The Significance of Wild Plants in the Evolutionary Ecology of Three Major Viruses Infecting Cultivated Sweetpotato in Uganda

The studies presented in this thesis contribute to the understanding of evolutionary ecology of three major viruses threatening cultivated sweetpotato (Ipomoea batatas Lam) in East Africa: Sweet potato feathery mottle virus (SPFMV; genus Potyvirus; Potyviridae), Sweet potato chlorotic stunt virus (SPCSV; genus Crinivirus; Closteroviridae) and Sweet potato mild mottle virus (SPMMV; genus Ipomovirus; Potyviridae). The viruses were serologically detected and the positive results confirmed by RT-PCR and sequencing. SPFMV was detected in 24 wild plant species of family Convolvulacea (genera Ipomoea, Lepistemon and Hewittia), of which 19 species were new natural hosts for SPFMV. SPMMV and SPCSV were detected in wild plants belonging to 21 and 12 species (genera Ipomoea, Lepistemon and Hewittia), respectively, all of which were previously unknown to be natural hosts of these viruses. SPFMV was the most abundant virus being detected in 17% of the plants, while SPMMV and SPCSV were detected in 9.8% and 5.4% of the assessed plants, respectively. Wild plants in Uganda were infected with the East African (EA), common (C), and the ordinary (O) strains, or co-infected with the EA and the C strain of SPFMV. The viruses and virus-like diseases were more frequent in the eastern agro-ecological zone than the western and central zones, which contrasted with known incidences of these viruses in sweetpotato crops, except for northern zone where incidences were lowest in wild plants as in sweetpotato. The NIb/CP junction in SPMMV was determined experimentally which facilitated CP-based phylogenetic and evolutionary analyses of SPMMV. Isolates of all the three viruses from wild plants were genetically similar to those found in cultivated sweetpotatoes in East Africa. There was no evidence of host-driven population genetic structures suggesting frequent transmission of these viruses between their wild and cultivated hosts. The p22 RNA silencing suppressor-encoding sequence was absent in a few SPCSV isolates, but regardless of this, SPCSV isolates incited sweet potato virus disease (SPVD) in sweetpotato plants co-infected with SPFMV, indicating that p22 is redundant for synergism between SCSV and SPFMV. Molecular evolutionary analysis revealed that isolates of strain EA of SPFMV that is largely restricted geographically in East Africa experience frequent recombination in comparison to isolates of strain C that is globally distributed. Moreover, non-homologous recombination events between strains EA and C were rare, despite frequent co-infections of these strains in wild plants, suggesting purifying selection against non-homologous recombinants between these strains or that such recombinants are mostly not infectious. Recombination was detected also in the 5 - and 3 -proximal regions of the SPMMV genome providing the first evidence of recombination in genus Ipomovirus, but no recombination events were detected in the characterized genomic regions of SPCSV. Strong purifying selection was implicated on evolution of majority of amino acids of the proteins encoded by the analyzed genomic regions of SPFMV, SPMMV and SPCSV. However, positive selection was predicted on 17 amino acids distributed over the whole the coat protein (CP) in the globally distributed strain C, as compared to only 4 amino acids in the multifunctional CP N-terminus (CP-NT) of strain EA largely restricted geographically to East Africa. A few amino acid sites in the N-terminus of SPMMV P1, the p7 protein and RNA silencing suppressor proteins p22 and RNase3 of SPCSV were also submitted to positive selection. Positively selected amino acids may constitute ligand-binding domains that determine interactions with plant host and/or insect vector factors. The P1 proteinase of SPMMV (genus Ipomovirus) seems to respond to needs of adaptation, which was not observed with the helper component proteinase (HC-Pro) of SPMMV, although the HC-Pro is responsible for many important molecular interactions in genus Potyvirus. Because the centre of origin of cultivated sweetpotato is in the Americas from where the crop was dispersed to other continents in recent history (except for the Australasia and South Pacific region), it would be expected that identical viruses and their strains occur worldwide, presuming virus dispersal with the host. Apparently, this seems not to be the case with SPMMV, the strain EA of SPFMV and the strain EA of SPCSV that are largely geographically confined in East Africa where they are predominant and occur both in natural and agro-ecosystems. The geographical distribution of plant viruses is constrained more by virus-vector relations than by virus-host interactions, which in accordance of the wide range of natural host species and the geographical confinement to East Africa suggest that these viruses existed in East African wild plants before the introduction of sweetpotato. Subsequently, these studies provide compelling evidence that East Africa constitutes a cradle of SPFMV strain EA, SPCSV strain EA, and SPMMV. Therefore, sweet potato virus disease (SPVD) in East Africa may be one of the examples of damaging virus diseases resulting from exchange of viruses between introduced crops and indigenous wild plant species. Keywords: Convolvulaceae, East Africa, epidemiology, evolution, genetic variability, Ipomoea, recombination, SPCSV, SPFMV, SPMMV, selection pressure, sweetpotato, wild plant species Author s Address: Arthur K. Tugume, Department of Agricultural Sciences, Faculty of Agriculture and Forestry, University of Helsinki, Latokartanonkaari 7, P.O Box 27, FIN-00014, Helsinki, Finland. Email: tugume.arthur@helsinki.fi Author s Present Address: Arthur K. Tugume, Department of Botany, Faculty of Science, Makerere University, P.O. Box 7062, Kampala, Uganda. Email: aktugume@botany.mak.ac.ug, tugumeka@yahoo.com

