THE I‘I‘EMAGGLU‘I’ININ OF INFECTIOUS BRONCHITIS VIRUS Thesis £1» $0 Dawn of DB. D. MICHIGAN STATE UNIVERSITY NiEambar Biswai 1965 ABSTRACT THE HEMAGGLUTININ OF INFECTIOUS BRONCHITIS VIRUS by Nilambar Biswal A hemagglutinin of infectious bronchitis virus (IBV) is isolated by anion exchange diethylaminoethyl-cellulose column chromatography and by diethyl ether. The hemagglutinin is a lipoprotein with traces of carbohydrate and RNA. Only chicken erythrocytes are known to be agglutinable. The isolated hemagglutinin is stable for at least 3 hours at 56 C. A pH range of 5 to 7.5 does not significantly affect the hemagglutinin. The size of the hemagglutinin is estimated to be around 60 mu. Neuraminidase-like activity is associated with the hemagglutinin which linearly liberates N-acetylneuraminic acid from N—acetylneuraminic acid-lactose in 50 minutes. Incubation for 4 hours at 57 C in 0.1M phosphate buffer, pH 6.5, is considered the optimum condition for the neuramini- dase-like activity of the viral hemagglutinin. This activity is appreciably destroyed in 45 minutes at 56 C. 'The hemagglutinin destroys receptors for trypsin- modified IBV, ether-treated IBV, and for itself. Receptors for PR-8 strain of influenza virus and NDV are not affected. 1 Nilambar Biswal Destruction of receptors by IBV neuraminidase and bacterial neuraminidase is proportionately related to the release of N-acetylneuraminic acid, and concomittant loss of electrons from the chicken erythrocytes. Neuraminidase from Vibrio cholerae destroyed the re- ceptors for trypsin-modified IBV, ether-treated IBV, IBV hemagglutinin, and PR-8 strain of influenza virus. Neuraminidase inhibited the growth of IBV-41 in chicken embryos possibly by destroying its substrates. Hemadsorption occurs when chicken erythrocytes are added to IBV-41 infected chicken embryo kidney cells. Anti-IBV hemagglutinin-rabbit serum inhibits hemag- glutination. Antigenically the hemagglutinin is related to IBV-41, and a single precipitin line is produced in agar gel diffusion test. In order to explain the behavior of the hemagglutinins and neuramindases of different viruses, it is hypothesized that (1) there may be several arrangements of the hemagglu- tinin and neuraminidase molecules of multiple configurations to constitute the ionogenic binding points on the virus, (2) the configuration of the sialic acids contained on the mucoproteins present on the erythrocytes may be dissimilar: so that only the mutual stereospecificities of the mole- cules or ionic groups can bind each other and reduce the Nilambar Biswal over all electric potential. This hypothesis may explain some of the observations that would otherwise deviate from the classical concept underlying "receptor gradient." THE HEMAGGLUTININ OF INFECTIOUS BRONCHITIS VIRUS BY Nilambar Biswal A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Microbiology and Public Health 1965 ACKNOWLEDGMENT S I am grateful to Dr. C. H. Cunningham, Professor of Microbiology and Public Health, for his valuable guidance throughout this investigation. I am thankful to Dr. L. C. Ferguson, Dr. V. H. Mallmann and Dr. H. A. Lillevik for their interest and helpful criticisms. I am indebted to Dr. J. J. Stockton, chairman of the Department of Microbiology and Public Health for his kind advice and encouragement. My sincere thanks are also due to Mrs. M. P. Spring and my fellow graduate students for their help and interest. This study was supported in part by the National Institutes of Health Grant No. AI 05549. ii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . 2 Erythrocyte. . . . . . . . . . . . . . . . . . . 2 Hemagglutination . . . . . . . . . . . . . . . . 2 Neuraminidase. . . . . . . . . . . . . . . . . . 9 Indicator virus. . . . . . . . . . . . . . . . 15 Infectious bronchitis virus. . . . . . . . . . . 17 MATERIALS AND METHODS. . . . . . . . . . . . . . . . . 21 Viruses. . . . . . . . . . . . . . . . . . . . . 21 Chicken embryo kidney cell culture . . . . . . . 22 Cultivation of virus in cell culture . . . . . . 25 Virus assay. . . . . . . . . . . . . . . . . . . 25< Normal allantoic fluid . . . . . . . . . . . . . 24 Anion exchange DEAR—cellulose column chroma- tography . . . . . . . . . . . . . . . . . 25 Preparation and purification of hemagglutinin. . 26 Purification of IBV hemagglutinin by diethyl ether. . . . . . . . . . . . . . . . . . . 27 Purification of PR-8 strain of influenza virus . 28 Chemical assay . . . . . . . . . . . . . . . . . 29 Neuraminidase assay. . . . . . . . . . . . . . . 51 Determination of N-acetylneuraminic acid . . . o 52 Paper partition chromatography of N-acetyl- neuraminic acid. . . . . . . . . . . . . . 52 Ultrafiltration. . . . . . . . . . . . . . . . . 55 Erythrocytes . . . . . . . . . . . . . . . . . . 55 Hemagglutination test. . . . . . . . . . . . 55 Hemagglutination inhibition (HI) test. . . . . . 54 Preparation of antisera. . . . . . . . . . . . . 55 Hemadsorption test . . . . . . . . . . . o . . 56 Treatment of erythrocytes for removal of sialic acids . . . . . . . . . . . . . . . . . . 56 Treatment of chicken embryos with neuraminidase for prevention of infection by IBV-41. . . 58 Electrophoretic measurements . . . . . . . . . . 58 Immunodiffusion and immunoelectrophoresis. . . . 40 iii TABLE OF CONTENTS - Continued Page RESULTS . . . . . . . . . . . . . . . . . . . . . . . 45 Diethylaminoethyl-cellulose column chromaw tography. . . . . . . . . . . . . . . . . 45 Chemical composition of IBV hemagglutinin . . . 46 Ultrafiltration . . . . . . . . . . . . . . . 47 Agglutinability of different species of etythrocytes by IBV hemagglutin . . . . . 47 Thermostability of IBV-hemagglutinin. . . . . . 48 Effect of pH on IBV hemagglutinin . . . . . . . 49 Hemagglutination inhibition test. . . . . . . 50 Effect of IBV hemagglutinin on the receptors on chicken erythrocytes. . . . . . . . . . . 50 Effect of neuraminidase on the receptors on chicken erythrocyte . . . . . . . . . . . 52 Effect of neuraminidase on the infectivity of IBV- 41. . . . . . . . . . . . . . . . . 55 Effect of IBV hemagglutinin on the electro- kinetic charge on chicken erythrocyte . . 55 Release of sialic acid from erythrocytes by IBV hemagglutinin . . . . . . . . . . . . . . 59 Characteristics of IBV neuraminidase. . . . . . 59 Hemadsorption due to IBV-41 . . . . . . . . . . 62 Antigenic relationship of IBV hemagglutinin to IBV-41 in allantoic fluid and NAF . . . . 62 DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 66 SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 77 BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . . 79 iv LIST OF TABLES TABLE Page 1. Procedure for Chemical Assay of IBV-Hemagglu- tinin. . . . . . . . . . . . . . . . . . . . 50 2. Absorption Spectra at 260 and 280 mu, and HA Activity of Fractions Eluted from IBV—41 Infected Allantoic Fluid by DEAE-cellulose Column Chromatography. . . . . . . . . . . . 44 5. Infectivity of Pooled Samples Collected at 0.1, 0.5, and 0.5 M NaCl from DEAE-cellulose Column Chromatography. . . . . . . . . . . . 46 4. Chemical Composition of IBV Hemagglutinin. . 46 5. Filterability of IBV-Hemagglutinin and Its Approximate Size . . . . . . . . . . . . . . 47 6. Agglutinability of Different Species of Erythrocytes of IBV Hemagglutinin. . . . . . 47 7. Effect of Temperature and Time on IBV Hemagglutinin . . . . . . . . . . . . . . . 48 8. Effect of pH on IBV Hemagglutinin. . . . . . 49 9. The Development of Hemagglutination Inhibit- ors Against IBV Hemagglutinin and IBV in Rabbit . . . . . . . . . . . . . . . . . . . 50 10. Effect of IBV Hemagglutinin on the Receptors on Chicken Erythrocytes. . . . . . . . . . . 52 11. Effect of Neuraminidase on the Receptors on Chicken Erythrocytes . . . . . . . . . . . . 52 12. Effect of Neuraminidase on the Growth of IBV-41 in Chicken Embryos. . . . . . . . . . 55 15. Electrophoretic Mobility of Erythrocytes Treated with IBV Hemagglutinin, PR-8 Strain of Influenza Virus and Neuraminidase, at pH 6.4 and F/2, 0.0726. . . . . . . . . . . . . 