Final Report Summary - RNAFLU (Effect of natural viral RNA sequence variation on influenza virus RNA function)
In this project we investigated if mutations introduced in the influenza virus ribonucleic acid (RNA) genome could affect the function of the viral RNA genome itself, irrespectively of the effect on the viral proteins. Mutations in the viral RNA genome could for example affect the stability of the viral RNA, the splicing efficiency or the translation efficiency. Such mutations might indirectly affect the ability of the virus to replicate in human cells, thereby affecting the pathogenic properties of the influenza virus or their ability to adapt to new host species. We speculated that there were mutations in influenza virus that affected the properties of the viral RNA genome and we designed experiments to identify such mutations in naturally occurring influenza virus strains. The identification of such mutations would be important as they could serve as markers for the appearance of novel influenza virus strains with altered pathogenic properties. To investigate if naturally occurring mutations affected the function of the influenza virus RNA itself we selected a number of important properties and functions of the viral RNA, such as messenger RNA (mRNA) structure, mRNA splicing, binding to cellular microRNAs and mRNA translation that could be monitored in various assays.
RNA molecules derived from influenza viruses with different pathogenic properties or with different species tropism had different properties. The secondary structure of the RNA molecules from different influenza viruses was different. Such differences in RNA structure might contribute to viral replication efficiency. Well conserved, putative binding sites for cellular microRNAs were identified in the influenza virus mRNAs and experimental data indicated that such cellular microRNAs inhibited influenza virus expression. Processing of the influenza viral mRNAs revealed substantial differences in mRNA splicing efficiencies, which might affect the ration between the NS1 and NS2 proteins and the M1 and M2 proteins. Furthermore, we showed that the translation of influenza virus mRNAs was modulated by the NS1 protein. Differences in NS1 expression and function were of particular interest as it was implicated in influenza virus pathogenesis.
RNA sequence variation seen in naturally occurring influenza virus strains with widely different pathogenic properties could affect the structure and function of the influenza virus mRNAs. We found that influenza RNA sequence differences affected influenza virus RNA secondary structure and RNA splicing efficiency in a virus-strain specific manner. Therefore, variation in influenza virus RNA sequence might affect the pathogenic properties of the virus without affecting viral protein sequences.
This was the first evidence to show that influenza virus sequence polymorphism affected the function of the viral RNA. Differences in viral mRNA function were found in influenza virus isolates with widely different pathogenic properties, leading to the testable hypothesis that pathogenicity of influenza virus might be affected by differences in viral RNA sequences that did not affect the function of the viral proteins. If this hypothesis was correct, identification of RNA sequence markers for viral pathogenicity would aid in early detection of highly pathogenic influenza viruses that suddenly appeared in the avian or human populations. Early detection of pathogenic influenza virus strains might limit the transmission of such strains in the human population. The results of this research might therefore be important for disease control around the world, in particular for strategic decisions taken at centres of disease control.
Pathogenicity of influenza virus can range from respiratory infections that are cleared within weeks to pneumonia and lower respiratory tract infections with high mortality such as Spanish flu or highly pathogenic bird flu. The sequences of the RNA genomes of the various influenza viruses show a high degree of variability due to the influenza virus' inherent ability to evolve at a high rate. Since it had not been clarified why some influenza virus types are highly pathogenic in spite of many studies on differences in influenza virus protein sequence, we investigated if sequence differences between low and high pathogenic influenza viruses affected the function of the viral RNA rather than viral protein. We found that RNA molecules derived from influenza virus strains with different pathogenic properties had different structural and functional properties. These results opened up for further research on the role of influenza virus RNA structure and function in virus pathogenesis.
