Browsing by master's degree program "Magisterprogrammet i mikrobiologi och mikrobibioteknik"

Human norovirus (HuNoV) is the leading cause of foodborne illness globally. Especially minimally processed goods such as berries and shellfish are common sources of HuNoV outbreaks. High pressure processing (HPP) is a relatively novel food processing technology that can both inactivate foodborne pathogens and extend the shelf-life of food items in cold temperatures. HPP is especially suitable for fresh purees, juices and sauces. In this study, we used murine norovirus strain MNV-1 to model the inactivation of HuNoV by HPP in three berry puree matrices and phosphate-buffered saline (PBS). We assessed the effect of HPP by cell-based TCID50 infectivity assay and real-time reverse transcription quantitative PCR (RT-qPCR) with RNase and porcine gastric mucin (PGM) binding assay. We strived to find if there were differences between distinct pressures (4500 and 6000 bars) and hold times (3, 6, and 9 minutes) to efficiently inactivate MNV-1 in berry purees. We observed that the matrix type affected the survival of MNV-1 significantly both during HPP and transportation. During transportation, MNV-1 survived better in PBS than in berry purees. MNV-1 was efficiently inactivated in PBS leading to >3-log10 reductions in the number of infectious particles (TCID50/ml) at both 4500 and 6000 bars. In berry puree matrices, MNV-1 was most efficiently inactivated in blackcurrant puree resulting in ≈3-log10 reductions in genome equivalents. The efficacy of pressures and hold-times could not be differentiated in any of the used matrices. MNV-1 in raspberry puree showed no infectivity in RAW 264.7 cells but displayed ≈2-log10 reductions in genome equivalents. MNV-1 in strawberry puree displayed <1-log10 reductions in RAW 264.7 cells. Our results imply that PGM binding assay and RNase as pre-RT-qPCR treatments have problems in selecting infectious MNV-1 particles for amplification. Hence when using these pre-treatments, concluding on MNV-1 infectivity should be done cautiously.

Antibiotic resistance is a growing threat to global health due to overuse and misuse of antibiotics leading to untreatable or difficult to treat infections. Natural environments are an important reservoir of antibiotic resistance. The release of antibiotics into the environment promotes the development of antibiotic resistant bacteria and environmental occurrence of antibiotic resistance genes (ARGs). ARGs are common in nature and prevalent in aquatic environments such as surface waters and effluent. Cyanobacteria are widely found in marine, freshwater, and terrestrial environments. Since their ubiquitous presence in water environments cyanobacteria are exposed to antibiotic pollution and are in contact with resistant bacteria. The role of cyanobacteria in the antimicrobial resistome and dissemination of ARGs has only been studied recently. This work aimed to evaluate the antibiotic susceptibilities of 51 cyanobacterial strains against different classes of antibiotics, using liquid batch cultures, antibiotic discs, and bioinformatics approaches. Cyanobacterial strains used in this work were sensitive to most of the tested antibiotics. However, majority of the strains also showed resistance against trimethoprim and novobiocin. Overall, there was little variation in the antibiotic resistances observed between strains but differences in sensitivity to different antibiotics was observed between species and strains with most differences seen with Nostoc spp. According to bioinformatic tools used (CARD database and BLASTp) FosA protein was found only in strains showing resistance against fosfomycin but not in any sensitive phenotypes and therefore fosA gene was selected as the most promising putative resistance gene for subsequent assays. To determine whether the fosA from cyanobacteria could confer resistance to fosfomycin, the fosA gene from Nostoc sp. XPORK 5A was cloned into pET28a(+) expression vector under the control of T7 promoter and subsequently native cyanobacterial promoter. The ability of Escherichia coli BL21 (DE3) carrying each plasmid constructs to grow in the presence of fosfomycin was determined with agar plates and growth curve assay. E. coli transformants containing the fosA gene and T7 promoter conferred high-level resistance to fosfomycin showing ability to grow at the highest concentrations tested (1mg/ml) on agar plates and (500 µg/ml) in growth curve assay. FosA protein expression from the native cyanobacterial promoter appeared to be weaker and conferred lower-level resistance to fosfomycin (≥ 10 µg/ml). The results of this study provide more information about the antibiotic susceptibility of cyanobacteria. In addition, replicating a horizontal transfer of the fosA gene from cyanobacteria to proteobacteria conferred resistance to fosfomycin and these results may indicate that also nonpathogenic cyanobacteria could act as a source of fosA antibiotic resistance genes.