55 LIST OF TABLES - Continued TABLE 14. 15. 16. 17. 18. Page Surface Charge Density and Electrons Lost from Chicken Erythrocyte Due to IBV Hemagglutinin, PR-8 Strain of Influenza Virus, and Neuraminidase. . . . . . . . . . . 57 Amount of NANA Released Due to IBV Hemagglu- tinin, PR-8 Strain of Influenza, and Neuraminidase . . . . . . . . . . . . . . . . 58 Comparison of the Physical and Chemical Changes on the Chicken Erythrocyte Treated with Various Reagents . . . . . . . . . . . . 58 Hydrolysis of NANA-L as a Function of Time. One mg. of NANA—L was Incubated with IBV- Hemagglutinin for Various Times at 57 C . . . 59 Effect of Heat on Neuraminidase Associated with IBV Hemagglutinin at 56 C. NANA-L was Used as a Substrate . . . . . . . . . . . . . 62 vi FIGURE 1. LIST OF FIGURES Absorption spectra at 260 and 280 mu, and HA activity of fractions eluted from IBV~41 in- fected allantoic fluid by DEAE-cellulose column chromatography. . . . . . . . . . Development of hemagglutination inhibitors against IBV-hemagglutinin and IBV in rabbit serum. . . . . . . . . . . . . . . . . . . . Inhibitory effect of neuraminidase (y, cholerae) on IBV-41. . . . . . . . . . . Hydrolysis of NANA—L as a function of time. One mg of NANA-L was incubated with IBV hemagglutinin for various times at 57 C. . Effect of pH for liberation of NANA from NANA-L due to neuraminidase activity associ- ated with IBV—hemagglutinin in a 0.1 M phos~ phate buffer adjusted to appropriate pH. . . Adsorption of chicken erythrocytes onto the chicken—embryo-kidney-cells infected with IBV-41 . . . . . . . . . . . . . . . . . . Diffusion of IBV hemagglutinin and IBV 1n~ fected allantoic fluid against antiwIBV hemagglutinin rabbit serum . . . . . . . . . Diffusion of IBV hemagglutinin, IBV infected allantoic fluid and NAF against anti-IBV rabbit serum . . . . . . . . . . . . . . Antigenic relation of IBV hemagglutinin, IBV infected allantoic fluid and NAF as demon‘ strated by immunoelectrophoretic test. . . . vii Page 45 51 54 60 61 65 64 64 65 INTRODUCTI ON Infectious bronchitis virus (IBV) does not cause direct hemagglutination. Trypsin-modified virus aggluti~ nates chicken and turkey erythrocytes but not those of certain other species (Corbo and Cunningham, 1959). Hemagglutination by trypsin-modified virus is attributed to either a structural reorganization of the protein coat of the virus that exfoliates some ionogenic group or to the destruction of some inhibitors, ordinarily present in the allantoic fluid, that mask the hemagglutinin, thus favoring its attachment to the surface of the erythrocyte. In the present study an attempt has been made to (i) isolate and characterize the hemagglutinin of infectious bronchitis virus, (ii) determine the hemagglutinin—erythro~ cyte interaction at the cellular and molecular level, and (iii) determine the antigenic relation of the hemagglutinin to the whole virus particle and to normal allantoic fluid. LITERATURE REVIEW Erythrocyte The surface of the erythrocyte offers a unique tool for observation of many biological phenomena. The cell stroma has an inner fibrous layer and an outer layer of plaques (Moskowitz and Calvin. 1952) consisting of long thin rods composed of a lipid-protein-carbohydrate complex (Hiller and Hoffman, 1955). The latter serves as the re- ceptor to which certain hemagglutinins from a variety of sources may become attached. Hemagglutinatign The ability of influenza virus to agglutinate chicken erythrocytes was a significant discovery in virology (Hirst, 1941; McClelland and Hare, 1941). Hirst (1942) demonstrated that influenza virus adsorbed onto the chicken erythrocyte and formed bridges between adjacent erythro- cytes with resulting agglutination. After a reaction period dependent upon time and temperature, the virus eluted from the surface of the erythrocyte. It was soon dis- covered that hemagglutination was not a characteristic of all viruses and that not all erythrocytes from mammals and birds were agglutinable. However, agglutination of various species of erythrocytes has been demonstrated for many members of most groups of viruses (Rosen, 1964). Hemadsorp— tion is a related phenomenon in which erythrocytes can absorb onto the surface of virus-infected cells (Shelokov gt_al., 1958; Hotchin et al., 1960) or single cells (Marcus, 1962). Indirect hemagglutination (Lenette, 1959), another related phenomenon, is due to the action of specific immune serum on the erythrocyte previously treated with tannic acid and to which viral or rickettsial antigen has been adsorbed. The hemagglutinin is described in this review as a biological entity associated with the virus. It may be associated with or separated from the infectious portion of the virus (Buckland and Tyrrell, 1965). Rosen (1964) has classified viruses causing hemagglutination into eight groups. 1. ArthrOpodmborne viruses. These viruses contain RNA. The virus particle itself is possibly the hemagglutinin (Sabin and Buescher, 1950), and very specific conditions are necessary for hemaggluti- nation. The commonly used erythrocytes are those from newly hatched chicks or adult geese (Rosen, 1964), and they are not known to contain any enzymatic activity (Anderson, 1959). Some viruses of this group cause hemadsorption (Buckley, 1959). Group A Arbor viruses possess an hemagglutinin associated with the infectious portion of the virus, but an additional hemagglutinin has also been separated from the virus particle by calcium phosphate column chromatography (Smith and Holt, 1961). Sindbis virus has two virus-specific hemagglutinins: one extractable by ether and the other associated with the infectious portion of the virus (Mussagay and Rutt, 1964). 2. Adenoviruses. These DNA containing (Pereira et al., 1965) viruses agglutinate erythrocytes, with the exception of the mouse and avian types (Rowe and Hartley, 1962). The receptors for adenoviruses are destroyed by receptor destroying enzyme (RDE) (Kasel et_al., 1960), but the enzymatic mechanism of destruction of receptors by the virus may be different. Philipson (1960) separated three components of adenovirus by anion exchange diethylaminoethyl-cellulose (DEAE-cellulose) column chromatography. Indirect hemagglutination with tanned sheep erythrocytes coated with adenovirus type 5 has been demonstrated (Lefkowitz and Sigel, 1965). 5. Enteroviruses. The hemagglutinin of Enteric Cytopathic Human Orphan (ECHO), Coxackie A and B viruses of man, the GDV 11 strain of Theiler's virus, encephalomyocarditis virus and simian adenoviruses (Rosen, 1964) is associated with the virus particle which contains ribonucleic acid (RNA). The hemagglutinin elutes from the erythrocyte. Receptor destroying enzyme does not inactivate ECHO virus receptors (Goldfield et al., 1959; Buckland and Tyrrell, 1965). Two kinds of hemagglutinating particles are produced by ECHO 12 virus. The particle with a buoyant density of 1.55 is associated with the infectious portion and the other is related to the non-infectious empty capsids (Halperen E£1§l , 1964). Simian enterovirus (SV2) agglutinates rhesus monkey erythrocytes at 4 C and room temperature but not at 57 C (Heberling and Cheever, 1965). 4. Poxviruses. With vaccinia, variola, ectromelia and a virus iso- lated from a pox disease of monkeys, the hemagglutinin is distinctly separate from the infectious portion, which conw tains DNA. Lecithinase inactivates the hemagglutinin which may be a lipoprotein. McCrea and O'Loughlin (1959) separated the vaccinia virus hemagglutinin from the infectious portion by anion exchange DEAE-cellulose column chromatography. Cassel (1957) isolated a nonwhemagglutinating mutant (HA7) from (HA+) neurovaccinia virus in cell culture. Oda (1964) described a plaque hemadsorption technique for genetic studies of the virus with regard to an HA marker. Both vaccinia and variola viruses are capable of hemadsorption (Shelokov gt_al,, 1958). 