Studies on relationships between human and porcine influenza

Bull. Org. mond. Sante 1972, 46, 721-728 Bull. Wid Hlth Org. Studies on relationships between human and porcine influenza 2. Immunological comparisons ofhuman A/Hong Kong/68 virus with influenza A viruses of porcine origin G. C. SCHILD,1 C. M. BRAND,1 J. W. HARKNESS,2 AND P. H. LAMONT 2 Antigenic comparisons were made between the human A/Hong Kong/68 (H3N2) virus and a collection of influenza A viruses of swine origin. Haemagglutination-inhibition and neuraminidase-inhibition tests were used in addition to immunoprecipitin tests with monospecific antisera prepared against purified haemagglutinin and neuraminidase pre- parations. The antigenic relationships revealed by the studies are summarized as follows: (J)swine/Taiwan/7310/70 virus contained envelope antigens that were antigenically indis- tinguishablefrom those ofA/Hong Kong/68 virus, (2) "classical " strains ofswine influenza- virus related to A/swine/lIowa/JS/30 (HswlNJ) isolated between 1930 and 1967 contained neuraminidase that was antigenically distinctfrom that ofA/Hong Kong/68 virus but related to that of human AO and Al viruses, and (3) the haemagglutinins of certain strains of " classical" influenza A virus appeared to show a minor antigenic relationship with the haemagglutinin of A/Hong Kong/68 virus. Immunoprecipitin tests suggested that this relationship was confined to only one of the two antigenic components of the haemagglutinin subunit. The antigenic relationships are discussed in respect of possible epidemiological relationships between human and swine influenza A viruses. It is proposed that the swinel Taiwan isolates be designated A/swine/Taiwan/70 (H3N2), indicating their antigenic identity with the human A/Hong Kong/1/68 virus. Kundin (1970) recently reported the isolation from swine in Taiwan of a number of influenzavirus strains antigenically closely related to the human A/Hong Kong/68 virus. It was also shown that swine sera collected in Taiwan during late 1969 and 1970 contained haemagglutination-inhibiting antibody for human A/Hong Kong/68 virus. Sur- veys of pig populations in Great Britain (Harkness et al., 1972) and several other countries have also indicated the presence of antibody for human A/ Hong Kong/68 virus in pig sera collected in 1969 or 1970. In view of the possible significance of these findings to the understanding of the epidemio- logy of human influenza it was of importance to carry out detailed immunological comparisons be- 1 WHO World Influenza Centre, National Institute for Medical Research, Mill Hill, London, England. 2 Ministry of Agriculture and Fisheries, Central Veterinary Laboratory, Weybridge, Surrey, England. tween the A/Hong Kong/68 virus and a collection of influenza A viruses of porcine origin. The first reported isolation of an influenza A virus from pigs was made by Shope in 1930 in the USA (Shope, 1931). Since Shope's original obser- vations, influenza A viruses have continued to infect swine populations and are frequently associated with epidemics of clinical influenza among swine in the USA. Virus strains isolated from swine in Great Bri- tain in 1939 and in North America from 1954 to 1967 are antigenically related to Shope's prototype strain, S15, and are regarded as belonging to the same anti- genic type although some antigenic variation among strains has been detected (see Pereira, 1969; Meier- Ewert et al., 1970). In the present study we wished to compare a number of isolates of swine virus obtained between 1930 and 1967 with the viruses isolated in 1970 from swine in Taiwan and with the human A/Hong Kong/68 virus. Since the influenza- 2850 721 G. C. SCHILD ET AL. virus contains two subtype-specific antigens in its outer envelope-namely, the haemagglutinin and the neuraminidase-tests were carried out to examine the immunological relationships of each of these