The orthomyxoviridae family of RNA viruses contains three human viruses named influenza A, B and C. These three viruses cause acute infections of the respiratory tract, normally giving rise to symptoms that are typical for influenza virus infection, such as headache, high fever, myalgia and coughing. Influenza virus type C appears to be less pathogenic than A and B. While all three viruses can cause local, seasonal epidemics of influenza, influenza virus A can also give rise to pandemics that involve the entire world, spreading across the globe within a year. Although influenza virus infections have been described many times in historical records, the most infamous pandemic we are aware of today is the Spanish flu that occurred in 1918, as well as the 1957 H2N2 and 1968 H3N2 pandemics. In-between pandemics, strains from the last pandemic circulate in the population and give rise to local epidemics and annual outbreaks of influenza. Different strains of influenza virus therefore display enormous difference in transmission efficiency.
Time has also shown that there is huge variation in the pathogenicity of various influenza virus strains, ranging from relatively mild symptoms caused by today's circulating H3N2 strain to the extreme mortality of the 1918 H1N1 strain, which gave rise to the infamous Spanish flu pandemic, in which an estimated 30 to 50 million people succumbed. The H1N1 virus apparently caused local infections of the respiratory tract, as all other influenza virus strains that have been isolated from humans, as respiratory failure was the main cause of death. However, these examples highlight the ability of influenza virus to change its pathogenic properties over a wide range.
Recent decades have also seen examples of influenza virus spreading from animals directly to humans, demonstrating the zoonotic properties of influenza A. One example is the transmission of the highly pathogenic H5N1 bird flu to humans in Hong Kong in 1997, killing 6 of 18 infected humans. Although it was generally believed that influenza virus from mammals such as pigs could be transmitted to humans and vice versa, it was not until recently we obtained evidence that influenza virus could also spread directly from avian hosts to humans, sometimes with devastating results. This was highly unexpected since influenza infection in birds is primarily an enteric infection, targeting the mucosal epithelium of the gut, whereas influenza virus infects respiratory epithelium in humans. Fortunately these viruses were poorly transmitted between humans and did not adopt a more efficient mode of transmission during the epidemics.
At the molecular level, influenza virus is a negative strand RNA virus, with a segmented genome. Mutations are frequently introduced in the viral genomes as a result of lack of proof-reading of the viral RNA dependent RNA polymerase. An additional mechanism for generating variability in the influenza virus population is genetic shift in which different influenza virus strains exchange one or more segments with a co-infecting strain. This mechanism is of particular importance for influenza virus type A, which has an extraordinary large gene pool available for genetic shifts that includes both avian and various mammalian influenza virus strains.
Within the project three properties of influenza A were of particular interest, namely antigenicity, transmission efficiency and pathogenicity. Although the antigenic properties and to a large extent the transmission efficiency were determined primarily by genetic changes of influenza virus that caused changes in the amino acid sequence of the viral H and N proteins, it was more difficult to identify a single viral proteins that was responsible for the pathogenic properties of highly pathogenic Spanish flu or bird flu. We speculated that pathogenic properties of an influenza virus could also be determined by differences in influenza virus RNA sequence that did not affect the protein sequence of the virus genome. In other words, we hypothesised that sequence variation in the influenza virus genome could also affect structure or function of the viral RNA, thereby contributing to the pathogenic properties of the influenza A virus, as well as to its ability to adopt to a new host.
All RNAs in a cell are associated with proteins and RNAs are dependent on these interactions to function efficiently. Interactions of RNA with proteins depend on the RNA sequence and secondary structure. The exact RNA sequence is therefore of paramount importance since it affects secondary structure and function and utilisation efficiencies of the viral RNAs. RNA sequence has a direct effect on influenza virus RNA splicing, mRNA stability and translation. This affects the replication efficiency of the influenza virus strain. It was thus reasonable to speculate that RNA sequence variation itself could affect virus pathogenicity.
The immediate goal of this project was to investigate if naturally occurring influenza virus RNA sequences displayed differences in virus RNA structure and function, such RNA splicing and translation. If so, these results would open up for further research on the effect of these effects on pathogenicity and tropism of various virus strains.