5. Psittacosis group. This group of viruses contains DNA. The meningOn pneumonitis strain of ornithosis virus, murine pneumonitis virus, feline pneumonitis virus and psittacosis virus (Rosen, 1964) possess a hemagglutinin that may be a lipo~ protein which agglutinates mouse and chicken erythrocytes (Barron gtmal., 1965). These viruses agglutinate best at 57 C and not at 4 C. 6. Reoviruses. These double stranded RNA-containing viruses (Gomatos and Tamm, 1965) agglutinate human 0 and sometimes bovine erythrocytes but not chicken or guinea pig erythrocytes (Macrae, 1962). Papain (Buckland and Tyrrell, 1965) trypsin and chymotrypsin (Newlin and McKee, 1965) remove the re- ceptors for reoviruses from the surface of human erythrocytes. Rhim §£_gl., (1965) have described hemagglutination by reovirus propagated in cell lines. 7. Myxoviruses. Members of this group of viruses contain RNA (Frommhagen §t_ai,, 1959). There are two primary subgroups, influenza viruses and paramyxoviruses (Wilner, 1964), though many other viruses antigenically related to these sub- groups, have been included in this group. All of them do not necessarily cause hemagglutination (Andrews, 1964). Hemagglutination by myxoviruses, especially by those of influenza, parainfluenza and NDV, has been intensively studied. They agglutinate erythrocytes from a variety of sources but chicken, guinea pig and human erythrocytes are most agglutinable by the virus at 4 C. The hemagglutinin is associated with the virus particle containing the enzyme, neuraminidase (Burnet g£_al., 1946). The interaction of influenza virus and the erythrocyte is similar to the formation of an enzyme substrate complex (Hirst, 1942). The initial attachment is ionic in nature and the hydroxyl group of a polysaccharide chain present on the substrate is responsible (Buzzell and Hanig, 1958). The virus finally elutes from the erythrocyte due to the action of neuraminidase on the receptor of the erythrocyte. According to Hanig (1948) there is a decrease of the net surface electric charge of the human erythrocyte and the electrophoretic mobility which may be attributed to any of the following: (1) loss of charged substances from the cell surfaces, (2) change in Spatial configuration of charged groupings on the cell surface, or (5) adsorption by the erythrocyte of charged elements of the solvent. Hanig also concluded that the reduced electrophoretic mobility was a result of the destruction of the virus receptors on the erythrocyte surface. Bateman §t_al., (1956) attributed the reduced electrophoretic mobility to the formation of addi- tional free, positively-charged groups on the surface of the erythrocyte. Ada and Stone (1950), Stone and Ada (1952) reported that the reduction in electrophoretic mobility after treat- ment of human erythrocytes with myxoviruses and RDE was in close approximation to the "receptor gradient" proposed by Burnet et al., (1946). However, Newcastle disease virus and swine influenza virus in allantoic fluid deviated from their normal position in the receptor gradient. This indi- cated that there may be some inhibitors for these viruses in the allantoic fluid, and that the loss of agglutin- ability was not a direct function of the total residual electric charge on the erythrocyte surface. Cook et al., (1961) demonstrated that the surface charge density on the human erythrocyte is reduced mainly because of the release of sialic acid due to neuraminidase action upon the human erythrocyte. Madoff and Eylar (1961) reported that the negative charge on the erythrocyte is mainly due to the presence of sialic acid of varying concen- tration and various kinds in different species of animals. There is a linear relationship between the per cent of total sialic acid released and the per cent reduction in the net surface charge on human erythrocyte (Eylar §t_al., 1962). Furchgott and Ponder (1941) found that the electro- kinetic behavior of human erythrocytes is mainly associated with the lipid fraction of the cell stroma and that the phosphoric acid group of cephalin may be responsible for such behavior. According to Bangham et a1. (1958) phOSphate groups on lipids are responsible for the negative electric charge on the erythrocytes. Heard and Seaman (1960) and Seaman and Heard (1960) consider the surface structure of human erythrocytes to be a macropolyanion imparting a definite electric charge which can be modified by opposite ion association and possibly by adsorption of hemolysate. The authors attribute the negative charge on the erythrocyte principally to a carboxyl or sulphate group or a mixture of both. At a physiological pH and ionic strength there is no positive grouping and evidence for the phosphate groups being mainly responsible for the net negative charge is circumstantial. Crystalline trypsin releases a mucopeptide containing sialic acid (Cook et al., 1960) which is thought to be related to the decrease in the electrophoretic mobility of the trypsin-treated erythrocyte (Ponder, 1951; Seaman and Heard, 1960). Only a small fraction of plasma membranewsituated N-acetylneuraminic acid contributes to the net negative charge on the HeLa-S5 cells (Marcus and Salb, 1965). Neuraminidase, The receptor destroying enzyme (RDE). Receptor destroying enzyme (Burnet and Stone, 1947) or neuraminidase (Gottschalk, 1957), or sialidase (Heimer and Mayer, 1956) is an induced (adaptive) exo-enzyme of Vibrio cholerae that has played an important role in a 10 better understanding of some of the biological concepts underlying the virus—erythrocyte receptor reaction, inhi- bition of hemagglutination, and inactivation of both receptor and inhibitor by virus. This enzyme is defined as the Specific a-glycosidase which cleaves the a—ketosidic link- age of a terminal N-acylated neuraminic acid to an adjacent sugar residue (Gottschalk, 1958). Neuraminidase can be pre- pared in crystalline form from a variety of bacteria (Schramm and Mohr, 1959; Ada and French, 1959), influenza virus (Mayron et al., 1961), and from a variety of mammalian and avian cells (Ada g£_al., 1965). Neuraminidase with Specific activity of 20 units per mg of protein can be pre- pared from culture filtrates of Clostridium perfringens. This fraction can be precipitated with ammonium sulfate at 60 to 75 per cent saturation (Johnson §£_al., 1964). Purified sialidase (neuraminidase) from the A2 strain of influenza virus is electrophoretically heterogenous (Seto and Hokama, 1964). Ada et al0 (1965) demonstrated the immunological relationship of the neuraminidase present in Lee strain of influenza virus and neuraminidase present in normal chorioallantoic membranes (CAM). Neuraminidase present in normal CAM has a sedimentation coefficient of 5 s and is susceptible to low concentrations of dodecyl sulfate. Lee virus neuraminidase is stable to this detergent, but the neuraminidase activity of MEL, BEL, and PR—8 strains of 11 influenza virus is susceptible to low concentrations of dodecyl sulfate (Laver, 1965). Lee virus grown in calf kidney cells has about twice the neuraminidase activity per HA unit of the same virus grown in chicken embryo (Laver, 1965). The sediment obtained from normal allantoic fluid has no neuraminidase activity, whereas analogous sediment of virus in allantoic fluid was regularly able to split N-acetylneuraminic acid from Collocalia mucoid (Howe §£_§l., 1961). Neuraminidase activity of the Asian and PR—8 strains of influenza virus can be dissociated by trypsin (Mayron et al., 1961) and chymotrypsin (Wilson and Rafelson, 1962). These results are in direct contrast to those obtained by Stone (1949), who reported that trypsin inactivated neuraminidase of y, cholerae and neuraminidase of Lee, influenza B, MEL, WSE and swine influenza viruses. A struc- tural difference among the various strains of virus was considered to be responsible for this discrepancy (Mayron et al., 1961). Neuraminidase has been considered to be involved at some stage or stages in the multiplication of influenza virus. Hoyle (1950) suggested that the enzyme attacks the cell membrane intracellularly and the newly formed virus particles are released. Cairus and Mason (1955), experi— mentally supported this view. On the other hand, Schlesinger and Karr (1956) suggested that the viral enzyme was in- volved in the synthesis of new viral material. 