The COVID-19 pandemic of 2019 has had a huge impact on the hospitality industry, decreasing production by 35.4% in Q4 of 2020. To keep the industry functional, new safety solutions have to be studied and developed for mitigation of the pandemic. In this study, airborne transmission of viruses in an indoor space was studied, and air purifiers and space dividers were tested as potential intervention methods against SARS-CoV-2 by using a non-pathogenic model virus phi 6. Filtered air purifiers were found to work as a possible solution for the mitigation of viruses spreading through aerosols in public spaces such as restaurants, however, the positioning of the devices is crucial, as the air flow to them may increase the concentration of viruses locally. Space dividers were found to increase the possibility of infection via aerosols. Other types of air purifiers were also tested: an ionizer prototype and a hydroxyl radical emitting unit, of which the ionizer prototype proved to be efficient in reducing the virus concentrations in the air. Most importantly, it was confirmed that enveloped viruses resembling coronaviruses are capable of spreading via aerosol transmission indoors.

OBJECTIVES and RESEARCH QUESTION. Human parechovirus 3 (HPeV3) is a (+)ssRNA icosahaedrally symmetric virus which causes meningoencephalitis and sepsis in children and neonates. As it causes the most severe symptoms among parechoviruses it is attracting more attention (4). Currently there are no approved broad treatment strategies against parechoviruses, however recent research by Rhoden et al., 2017, reported the antiviral activity of posaconazole (PSZ) against HPeV3 in cell culture. Posaconazole is an antifungal drug approved for use against Candida and Aspergillus infections. It targets lanosterol-14alpha-demethylase and prevents the production of ergosterol, a lipid vital for fungal membranes not present in mammalian cells (24). In mammalian cells PSZ accumulates at the endoplasmic reticulum (ER) and binds to the oxysterol-binding protein (OSBP) and Niemann-pick type C1 (NPC1) (59, 28, 30). The drug may affect cellular components and thusly block parechoviral infection or could bind to the viral capsid. METHODS. To test viral capsid-binding hypothesis PSZ activity was tested in a range of concentrations against two HPeV3 isolates and HPeV1 Harris in Vero and HT29 cell lines. HPeV3 isolate 152037 was purified on a CsCl step gradient and imaged by cryo electron microscopy (cryo-EM). Single particle analysis was done in Scipion (40) and acquired density maps visualized in UCSF Chimera (49). Atomic model of a different isolate of HPeV3 (PDB ID: 6GV4, 16) was changed at 6 sites and fitted to density maps from this work in Coot (52). Maps were subtracted in search of density that would represent PSZ. RESULTS. PSZ was effective against both HPeV3 isolates at 1 μM in Vero cells when added to the virus prior to infection, however not in HT29 cells. At higher concentrations (>10 μM) PSZ formed crystals which limited the concentration that can be used for cryo-EM. In order to test the hypothesis of PSZ being a capsid binder 3 datasets were collected, HPeV3 control, HPeV3+DMSO and HPeV3+PSZ (4 μM) with final resolutions after single particle analysis of 3.3 Å, 3.9 Å and 3.4 Å respectively. Subtraction of maps yielded no difference that would represent PSZ. DISCUSSION and CONCLUSION. PSZ does not appear to be a capsid binder although it appears to work early in the infection. Absence of PSZ density in HPeV3+PSZ density map could be due to low saturation and images containing PSZ were filtered out in image processing. Another possibility is low affinity of PSZ for the capsid. As PSZ binds various membranes it is possible that it blocks HPeV3 infection by targeting cell components. Additional experiments could be performed in the future in order to provide insight into which stages of infection PSZ affects.

Fecal microbiota transplantation (FMT) is used to treat recurrent Clostridioides difficile infection (rCDI), and its potential as a treatment for other inflammatory conditions, like inflammatory bowel diseases (IBD), or irritable bowel syndrome (IBS), has been extensively studied lately. It has been noticed that some bacteria in fecal transplants do not require physical contact with intestinal epithelium to alleviate inflammation, and extracellular vesicles (EVs) have been proposed to carry the anti-inflammatory properties of those beneficial bacteria. In this thesis project, an isolation protocol was set up to isolate EVs from two fecal-originated Bacteroides isolates, Bacteroides ovatus and Bacteroides vulgatus, which had shown anti-inflammatory potential in previous studies. Isolation of EVs succeeded, and both isolates were confirmed to produce EVs. To study the anti-inflammatory potential, human colon epithelial cells (HT-29) were treated with several dilutions of isolated EVs, and then challenged with lipopolysaccharide (LPS) to induce inflammation. Amount of produced interleukin (IL-) 8 was measured as a marker of inflammation. EVs of both Bacteroides isolates continuously showed anti-inflammatory potential, but statistically significant conclusions could not be made. EVs have a potential to be used as a treatment in different inflammatory conditions and as adjuvant factors in synthetic FMT. To study the immunomodulatory potential of EVs of Bacteroides species more, proteomic analysis of contents of EVs, as well as potential to improve intestinal barrier are suggested. Also, testing the ability to alleviate production of other inflammatory markers could reveal more anti-inflammatory potential.