Due to the high variability of influenza virus and its ability to cross species barriers, new influenza virus types with different pathogenic properties might arise quickly. It is of paramount importance to quickly identify influenza viruses that are highly pathogenic or are efficiently transmitted between humans at an early stage of an epidemic. Knowledge of viral properties that determine its pathogenesis is therefore necessary. It is generally believed that differences in pathogenicity are caused by mutations in the viral genome that affect structure and functions of the viral proteins. RNAFLU investigated if mutations in the viral genome could affect the structure and functions of the viral RNAs. If so, such mutations could potentially affect the pathogenic properties of the influenza virus.
We investigated if RNA sequence differences between highly pathogenic influenza viruses and low pathogenic influenza viruses affected virus RNA structure, primary structure and microRNA binding sites and secondary structure, and function, i.e. translation, stability and splicing, and we investigated if these sequence differences correlated with the pathogenic properties of the virus.
We found that differences in RNA sequence between highly pathogenic and low pathogenic influenza virus strains affected both structure and function of the viral RNAs. We speculated that such differences in viral RNA function might affect the pathogenic properties of the virus. The obtained results paved the way for experiments to test if mutations that affected viral RNA structure and function also affected the pathogenic properties of the virus. The contribution of the viral RNA sequence to the virus pathogenic properties was not previously studied and might have been overlooked.
This knowledge might lay the ground for new diagnostic tools that could quickly identify a highly pathogenic influenza virus strain that has entered the human population, thereby contributing to measures that would stop the transmission of the virus in the population.
The following scientific articles were produced:
1. 'A family of non-classical pseudoknots in influenza A and B viruses', Gultyaev AP, Olsthoorn RC, RNA Biol. 2010 Mar 25;7(2)
2. 'Attenuated strains of influenza A viruses do not induce degradation of RNA polymerase II', A. Rodriguez, A. Pérez-González, M. Jaber Hossain, L-M. Chen,T. Rolling, P. Pérez-Breña, R. Donis, G. Koch and A. Nieto, in Journal of Virology (2009) 83, 11166-11174.
Results were also presented at:
1. the national virology conference in Salmanca, Spain, 2009
2. the Oxford virology symposium 'From theory to therapeutics' on 16 April 2010
3. the RNA 2010 annual meeting in Seattle, United States of America, from 22 to 26th June 2010
4. the negative strand virus meeting 2010 held in Brugge, Belgium, in June 2010.
Further information about the project could be obtained at 'http://www.imbim.uu.se/RNAFLU/index.html'.
RNA molecules derived from influenza viruses with different pathogenic properties or with different species tropism had different properties. The secondary structure of the RNA molecules from different influenza viruses was different. Such differences in RNA structure might contribute to viral replication efficiency. Well conserved, putative binding sites for cellular microRNAs were identified in the influenza virus mRNAs and experimental data indicated that such cellular microRNAs inhibited influenza virus expression. Processing of the influenza viral mRNAs revealed substantial differences in mRNA splicing efficiencies, which might affect the ration between the NS1 and NS2 proteins and the M1 and M2 proteins. Furthermore, we showed that the translation of influenza virus mRNAs was modulated by the NS1 protein. Differences in NS1 expression and function were of particular interest as it was implicated in influenza virus pathogenesis.
RNA sequence variation seen in naturally occurring influenza virus strains with widely different pathogenic properties could affect the structure and function of the influenza virus mRNAs. We found that influenza RNA sequence differences affected influenza virus RNA secondary structure and RNA splicing efficiency in a virus-strain specific manner. Therefore, variation in influenza virus RNA sequence might affect the pathogenic properties of the virus without affecting viral protein sequences.