12 Padgett and Walker (1964) isolated a variant of influenza B strain of Lee virus which differs from the parent population only in characteristics related to neuraminidase. The enzymatic activity was manipulated by changing the temperature and calcium ion concentration, and they suggest that the enzyme does not participate in the adsorption, penetration and eclipse phase. It helps release the virus to the medium. Neuraminidase prevents infection of the chicken embryo (Stone, 1948) and cells from the chicken embryo (Johnson et al., 1964) against a variety of strains of influenza virus. The results support the proposition that Specific receptor sites containing sialic acid on suscept~ ible cells are required for adsorption and attack by in- fluenza virus, and that different viruses vary in the extent to which they depend on the intactness of these receptor Sites (Gottschalk, 1960). This protective effect is, however, a temporary one, since the receptor sites tend to regenerate over a period of time (Fazekas de St. Groth, 1948; Johnson et al., 1964). Sialidase antiserum did not inhibit the embryo adapted Asian strain of influenza virus but inhibited the mouse—adapted strain of the same virus (Smith and Rasmussen, Jr., 1965). Burnet (1942), proposed a "receptor gradient," or the order in which the myxoviruses may be graded according to their receptors. According to Hirst (1959), there are 15 two hypotheses to explain the receptor gradient: (1) several different kinds of receptors are on the erythro- cyte surface and different kinds of viral enzymes may be involved, although there is yet no evidence for such multiplicities, and (2) accessibility of similar receptors on the erythrocyte surface for viral enzymatic action. This does not explain the situation where the hemaggluti~ nation is not a direct function of the total residual electric charge on the erythrocyte surface. Two influenza A2 strains are known to behave as "+” viruses, i;§., 'V. cholerae filtrate could not destroy their receptors (Takatsy and Barb, 1959; Choppin and Tamm, 1960; Buckland and Tyrrell, 1965). Inhibitors (Ada and Stone, 1950) or mucoprotein receptor substances having different configur- ations (Howe gt_al., 1961) may be responsible for such discrepancies. Indicator virus. Francis (1947) reported that influenza virus heated at 56 C for 50 minutes agglutinates erythrocytes but does not elute from them. Francis identified this as ”indicator virus." Since the inception of this term the "indicator" state of viruses has been correlated with the loss of "enzyme activity." Neuraminidase activity is equated with the receptor destroying capacity and eluting ability and, accordingly, different members of the myxoviruses can be 14 classified (Howe §£_al,, 1961) as those in which (1) both hemagglutinating activity and enzyme activity are lost after only a few minutes at 56 C, e. ., NDV and the NWS strain of influenza virus; (2) the hemagglutinin is rela- tively stable but the enzymatic activity is lost after 15 minutes at 56 C (GL, FMI, Lee strain of influenza virus); (5) the hemagglutinin is thermostable, neuraminidase is fairly thermostable but disappears well before the hemag- glutinin (PR-8, PR 501, and swine influenza); (4) both hemagglutinin and neuraminidase Show much thermostability and in certain cases neuraminidase is more stable than the hemagglutinin, 343., A 2/Jap., Parainfluenza (Darrell and Howe, 1964). Purified sialidase (neuraminidase) from A2 strain of influenza, at 0 to 5 C dissociated and simul- taneously lost 90 per cent of its activity (Seto and Hokama, 1964). Mayron §£_gl., (1961) and Howe et al., (1961) sug- gested that the hemagglutinin and neuraminidase are two separate entities of the same influenza virus particle, and reside in separate covalently bonded structures (Laver, 1965). Normal serum strongly inhibits hemagglutination by indicator virus but not by non—heated virus. The inhibi— tors identified thus far are mucoproteins (Hirst, 1959) and some mucolipids (Rosenberg et al., 1956). MuCOproteins are defined as conjugated proteins with multiple hexosamine 15 containing oligosaccharides or small polysaccharides as the prosthetic groups. The prosthetic groups are covalently linked to the protein core (Gottschalk, 1954). Inhibitory mucoproteins are present in a variety of sources such as human and rabbit serum, ovarian cysts, sheep salivary gland (McCrea, 1948, 1955a), normal human urine (Tamm and Horsfall, 1952) , tissue extracts (Hirst, 1959), sputum mucoid, brain mucoid and Collocalia mucoid (Howe et al., 1961), meconium from infants (Curtain et al., 1955; Pye, 1955; Zilliken et al., 1957), and erythrocyte stroma (Howe, 1951; Howe et al., 1957; McCrea, 1955b). Bayer (1964) reported that hemagglutination inhibi- tory mucoprotein from human urine is filamentous with a diameter of 40 A to 240 A composed of small fibrils. The interaction between influenza virus and inhibitory muco— protein consists of attachment of a molecular fiber to the projections of the virus at several sites and frequently on more than one virus particle. All these known inhibitors, however, lose their activity when treated with intact myxoviruses, neuramini— dase, trypsin or periodate (0.001 M) (Gottschalk, 1960). Sialic acid, a group name for acylated neuraminic acids, is released from the inhibitors through the action of neuraminidase. Neuraminic acid is the basic unsubstituted structure, C9H1708N, common to all the inhibitors. Bentonite or Kaolin can remove non-Specific inhibitors for 16 certain viruses present in the sera (Bussell et al., 1962). Schmidt et al. (1964) reported that an enzyme-like sub- stance from a psychrophilic psuedomonad can inactivate non- specific HA inhibitors for certain strains of viruses in human and animal sera. Not all sialomucoproteins are inhibitory. To qualify as a virus hemagglutinin inhibitor, the mucoprotein must have a substrate for neuraminidase, and compete success- fully with the receptor on the cell surface for the virus. The successful competitor, either the erythrocyte receptor or the mucoprotein, exerts the more attractive force and has the relatively greater number of functional groups for the virus particle (Gottschalk, 1960). Bogoch et al. (1962) demonstrated that a novel com- pound, sialoresponsin, accumulates in the allantoic fluid during the first few minutes, in response to influenza virus infection. This compound contains neuraminic acid but is distinct from free N-acetylneuraminic acid and interferon. Viral neuraminidase has to be inactivated in order to detect sialoresponsin (Bogoch and Kaufman, 1965). Mucolipids from ox brain (Folch et al., 1951) and human brain have a low degree of inhibitory capacity towards PR—8 strain of influenza (Rosenberg et al., 1956) and PR-501 (Howe et al., 1961) indicator virus. Diethyl ether has been successfully used to disinte- grate influenza viruses (Hoyle, 1952; Lief and Henle, 1956; 17 Paucker et al., 1959; Hoyle et al., 1961) and to purify the hemagglutinin (Mitzutani et al., 1962). The hemagglu— tination activity of the virus is greater after treatment with sodium deoxycholate (Laver, 1965). Inhibition of hemagglutination can be increased Sixteenfold over that observed with untreated virus and chicken erythrocytes by using ether treated equine influenza virus and pigeon erythrocytes (Berlin §£_al., 1965). Measles, canine distemper and rinderpest virus are members of the somewhat heterogenous group of myxoviruses (Andrews, 1964). They have an RNA core, are sensitive to ether, and are similar in size, structure and cytopatho- genicity (Gillespie, 1962; Karzon, 1962). Measles virus has large and small "populations” of hemagglutinin particles. The large hemagglutinins are usually associated with the infectious particle (Norrby, 1965). The virus agglutinates and lyses monkey