antigenic components for the group of viruses under test. MATERIALS AND METHODS Influenzaviruses These were from the collection of the World Influenza Centre, London, or have been described by Harkness et al. (1972). Swine/Taiwan/7251/70 and 7310/70 viruses were provided by Dr W. R. Dowdle, International Influenza Centre for the Americas, Centre for Disease Control, Atlanta, Ga., USA. Immunological techniques Haemagglutination-inhibition tests, neuramini- dase-inhibition tests, and immunoprecipitin tests were performed as described by Harkness et al. (1972). Preparation of nonspecific antisera Rabbit antiserum against purified N2 neura- minidase was prepared as described by Laver & Webster (1966) using a recombinant virus X-7 (FI) as a source of purified neuraminidase. Antiserum against the purified haemagglutinin of A/Hong Kong/68 virus was prepared as described by Schild et al. (1971). RESULTS Haemagglutination-inhibition tests The cross-reactions between the human A/Hong Kong/68 virus and the various swine influenza A viruses in haemagglutination-inhibition tests with post-infection ferret sera are indicated in Table 1. The reactions of swine/Taiwan/70 isolates were close- ly similar to those of the prototype A/Hong Kong/ 68 virus. It is of interest that there was a less close relationship between the swine/Taiwan/70 viruses and A/England/878/69, a virus strain showing a minor degree of antigenic variation from the proto- type A/Hong Kong/68 (Pereira & Schild, 1971). Minor cross-reactions in haemagglutination-inhibi- tion tests between A/Hong Kong/68 and the former Asian viruses, including A/Singapore/57 and A/Eng- land/12/64 viruses, were observed. Such cross- reactions have also been reported in previous studies (Coleman et al., 1968), although their significance remains uncertain. The swine/Taiwan/70 viruses showed the same minor cross-reactions with these former Asian virus strains as did A/Hong Kong/68. An unexpected finding was the demonstration of reciprocal cross-reactions between A/Hong Kong/ 68 and certain strains of classical swine influenza A virus-namely, swine/Cambridge/39, swine/Wiscon- sin/1/61, and swine/Wisconsin/2/66. These cross- reactions were consistently observed in several experiments but were of a lower order than the cross- reactions observed between swine/Taiwan/70 and A/Hong Kong/68. A number of other classical swine influenzaviruses, including A/swine/Iowa/1 5/30 (HswlNl), did not show such cross-reactions with A/Hong Kong/68. In order to investigate further the nature of the antigenic relationships between the haemagglutinins of the various virus strains, studies were carried out with potent rabbit antiserum prepared against the purified haemagglutinin (H3) of A/Hong Kong/ 68 virus (Table 2). In haemagglutination-inhibition tests this antiserum reacted to equally high titres with the homologous virus and with swine/Taiwan/70. An unexpected finding, in view of the cross-reactions observed with ferret antisera, was the fact that anti- serum against purified haemagglutinin failed to react with former Asian viruses, A/Singapore/1/57, A/England/12/64, and A/Tokyo/3/67. In con- trast, inhibition reactions were detected with three classical swine influenza viruses, swine/Cambridge/ 39, swine/Wisconsin/61, and swine/Wisconsin/66, although the inhibition titres were not greater than 1 % of that observed with the homologous virus. These studies thus suggested that the haemagglutinin of A/Hong Kong/68 virus showed minor immuno- logical relationships to that of certain classical swine influenzaviruses but failed to confirm any such relationship with the haemagglutinins ofother human Asian viruses. In addition, antiserum against puri- fied haemagglutinin inhibited A/equine/Miami/63 (Heq2Neq2) virus, thus confirming the evidence of minor relationships between the haemagglutinins of the equine virus and A/Hong Kong/68 des- cribed in previous studies (Coleman et al., 1968). Confirmation of these findings was obtained by the cross-reaction of the viruses in the immuno- diffusion tests described in this paper. Neuraminidase-inhibition tests The results of neuraminidase-inhibition tests with immune rabbit sera are shown in Table 3 The neura- 722 RELATIONSHIPS BETWEEN HUMAN AND PORCINE INFLUENZA. 