Brewers’ spent grains (BSG) are by-products of the brewing industry. Utilization of BSG in food applications is challenging, due to its poor technological characteristics. Because of their water retaining properties, interactions with matrix components and impact on texture formation, bacterial exopolysaccharides (EPS) represent a promising tool for improvement of BSG properties. Among bacterial exopolysaccharides, dextran produced in situ by lactic acid bacteria (LAB) during fermentation has shown major improvements in technological and sensorial features of products prepared from various types of plant materials. The nutritious composition of BSG may support the growth of LAB and enable in situ dextran production. The aim of this study was to establish and examine the synthesis of dextran by LAB in BSG. Sixteen dextran producing LAB strains were screened for viscosity formation in BSG fermentation. The strains showing the highest viscosity formation were further assessed for fermentation performance. The more suitable fermentation temperature was traced by comparing the viscosifying performance of selected starters at 20 and 25 °C. Dextran amount was determined semi-quantitatively from selected fermented samples showing optimal results, and the presence of oligosaccharides was assessed. Sucrose, glucose, maltose and fructose amounts were analyzed to observe the relation between sugar consumption and dextran and oligosaccharides formation. Weissella confusa strains A16 and 2LABPTO5 and Leuconostoc pseudomesenteroides strain DSM20193 appeared the most promising starters for viscosity formation and thus dextran synthesis in this matrix. From the examined fermentation temperatures, strains showed the highest potential for dextran synthesis at 25 °C. The amount of synthesized dextran ranged from 1.1 to 1.7 % w/w (of the wet weight of the whole sample matrix). The rheological properties of BSG were modified via LAB fermentation and dextran synthesis, resulting in more viscous texture, and its applicability in food systems was thus potentially enhanced.

The objective of the study was to demonstrate proof-of-concept for ResistApp – a newly developed digital platform for antibiotic resistance monitoring in hospital wastewater. ResistApp combines culture-independent, high throughput gene quantification with automated data analysis to synthesise and visualise monitoring data in an interactive dashboard. To do this, wastewater of two hospitals in Helsinki, Finland (HUS1 and HUS2) were monitored for over nine weeks (weeks 25-33 in 2020) for a total of 216 antibiotic resistance genes (ARGs), mobile genetic elements (MGEs), integrons, and taxonomy of bacteria, including bacteria causing hospital acquired infections, and the 16S rRNA gene using high-throughput quantitative PCR. The data from HT-qPCR was analysed and visualised using ResistApp. A higher number of ARGs and MGEs were detected at both hospitals in weeks 27-30 compared to other sampling weeks, with weeks 27-30 grouped separately from other sampling weeks by non-metric multidimensional scaling (NMDS)-ordination analysis. The NMDS ordination also indicated that the two hospitals, which use different amounts of antibiotics, had distinct resistance profiles. The study found that blaGES was the most abundant and prevalent carbapenem resistance gene in both hospitals throughout the sampling period. Low abundances of HAI-bacteria were detected in both hospitals. A correlation analysis was done, which revealed a positive association between blaGES and MGEs in both hospitals. Moreover, substantially more positive associations between carbapenem resistance genes and MGEs were found in HUS1 than HUS2, as well as a strong positive association between blaKPC and Klebsiella pneumoniae in the wastewater of HUS1. Wastewater monitoring with high-throughput qPCR is a promising tool for wastewater-based epidemiology, and it has been successfully used for the surveillance of SARS-CoV-2 -virus. Routine monitoring using ResistApp can capture both the impact of antibiotic use on resistance profiles and the dynamics of these profiles in hospital wastewater. In addition, ResistApp can simplify the analysis of HT-qPCR data considerably, compared to processing large amounts of raw data by hand.