erythrocytes. Hemadsorption in measles- infected cells has been demonstrated (Karzon, 1962). Rinderpest virus is not known to cause hemagglutination (Plowright, 1962), and canine distemper virus gives irregu- lar results (Gillespie, 1962). Infectious bronchitis virus. Infectious bronchitis virus is the etiological agent of a respiratory disease of chickens (Schalk and Hawn, 1951). The virus is a.sphere with a diameter of 80 to 120 18 millimicrons (Reagan and Brueckner, 1952; Nazerian, 1960; and Berry §£_al., 1964). The virus exists in two phases, the thermolabile D phase and the thermostabile 0 phase (Singh, 1960). The optimal stability of viral infectivity is at pH 7.8. The isoelectric point is about pH 4.05 (Cunningham and Stuart, 1947). The approximate density of the virus is 1.15 (Buthala, 1956). In cesium chloride density gradient, the virus has a density of 1.25 (Tevethia, 1964). The virus can readily be cultivated in chicken embryos. The viru- lence of the virus for chicken embryos is increased through serial passage but the antigenicity for chickens is decreased (Beaudette and Hudson, 1957; Cunningham, 1957). The Beaudette strain or egg-adapted strain of IBV can be cultivated in chicken embryo kidney cells (CEKC), chicken embryo fibroblasts (Spring, 1960), whole embryo cell culture (Mallmann, 1960), and in the isolated choricallantoic membrane (Ferguson, 1958; Ozawa, 1959). The virus forms plaques on CEKC cultures (Wright and Sagik, 1958; Cunningham and Spring, 1965). The virus contains RNA, and forms syncytia within 24 hours after infection of CEKC. DL-p—fluorOphenylalanine inhibits the formation of syncytia and viral synthesis, whereas, aminopterin does not affect the viral development (Akers, 1965). 19 Fluorescent antibody studies reveal viral antigen in the cytoplasm in 2é-hours (Stultz, 1962), 56 to 48 hours and in the nucleus, 7 hours post-inoculation (Mohanty et al., 1964). The virus is ether sensitive (Akers, 1965; Mohanty et al., 1964). The virus in allantoic fluid has three antigens designated as 1, 2, and 5. Ether, sodium dodecyl sulphate, and heat at 100 C disintegrate the virus into these components, whereas DEAE-cellulose column chroma- tography and cesium chloride density gradient ultracentri- fugation can be used to separate the three antigens (Tevethia, 1964). The immunodiffusion technic cannot be used to differentiate strains of IBV, but cross neutrali- zation test can be used for this purpose (Tevethia, 1964). Infectious bronchitis virus does not cause direct hemagglutination. Modification of the virus with trypsin induces hemagglutination (Corbo and Cunningham, 1959) of erythrocytes from turkeys and from chickens older than three weeks. The same receptors on the chicken erythrocyte may be involved with both influenza and infectious bron- chitis virus (Muldoon, 1960). The hemagglutinin is associated with the virus particle (Nazerian, 1960). Specific inhibition of hemagglutination by anti-IBV serum has not been accomplished (Corbo and Cunningham, 1959; Muldoon, 1960). Trypsin, sodium or potassium periodate (0.9 to 0.1 M), zymosan, and RDE do not remove inhibitors 20 of the hemagglutinin present in the normal and immune chicken sera (Muldoon, 1960). Only trypsin-modified IBV adsorbs to and agglutinates chicken erythrocytes according to indirect fluorescent anti- body technic (Stultz, 1962) and electron microscopy (Nazerian, 1960). Trypsin-modified IBV in allantoic fluid reduced the electrophoretic mobility of chicken and turkey erythrocytes by 21.4 per cent and 17.4 per cent, respectively. Virus not treated with trypsin, normal allantoic fluid and trypsin- modified normal allantoic fluid do not affect the electro— phoretic mobility of the chicken and turkey erythrocytes. Trypsin does not destroy the receptors for PR-8 strain of influenza virus, Newcastle disease virus, RDE and trypsin- modified IBV. The electrophoretic mobility of RDE-treated chicken erythrocytes is not appreciably changed when further treated with trypsin-modified IBV or PR-8 strain of influenza virus (Biswal, 1965). Neuraminidase-treated virus does not induce aggluti- nation of chicken erythrocytes. Ether disintegrates the virus, and sodium dodecyl sulphate removes the spikes of the virus at lower concentration. Disrupted virus does not cause hemagglutination (Berry §£_al., 1964). Brown et al. (1962) reported that horse erythrocytes can be modified with tannic acid to give an indirect hemagglutination test for two strains of IBV. MATERIALS AND METHODS Viruses Infectious bronchitis virus strains 41 and 42 (IBV-41 and IBV-42), the Michigan State University code for the Massachusetts and Beaudette strains, respectively, Newcastle disease virus (NDV) and PR-8 strain of influenza virus were used. The viruses were cultivated in ten-day-old embryo- nating chicken eggs, inoculum 0.1 ml per egg. The allantoic fluid from the IBV-41 infected eggs were harvested 72 hours postinoculation, whereas IBV-42, NDV, and PR-8 strain of influenza virus were harvested 48 hours postinoculation. The pooled viruses were stored at -72 C in sterile screw cap vials. All cultures were tested for bacteriological sterility in Brewer's thioglycolate medium (Difco). The embryo infectious dose, 50 per cent end point (EIDSO) of the viruses were as follows; IBV-41, 106‘6 to 107°6 ; IBV-42, 107 to 108; NDV, 107 to 108; PR-8 strain of influ- enza virus, 107'5 to 108's. At the time of use the viruses were thawed at room temperature, centrifuged at 1400 g for ten minutes. The supernatant fluid was then removed and used for experi— mental purposes. Infectious bronchitis virus—41 21 22 7th passage) and IBV-42 (111th passage) were also cultivated in chicken embryo kidney cell culture. Chicken embryo kidney cell culture. The procedure was that described by Cunningham (1965)- The kidneys from 16 to 17 day old chicken embryos were carefully removed and washed several times with Hank's balanced salt solution (BSS) containing phenol red. The kidneys were then cut into small pieces, and the blood clots and other tissue debris were removed. The washed kidney tissues were transferred to a 500 ml trypsinizing flask containing a Teflon covered magnet. A 0.25 per cent sus- pension of trypsin (1:250) (Difco) in BSS, pH 7.8 to 8.2, was added to the trypsinizing flask at the rate of 10 ml per pair of kdineys, and the contents were stirred Slowly for 1 hour at room temperature. The cell suspension was filtered through two layers of cheese cloth and then centrifuged at 200 g for 6 to 8 minutes. The supernatant fluid was discarded and the cells were suspended in fresh B88 and centrifuged again. This process was repeated twice. One ml of packed cells was suspended in 100 ml of cultural medium 199 containing vitamins, amino acids in- cluding L-glutamine, 0.1 per cent sodium bicarbonate, 100 units of penicillin, 100 ugm of streptomycin and 50 units of mycostatin per ml. The cell suspension was filtered through cheese cloth and newborn calf serum was added to a 25 final concentration of 5 per cent. The final concentration of cells was approximately 107 cells per ml. Four milliliters of the cell suspension were then plated per plastic tissue culture petri dish (15 mm x 60 mm). The cells were incubated in an atmOSphere of 8 per cent C02 and 80 to 85 per cent relative humidity at 57CL 2 atmospheric changes per hour. A monolayer of cells was formed by 48 to 60 hours. Cultivation of virus in cell culture. Infectious bronchitis virus-41 (9th passage), and IBV- 42 (112th passage) CEKC culture adapted strains were used. The monolayer of cells was washed once with 5 ml of B88 without phenol red and inoculated with 0.5 ml of the stock virus (IBV-42) diluted to contain 106 to 107 plaque forming units (PFU) per ml. The virus was allowed to adsorb to the cells for 2 hours at 57 C. The fluid was then drained out of the petri dish and 4 ml of the cultural medium without calf serum was added to each dish. Virus assay. 