Dengue virus infection : Diagnostics and molecular epidemiology

Dengue is a mosquito-borne viral disease caused by the four dengue virus serotypes (DENV-1-4) and is currently considered as the most important arthropod-borne viral disease in the world. Nearly half of the human population lives in risk areas, and 50-100 million infections occur yearly according to World Health Organization. The disease can vary from a mild febrile disease to severe haemorrhagic fever and shock. A secondary infection with heterologous serotype increases the risk for severe disease outcome. During the last three decades the impact of dengue has dramatically increased in the endemic areas including the tropics and subtropics of the world. The current situation with massive epidemics of severe disease forms has been associated with socio-ecological changes that have increased the transmission and enabled the co-circulation of different serotypes. Consequently, an increase of dengue has also been observed in travelers visiting these areas. Currently approximately 30 cases are diagnosed yearly in Finnish travelers. In travelers dengue is rarely a life-threatening disease, however in the current study, a fatality was documented in a young Finnish patient who experienced a prolonged primary dengue infection. To improve particularly early laboratory diagnostics, a novel real-time RT-PCR method was developed for the detection of DENV-1-4 RNA based on TaqMan chemistry. The method was shown to be sensitive and specific for detecting DENV RNA and suitable for diagnostic use. The newly developed real-time RT-PCR was compared to other available early diagnostic methods including IgM and NS1 antigen detection using a panel of selected patient samples. The results suggest that the best diagnostic rates are achieved by a combination of IgM with RNA or NS1 detection. The dengue virus strains studied here included the first DENV strains isolated from serum samples of Finnish travelers collected in 2000-2005. The results of sequence analysis demonstrated that the 11 isolates included all four DENV serotypes and presented a global sample of DENV strains from different geographical areas including Asia, Africa and South America. In the present study sequence analysis was also carried out for a collection of 23 novel DENV-2 isolates from Venezuelan patients collected in 1999-2005. The Venezuelan DENV-2 exclusively represented the American-Asian genotype, suggesting that no foreign DENV-2 lineages have recently been introduced to the country. The results also suggest that the DENV-2 viruses detected earlier from Venezuela have been maintained in the area where they have evolved into several lineages. This is in contrast to the pattern observed in some other dengue endemic areas, where introductions of novel virus types and lineages are frequently detected.

