Influenza A viruses are significant human pathogens causing yearly epidemics and occasional pandemics. Past pandemics have resulted in significant morbidity and mortality. The 1918 influenza pandemic was thought to have resulted in the death of at least 675,000 people in the U.S., and 40 million people worldwide. Pandemics in 1957 and 1968, while less severe, were also of major public health importance. In 2009, a novel pandemic emerged and its impact is not yet fully understood. Understanding the molecular basis for the formation of pandemic influenza strains is critical. The 1957 and 1968 pandemics were human-avian reassortant viruses in which two or three influenza gene segments from the then circulating human influenza viruses were replaced with genes from an avian source. Sequence and phylogenetic analysis of the 1918 pandemic virus suggested that it was derived from an avian-like virus in toto. The 1918 pandemic virus caused several epidemiologically distinct waves. The so-called first wave, in the summer months of 1918, may have represented an early form of the virulent second wave. To understand how this pandemic virus emerged and to model its virulence, it is important to place this virus in the context of human influenza viruses circulating before 1918 and to follow the early evolution of human H1N1 viruses after 1918. Because human influenza isolates are not available earlier than 1933, the only way to characterize these viruses is by identification of influenza RNA fragments preserved in formalin-fixed, paraffin-embedded autopsy tissues. With our collaborator, Professor John S. Oxford, post-mortem records of fatal pneumonia cases from 1908-1928 were screened to identify putative influenza cases based on clinical history and post-mortem pathology findings. Promising cases were sectioned for subsequent molecular screening. Sections of post-mortem lung tissues are currently being examined for the presence of influenza A virus RNA fragments by RT-PCR methods previously used to characterize the 1918 pandemic virus from similar samples. 1. Determining the complete genomic sequence of the 1918 influenza virus: Because the 1918 virus sequence was determined by designing degenerate, overlapping RT-PCR primer sets, the terminal segment sequences, reflecting the 5 and 3 untranslated regions (UTRs), could not be determined using this method. In experimental pathogenesis experiments rescued influenza viruses containing 1918 gene segments have been chimeras with UTR sequences from the mouse-adapted A/WSN/33 H1N1 virus. It is possible that differences in the native 1918 virus may affect the results of pathogenesis studies. Thus, we have sought to use novel modifications of the rapid amplification of cDNA ends (RACE) methodologies to determine the native 1918 segment UTR sequences from the A/Brevig Mission/1/1918 strain. Additionally, 7 of the coding sequences of the 1918 virus were previously determined from this strain, but the full-length HA sequence was only determined from A/South Carolina/1/1918. Thus, we also sought to complete the HA2 domain sequence of A/Brevig Mission/1/1918 in addition to the UTRs to give a complete genomic sequence from one pandemic case. The HA2 sequence was completed and found to match the A/South Carolina/1/1918 sequence exactly. Using novel RACE approaches, the complete UTR sequences of the 1918 virus were determined. Several differences between 1918 and WSN33 have been observed in the UTRs of the gene segments. 2. Identification of pre-1918 and post-1918 influenza A virus cDNA-positive human pneumonia autopsy cases: Autopsy case records from the Royal London Hospital from 1907-1935 were screened for possible influenza pneumonia cases in collaboration with John Oxford, and the 20 most promising cases per year were sampled for molecular analysis. Additional 1918 autopsy cases from the AFIP were also screened. These efforts involve archevirologic surveys for pre-1918 human IAV RNA-positive autopsy cases to determine: 1) which subtype(s) of influenza A virus circulated in humans prior to the H1N1 pandemic in 1918;and 2) whether the 1918 pandemic virus retained any gene segment(s) from the previously circulating strain. Examination of post-1918 cases would allow study of the early evolution of human H1N1 viruses in the post-pandemic era before viral isolation in 1933, and allow characterization of virulence factors in the 1918 virus by comparison with less pathogenic but highly related viruses from the mid-1920s. Pre-1918 cases: Several pre-1918 cases were positive in an initial RT-PCR screen for the matrix gene, from 1910-1917, and three 1913 cases were RT-PCR positive after amplification with two degenerate H3 primer sets. Sequence analysis of these small fragments confirmed H3, but intriguingly, had no perfect matches in GenBank. Additional influenza RNA-positive cases from 1915-1917 have also been identified. 1918 second wave cases: During an initial screening by RT-PCR for the matrix gene, an additional 26 second wave (fall wave) cases were screened by RT-PCR and 5 new positive cases were identified. Sequence analysis of the HA1 domain of the hemagglutinin gene is being performed on these cases. Post-1918 cases: Several positive cases were identified from 1923, 1926, and 1927. Fragmentary sequence of the HA1 domain of a 1923 case demonstrated several coding differences with the 1918 virus. 3. Re-evaluation of the cause of death in 1918 influenza cases: We performed a detailed histopathologic review of available 1918, 1957, and 1968 pandemic influenza autopsy cases from the archives of the AFIP. The histopathology of fatal influenza virus pneumonias strikingly shows that the spectrum of pathologic changes described in the 1918 influenza pandemic is not significantly different from the histopathology observed in other less lethal pandemics or even in deaths occurring during seasonal influenza outbreaks. We next examined relevant information from the most recent influenza pandemic that appeared in the pre-antibiotic era, the 1918 pandemic. We examined lung tissue sections from 58 autopsies and reviewed pathologic and bacteriologic data from 118 published autopsy series describing in total 8,398 individual autopsy investigations. The 1918 influenza autopsies examined uniformly exhibited severe changes of secondary bacterial pneumonia. Bacteriologic and histopathologic results from published autopsy series also clearly and consistently implicated secondary bacterial pneumonia caused by common upper respiratory flora in most influenza fatalities. Thus, the majority of deaths in the 1918-1919 pandemic likely resulted directly from secondary bacterial pneumonia caused by common upper respiratory flora. Data from the subsequent 1957 and 1968 pandemics are consistent with these findings. If severe pandemic influenza is largely a problem of viral-bacterial co-pathogenesis, pandemic planning needs to go beyond addressing the viral cause alone (e.g., influenza vaccines, antivirals). Prevention, diagnosis, prophylaxis, and treatment of secondary bacterial pneumonias, as well as stockpiling of antibiotics and bacterial vaccines, should also be high among pandemic planning priorities.
Showing the most recent 10 out of 27 publications
