inform clinical decisions in CRC.

Lung cancer remains the leading cause of cancer death worldwide. Cancer immunotherapies have changed the treatment path in some cancers and even led to favorable clinical outcomes in previously incurable cancer types. However, only a fraction of patients benefit from the current immunotherapies. Even though immuno-oncology has great potential, it is facing several challenges including the lack of biomarkers, unknown mechanisms of therapy resistance, complexity of the tumor-immune interactions, and involvement of the complex tumor microenvironment that significantly affects therapeutic efficacy. It remains a great challenge to predict which patients will benefit from immunotherapies, and current immunotherapies are deemed expensive when compared to the more traditional therapeutic modalities. In this work, we aimed to develop platform to study responses to immunotherapy ex vivo in a personalized manner. The platform could enable the study of immune effects and T cell mediated tumor killing in the absence and presence of immunotherapy and other selected drugs. We demonstrate the utility of our ex vivo platform that has potential for personalized drug testing.

The infant gut microbiota maturation is central to infant health and well-being and has been suggested to have important health impacts in adulthood. While most of the previous research and description of the dynamics of the gut microbiota acquisition and maturation focused on the prokaryotic community; recent studies have suggested the importance of other microbial members in the community. Viruses, specifically bacteriophages (phages), are found in high abundances in the gut microbiota, and may influence prokaryotic composition and the microbiota’s trajectory during infancy. Phages can shape the bacterial community by killing their cellular host but also by modulating their bacterial host fitness in the ecosystem. Additionally, phages can carry genes which have no role in the viral replication machinery, but instead modulate the cellular host metabolism. These genes, termed auxiliary metabolic genes (AMGs), are largely uncharacterized in the human gut, in particular in the context of the infant gut microbiota maturation. In this thesis, we explored the diversity and persistence of AMGs in infant gut microbiota from 3 months to 2 years old and identified and characterized changes in the AMGs repertoire during the infant gut microbiota maturation. This project utilized a subset of the broader Finnish Health and Early Life Microbiota (HELMi) birth cohort study, a Finnish prospective cohort on early life microbiota and health. In this study, we leveraged whole genome shotgun metagenomes from faecal samples of 475 infants collected at four time points, as well as samples from their mothers (n = 304) and fathers (n = 123). The viral sequences in these metagenomes were then identified, annotated, and characterized, allowing us to build a large catalogue of bacteriophage sequences, called the HELMi Bacteriophage Catalogue (HBaC). Next, we assessed the presence of putative AMGs in this phage collection and determined their prevalence in the viral community and dynamics during the gut microbiota maturation. The HBaC contains 145,818 unique species-like viral OTUs (vOTUs) of which a majority are temperate phages, classified as Caudoviricetes. Notably, we observed an increase in phage diversity and richness during the infant gut microbiota maturation and an increase in relative abundance of virulent phages. Further, the viral community composition strongly associated with the observed prokaryotic faecal community types (FCTs). In our phage catalogue, 8 % of the vOTUs carried at least one putative AMGs and the Demerecviridae, Autographiviridae and Herelleviridae families proportionally carried the most. The most common metabolic pathways encoded by these putative AMGs found in HBaC were organic nitrogen metabolism, transport functions and carbon utilization. Interestingly, we observed a change in putative AMGs abundance and function during the infant gut microbiota maturation. In the future, additional analysis exploring the influence of early life exposures on phage and AMGs prevalence and dynamics could help unravel the complex interaction between phages and their bacterial host in the dynamic infant gut microbiota.

The prevalence of major depressive disorder is increasing despite the increased standard of living. The prevailing hypothesis to explain depression is that there is an unbalance in information processing in relevant brain networks. Antidepressants (SSRIs, SNRIs) have been shown to induce a juvenile-like plasticity state (iPlasticity) in the brain that helps in rewiring the affected neuronal networks when combined with beneficial environmental stimuli (e.g. psychotherapy). However, it takes weeks to see the beneficial effects of conventional antidepressants on mood and they bring relief only to approximately two-thirds of the patients. There is an urgent need for more efficient and rapid-acting antidepressants. Preliminary data suggests that psychedelics may have potential to respond to this need. It is thought that the therapeutic effect of psychedelics rises from the molecular effects leading to structural and functional plasticity and behavioral changes. Molecular effects of psychedelics are believed to arise from the activation of serotonin 2A (5-HT2A) receptors. It is well established that serotonin 2A (5-HT2A) receptor activation lies behind the hallucinogenic effects of psychedelics, but its role in drug-induced plasticity is currently under debate. Signaling of brain-derived neurotrophic factor (BDNF) through its receptor TrkB has been proposed to underlie the plasticity-promoting effects of psychedelics. However, the mechanisms leading to increased BDNF/TrkB signaling after psychedelic administration are poorly understood. This thesis aimed to study the molecular mechanisms associated with psychedelic-induced plasticity in cortical neuronal cultures. The timeline of the effects of LSD was studied by analyzing the phosphorylation of neurotrophic signaling markers downstream of TrkB (mTOR and ERK) in primary neuronal cultures using Western blot. The role of the 5-HT2A receptor was assessed by combining 5-HT2A antagonist M100907 pretreatment with LSD treatment, followed by Western blot analyses of the same signaling markers mTOR and ERK. The degree of molecular effects of psychedelics was compared to the effects of classical antidepressant fluoxetine. Protein-fragment complementation assay (PCA) was used to evaluate the dimerization of the TrkB receptor in the presence of psychedelics and classical antidepressants. In this context, Western blot was also used to assess the phosphorylation of the plasticity-related BDNF signaling markers ERK and two tyrosines of TrkB receptor (Y515 and Y816) that mediate recruitment of neurotrophic signaling pathways. We found that psychedelic treatment promoted phosphorylation of mTOR and ERK significantly. These effects were not affected by pretreatment with M100907, indicating activation of BDNF/TrkB signaling by psychedelics is independent from 5-HT2A activation. Psychedelics were also shown to cause a significant increase in dimerization of TrkB whereas increase caused by fluoxetine was not significant. Lastly, psychedelics were shown to cause increase in phosphorylation of TrkB and ERK that were comparable to those induced by fluoxetine. These results highlight the potential of psychedelics to promote BDNF-mediated neurotrophic signaling associated with juvenile-like plasticity. Interestingly, the results show recruitment of BDNF/TrkB downstream signaling independently from 5-HT2A activation, which suggests that plasticity-promoting effects of psychedelics might be detached from their hallucinogenic effects.