Äidinmaito - terveysjuomaa ja normaalibakteereita

The chemical composition of breast milk has been studied in detail in the past decades. Hundreds of new antibacterial and antiviral components have been found. Several molecules have been found to promote the proper function of neonatal intestine. However, microbiological studies of breast milk have been, until recently, focused mainly on detecting harmful and pathogenic bacteria and viruses. Natural microbial diversity of human milk has not been widely studied before the work reported in this thesis. This is mainly because breast milk has traditionally been thought to be sterile - even if a certain amount of commensal bacteria have usually been detected in milk samples. The first part of this licentiate thesis contains a short literature review about the anatomy and physiology of breast feeding, human milk chemical and microbiological composition, mastitis, intestinal flora and bacteriocins. The second part reports on the experiments of the licentiate work, concentrating on the microbial diversity in the milk of healthy breast-feeding mothers, and the ability of these bacteria to produce antibacterial substances against pathogenic bacteria. The results indicate that human milk is a source of commensal bacteria for infant intestine. 509 random isolates from 40 breast milk samples were isolated and identified by 16S rRNA sequencing. Median bacterial count was about 600 colony forming units per milliliter. Over half of the isolates were staphylococci, and almost one third streptococci. The most common species were skin bacteria Staphylococcus epidermidis and oral bacteria Streptococcus salivarius and Streptococcus mitis. Lactic acid bacteria, identified as members of Lactobacillus-, Lactococcus- and Leuconostoc -genera, were found in five milk samples. Enterococci were found in three samples. A novel finding in this study is the capability of these commensal bacteria to inhibit the growth of pathogens. In 90 precent of the milk samples commensal bacteria inhibiting the growth of Staphylococcus aureus were found. In 40 precent of samples the colonies could block the growth completely. One fifth of the isolated Staph. epidermidis strains, half of Str. salivarius strains, and all lactic acid bacteria and enterococci could inhibit or block the growth of Staph. aureus. In further study also Listeria innocua- and Micrococcus luteus active isolates were found in 33 and 11 precent of milk samples (out of 140). Furthermore, two Lactococcus lactis isolates from the breast milk were shown to produce bacteriocin nisin, which is an antimicrobial molecule used as a food preservative. The importance of these human milk commensal bacteria in the development of newborn intestinal flora and immune system, as well as in preventing maternal breast infections, should be further explored.

Computational Methods for Detecting Large-Scale Chromosome Rearrangements in SNP Data

Large-scale chromosome rearrangements such as copy number variants (CNVs) and inversions encompass a considerable proportion of the genetic variation between human individuals. In a number of cases, they have been closely linked with various inheritable diseases. Single-nucleotide polymorphisms (SNPs) are another large part of the genetic variance between individuals. They are also typically abundant and their measuring is straightforward and cheap. This thesis presents computational means of using SNPs to detect the presence of inversions and deletions, a particular variety of CNVs. Technically, the inversion-detection algorithm detects the suppressed recombination rate between inverted and non-inverted haplotype populations whereas the deletion-detection algorithm uses the EM-algorithm to estimate the haplotype frequencies of a window with and without a deletion haplotype. As a contribution to population biology, a coalescent simulator for simulating inversion polymorphisms has been developed. Coalescent simulation is a backward-in-time method of modelling population ancestry. Technically, the simulator also models multiple crossovers by using the Counting model as the chiasma interference model. Finally, this thesis includes an experimental section. The aforementioned methods were tested on synthetic data to evaluate their power and specificity. They were also applied to the HapMap Phase II and Phase III data sets, yielding a number of candidates for previously unknown inversions, deletions and also correctly detecting known such rearrangements.

Evolutionary Genomics of Prokaryotic Viruses

Evolutionary history of biological entities is recorded within their nucleic acid sequences and can (sometimes) be deciphered by thorough genomic analysis. In this study we sought to gain insights into the diversity and evolution of bacterial and archaeal viruses. Our primary interest was pointed towards those virus groups/families for which comprehensive genomic analysis was not previously possible due to the lack of sufficient amount of genomic data. During the course of this work twenty-five putative proviruses integrated into various prokaryotic genomes were identified, enabling us to undertake a comparative genomics approach. This analysis allowed us to test the previously formulated evolutionary hypotheses and also provided valuable information on the molecular mechanisms behind the genome evolution of the studied virus groups.