2 0 N 0 N 0 v- 0 lt 0 0) 0' az ol I 0 IC I) 0 0 C- 0 0 co Co m aw 0 0 t oco CD 0000 Co N 0) 00~ N co N N 00co 0c 0 N 0 cc 0 co 0 0) 0 N 0 co 0 C 0 to , 0 0 CN 0o co N 0 co 000 N Co 0 CN 0 0N 0 co T-co M CoD CO 0 - - _- C c- c-, -c c 0- -. 0 0 0 0 0 0 .- - D- ° °- .3 .3 .31 ux: am o o 0 n co2 723 I CD 0 co C14I 0 N0) o Nw0) Co 0 0) co 0 N 0 co 0 N 0) m w- 0 co N 0 0 0 0 CN co 00 U-C cc C .Ec 00 0cc.CD cc, r- 00 cco uj00 co~ .C En w oo 0D ICo Cor,<c- Co 0(D C 0 0 U- Co -o cc E cc 0) 0._ CoQ cn Co Co n0 Co 0) Co .c_ N C 0 0 CN 0 0 co w 0 0 co co N N 0 0 cc cc CD N L Co Co r- r- Co o Co o Co 00 m-0 C o N (C) ._ ._C 0 QaC CX _ o~~~~~~0 0 Lo 0 co N 0o o 0(C) 0 0 N 0 CN ol Co 0 N4 cC 0 0 co 0 o 0) co 0 (N Z I o0 co - -- co - cc C C C 0 C w 1w 0 TN o10, 0Co No z I 1- CN C,-LO 0 n r- n r- onm > C E I3 (- Co r-co .00 0 G. C. SCHILD ET AL. Table 2. Haemagglutination-inhibition reactions of rabbit antiserum (treated with receptor-destroying enzyme) against pure A/Hong Kong haemagglutinin Virus strain HI titre A/Singapore/I /57 20 A/England/i 2/64 <20 A/Tokyo/3/67 <20 A/Hong Kong/i /68 25000 A/England/878/69 6 400 A/swine/Taiwan/7310/70 50 000 A/swine/Taiwan/7251/70 50 000 A/swine/lowa/30 <20 A/swine/lowa/54 <20 A/swine/lowa/58 <20 swine/Cambridge/39 120 swine/Wisconsin/i 1/61 240 swine/Wisconsin/i 2/66 80 swine/Manitoba/647/67 <20 A/equine/Miami/63 480 minidase of the classical swine influenzaviruses, in- cluding strains isolated as distantly as 1930 (Iowa/ 30) and 1967 (swine/Manitoba/67), contained im- munologically related neuraminidases. The neura- minidases of these strains showed immunological relationships with those of the human AO and Al viruses used in the tests but were unrelated to those of human A2 virus strains. In contrast, the neura- minidases of the swine/Taiwan/70 viruses were immunologically closely related to that of the A/Hong Kong/68 virus. This relationship could also be clearly demonstrated using antiserum against a recombinant virus, FPV-A/Hong Kong, contain- ing the neuraminidase of A/Hong Kong/68 and the haemagglutinin of fowl plague virus. Antiserum against this recombinant inhibited the neuramini- dases of both A/Hong Kong/68 and swine/Tai- wan/70. In neuraminidase-inhibition tests only minor degrees of cross-inhibition have been demonstrated between the neuraminidases of human A2 viruses isolated in 1957 (A/Singapore/1/57) and A/Hong Kong/68 virus (Schild & Newman, 1969). The results of the present studies indicated that the neuramini- dase of swine/Taiwan/70 virus, like that of A/Hong Table 3. Cross-reactions of human and porcine influenza A viruses in neuraminidase-inhibition tests Immune rabbit sera against: Source of Iowa Swine/ purified FPV-A/HKswnTa/ neuraminidase A/PR8 A/BEL A/FM1i owa Camb./39 A/57 A/H K/68 recombi- swine/Tai0 190 virus nidasmi nant virus 73i10/70 (N2) A/PR8/34 3 000 b 1 000 500 300 500 NTC 30 20 20 A/BEL/42 1 000 2 500 500 100 300 NT d d d A/FM1/47 750 1 000 750 100 1 000 NT d d d A/swine/lowa/30 1 000 350 500 500 2 000 NT d d d swine/Cambridge/39 750 350 500 500 3 000 NT d d d swine/Wisconsin/i /61 500 200 300 350 - NT d d d swine/Wisconsin/2/66 350 200 100 200 1 500 NT d d d swine/Manitoba/67 500 200 150 100 1 500 NT d d d A/Singpore/i /57 d d d d d 1 500 50 75 20 A/Hong Kong/68 20 d d d d 50 1 000 1 000 500 FPV-A/Hong Kong rec. d d d d d NT 2 000 1 500 500 swine/Taiwan/731 0/70 30 d d d 20 NT 1 500 1 000 750 a Virus preparations treated with pronase. b Serum dilution inhibiting 50 % of neuraminidase activity. c NT: not tested. d Titres less than 1 :20. 