JLE

Virologie

MENU

On the trail of Spanish influenza virus Volume 5, issue 1, Janvier - Février 2001

Figures

See all figures

Author
25, rue Périer, 92120 Montrouge

The « Spanish » influenza pandemic killed at least 20 millions people in 1918-1919, much more than the Second World War, making it the worse infectious pandemic in human history. A first wave, during the spring of 1918 was widespread but relatively benign ; a second wave at 1918 fall was exceptionally severe with a high morbidity and a spectacular morbidity ; a third wave at the beginning of 1919 was less severe. Understanding the origins of « Spanish influenza » and the basis for its exceptional virulence may help to predict future influenza pandemics. It is indeed certain that there will be other influenza pandemics and for many authors, it is likely that the virus will emerge from aquatic bird reservoirs and involve reassortment with human viruses, after additional mutations or recombinations that will permit spread and pathogenicity among humans. This reassortment could occur in pigs which possess cellular receptors both for avian and human viruses. So far, only H1, H2 and H3 subtypes have infected humans, but the possibility of infections with appropriate virus belonging to other subtypes remains open, for instance H5, H7 or H9. Studies conducted during the 1918 epidemic first indicated that bacteria could be the etiologic agents, but this observation was not confirmed. However, other groups demonstrated the possibility of transmission by filtrates of the infectious material, but only in 1930 (swine virus) and 1933 (human) the isolation of the virus in ferrets could first demonstrate unequivocally the viral origin of the disease and allow studies on the nature and properties of the virus. It was then possible to titrate the agent, demonstrate the presence of antibodies and identify the antigenic characteristics of the virus and its variants. The first serological findings indicated a relationship between antibodies found in survivors and the swine agent isolated in 1930. Modern techniques have permitted a reconstitution of some parts of the genome of the 1918 agent by amplifying fragments of viral RNA obtained from different sources. One of them was anatomopathologic samples of lungs from patients who died of the disease in 1918. These samples were kept embedded in paraffin after formalin fixation and could be recovered in the archives of the Armed Forces Institute of Pathology in Washington. Other samples were obtained after the exhumation of victims of Spanish flu in Alaska and in Svalberg whose bodies had been buried in permafrost ground. Sequences were obtained by genic amplification of viral RNA extracted from these fragments and were compared to recent human and animal viruses. The comparison showed that the hemagglutin of the 1918 virus was of the H1 subtype belonging to a subgroup of strains infecting human and pigs, but also sharing avian determinants. Sequence analysis indicates that many avian characteristics are present in critical locations of the hemagglutinin gene such as receptor, antigenic and glycosylation sites suggesting an avian relationship. However, the virus is closely related to human and swine viruses. Equivalent findings were obtained from the study of the neuraminidase gene : the enzymatic site is preserved but avian characteristics are found in antigenic and glycosylation sites. These results suggest that the 1918 virus borrowed determinants from avian strains but was already present in mammals for a prolonged period before the pandemic started. The exceptional virulence of this virus cannot be explained by changes at the cleavage site of the hemagglutinin nor by neuraminidase characters.