This was the first evidence to show that influenza virus sequence polymorphism affected the function of the viral RNA. Differences in viral mRNA function were found in influenza virus isolates with widely different pathogenic properties, leading to the testable hypothesis that pathogenicity of influenza virus might be affected by differences in viral RNA sequences that did not affect the function of the viral proteins. If this hypothesis was correct, identification of RNA sequence markers for viral pathogenicity would aid in early detection of highly pathogenic influenza viruses that suddenly appeared in the avian or human populations. Early detection of pathogenic influenza virus strains might limit the transmission of such strains in the human population. The results of this research might therefore be important for disease control around the world, in particular for strategic decisions taken at centres of disease control.
Pathogenicity of influenza virus can range from respiratory infections that are cleared within weeks to pneumonia and lower respiratory tract infections with high mortality such as Spanish flu or highly pathogenic bird flu. The sequences of the RNA genomes of the various influenza viruses show a high degree of variability due to the influenza virus' inherent ability to evolve at a high rate. Since it had not been clarified why some influenza virus types are highly pathogenic in spite of many studies on differences in influenza virus protein sequence, we investigated if sequence differences between low and high pathogenic influenza viruses affected the function of the viral RNA rather than viral protein. We found that RNA molecules derived from influenza virus strains with different pathogenic properties had different structural and functional properties. These results opened up for further research on the role of influenza virus RNA structure and function in virus pathogenesis.
The orthomyxoviridae family of RNA viruses contains three human viruses named influenza A, B and C. These three viruses cause acute infections of the respiratory tract, normally giving rise to symptoms that are typical for influenza virus infection, such as headache, high fever, myalgia and coughing. Influenza virus type C appears to be less pathogenic than A and B. While all three viruses can cause local, seasonal epidemics of influenza, influenza virus A can also give rise to pandemics that involve the entire world, spreading across the globe within a year. Although influenza virus infections have been described many times in historical records, the most infamous pandemic we are aware of today is the Spanish flu that occurred in 1918, as well as the 1957 H2N2 and 1968 H3N2 pandemics. In-between pandemics, strains from the last pandemic circulate in the population and give rise to local epidemics and annual outbreaks of influenza. Different strains of influenza virus therefore display enormous difference in transmission efficiency.
Time has also shown that there is huge variation in the pathogenicity of various influenza virus strains, ranging from relatively mild symptoms caused by today's circulating H3N2 strain to the extreme mortality of the 1918 H1N1 strain, which gave rise to the infamous Spanish flu pandemic, in which an estimated 30 to 50 million people succumbed. The H1N1 virus apparently caused local infections of the respiratory tract, as all other influenza virus strains that have been isolated from humans, as respiratory failure was the main cause of death. However, these examples highlight the ability of influenza virus to change its pathogenic properties over a wide range.
Recent decades have also seen examples of influenza virus spreading from animals directly to humans, demonstrating the zoonotic properties of influenza A. One example is the transmission of the highly pathogenic H5N1 bird flu to humans in Hong Kong in 1997, killing 6 of 18 infected humans. Although it was generally believed that influenza virus from mammals such as pigs could be transmitted to humans and vice versa, it was not until recently we obtained evidence that influenza virus could also spread directly from avian hosts to humans, sometimes with devastating results. This was highly unexpected since influenza infection in birds is primarily an enteric infection, targeting the mucosal epithelium of the gut, whereas influenza virus infects respiratory epithelium in humans. Fortunately these viruses were poorly transmitted between humans and did not adopt a more efficient mode of transmission during the epidemics.
At the molecular level, influenza virus is a negative strand RNA virus, with a segmented genome. Mutations are frequently introduced in the viral genomes as a result of lack of proof-reading of the viral RNA dependent RNA polymerase. An additional mechanism for generating variability in the influenza virus population is genetic shift in which different influenza virus strains exchange one or more segments with a co-infecting strain. This mechanism is of particular importance for influenza virus type A, which has an extraordinary large gene pool available for genetic shifts that includes both avian and various mammalian influenza virus strains.