724 RELATIONSHIPS BETWEEN HUMAN AND PORCINE INFLUENZA. 2 Fig. 1. Immunoprecipitin reactions of rabbit antiserum against purified A/Hong Kong/68 haemagglutinin. A/Hong Kong/68 and swine/Taiwan/70 virus shared a common precipitin line corresponding to virus haemagglutinin. A/Singapore/57 virus did not react. Under the conditions of test (1 % agar) a single preci- pitin line corresponding to haemagglutinin was detected; under other conditions (1.5 % agar), however, two precipitin lines corresponding to haemagglutinin are readily distinguished (see Fig. 3). (Antigens and antisera: see legend to Fig. 2.) Kong/68 virus, showed only minor cross-reactions with A/57 neuraminidase. Immunoprecipitin tests The antigenic compositions of A/Hong Kong/68 and swine/Taiwan/70 viruses were compared in pre- cipitin tests using monospecific rabbit sera prepared against the purified haemagglutinin of A/Hong Kong/68 and against purified N2 neuraminidase. Swine/Taiwan/70 virus (Fig. 1, 2) gave precipitin lines that showed identity with the precipitin lines corresponding to the haemagglutinin and the neura- minidase of A/Hong Kong/68 virus. These studies thus provided strong support for the antigenic iden- tity of A/Hong Kong/68 and swine/Taiwan/70 viruses. Precipitin tests were also used to examine the antigenic relationships between the haemagglutinins ofA/Hong Kong/68 and swine/Cambridge/39 viruses for which minor cross-reactions were observed in haemagglutination-inhibition tests. It has been shown in previous studies (Schild et al., 1971) that Fig. 2. Immunoprecipitin reactions of rabbit antiserum against purified A2 neuraminidase. A/Hong Kong/68 (H3N2), swine/Taiwan/70, and A/Singapore/i/57 viruses gave a common precipitin line corresponding to A2 neuraminidase. Swine/Cambridge/39 virus gave no reaction. Antigens: wells marked A/57, A/67, A/HK, S/T/70, S/C/39 and Eq-63 contained paired, concentrated preparations of A/Singapore/i /57 (H2N2) A/Tokyo/ 3/67 (H2N2), A/Hong Kong/i /68, swine/Taiwan/ 7310/70, swine/Cambridge/39, and equine/Miami/ 63 virus. Antisera: Wells marked anti-A2N and anti-H K HA (H3) contain monospecific rabbit antisera for N2 neuraminidase and A/Hong Kong/68 haemagglu- tinin, respectively. under certain test conditions antiserum against puri- fied A/Hong Kong haemagglutinin produces two distinct precipitin lines. In precipitin tests with purified swine/Cambridge/39 virus, antiserum against the purified A/Hong Kong haemagglutinin gave only a single precipitin line-which, however, was continuous with one of the two lines given by A/Hong Kong/68 virus (Fig. 3). A preparation of purified A/equine/Miami/63 virus was included in the test since this virus also shows minor cross- reactions with A/Hong Kong/68 virus in haemag- glutination-inhibition tests (Coleman et al., 1968). The equine virus also gave a single precipitin line that was continuous with one of the lines given by A/Hong Kong virus. It is of interest that although both swine/Cambridge/39 and equine virus appear- ed to share the same precipitin line with A/Hong Kong/68 (i.e., the line nearest the well containing 725 G. C. SCHILD ET AL. Fig. 3. Diagrammatic representation of reactions of rabbit antiserum against A/Hong Kong/68 (H3N2) hae- magglutinin. In precipitin tests in 1.5 % agar two distinct precipitin lines were produced with homologous A/Hong Kong/68 virus. A/Singapore/i /57 (H2N2) and A/England/i 2/64 (H2N2) gave no reaction. Swine/ Cambridge/39 and A/equine/Miami/63 (Heq2Neq2) virus each gave a single precipitin line that was conti- nuous with one of the lines given by A/Hong Kong/68, indicating a partial immunological relationship of the haemagglutinins of these viruses with that of A/Hong Kong/68. Note that swine/Cambridge/39 and equine viruses both shared the same one of the two lines given by A/Hong Kong/68 but that the precipitin lines given by these two