Characterization of New Viruses from Hypersaline Environments

Viruses of Archaea are the least studied group of viruses. Fewer than 50 archaeal viruses have been reported which constitutes less than one percent of all the isolated prokaryotic viruses. Only about one third of the isolated archaeal viruses infect halophiles. The diversity of haloviruses, virus ecology in highly saline environments and the interactions of haloviruses with their hosts have been little studied. The exiguous knowledge available on halophilic systems is not only due to inadequate sampling but also reflects the extra challenge highly saline systems set on biochemical studies. In this study six new haloviruses were isolated and characterized. Viruses included four archaeal viruses and two bacteriophages. All of the other isolates exhibited head-tail morphology, except SH1 which was the first tailless icosahedral virus isolated from a high salt environment. Production and purification procedures were set up for all of these viruses and they were subjected to stability determinations. Archaeal virus SH1 was studied in more detail. Biochemical studies revealed an internal membrane underneath the protein capsid and a linear dsDNA genome. The overall structure of SH1 resembles phages PRD1, PM2 and Bam35 as well as an archaeal virus STIV. SH1 possesses about 15 structural proteins that form complexes under non-reducing conditions. Quantitative dissociation provided information about the positions of these proteins in the virion. The life cycle of SH1 was also studied. This lytic virus infects Haloarcula hispanica. Adsorption to the host cells is fairly inefficient and the life cycle rather long. Finally, virus responses in a variety of ionic conditions were studied. It was discovered that all of the studied viruses from low salt, marine and high salt environments tolerated larger range of salinities than their bacterial or archaeal hosts. The adsorption efficiency was not determined by the natural environment of a virus. Even though viruses with the slowest binding kinetics were among the haloviruses, fast binders were observed in viruses from all environments. When the salinity was altered, the virus adsorption responses were diverse. Four different behavioral patterns were observed: virus binding increased or decreased in increasing salinity, adsorption maximum was at a particular salt concentration or the salinity did not affect the binding. The way the virus binding was affected did not correlate with the environment, virus morphology or the organism the virus infects.

Microarrays in the Diagnosis of Human Herpesvirus infections

Currently, there are nine known human herpesviruses and these viruses appear to have been a very common companion of humans throughout the millenia. Of human herpesviruses, herpes simplex viruses 1 and 2 (HSV-1, HSV-2), causative agents of herpes labialis and genital herpes, and varicella-zoster virus (VZV), causative agent of chicken pox, are also common causes of central nervous system (CNS) infections. In addition, human cytomegalovirus (CMV), Epstein-Barr virus (EBV) and human herpesviruses 6A, 6B, and 7 (HHV-6A, HHV-6B, HHV-7), all members of the herpesvirus family, can also be associated with encephalitis and meningitis. Accurate diagnostics and fast treatment are essential for patient recovery in CNS infections and therefore sensitive and effective diagnostic methods are needed. The aim of this thesis was to develop new potential detection methods for diagnosing of human herpesvirus infections, especially in immunocompetent patients, using the microarray technique. Therefore, methods based on microarrays were developed for simultaneous detection of HSV-1, HSV-2, VZV, CMV, EBV, HHV-6A, HHV-6B, and HHV-7 nucleic acids, and for HSV-1, HSV-2, VZV, and CMV antibodies from various clinical samples. The microarray methods developed showed potential for efficiently and accurately detecting human herpesvirus DNAs, especially in CNS infections, and for simultaneous detection of DNAs or antibodies for multiple different human herpesviruses from clinical samples. In fact, the microarray method revealed several previously unrecognized co-infections. The microarray methods developed were sensitive and provided rapid detection of human herpesvirus DNA, and therefore the method could be applied to routine diagnostics. The microarrays might also be considered as an economical tool for diagnosing human herpesvirus infections.

