Professor and Chair, Department of Microbiology, Mount Sinai School of Medicine, New York
|INFLUENZA PANDEMICS: PAST AND FUTURE|
Influenza viruses are RNA viruses, belonging to the group which includes measles, Ebola and Lassa fever viruses, among many others. In contrast to these viruses, influenza viruses undergo dramatic changes over short periods of time. Specifically, the surface proteins, hemagglutinin (H) and neuraminidase (N), change by accumulating mutations in a process termed antigenic drift, which allows the virus to evade our immune system. Thus, after a period of only three or four years, humans can become infected by the changed virus even though they still have protective immunity against the earlier strain. The virus can also undergo a more dramatic antigenic change (known as antigenic shift) which results from acquiring a novel H gene from an animal influenza virus. In the last century, this happened three times: in 1918 (Spanish flu), 1957 (Asian flu) and 1968 (Hong Kong flu). In each case, a virus with a new H caused a worldwide epidemic (pandemic) because the population had little or no immunity to the new viruses.
The Pandemic of 1918
The influenza pandemic of 1918/1919 was a unique event in recorded history, costing on the order to 50 million lives in less than a year (Johnson and Mueller, 2002). Even though this number of deaths was staggeringly high, the overall mortality rate for those infected in the U.S. and in Europe was only about 2%. The extent of the impact of the 1918 pandemic can be seen in the drop of the life expectancy in the U.S. by about 11 years. This sharp decline was largely the result of death in the young adult population, which was disproportionately affected by the pandemic. In addition, a very large percentage of the two billion world population in 1918/19 became infected and suffered some degree of illness. Not only was the 1918/19 pandemic unique in the number of people affected, but also in the shape of its mortality curve. The very young (less than one year) died at a high rate as a result of the infection, there was a paucity of fatality cases in the 5-14 year age group, and a peak among young adults aged 25-35 years. Interestingly, the mortality rate dropped beyond about age 35, only to again rise in people older than 65 years. This W-shaped curve is unusual and was not seen in pre-1918 influenza epidemics and has not been seen in those since. It is not clear why there is a valley of low fatality in the 5-14 year group, but the same phenomenon has been observed (in immunologically naïve populations) with other infectious diseases, including smallpox, measles and poliomyelitis. This first drop in overall fatalities may be associated with an enhanced innate immunity in the 5-14 age group. We postulate that the second drop, in the 35-65 year old population, was the result of exposure to an H1-like influenza virus which circulated before 1889 in the human population.
If this interpretation is correct, we can make two points which may influence our thinking about future pandemics: (1) Influenza pandemic viruses might be even more devastating than the 1918 virus if there is no cross protection in the human population because of exposure to antigenically related viruses circulating earlier in the population. (2) There appear to be intrinsic differences in the virulence of pandemic influenza viruses (this is exemplified by the much lower fatality numbers for the H2 pandemic in 1957, even thought there was no evidence for a prior exposure of the population to H2 viruses).
Reconstruction of the 1918 Influenza Virus
The advent of reverse genetics techniques for negative strand RNA viruses makes it possible to introduce DNA-derived sequences/genes into influenza viruses (Luytjes et al., 1989) and to reconstruct infectious influenza viruses entirely from commercially available oligonucleotides (Fodor et al., 1999; Neumann et al., 1999). Based on the nucleotide sequence that was obtained from RNA fragments present in lung samples of victims of the 1918 influenza virus (Taubenberger et al., 2005), we succeeded in reconstructing the extinct pandemic virus (Tumpey et al., 2005). This virus turns out to be highly virulent in the mouse model, moreso than any other human (i.e., non mouse-adapted) influenza virus strain tested. It is also highly pathogenic for chicken embryos, capable of killing these embryos at very low doses (less than 100 tissue culture infectious particles). Finally, the virus is also able to grow in human tissue culture cells to high titers (almost 109 plaque-forming units/ml, which is a ten times higher titer than that observed for other human influenza viruses), and it can replicate in cells in the absence of trypsin, which may also indicate high virulence (Tumpey et al., 2005).
Furthermore, we were able to show (in mice) that the 1918 virus is sensitive to neuraminidase inhibitors as well as to the M2 ion channel blocker amantadine. These are FDA-approved antivirals against influenza which appear to be just as effective against the most virulent of the human influenza viruses as against the milder forms. Clearly, this is a most encouraging finding in light of the likelihood of new pandemics in the future.
Finally, we found that vaccines made against the 1918 virus protected mice against a high challenging dose of the 1918 virus. Another very interesting result was that mice could be partially protected against challenge with the 1918 virus when vaccinated with the H1 vaccine component of the 2006/2007 influenza virus vaccine. This suggests that H1 viruses circulating at the present time elicit a partially cross-reactive immune response against the 1918 virus (the vaccine component reflects the H1 strains currently circulating). Thus, the 1918 virus would hardly be the best pathogen to use as a biothreat agent. It would not cause nearly the amount of morbidity and mortality that the 1918 virus did because we have all been exposed to H1 descendents of the 1918 virus and consequently we all have at least partial immunity against the 1918 virus.