viruses showed crossing, indicating that their antigenic relationships with A/Hong Kong involved different determinants in each strain. antiserum in Fig. 3), there was crossing of the lines given by the swine and equine viruses, indicating that the shared antigenic component with A/Hong Kong virus was not identical for the two viruses. Antiserum against A/Hong Kong haemagglutinin failed to give precipitin lines when tested against a variety of human Asian viruses (A/Singapore/57, A/England/12/64, and A/Tokyo/3/67), against hu- man AO or Al viruses (A/PR8, A/BEL, A/FM1), or against fowl plague virus (Dutch strain). The findings, like the results of haemagglutination-inhi- bition tests with antiserum against purified haemag- glutinin, suggested that the haemagglutinins of A/ Hong Kong/68 virus and those of the former Asian viruses are unrelated. However, both neuramini- dase-inhibition tests and immunodiffusion tests (Schild & Newman, 1969) have indicated that the former Asian virus contains neuraminidases anti- genically related to that of A/Hong Kong/68. The properties of antiserum prepared against the purified haemagglutinin of A/BEL virus have been described previously (Schild, 1970). This antiserum gave precipitin lines with all AO and Al viruses used in the tests, indicating that the haemagglutinins of these viruses were antigenically related. Concen- trates of swine influenzaviruses Iowa/30 and swine/ Cambridge/39 failed to give precipitin lines with the antiserum against HO haemagglutinin. However, that these two swine viruses contained neuramini- dase (NI) related to that ofhuman AO and Al viruses was confirmed by precipitin tests with antiserum against purified A/BEL virus. Precipitin lines corres- ponding to the neuraminidase (NI) of AO and Al virus were detected in tests with this antiserum (Schild et al., 1969). Swine/Cambridge/39 and swine/ Iowa/30 viruses gave precipitin lines that were con- tinuous with the line corresponding to the neuramini- dase of A/BEL virus. These results thus supported the results of the neuraminidase-inhibition tests, which indicated that " classical " swine viruses con- tained neuraminidase related to that of human AO and Al viruses. In contrast, swine/Taiwan/70 virus failed to give a precipitin line corresponding to NI neuraminidase. The results of the precipitin reactions on the test strains of influenzavirus are summarized in Table 4. DISCUSSION Evidence has been presented for the complete antigenic identity of swine/Taiwan/70 and A/Hong Kong/68 viruses based on haemagglutination-inhi- bition and neuraminidase-inhibition tests and on precipitin reactions with monospecific antihaemag- glutinin and antineuraminidase sera. The swine/ Taiwan/70 virus should therefore be designated A/ swine/Taiwan/70 (H3N2). The swine/Taiwan/70 virus is of considerable interest, representing as it does the first example of an animal influenza A virus showing antigenic identity with a human pan- demic strain. The isolation of swine/Taiwan/70 virus following a human A/Hong Kong/68 epidemic in Taiwan (Kundin, 1970) and the demonstration of a high frequency of antibody to A/Hong Kong/68 virus in swine sera collected after, but not before, the appearance of the A/Hong Kong/68 virus in man in Taiwan (Kundin, 1970) and in Great Britain 726 RELATIONSHIPS BETWEEN HUMAN AND PORCINE INFLUENZA. 