Electrocardiographic repolarization variables in detecting myocardial infarction and ischemic injury : From body surface potential mapping to a single lead

The aim of the studies was to improve the diagnostic capability of electrocardiography (ECG) in detecting myocardial ischemic injury with a future goal of an automatic screening and monitoring method for ischemic heart disease. The method of choice was body surface potential mapping (BSPM), containing numerous leads, with intention to find the optimal recording sites and optimal ECG variables for ischemia and myocardial infarction (MI) diagnostics. The studies included 144 patients with prior MI, 79 patients with evolving ischemia, 42 patients with left ventricular hypertrophy (LVH), and 84 healthy controls. Study I examined the depolarization wave in prior MI with respect to MI location. Studies II-V examined the depolarization and repolarization waves in prior MI detection with respect to the Minnesota code, Q-wave status, and study V also with respect to MI location. In study VI the depolarization and repolarization variables were examined in 79 patients in the face of evolving myocardial ischemia and ischemic injury. When analyzed from a single lead at any recording site the results revealed superiority of the repolarization variables over the depolarization variables and over the conventional 12-lead ECG methods, both in the detection of prior MI and evolving ischemic injury. The QT integral, covering both depolarization and repolarization, appeared indifferent to the Q-wave status, the time elapsed from MI, or the MI or ischemia location. In the face of evolving ischemic injury the performance of the QT integral was not hampered even by underlying LVH. The examined depolarization and repolarization variables were effective when recorded in a single site, in contrast to the conventional 12-lead ECG criteria. The inverse spatial correlation of the depolarization and depolarization waves in myocardial ischemia and injury could be reduced into the QT integral variable recorded in a single site on the left flank. In conclusion, the QT integral variable, detectable in a single lead, with optimal recording site on the left flank, was able to detect prior MI and evolving ischemic injury more effectively than the conventional ECG markers. The QT integral, in a single-lead or a small number of leads, offers potential for automated screening of ischemic heart disease, acute ischemia monitoring and therapeutic decision-guiding as well as risk stratification.

Detecting Earnings Management Using Neural Networks

Detecting Earnings Management Using Neural Networks. Trying to balance between relevant and reliable accounting data, generally accepted accounting principles (GAAP) allow, to some extent, the company management to use their judgment and to make subjective assessments when preparing financial statements. The opportunistic use of the discretion in financial reporting is called earnings management. There have been a considerable number of suggestions of methods for detecting accrual based earnings management. A majority of these methods are based on linear regression. The problem with using linear regression is that a linear relationship between the dependent variable and the independent variables must be assumed. However, previous research has shown that the relationship between accruals and some of the explanatory variables, such as company performance, is non-linear. An alternative to linear regression, which can handle non-linear relationships, is neural networks. The type of neural network used in this study is the feed-forward back-propagation neural network. Three neural network-based models are compared with four commonly used linear regression-based earnings management detection models. All seven models are based on the earnings management detection model presented by Jones (1991). The performance of the models is assessed in three steps. First, a random data set of companies is used. Second, the discretionary accruals from the random data set are ranked according to six different variables. The discretionary accruals in the highest and lowest quartiles for these six variables are then compared. Third, a data set containing simulated earnings management is used. Both expense and revenue manipulation ranging between -5% and 5% of lagged total assets is simulated. Furthermore, two neural network-based models and two linear regression-based models are used with a data set containing financial statement data from 110 failed companies. Overall, the results show that the linear regression-based models, except for the model using a piecewise linear approach, produce biased estimates of discretionary accruals. The neural network-based model with the original Jones model variables and the neural network-based model augmented with ROA as an independent variable, however, perform well in all three steps. Especially in the second step, where the highest and lowest quartiles of ranked discretionary accruals are examined, the neural network-based model augmented with ROA as an independent variable outperforms the other models.