The 1957 and 1968 Pandemics
While the 1918 influenza pandemic was certainly unique, the pandemics of 1957 and 1968 confirmed observations made in previous centuries that, on average, two to three pandemics occur every 100 years. In the case of the 1957 virus, three genes were acquired from avian influenza viruses (the H2, N2 and the PB1 genes) and the 1968 strain received two genes (the H3 and the PB1 genes) from an avian virus by reassortment. Both the 1957 and 1968 viruses had a much lower fatality rate among those infected than did the 1918 virus. Compared to the latter virus 2% fatality rate, these had rates of 0.02% and 0.01%, respectively. Despite the lower mortality rates, these viruses caused extensive disease which was only mitigated (in developed countries) by the use of vaccines.
Based on historical precedent, it is highly likely that new pandemics will emerge in the future. Since 1997, our awareness has been heightened that H5 viruses (particularly H5N1) can cause massive and devastating outbreaks in poultry and occasional fatalities in humans. While only eight countries were affected with outbreaks of H5N1 virus in poultry in 2004, this number has risen to more than 50 countries with serious H5N1 epizootics in 2006. Thus, H5N1 outbreaks in animals are of major concern. However, despite more than 130 confirmed human deaths, mostly in Indonesia and Vietnam, there is no evidence that these H5N1 viruses are easily transmitted from human to human or that they will be responsible for a new pandemic. In fact, these viruses have been around for decades (H5 chicken viruses were isolated in the 1950s and probably were circulating even earlier). Questions then arises: If H5 viruses have been circulating in different avian species for so long, why has the virus not been successful in jumping into humans? How many people have in fact been infected in a subclinical or aclinical manner? Why has this virus not been transmitted more readily from human to human? These are important research issues which will have to be addressed. Also, mammalian transmission models, including the ferret and the guinea pig, will have to be used in order to define the molecular basis for transmission (or lack thereof) of influenza viruses. Since there are 16 H subtypes and 9 N subtypes of influenza viruses circulating in animal populations, these transmission studies need to be expanded to include viruses other than H1, H2, H3 or H5 subtypes. Viruses belonging to the other 12 H subtypes are potential candidates for human pandemic strains as well, and it behooves us to find out what the real potential is for viruses of these other subtypes to cause pandemics in humans.
The influenza pandemic of 1918/1919 was a unique event, and reconstruction the virus has helped us to begin understanding why the virus was extraordinarily virulent and what contributed to the unusual mortality pattern in the population. Questions still need to be answered about how the 1918 virus and other influenza viruses interact with the innate immune system, and new technologies will have to be developed for quantitatively measuring immune parameters after infections. These include both cellular and humoral immune responses. The availability of the 1918 virus will help us to understand the mechanisms by which pandemic influenza viruses are transmitted from human to human and from one species to another. We also need to increase our efforts to learn more about other subtype viruses, including the H5 viruses, and to study the molecular signatures which define pandemic potential. Efforts underway in many laboratories will effectively expand our knowledge of the biological and molecular properties of pandemic influenza viruses, and this research will provide us with better prevention and treatment strategies for the future.
This work was supported by National Institutes of Health grants P01 AI058113, U19 AU062623, and P01 AI052106.
Fodor, E., Devenish, L., Engelhardt, O. G., Palese, P., Brownlee, G. G., and Garcia-Sastre, A. (1999). Rescue of influenza A virus from recombinant DNA. J Virol 73, 9679-9682.
Johnson, N. P., and Mueller, J. (2002). Updating the accounts: global mortality of the 1918-1920 "Spanish" influenza pandemic. Bull Hist Med 76, 105-115.
Luytjes, W., Krystal, M., Enami, M., Pavin, J. D., and Palese, P. (1989). Amplification, expression, and packaging of foreign gene by influenza virus. Cell 59, 1107-1113.
Neumann, G., Watanabe, T., Ito, H., Watanabe, S., Goto, H., Gao, P., Hughes, M., Perez, D. R., Donis, R., Hoffmann, E., et al. (1999). Generation of influenza A viruses entirely from cloned cDNAs. Proc Natl Acad Sci U S A 96, 9345-9350.
Taubenberger, J. K., Reid, A. H., Lourens, R. M., Wang, R., Jin, G., and Fanning, T. G. (2005). Characterization of the 1918 influenza virus polymerase genes. Nature 437, 889-893.
Tumpey, T. M., Basler, C. F., Aguilar, P. V., Zeng, H., Solorzano, A., Swayne, D. E., Cox, N. J., Katz, J. M., Taubenberger, J. K., Palese, P., and Garcia-Sastre, A. (2005). Characterization of the reconstructed 1918 Spanish influenza pandemic virus. Science 310, 77-80.