2 Table 4. Summary of immunodiffusion reactions of influenzavirus concentrates Production of precipitin lines corresponding to the following influenza antigens: Virus strain a HO A/Hong Kong/68 N2 Nihaemagglutinin haemagglutinin neuraminidase neuram nidase(anti-pure AO/BEL HA) (anti-pure-H3) (anti-pure A2/57 Nase) (anti-AO/BEL purifiedvirus) A/BEL/43 + - + A/FM1/46 + - - + A/Singapore/i /57 - + A/Hong Kong/68 - + b + A/swine/Taiwan/70 - + b + A/swine/Cambridge/39 - + C _ + A/equine/Miami/63 - + c a Purified concentrated virus disrupted by 1-2 % sodium dodecyl sulfate. b Production of two precipitin lines. c Partial relationship; single precipitin line showing continuity with one of the two lines corresponding to Hong Kong haemagglutinin. (Harkness et al., 1972) suggest that the A/Hong Kong/68 virus is capable of crossing species bound- aries in nature. Moreover, the swine/Taiwan/70 iso- late has been shown to be capable of infecting and inducing respiratory symptoms in human volunteers (Beare et al., 1971). The above findings have obvious epidemiological implications for influenza, as dis- cussed by Harkness et al. (1972). The present studies support the findings of Meier- Ewert et al. (1970) that the strains of swine influenza- virus isolated prior to 1970 form an antigenically homogeneous group of agents antigenically related to the classical swine/Iowa/30 strain. These " classi- cal" strains of swine influenzavirus were clearly antigenically distinct from the A/Hong Kong/68 or swine/Taiwan/70 virus and it is of interest that all the strains examined in the present study contained neuraminidase antigenically related to that of human AO viruses, thus confirming and extending the ori- ginal observations of Paniker (1968). This antigenic relationship of the neuraminidases of swine and human AO virus was detected using two independent methods, neuraminidase-inhibition and immuno- diffusion (see Table 4). Although previous studies have not revealed the existence of complete identity on the subtype-spe- cific antigens of human and animal influenzaviruses, partial relationships have been frequently reported. For example, the neuraminidases, but not haemagglu- tinins, of certain avian influenza A viruses are anti- genically related to that of human Asian viruses (Pereira et al., 1967; Webster & Pereira, 1968; Schild & Newman, 1969), while other avian viruses have been shown to contain neuraminidase related to that of human AO and Al strains (Schild et al., 1969). The haemagglutinin, but not the neuramini- dase, of equine/Miami/63 virus is antigenically re- lated to that of the human A/Hong Kong/68 virus (Coleman et al., 1968). A further partial relation- ship described for the first time in the present paper is that between the haemagglutinin of certain classical swine influenza strains and the A/Hong Kong/68 virus. The demonstration of such partial relationships has stimulated considerable speculation on the possible epidemiological relations between influenza A viruses of man and of other vertebrates (Laidlaw, 1935; Schild & Stuart-Harris, 1965; Tumova & Pereira, 1968; Pereira, 1969; Kaplan, 1969; Tumova & Easterday, 1969) but the significance of these findings still remains uncertain. In the present study immunodiffusion tests were employed in attempts to confirm the existence of antigenic relationships between the envelope anti- gens of different virus strains. This technique has the advantage that it is not susceptible to non- specific factors, such as those that affect haemaggluti- nation-inhibition tests and-particularly when mono- specific antisera are available-it enables the shared antigenic component to be identified unequivocally as haemagglutinin or as neuraminidase. Further- more, cross-reaction of influenza antigens in im- munodiffusion tests with antihaemagglutinin and antineuraminidase sera are broader than those de- tected by haemagglutination-inhibition (Schild, 1970) 727 728 G. C. SCHILD ET AL. or by neuraminidase-inhibition tests (Schild & New- man, 1969) and immunodiffusion tests may thus reveal antigenic relationships not detected by the other methods. In general, the precipitin reactions described in this paper confirmed cross-reactions detected by other methods. Of particular interest was the observation that in the case of two strains of virus that showed minor cross-reactions in haemag- glutination-inhibition tests with A/Hong Kong/68 (i.e., swine/Cambridge/39 and