Bacteria colonizing paper machines

Bacteria growing in paper machines can cause several problems. Biofilms detaching from paper machine surfaces may lead to holes and spots in the end product or even break the paper web leading to expensive delays in production. Heat stable endospores will remain viable through the drying section of paper machine, increasing the microbial contamination of paper and board. Of the bacterial species regularly found in the end products, Bacillus cereus is the only one classified as a pathogen. Certain B. cereus strains produce cereulide, the toxin that causes vomiting disease in food poisonings connected to B. cereus. The first aim of this thesis was to identify harmful bacterial species colonizing paper machines and to assess the role of bacteria in the formation of end product defects. We developed quantitative PCR methods for detecting Meiothermus spp. and Pseudoxanthomonas taiwanensis. Using these methods I showed that Meiothermus spp. and Psx. taiwanensis are major biofoulers in paper machines. I was the first to be able to show the connection between end product defects and biofilms in the wet-end of paper machines. I isolated 48 strains of primary-biofilm forming bacteria from paper machines. Based on one of them, strain K4.1T, I described a novel bacterial genus Deinobacterium with Deinobacterium chartae as the type species. I measured the transfer of Bacillus cereus spores from packaging paper into food. To do this, we constructed a green fluorescent protein (GFP) labelled derivative of Bacillus thuringiensis and prepared paper containing spores of this strain. Chocolate and rice were the recipient foods when transfer of the labelled spores from the packaging paper to food was examined. I showed that only minority of the Bacillus cereus spores transferred into food from packaging paper and that this amount is very low compared to the amount of B. cereus naturally occurring in foods. Thus the microbiological risk caused by packaging papers is very low. Until now, the biological function of cereulide for the producer cell has remained unknown. I showed that B. cereus can use cereulide to take up K+ from environment where K+ is scarce: cereulide binds K+ ions outside the cell with high affinity and transports these ions across cell membrane into the cytoplasm. Externally added cereulide increased the growth rate of cereulide producing strains in medium where potassium was growth limiting. In addition, cereulide producing strains outcompeted cereulide non-producing B. cereus in potassium deficient environment, but not when the potassium concentration was high. I also showed that cereulide enhances biofilm formation of B. cereus.

Detection, epidemiology and host spectrum of cowpox and Borna disease virus infections

Several orthopoxviruses (OPV) and Borna disease virus (BDV) are enveloped, zoonotic viruses with a wide geographical distribution. OPV antibodies cross-react, and former smallpox vaccination has therefore protected human populations from another OPV infection, rodent-borne cowpox virus (CPXV). Cowpox in humans and cats usually manifests as a mild, self-limiting dermatitis and constitutional symptoms, but it can be severe and even life-threatening in the immunocompromised. Classical Borna disease is a progressive meningoencephalomyelitis in horses and sheep known in central Europe for centuries. Nowadays the virus or its close relative infects humans and also several other species in central Europe and elsewhere, but the existence of human Borna disease with its suspected neuropsychiatric symptoms is controversial. The epidemiology of BDV is largely unknown, and the present situation is even more intriguing following the recent detection of several-million-year-old, endogenized BDV genes in primate and various other vertebrate genomes. The aims of this study were to elucidate the importance of CPXV and BDV in Finland and in possible host species, and particularly to 1) establish relevant methods for the detection of CPXV and other OPVs as well as BDV in Finland, 2) determine whether CPXV and BDV exist in Finland, 3) discover how common OPV immunity is in different age groups in Finland, 4) characterize possible disease cases and clarify their epidemiological context, 5) establish the hosts and possible reservoir