Brewer's yeasts metabolize sugars and produce ethanol and CO2. This study aimed to investigate the relation between the assimilation of sugars in all-malt wort and isotopic signature of carbon and oxygen in the evolved CO2 from brewery fermentations. The isotopic composition of CO2 was measured by a tunable diode laser absorption spectrometer. The isotopic data obtained with automatic sampling, on-line, and in real-time. Wort samples were collected with 3 h intervals to quantify the residual sugars by high-performance liquid chromatography. Patterns of changes in δ13C and δ18O values were unique to experiments with each yeast type. The common overall ascending trend in δ13C and δ18O values in all experiments can be described by kinetic fractionation of isotopes, which explains that in a bioreaction the lighter isotopes participate in the reaction more readily than the heavier ones. Therefore, the early emergence of light isotopologues of CO2 may be a consequence of the fermentation of light isotopologues of sugars. A sudden decrease and then increase in delta values were observed in all experiments before the residual concentrations of glucose and fructose reach their lowest levels. This can be an indicator of the complete assimilation of monosaccharides by yeast. In the fermentations that yeast was able to consume maltose, δ13C and δ18O values raised considerably in a short period. Concurrently, maltose approached its eventual residual concentrations indicating an endpoint for its utilization by yeast. Our results confirm the hypothesis of a connection between the assimilation of sugars and the isotopic signature of evolved CO2 during brewery fermentations. The findings support the potential of off-gas isotopic analysis to monitor sugar assimilation in brewery fermentations.

Rift Valley fever virus (RVFV) is a mosquito-borne virus of the order Bunyavirales with a tripartite (-)ssRNA genome. It infects humans and cattle, causing a febrile disease with symptoms ranging from mild to severe. Safe and efficient human vaccines have not yet been developed, which underlines the importance of gaining a clear understanding of the viral antigenic surface. One significant challenge for RVFV research is posed by the costly and time-consuming biosafety precautions warranted by the pathogenicity of the virus. The surface of RVFV, formed by the two viral envelope glycoproteins Gn and Gc, could also be studied with a non-pathogenic model, such as a virus-like particle (VLP). A VLP is a macromolecular complex that resembles the virus, especially with respect to its outer structure, but lacks the viral genome. In this work, RVFV VLPs were produced by transient transfection of mammalian cells with genes encoding RVFV glycoproteins Gn and Gc. The objective was to design an optimized production and purification pipeline for RVFV VLPs to elucidate their structure by cryo-electron microscopy. To optimize the VLP production and purification, the effect of sample harvest times and DNA-to-cell ratios of transfection on RVFV glycoprotein expression was examined. Several methods were tested for VLP sample concentration and purification. VLPs were successfully detected in the purified samples by immuno-electron microscopy. Despite some challenges related to sample purity and scarcity of VLPs in samples, which prevented analyses by cryo-electron microscopy, the expression system described in this thesis has great potential to streamline RVFV VLP sample preparation for electron microscopy and to accelerate vital research into the structural properties of this emerging pathogen.

Rhinovirus infections cause the common cold, for which there is no cure. The consequences of the disease are mainly seen as absence from school or work, as the disease is usually mild, self-limiting and symptoms disappear in a few days, but infections can also lead to severe asthma exacerbations and hospitalization. This research is based on earlier findings of benzene sulfonamide derivative compounds that inhibited a broad range of enteroviruses by binding to a newly identified capsid pocket. Here, four promising compounds were tested against four different rhinoviruses belonging to Rhinovirus A and B species in a cell-based inhibition assay. After identification of a potential inhibitory compound against rhinovirus A9 and rhinovirus B14, a rhinovirus A9 homology model was generated, and then the binding was predicted with computational analysis using the rhinovirus A9 homology model and the published structure for rhinovirus B14. As the same compound has previously been shown to bind and inhibit coxsackievirus B4, this virus was used a control. Thermal stability testing of rhinovirus A9 revealed that the virus survives heat treatment up to at least 58°C, in contrast to previously published results. An efficient rhinovirus A9 purification protocol was established and high- quality cryo electron microscopy data were collected for structure determination of rhinovirus A9, which resulted in a 3.1 Å resolution map. The map was used for building the first atomic model of rhinovirus A9. The atomic model revealed that rhinovirus A9 has the expected T=1, pseudo T=3 capsid structure composed primarily of the three β-barrel proteins VP1, VP2 and VP3, with VP4 between the inside of the capsid and the RNA genome. The model confirmed the presence of the interprotomer pocket and VP1 pocket, although density for lipid factor was not detected. The final atomic model was compared to the homology model and used for comparative docking of the inhibitory compound to the pocket. The results revealed that empirical structure determination is still more accurate for large macromolecule complexes than modelling and that empirical data of the binding is required for reliable computational work.