equine/Miami/63), mmunoprecipitin tests revealed that only one of the two antigenic components of the haemagglutinin subunit was shared between A/Hong Kong/68 and the other viruses. It is worthy of note that the pre- viously described minor antigenic relationships be- tween the haemagglutinins of " classical " swine influenzaviruses and human AO virus (Andrewes et al., 1935; Francis & Shope, 1936) could not be confirmed by immunodiffusion tests with potent antiserum against purified HO haemagglutinin, while the sharing of neuraminidase antigens by these viruses was confirmed by immunodiffusion. RE~SUME ETUDE DES RAPPORTS ENTRE LA GRIPPE HUMAINE ET LA GRIPPE PORCINE: 2. COMPARAISONS IMMUNOLOGIQUES ENTRE LE VIRUS HUMAIN A2/HONG KONG/68 ET DES VIRUS GRIPPAUX A D'ORIGINE PORCINE On a recherche les affinites antigeniques entre le virus A/Hong Kong/68 de la grippe humaine et une serie de virus grippaux A d'origine porcine en recourant a diverses epreuves: inhibition de l'hemagglutination, inhibition de la neuraminidase et immunoprecipitation A&l'aide de serums monospecifiques anti-hemagglutinine et anti-neuraminidase prepares chez le lapin. Ces etudes ont permis un certain nombre de consta- tations: a) le virus swine/Taiwan/70 renferme une hemaggluti- nine et une neuraminidase indiscernables de celles du virus A/Hong Kong/68. Cette similitude antigenique est con- firmee par l'identite des reactions fournies par les deux souches en epreuves d'immunoprecipitation utilisant des serums anti-hemagglutinine et anti-neuraminidase; b) les souches ((classiques # de virus de la grippe porcine isolees entre 1930 (souche Shope S15) et 1967 (souche swine/Manitoba/67) renferment une neura- minidase depourvue de lien antigenique avec celle du virus A/Hong Kong/68, mais etroitement apparentee a celle des virus humains AO et Al; c) l'hemagglutinine de certaines souches classiques * de virus de la grippe porcine presente une faible affinite antigenique avec l'hemagglutinine du virus A/Hong Kong/68. Les epreuves d'immunoprecipitation effectu6es a l'aide de serum anti-hemagglutinine montrent que cette affinite est limitee a l'une des deux lignes de preci- pitation fournies par le virus A/Hong Kong/68, lignes correspondant probablement a deux composants dis- tincts de l'hemagglutinine; d) en immunoprecipitation avec un serum anti- hemagglutinine, l'hemagglutinine du virus A/Hong Kong/68 apparait antigeniquement differente de celle des virus A2 de la grippe humaine isoles entre 1957 et 1967. Les donnees recueillies au cours de cette etude sont examinees sous l'angle des rapports epidemiologiques entre la grippe humaine et la grippe porcine. REFERENCES Andrewes, C. H. et al. (1935) Brit. J. exp. Path., 16, 566 Beare, A. S. et al. (1971) Lancet, 1, 305 Coleman, M. T. et al. (1968) Lancet, 2, 1384 Francis, T., Jr & Shope, R. E. (1936) J. exp. Med., 63, 645 Harkness, J. W. et al. (1972) Bull. Wid Hlth Org., 46, 709 Kaplan, M. M. (1969) Bull. Wid Hlth Org., 41, 485 Kundin, W. D. (1970) Nature (Lond.), 228, 857 Laidlaw, P. P. (1935) Lancet, 1, 1118 Laver, W. G. & Webster, R. G. (1966) Virology, 30, 104 Meier-Ewert, M. et al. (1970) J. gen, Virol., 6, 409 Paniker, C. K. J. (1968) J. gen. Virol., 2, 385 Pereira, H. G. (1969) Progr. med. Virol., 11, 46 Pereira, M. S. & Schild, G. C. (1971) J. Hyg. (Lond.), 69, 99 Pereira, H. G. et al. (1967) Nature (Lond.), 215, 982 Schild, G. C. (1970) J. gen. Virol., 9, 191 Schild, G. C. & Newman, R. W. (1969) Bull. Wid Hlth Org., 41, 437 Schild, G. C. & Stuart-Harris, C. H. (1965) J. Hyg. (Lond.), 63, 476 Schild, G. C., et al. (1969) Nature (Lond.), 222, 1299 Schild, G. C. et al. (1971) Bull. Wld Hith Org., 45, 465 Shope, R. E. (1931) J. exp. Med., 54, 373 Tumova, B. & Easterday, B. C. (1969) Bull. WldHlth Org., 41, 429 Tumova, B. & Pereira, H. G. (1968) Bull. Wld Hlth Org., 38, 415 Webster, R. G. & Pereira, H. G3. (1968) J. gen. Virol., 3, 201