1. In chicken embryos. The EIDSO of the viruses was determined by inoculating 10 day old embryonating chicken eggs with serial 10-fold dilutions of the viruses in sterilized nutrient broth. Five embryos were inoculated per dilution via the allantoic cavity, 24 and were incubated for 2 to 5 days at 57 C, depending upon the particular virus. Mortality, curling and dwarfing and other pathological evidences including urate deposits in the mesonephros were the criteria used to calculate the 50 per cent infective end point according to the method of Reed and Muench (1958). 2. Plaque assay. Only infectious bronchitis virus—42 was used. The monolayer of CEKC was washed once with 5 ml of 858 without phenol red and then inoculated with 0.5 ml of serial 10~fold diluted virus. Usually four cell cultures were used per dilu- tion. After incubation for 90 minutes the fluid was drained off the cells and the monolayer was overlayed with 4 ml of 0.9 per cent Difco Noble agar in cultural medium. After the agar had hardened the cultures were incubated for about 72 hours. One-half ml of a 0.1 per cent solution of neutral red in phosphate-buffered saline was added to the cultures which were then incubated at 57 C for an hour followed by 2 hours at 4 C. The titer of the virus was expressed as plaque forming units (PFU) per ml. Normal allantoic fluid. Normal allantoic fluid (NAF) was collected from 15-day- old embryonating chicken eggs to serve as the control for the virus contained in the allantoic fluid. 25 Anion exchange DEAE-cellulose column chromatography. Anion exchange DEAEwcellulose (0.78 meq/g; Cellex D, California Corporation for Biochemical Research, Los Angeles, California) was suspended in 0.0067 M phosphate buffer, pH 7.4 (initial buffer) and the fine particles removed by re- peated decantation at room temperature. Preliminary investi- gations were conducted with glass columns (2 cm x 20 cm) fitted with fritted glass discs and some glass beads. The column was packed with the adsorbent under positive pressure (1 to 5 pounds pressure of dry nitrogen) so as to give a flow rate of 15 ml per hour. The adsorbent column was further washed with the initial buffer until the effluent was pH 7.4. Sometimes it was necessary to wash the column with 0.01 N NaOH followed by the initial buffer to maintain a constant pH. Infectious bronchitis virus-41 in allantoic fluid was dialyzed against large volumes of the initial buffer at 4 C for 24 to 48 hours. Twenty ml of the dialysate was then carefully layered on top of the column. Fractions from the column were eluted with 0.01 M, 0.05 M, to 1.0 M in 0.05 M increments NaCl in 0.0067 M phosphate buffer, pH 7.4. Five 4-ml fractions for each molarity of sodium chloride were collected by an automatic fraction collector. Each fraction was analyzed for absorbancy at 260 and 280 mu by a Beckman DB spectrophotometer. The fractions 26 were dialyzed against phosphate buffered saline (0.0067 M Sorensen's phosphate buffer containing 0.145 M NaCl), pH 7.4, for 24 hours at 4 C, and analyzed for hemagglutinating activity. Fractions with maximum absorbancy at 260 and 280 mu were tested for infective virus. Preparation and purification of hemagglutinin. Preliminary investigation established that the eluate at 0.1 M NaCl in initial buffer contained maximum HA activity. The techniques were developed to prepare and purify hemagglu— tinin of IBV in large quantities. Glass column, 4 x 60 cm, fitted with fritted glass disk and glass beads was packed with DEAR-cellulose as previously described, but to give a flow rate of 45 to 65 ml per hour. After the column was flushed with the initial buffer contain- ing 0.02 M NaCl, one volume of the virus preparation was then added to the column. Two volumes of the initial buffer containing 0.02 M NaCl was then passed through the column. Much of the extraneous protein material was removed this way. When no more protein could be eluted, initial buffer containing 0.15 M NaCl was passed through the column. Three to 4 volumes of this buffer usually assured elution of all the hemagglutinin from the column. The eluates were dialyzed against large volumes of the initial buffer for 24 to 48 hours at 4 C. A sample of the dialysate was tested for hemagglutinating activity and the 27 remaining portion was passed through a 450 mu pore size Millipore filter. The filtrate was freezeudried and stored at 4 C until further use. At the time of use, the material was reconstituted to at least 2 mg protein per ml diluent appropriate for the specific tests to be performed. This was considered the stock solution, and cOntained 2000 1.100 HA units per mg protein. When necessary it was further diluted for specific experiments to be performed. Purification of IBV hemagglutinin by diethyl ether. The method followed was according to Mizutani QEmEI. (1962). The allantoic fluid containing the virus was concen~ trated by dialysis against polyethylene glycol at 4 C for 24 hours. The dialysate was mixed with three volumes of diethyl ether and stirred with a magnetic stirrer at 4 C for 24 hours. Excess ether was removed by bubbling dry nitrogen through the solution. The aqueous phase and the aqueous~ether interphase were dialyzed against large volumes of 0.15 M sodium acetate, pH 7.6, at 4 C for 48 hours. The dialysate was mixed with an equal volume of 4 per cent fresh lanthanum acetate solution in 0.15 M sodium acetate buffer, pH 6.8, and was let stand for 2 hours at 4 C. The suspension was centrifuged at low speed for 50 minutes and the precipitate was discarded. The supernatant fluid 28 was dialyzed against 0.25 M sodium citrate, pH 7.4, for 24 hours and then tested for HA activity. Purification of PR-8 strain of influenza virus. The red cell method (Frommhagen and Knight, 1959) was slightly modified as follows: Allantoic fluid containing the virus was centrifuged at 2000 g to remove extraneous materials. The supernatant fluid was again centrifuged at 109,000 g. for 50 minutes at 5 C in a Spinco model L preparative ultracentrifuge. The resulting pellet was redissolved in the minimum quantity not exceeding 5 ml of hemagglutinination buffer (HA buffer) (Difco), pH 7.5, which contained 1 part of Sorensengs phos~ phate buffer and 9 parts of 0.145 M NaCl solution. One volume of the virus was added to 5 volumes of a 4 per cent suSpen~ sion of washed chicken erythrocytes. The virus-erythrocyte suspension was allowed to settle at 4 C for about 2 hours and then most of the supernate was carefully decanted. The remaining supernate and erythrocytes were separated by centrifugation at 2000 g for ten minutes. The packed erythro‘ cytes were resuspended in about 5 times their volume of HA buffer at 4 C, and then incubated for 4 hours at 57 C. After centrifugation in a clinical centrifuge to sediment the erythrocytes, the supernate containing the eluted virus was recentrifuged at 109,000 g for 50 minutes. The pellet was reconstituted in HA buffer, pH 7.5. This suspension of the virus contained about 105°01 HA units per ml. 29 Chemical assay. Five hundred milligrams of the freeze-dried material was reconstituted in 20 ml of déonized distilled water (Table 1). Ten ml of the suspension was dried in a hot air oven to a constant weight, and the remaining 10 ml was tested for carbohydrates, lipid, protein, and nucleic acids as described below. Carbohydrate The quantitative anthrone reaction (Morris, 1948) was used. Standards were made with 10 to 500 ug of glucose (Reagent grade), (Baker Chemical Co., Phillipsburg, N. J.) per ml of the aqueous solution. _§roteip The hot TCA extract for protein assay by the Folin- (Ziocalteu Phenol reagent was modified by Lowry et al. (3.951) was used. Standards were prepared with 10 to 500 L53 of bovine serum albumin (National Biochemical Corp., Clxaveland, Ohio). Riloonucleic acid (RNA). The RNA was assayed with the Orcinol reagent accord- irug to Schneider (1945). Standards were prepared with 10 to 20C) ug of yeast RNA (General Biochemicals, Laboratory Park, Chakgrin Falls, Ohio) in each ml of 0.1 per cent of NaECOS. 