Within the project three properties of influenza A were of particular interest, namely antigenicity, transmission efficiency and pathogenicity. Although the antigenic properties and to a large extent the transmission efficiency were determined primarily by genetic changes of influenza virus that caused changes in the amino acid sequence of the viral H and N proteins, it was more difficult to identify a single viral proteins that was responsible for the pathogenic properties of highly pathogenic Spanish flu or bird flu. We speculated that pathogenic properties of an influenza virus could also be determined by differences in influenza virus RNA sequence that did not affect the protein sequence of the virus genome. In other words, we hypothesised that sequence variation in the influenza virus genome could also affect structure or function of the viral RNA, thereby contributing to the pathogenic properties of the influenza A virus, as well as to its ability to adopt to a new host.
All RNAs in a cell are associated with proteins and RNAs are dependent on these interactions to function efficiently. Interactions of RNA with proteins depend on the RNA sequence and secondary structure. The exact RNA sequence is therefore of paramount importance since it affects secondary structure and function and utilisation efficiencies of the viral RNAs. RNA sequence has a direct effect on influenza virus RNA splicing, mRNA stability and translation. This affects the replication efficiency of the influenza virus strain. It was thus reasonable to speculate that RNA sequence variation itself could affect virus pathogenicity.
The immediate goal of this project was to investigate if naturally occurring influenza virus RNA sequences displayed differences in virus RNA structure and function, such RNA splicing and translation. If so, these results would open up for further research on the effect of these effects on pathogenicity and tropism of various virus strains.
Due to the high variability of influenza virus and its ability to cross species barriers, new influenza virus types with different pathogenic properties might arise quickly. It is of paramount importance to quickly identify influenza viruses that are highly pathogenic or are efficiently transmitted between humans at an early stage of an epidemic. Knowledge of viral properties that determine its pathogenesis is therefore necessary. It is generally believed that differences in pathogenicity are caused by mutations in the viral genome that affect structure and functions of the viral proteins. RNAFLU investigated if mutations in the viral genome could affect the structure and functions of the viral RNAs. If so, such mutations could potentially affect the pathogenic properties of the influenza virus.
We investigated if RNA sequence differences between highly pathogenic influenza viruses and low pathogenic influenza viruses affected virus RNA structure, primary structure and microRNA binding sites and secondary structure, and function, i.e. translation, stability and splicing, and we investigated if these sequence differences correlated with the pathogenic properties of the virus.
We found that differences in RNA sequence between highly pathogenic and low pathogenic influenza virus strains affected both structure and function of the viral RNAs. We speculated that such differences in viral RNA function might affect the pathogenic properties of the virus. The obtained results paved the way for experiments to test if mutations that affected viral RNA structure and function also affected the pathogenic properties of the virus. The contribution of the viral RNA sequence to the virus pathogenic properties was not previously studied and might have been overlooked.
This knowledge might lay the ground for new diagnostic tools that could quickly identify a highly pathogenic influenza virus strain that has entered the human population, thereby contributing to measures that would stop the transmission of the virus in the population.
The following scientific articles were produced:
1. 'A family of non-classical pseudoknots in influenza A and B viruses', Gultyaev AP, Olsthoorn RC, RNA Biol. 2010 Mar 25;7(2)
2. 'Attenuated strains of influenza A viruses do not induce degradation of RNA polymerase II', A. Rodriguez, A. Pérez-González, M. Jaber Hossain, L-M. Chen,T. Rolling, P. Pérez-Breña, R. Donis, G. Koch and A. Nieto, in Journal of Virology (2009) 83, 11166-11174.
Results were also presented at:
1. the national virology conference in Salmanca, Spain, 2009
2. the Oxford virology symposium 'From theory to therapeutics' on 16 April 2010
3. the RNA 2010 annual meeting in Seattle, United States of America, from 22 to 26th June 2010
4. the negative strand virus meeting 2010 held in Brugge, Belgium, in June 2010.
Further information about the project could be obtained at 'http://www.imbim.uu.se/RNAFLU/index.html'.