species of these viruses and their geographical distribution in wild rodents, and 6) elucidate the infection kinetics of BDV in the bank vole. An indirect immunofluorescence assay and avidity measurement were established for the detection, timing and verification of OPV or BDV antibodies in thousands of blood samples from humans, horses, ruminants, lynxes, gallinaceous birds, dogs, cats and rodents. The mostly vaccine-derived OPV seroprevalence was found to decrease gradually according to the year of birth of the sampled human subjects from 100% to 10% in those born after 1977. On the other hand, OPV antibodies indicating natural contact with CPXV or other OPVs were commonly found in domestic and wild animals: the horse, cow, lynx, dog, cat and, with a prevalence occasionally even as high as 92%, in wild rodents, including some previously undetected species and new regions. Antibodies to BDV were detected in humans, horses, a dog, cats, and for the first time in wild rodents, such as bank voles (Myodes glareolus). Because of the controversy within the human Borna disease field, extra verification methods were established for BDV antibody findings: recombinant nucleocapsid and phosphoproteins were produced in Escherichia coli and in a baculovirus system, and peptide arrays were additionally applied. With these verification assays, Finnish human, equine, feline and rodent BDV infections were confirmed. Taken together, wide host spectra were evident for both OPV and BDV infections based on the antibody findings, and OPV infections were found to be geographically broadly distributed. PCR amplification methods were utilised for hundreds of blood and tissue samples. The methods included conventional, nested and real-time PCRs with or without the reverse transcription step and detecting four or two genes of OPVs and BDV, respectively. OPV DNA could be amplified from two human patients and three bank voles, whereas no BDV RNA was detected in naturally infected individuals. Based on the phylogenetic analyses, the Finnish OPV sequences were closely related although not identical to a Russian CPXV isolate, and clearly different from other CPXV strains. Moreover, the Finnish sequences only equalled each other, but the short amplicons obtained from German rodents were identical to monkeypox virus, in addition to German CPXV variants. This reflects the close relationship of all OPVs. In summary, RNA of the Finnish BDV variant could not be detected with the available PCR methods, but OPV DNA infrequently could. The OPV species infecting the patients of this study was proven to be CPXV, which is most probably also responsible for the rodent infections. Multiple cell lines and some newborn rodents were utilised in the isolation of CPXV and BDV from patient and wildlife samples. CPXV could be isolated from a child with severe, generalised cowpox. BDV isolation attempts from rodents were unsuccessful in this study. However, in parallel studies, a transient BDV infection of cells inoculated with equine brain material was detected, and BDV antigens discovered in archival animal brains using established immunohistology. Thus, based on several independent methods, both CPXV and BDV (or a closely related agent) were shown to be present in Finland. Bank voles could be productively infected with BDV. This experimental infection did not result in notable pathological findings or symptoms, despite the intense spread of the virus in the central and peripheral nervous system. Infected voles commonly excreted the virus in urine and faeces, which emphasises their possible role as a BDV reservoir. Moreover, BDV RNA was regularly reverse transcribed into DNA in bank voles, which was detected by amplifying DNA by PCR without reverse transcription, and verified with nuclease treatments. This finding indicates that BDV genes could be endogenized during an acute infection. Although further transmission studies are needed, this experimental infection demonstrated that the bank vole can function as a potential BDV reservoir. In summary, multiple methods were established and applied in large panels to detect two zoonoses novel to Finland: cowpox virus and Borna disease virus. Moreover, new information was obtained on their geographical distribution, host spectrum, epidemiology and infection kinetics.