50 Table 1. Procedure for Chemical Assay of Hemagglutinin 500 mg hemagglutinin in 20 ml water ~v , T 10 ml suspen31on 10 ml suspen51on (dry in hot air oven to constant weight) 10 ml 10% cold tri~ chloroacetic acid (TCA) Centrifuge at 15,000 g & for 15 minutes V . 4% Supernatent Prec1p1tate Determine total carbohydrate (Morris, 1948) 20 ml ethanol; ether (5/1) (Zwartouw, 1964) Incubate for 15 minutes at 50 C under hood Centrifuge at 15,000 g for 15 minutes Superngtant ReSIdue Evaporate for total lipid 10 ml of 5% TCA and incubate for 50 minutes in boiling water Centrifuge at 15,000 g A Supaénatant Pregipitate Determine RNA (Schneider, 1945) and DNA (Keck, 1956) Suspend in 10 ml of distilled water Determine total protein (Lowry et al., 1951) 51 Deoxyribonucleic acid (DNA). The diphenylamine reaction (Keck, 1956) was used for quantitative assay. Standards were prepared with calf thymus DNA (General Biochemicals, Ohio) in 5 mM NaOH. Dilutions were made with 0.5 N HC104. Neuraminidase assay. Neuraminidase was assayed according to Johnson 3; al. (1964). N-acetylneuraminic acid-lactose (NANA-L), (General Biochemicals, Ohio), 0.2 mg, was dissolved in 0.4 ml of 0.1 M phosphate buffer, pH 6.5. Onemtenth ml of the viral preparation was added and the mixture was incubated at 57 C for 1 hour. With neuraminidase from V. ghglggag (General Biochemi» cals, Ohio), instead of phosphate buffer, 0.1 M acetate buffer containing 0.01 M calcium acetate was used, Fiveztenth m1 of ice cold 5 per cent phosphotungstic acid in 0-1 M HCl was added to stop the reaction and the N-acetylneuraminic acid (NANA) released was determined according to Warren (1959). Controls were prepared (1) without substrate, (2) withm out the enzyme or the enzyme source. One unit of neuraminidase is defined as the amount liberating 1 ug of NANA from NANA-L in 1 minute under the Specified conditions. 52 Refsrmigsfiga_of N-acetylasgrsminis acid. N-acetylneuraminic acid was assayed by the thiobarbi- turic acid method of Warren (1959). Standards were prepared with 5 ug to 50 ug of NANA (Sigma Chemical Co., St. Louis, Mo.). ‘329§£_22£titigulshrgnéigszeshx_9£ E:§EEEXlQ§BE§TlflIS_39£§v Identification of the NANA was made by descending paper chromatography according to Svennerholm and Svennerhclm (1958). Whatman no- 1 chromatography papers were washed with chloroformzmethanol (2:1, v/v). The papers were dried and aboutflfiul of NANA was spotted on the starting line- A con‘ trol of standard NANA was also placed at one end of the paper. The solvent system consisted of n~butanolsnwpr0panol:0.1 N HCl (122:1, v/v)- Descending chromatography was performed for about 18 hours and the strips were dried at room temperature. The position of the NANA was determined by spraying Ehrlich S reagent (0.5 g of pmdimethylaminobenzal~ dehyde and 5.0 g of TCA dissolved in 20 ml of ethanol: water, 1:1, and then diluted with 60 ml of n butanol) (Svennerholm and Svennerholm, 1958). The paper strips were dried at 100 C for 10 to 15 minutes. 55 Ultrafiltration. Millipore filters of 450, 500, 100, 50 and 10 mu average pore diameter were used. Hemagglutinating buffer, 5 ml, was passed through the filters to satisfy the adsorp~ tion capacity of the filters. Then 5 ml of the hemagglu~ tinating fraction of the virus was passed through the filters, and the filtrates were analyzed for HA activity, Erythrocytes, Blood from Single Comb White Leghorn Cockerel, human (type 0), cow, horse, sheep, dog, and rabbit was used. The blood was collected in tubes containing 1 ml of a 2 per cent (w/v) sodium citrate solution for each 6 ml of blood, centrifuged immediately. and the buffy coat and plasma were removed. One volume of the packed erythrocytes was washed three times by centrifugation for 10 minutes per wash uSing about 50 volumes of HA buffer- After the last wash, the buffer was removed from the packed erythrOm cytes which were then stored at 4 C for as long as 2 days for electrophoretic tests and for 4 days for HA tests. Esmsgs 122192319139. Stu Modification of IBV in allantoic fluid by trypsin was based on the procedure described by Muldoon (1960). Allantoic fluid containing the virus was thawed at room temperature, centrifuged at 1400 g for 10 minutes and the 54 supernatant fluid collected. To 2 volumes of the super- natant fluid, one volume of 1 per cent trypsin was added. After the mixture was incubated in a water bath at 56 C for 50 minutes, one volume of 1 per cent egg white trypsin inhibitor (ETI) was added, and the mixture was incubated at room temperature for at least 15 minutes. This prep: aration of the virus served as a control for its hemagglutin~ ability. The same procedure was used for all hemagglutination tests (Cunningham, 1965). Serial two»fold dilutions of the virus were prepared in HA buffer. To each of a series of 12 x 75 mm tubes was added 0.25 ml each of diluted virus, HA buffer, and 0.5 per cent erythrocytes in HA buffer. The tubes were shaken for about 10 seconds and then incubated for 1 hour at room temperature- The HA titer expressed as HA units was the reciprocal of the highest dilution of the virus in which the hemagglutination was complete. Hemagglutination Inhibition (HI) test. Two procedures were used (Cunningham, 1965): Decreasing virus, constant serum method- This procedure was similar to that described for the HA test except that 0.25 ml of serum, diluted to 1:5 or 1:10, was substituted for 0.25 ml of HA buffer. The control tube contained 0.25 ml of serum, 0.25 ml HA buffer, and 55 0.25 ml of 0.5 per cent erythrocyte suspension. The serum titer was the reciprocal of the lowest dilution of the virus in which the hemagglutination was completely inhibited. The HI titer of the serum was computed as follows: ‘11 £251.12} fer. . x dilution of the serum. Serum titer Constant virus, degreasing_§erum meghgg. A constant amount, 10 HA units, of virus was used for each decreasing concentration of serum ranging from 1:5 through 1:2560. The same proportions of the reagents were used and the procedure was the same as described for the other method. The serum titer was the reciprocal of the highest dilution of the serum in which hemagglutination was completely inhibited. The HI titer was computed as follows: Serum titer x number of HA units. Preparation of antisepa. A preparation of purified hemagglutinin containing about 5 mg protein per ml and 2000 i.100 HA units per mg protein was injected intramuscularly into two rabbits, inoculum 1 ml per rabbit. Two other rabbits were inoculated with the untreated virus in allantoic fluid. Two rabbits served as controls. Blood was collected from the rabbits on alternate days through the 20th day and the sera were used immediately for H1 tests. The remaining sera was stored at —20 C until used. 56 Hemadsorption test. The cultural medium was removed from the CEKC culture petri dishes and replaced with 4 ml of a 0.1 to 0.4 per cent suspension of chicken erythrocytes at 4 C. After 50 to 45 minutes at 4 C, which was usually sufficient for hemadsorption, cells were washed gently with copious amounts of cold phosphate buffered saline, pH 6.1 (Marcus, 1962). The cells were dried with ethyl alcohol, stained with wright's stain, and microscopic examinations were made to determine if hemadsorption had occurred. Treatment of erythrocytes for removal of sialic acids. 