Swine Influenza HistologyInfluenza is an acute respiratory disease affecting a variety of mammalian speciesincluding swine, humans, and horses. The major reservoir of influenza virus, however, isthe aquatic birds of the world. Although fortuitously an infrequent event, it is from theseaquatic birds that viruses sporadically transmit to other avian and mammalian species.The presence of this vast reservoir of influenza viruses in the aquatic birds of the worldprecludes this viral disease from eradication. Thus, the quest for control of influenza inanimal and human health sectors lies heavily in the area of prophylactic treatment withvaccines.
Despite our ability to produce vaccines of proven efficacy, the influenza virus hasdeveloped avenues of counter assault that make vaccination a challenging and constantlychanging task. In addition to its vast reservoirs, the influenza virus has alternativemechanisms to promote constant evolution and evasion of the hosts' immune response.The ability of influenza to undergo genetic changes drives the continual emergence ofantigenically and genetically novel viruses. It is the aim of this report to summarize ourunderstanding of the current influenza viruses circulating in the swine populations of theworld with particular emphasis on current situations in North America.
The ability of the influenza virus to continually evolve resides in the fundamentalproperties of the virus particle itself. Through the genetic processes termed antigenic driftand antigenic shift, the virus has the ability to constantly sidestep the immune responseand sporadically cause pandemic disease of noteworthy proportions.
Antigenic drift, which is driven by the infidelity of the virally encoded polymerase,results in point mutations in the viral hemagglutinin (HA) and neuraminidase (NA)glycoproteins. The HA molecule is the major viral antigenic determinant, and theselection applied by the host immune system constantly selects for drift variants that canno longer be neutralized by circulating antibodies. In this way, influenza emergesseasonally as an endemic disease and can reemerge in populations that have considerableimmunity from previous exposures.
Less frequent, but potentially of far greater concern, is the process of antigenic shift. Theinfluenza A genome is composed of eight single-stranded negative-sense RNAmolecules. Infection of a single cell by two different influenza viruses can result in theproduction of progeny viruses containing a mixture of RNA segments from the parentalviruses. Such reassortment has the potential to completely change the antigenic nature ofthe circulating virus and, as such, allow unimpeded spread through a host population.
It has been postulated that swine play a central role in the ecology of influenza. Inaddition to being a natural host for a limited number of viral subtypes (see below) there isconvincing evidence that pigs can act as an intermediate host for human disease. Thelimiting factor in the emergence of pandemic influenza in humans is the inability of manyviruses from aquatic birds to replicate effectively in the respiratory tract of primates.1,19Likewise, human viruses inoculated using natural routes of infection replicate poorly inwaterfowl12 In contrast pigs seem to be readily infected by human viruses3,7 and most, ifnot all, avian HA subtypes are capable of replicating to some extent in swine.16 This traithas led to the hypotheses that pigs act as the mixing vessel for human, swine, and avianviruses with the resulting potential for reassortment and generation of novel viruses.Current theories on the source of the 1957 and 1968 human influenza pandemics are thatthe causative viruses were derived through reassortment in pig populations.
The molecular features responsible for the permissive nature of swine as a host ofinfluenza reside in the nature of the viral receptors. Avian and human influenza virusesbind to different sialic acid moieties on the surface of target cells. The preference for thedifferent receptors directly reflects the relative abundance of these receptors in the host.Avian cells contain primarily the receptors recognized by avian viruses and human cellscontain primarily those recognized by human viruses. In comparison, the cells lining therespiratory tract of swine contain both types of receptor allowing attachment of bothavian and human viruses.13
In addition to the potential pigs have as the mixing vessel for mammalian and swineviruses, there have also been numerous reports of human infection with swine viruses.Although some of these infections have been fatal23,25 the infections have all been selflimiting and there has been little or no human-to-human spread.
Since the first influenza virus was isolated from a swine in 1930, only two HA (H1 andH2) and two NA (N1 and N2) subtypes have formed stable lineages in swine populations.Reports appear sporadically that describe other subtypes infecting swine, such as H1N7,5H4N6,14 and H9N2,24 but these events have so far remained isolated cases and none havebecome established in swine populations. Endemic influenza in swine is restricted tothree subtype combinations: H1N1, H3N2, and H1N2. Although only three establishedviral subtype combinations are found, the different geographical populations arereservoirs for a much larger number of distinct viral lineages.