Molecular Characterization of Viruses Causing the Cassava Brown Streak Disease Epidemic in Eastern Africa

Cassava brown streak disease (CBSD) was described for the first time in Tanganyika (now Tanzania) about seven decades ago. Tanganyika (now Tanzania) about seven decades ago. It was endemic in the lowland areas of East Africa and inland parts of Malawi and caused by Cassava brown streak virus (CBSV; genus Ipomovirus; Potyviridae). However, in 1990s CBSD was observed at high altitude areas in Uganda. The causes for spread to new locations were not known.The present work was thus initiated to generate information on genetic variability, clarify the taxonomy of the virus or viruses associated with CBSD in Eastern Africa as well as to understand the evolutionary forces acting on their genes. It also sought to develop a molecular based diagnostic tool for detection of CBSD-associated virus isolates. Comparison of the CP-encoding sequences of CBSD-associated virus isolates collected from Uganda and north-western Tanzania in 2007 and the partial sequences available in Genbank revealed occurrence of two genetically distinct groups of isolates. Two isolates were selected to represent the two groups. The complete genomes of isolates MLB3 (TZ:Mlb3:07) and Kor6 (TZ:Kor6:08) obtained from North-Western (Kagera) and North-Eastern (Tanga) Tanzania, respectively, were sequenced. The genomes were 9069 and 8995 nucleotides (nt), respectively. They translated into polyproteins that were predicted to yield ten mature proteins after cleavage. Nine proteins were typical in the family Potyviridae, namely P1, P3, 6K1, CI, 6K2, VPg, NIa-Pro, NIb and CP, but the viruses did not contain HC-Pro. Interestingly, genomes of both isolates contained a Maf/HAM1-like sequence (HAM1h; 678 nucleotides, 25 kDa) recombined between the NIb and CP domains in the 3’-proximal part of the genomes. HAM1h was also identified in Euphorbia ringspot virus (EuRSV) whose sequence was in GenBank. The HAM1 gene is widely spread in both prokaryotes and eukaryotes. In yeast (Saccharomyces cerevisiae) it is known to be a nucleoside triphosphate (NTP) pyrophosphatase. Novel information was obtained on the structural variation at the N-termini of polyproteins of viruses in the genus Ipomovirus. Cucumber vein yellowing virus (CVYV) and Squash vein yellowing virus (SqVYV) contain a duplicated P1 (P1a and P1b) but lack the HC-Pro. On the other hand, Sweet potato mild mottle virus (SPMMV), has a single but large P1 and has HC-Pro. Both virus isolates (TZ:Mlb3:07 & TZ:Kor6:08) characterized in this study contained a single P1 and lacked the HC-Pro which indicates unique evolution in the family Potyviridae. Comparison of 12 complete genomes of CBSD-associated viruses which included two genomes characterized in this study, revealed genetic identity of 69.0–70.3% (nt) and amino acid (aa) identities of 73.6–74.4% at polyprotein level. Comparison was also made among 68 complete CP sequences, which indicated 69.0-70.3 and 73.6-74.4 % identity at nt and aa levels, respectively. The genetic variation was large enough for dermacation of CBSD-associated virus isolates into two distinct species. The name CBSV was retained for isolates that were related to CBSV isolates available in database whereas the new virus described for the first time in this study was named Ugandan cassava brown streak virus (UCBSV) by the International Committee on Virus Taxonomy (ICTV). The isolates TZ:Mlb3:07 and TZ:Kor6:08 belong to UCBSV and CBSV, respectively. The isolates of CBSV and UCBSV were 79.3-95.5% and 86.3-99.3 % identitical at nt level, respectively, suggesting more variation amongst CBSV isolates. The main sources of variation in plant viruses are mutations and recombination. Signals for recombination events were detected in 50% of isolates of each virus. Recombination events were detected in coding and non-coding (3’-UTR) sequences except in the 5’UTR and P3. There was no evidence for recombination between isolates of CBSV and UCBSV. The non-synonomous (dN) to synonomous (dS) nucleotide substitution ratio (ω) for the HAM1h and CP domains of both viruses were ≤ 0.184 suggesting that most sites of these proteins were evolving under strong purifying selection. However, there were individual amino acid sites that were submitted to adaptive evolution. For instance, adaptive evolution was detected in the HAM1h of UCBSV (n=15) where 12 aa sites were under positive selection (P< 0.05) but not in CBSV (n=12). The CP of CBSV (n=23) contained 12 aa sites (p<0.01) while only 5 aa sites in the CP gene of UCBSV were predicted to be submitted to positive selection pressure (p<0.01). The advantages offered by the aa sites under positive selection could not be established but occurrence of such sites in the terminal ends of UCBSV-HAMIh, for example, was interpreted as a requirement for proteolysis during polyprotein processing. Two different primer pairs that simultaneously detect UCBSV and CBSV isolates were developed in this study. They were used successfully to study distribution of CBSV, UCBSV and their mixed infections in Tanzania and Uganda. It was established that the two viruses co-infect cassava and that incidences of co-infection could be as high as 50% around Lake Victoria on the Tanzanian side. Furthermore, it was revealed for the first time that both UCBSV and CBSV were widely distributed in Eastern Africa. The primer pair was also used to confirm infection in a close relative of cassava, Manihot glaziovii (Müller Arg.) with CBSV. DNA barcoding of M. glaziovii was done by sequencing the matK gene. Two out of seven M. glaziovii from the coastal areas of Korogwe and Kibaha in north eastern Tanzania were shown to be infected by CBSV but not UCBSV isolates. Detection in M. glaziovii has an implication in control and management of CBSD as it is likely to serve as virus reservoir. This study has contributed to the understanding of evolution of CBSV and UCBSV, which cause CBSD epidemic in Eastern Africa. The detection tools developed in this work will be useful in plant breeding, verification of the phytosanitary status of materials in regional and international movement of germplasm, and in all diagnostic activities related to management of CBSD. Whereas there are still many issues to be resolved such as the function and biological significance of HAM1h and its origin, this work has laid a foundation upon which the studies on these aspects can be based.