Classical-swine H1N1, which is phylogenetically related to the virus responsible for the1918 human Spanish flu pandemic, circulates predominantly in North America andAsia.20 In Europe, H1N1 viruses also circulate, but this lineage is derived from a whollyavian-like virus that was first detected in the pig population in 1979.22 This virussuperceded the classical-swine viruses circulating at the time and is the current H1N1throughout Europe. In addition a distinct lineage of avian H1N1 virus has been reportedin China, although it is uncertain to what degree, if at all, it has spread.10
H3N2 viruses were first detected in swine in 1970, shortly after the emergence of similarviruses in humans.17 Since this time, human-like H3N2 viruses have been isolated fromswine throughout Europe, Asia, and the Americas and analogously to the situation inhumans these viruses continue to cocirculate with H1N1 viruses in most parts of theworld. For reasons unknown, H3N2 viruses did not emerge in the United States swinepopulation until late 1998. The gene segments of these viruses, which have since becomeestablished, are of mixed origin and contain human virus (HA, NA, PB1), swine virus(NP, M, NS), and avian virus (PA and PB2) genes.30
Reassortant H1N2 viruses of different lineages have been identified in various swinepopulations and are becoming more prominent. A reassortant H1N2 virus containingavian-like swine and human genes has become a significant problem in the UnitedKingdom, and this virus seems to have now spread to continental Europe.4,27 ReassortantH1N2 viruses derived from classical H1N1 and various H3N2 viruses also have beenisolated in France, Japan, and the United States.9,15,26
Historical dogma has us believe that swine viruses do not evolve as quickly as humanviruses. In addition, it seems that viruses can be maintained for prolonged periods inswine without any marked change in antigenic structure.2,21 Many consider this reduceddrift rate to be due to the continual availability of immunologically naive animals inswine populations. The lack of immunologic pressure means that changes in the swineHA tend to be evenly distributed throughout the HA molecule, whereas changes inhuman virus HA genes frequently occur at or around antigenic sites.3 Although thereduced antigenic drift in swine may historically be true, recent events in both Europeand North America demonstrate that the swine populations of the world are becomingreservoirs for a very genetically diverse pool of viruses.
Recent influenza activity in European swine populations includes reassortment betweenH1N2 and H1N1 and/or H3N2 viruses, the isolation of antigenically distinct H1N1viruses, and the isolation of contemporary human-like H3N2 viruses.18 Sequence analysisof HA genes has shown that antigenic drift does occur in both European H1N1 and H3N2viruses of swine6,8 raising concerns from some investigators that vaccines in swine mayneed to be continually updated as in human populations. Heinen and colleagues11 haveshown, however, that vaccination with A/Port Chalmers/1/73 (H3N2) was sufficient tostop the development of fever and transmission upon challenge with a recent field strain.
The vaccine was not able, however, to completely stop viral shedding from thechallenged animal. Similar studies with the European H1N1 viruses have revealed similarresults in that heterologous virus vaccination can protect from clinical disease and reduceviral load although not viral replication.28
Similar levels of viral reassortment have been recently identified in the U.S. swinepopulation. Prior to the emergence of H3N2 viruses in 1998, swine influenza in theUnited States was caused exclusively by H1N1 viruses. By the end of 1999, H3N2viruses had spread throughout the United States. Of particular concern was theidentification of three antigenically distinct virus groups, each having a different HAgene obtained from contemporary human H3 viruses.29 Shortly after the identification ofthe H3N2 viruses the first generation of H1N1/H3N2 reassortments were identified.15These reassortant viruses were H1N2 viruses containing seven H3N2 genes and the HAfrom a classical H1N1 virus. H1N2 viruses have continued to spread and phylogeneticanalysis suggests that multiple independent reassortment events have resulted in theirgenesis. A further reassortment event between the H3N2 and H1N1 viruses has resultedin the emergence of yet another variant of virus. These viruses contain the HA and NA ofthe classical swine virus but all other genes from the H3N2 viruses. Our ongoing researchsuggests that these may be becoming one of the dominant viral genotypes in the U.S.swine population. It is also interesting to note that a virus of this genotype has beenisolated from a human with a nonfatal respiratory disease.
The current swine and human commercial vaccines are both killed vaccines in whichprotection is afforded by the development of neutralizing antibodies primarily to the HAmolecule. In such circumstances the amount of juggling of the other gene segments is ofno consequence. Unfortunately, at least in the case of the North American situation, thereassortment of viral gene segments has been followed by a concomitant change in theHA molecule. The recent H1 molecules seem to be gathering mutations at an increasedrate. The amount of sequence divergence between certain 2001 isolates is as much as thedifference between classical H1N1 viruses isolated in the 1960s and those isolated in theearly 1990s. Studies similar to those described above in Europe are needed to assess thecross protection potential of the current vaccines against all antigenic variants.
Taken together, these data show the huge impact that the introduction of a single newvirus into a swine population can have on the diversity of viral genotypes. The U.S. swinepopulation has thus gone from a reservoir containing a single virus to one where H1N2,two antigenically distinct H3N2, and two distinct genotypes of H1N1 co-circulate (Figure1). Which of these viral lineages will eventually predominate will only become apparentif surveillance is intensified and centralized.
Influenza activity in recent years in the human population has been relatively mild interms of disease and viral evolution. The last major human drift variant was the 1997A/Sydney/1/97-like H3N2 viruses. In contrast, the last few years has seen major activityin influenza viruses in global swine populations, particularly in the United States. Theresulting increase in genetic diversity of swine influenza viruses is of concern for bothhuman and animal health. The likelihood is that both H1 and H3 viruses will continue toevolve and cocirculate in swine populations and that the key to managing this situation issurveillance. The challenge for the swine industry is to develop a surveillance system thatincorporates genetic and antigenic characteristics of circulating viruses. Such a systemwill be indispensable for ensuring the efficacy of vaccines and for the early detection ofnovel and potentially devastating viruses.
The work described in this report is part of an ongoing collaboration between thelaboratories of Dr Robert Webster at St Jude Children抯 Research Hospital, Dr抯 KurtRossow and Sagar Goyal from the University of Minnesota, St Paul, and Dr GeneErickson at the Rollins Animal Disease Diagnostic Laboratory, NC. This project has beensupported by the National Institutes of Health and the American Lebanese and SyrianAssociated Charities.