1. With IBV hemagglutinin and PR-8 strain of influenza virus: Chicken erythrocytes were washed three times with phosphate buffered saline (PBS), pH 6.4. The packed erythro~ cytes were suspended in an equal volume of PBS after the third cycle of centrifugation. The total volume of erythro— cytes was determined with a Wintrobe hematocrit. The total number of erythrocytes was determined with a Neubauer counting chamber. To the 50 per cent suspension of the erythrocytes, 5 volumes of the purified PR-8 strain of influenza virus or IBV hemagglutinin, each containing 1000 HA units per 0.5 ml, was added. The mixture was incubated for 2 hours at 57 4 C and then for 2 hours at 57 C. The incubation process was repeated. The erythrocytes were centrifuged and the supernatant fluid was saved for determination of NANA. A portion of the packed erythrocytes after incubation was further washed once with the PBS, pH 6.4, and used for electrophoretic measurements. Another portion of the packed erythrocytes treated with IBV hemagglutinin only, was further washed three times with the HA buffer, pH 7.5, and a 0.5 per cent suspension was made in HA buffer. These erythrocytes were saved for HA titrations of the PR-8 strain of influenza virus, NDV, trypsin-modified IBV, ether-treated IBV, and the IBV hemagglutinin itself. The supernatant fluid containing the Sialic acid almost invariably contained some hemoglobin. The hemoglobin was deproteinized by the addition of an equal volume of 10 per cent (w/v) trichloroacetic acid (TCA) solution (Cook, et al., 1961). The mixture was centrifuged at 2000 g for 10 minutes at 4 C. The supernatant fluid was assayed for NANA, by the thiobarbituric acid method (Warren, 1959) and was identified by paper partition chromatography (Svenner- holm and Svennerholm, 1958). 2. With neuraminidase from Vibrio cholerae: One volume of the 50 per cent erythrocyte suspension was mixed with 5 volumes of neuraminidase, 100 units of the enzyme per 0.5 ml of a buffer containing 0.145 M NaCl, 58 0.005 M CaClg buffered to pH 7.0 with 0.5 M aqueous NaHC03 (Cook, et_§l,, 1961) and incubated for 2 hours at 57 C. The erythrocytes were used for electrophoretic measure— ments and for HA tests. Treatment of chicken embryos with neuramini- dase foryprevention of infection by IBV-41. One-tenth ml of the appropriate concentration of neuraminidase from y, cholerae was injected into the allantoic cavity of 10-day-old chicken embryos. Two hours after incu- bation at 57 C, 0.1 ml of IBV-41, EIDSO 104's, in allantoic fluid was injected into the allantoic cavity of 10 embryos for each set of tests. The embryos were incubated for 48 hours at 57 C and the allantoic fluid was harvested. Hemagglutination tests were performed with the pooled allan- toic fluid from each set. The percentage protection was expressed as follows: HA titer of the control—HA titer of the test groups x 100 HA titer of the control ' Electrophoretic measurements. The electrophoretic mobility of the erythrocytes was measured by direct microscopic observation in a Northrop- Kunitz flat, horizontal type cell apparatus (Arthur H. Thomas Co.) using a Vokam Power Pack (Shandon Scientific Co., Ltd., London). The optical system consisted of a monocular microscope with 4 mm, 0.66 NA high dry objective 59 and 10 x eye piece with a micrometer disc. The zinc elec~ trodes were cleaned with water to remove the deposition on ' them after being used. The electrophoresis cell was thoroughly washed with acid cleaning solution, rinsed with double distilled water. It was then coated with 1 per cent aqueous gelatin solution, followed by the buffer to be employed before each individual electrophoretic test. The buffer (Eylar et_al., 1962) con- sisted of 0.0059 M NagHPO4, 0.0108 M of NaHgPO4, 0.0441 M of NaCl, and 0.201 M of sucrose per liter, with an ionic strength of 0.0726, pH 6.4. The two stationary levels of the electrophoresis cell were determined after the interior of the cell was coated with the gelatin solution (Abramson et al., 1942). Measurements made at the two levels were similar. After several preliminary electrophoretic tests in which the measurements were essentially the same, it was more convenient to make readings at only one of the two levels. Usually ten random samples were counted for each test and the results were averaged. Sometimes the direction of the current was reversed to reduce the convection effect. Tests were performed first with normal erythrocytes. The erythrocytes were washed twice with the buffer and were introduced into the cell as a 1 per cent suspension. All measurements were made within 1 to 2 minutes. The electrophoretic mobility, V, of the erythrocyte is expressed as the velocity per electric field Strength 40 (X) in volts per cm. distance traveled V = veloc1ty / X = time / x X = I/ q KS, where I = current in amperes; q = cross sectional area of the electrophoresis cell in cm2; and KS = specific conduct- ance of the buffer. In this experiment the electric field Strength, X, was 9.4 volts per cm. Immunodiffusion and immunoelectrophoresis. The method of Hirschfeld (1960 was used. Preparation of agar gel. Two per cent Noble agar (Difco) was prepared in dis- tilled water and stored at 4 C until used- Buffers A discontinuous buffer system was used. Buffer in the Buffer in the Constituents electrode chamber agar layer Diethylbarbituric acid 1.58 g. 1.66 g Sodium barbiturate 8.76 10.51 Calcium lactate 0.584 1.556 Distilled water to make 1 liter 1 liter The pH of the buffer was adjusted to 8.1. Preparation of the immunoelectro» phoresis slides. Two parts of the buffer was diluted with one part of distilled water. The agar was melted and mixed with the 41 diluted buffer (1:1, v/v). Two ml of the melted agar at 60 C was poured onto clean 26 mm x 76 mm microscope slides and allowed to harden 50 minutes at room temperature, and for at least 4 hours at 4 C in a humid chamber. Antigen wells were punched by a Pasteur pipette having a bore diameter of 1.2 mm at the tip. An antibody trough, 2 mm wide, was cut with a single-edged razor blade. Aliquotes from the virus preparations were placed on the antigen wells on each of the immunoelectrophoresis slides. Controls con” sisted of normal allantoic fluid. The slides were then placed in the electrophoretic apparatus and a constant current of 1.5 mA per slide was applied for 115 minutes at 4 C. Immunodiffusion. After electrophoresis, rabbit antiserum collected on 10th day, was introduced into the antibody trough with a Pasteur pipette. After diffusion at 4 C in a humid chamber for 48 hours, a diagram of the precipitin bands was made and the slides were dialyzed against large volume of HA buffer, pH 7.5, at 4 C for 48 hours. The slides were then dialyzed against large volumes of distilled water at 4 C for 12 hours with 5 changes of water. The slides were allowed to dry at room temperature under a moistened filter paper. The slides were stained with triple stain for 5 minutes. Excess stain was removed with 10 per cent (w/v) 42 TCA. The slides were dried at 57 C and the immunoelectro- phoretic patterns were recorded. Immunodiffusion on plates. Ten ml of the 2 per cent melted agar mixed with an equal volume of HA buffer, pH 7.5, was poured on Kodak lantern cover glass (10 cm x 8 cm). After the agar had solidified at room temperature, the plates were incubated at 4 C for 4 hours and 0.5 cm diameter wells were made. The bases of the wells were sealed with a drop of 1 per cent agar. Antibody, 0.5 ml, was placed in the center well and 0.5 ml of the antigens was placed in the side wells. After about 4 to 6 hours, another 0.5 ml of the samples were placed in their respective wells. Diffusion was carried out for at least 5 days at 4 C for the development of precipitin lines. RESULTS The HA titers of the trypsin-modified and ether treated IBV—41 were 1280 to 2560, and 2560 to 5120, respectively. When IBV-42 and NAF were treated similarly, hemagglutination did not occur. Diethylaminoethyl-cellulose column chromatography. Table 2 shows the ultraviolet absorption spectra at 260 mu and 280 mu and the hemagglutination of the fractions eluted at different molarities of NaCl. Hemagglutination occurred with fractions collected at 0.05 M NaCl through 0.15 M NaCl. Maximum HA occurred when 0.1 M NaCl was used (Fig. 1). Maximum absorbancy of ultraviolet light occurred with th fraction (0.5 M NaCl) the 14th fraction (0.1 M NaCl), 54 and with the 54th fraction (0.5 M NaCl). In order to test for infectivity, fractions 15, 14 and 15; 55, 54 and 55; and 52, 55 and 54 were pooled to represent each of the above respective molarities of NaCl. These pooled samples were dialyzed for 24 hours at 4 C against large volumes of the initial buffer, passed through a 450 mu diameter Millipore filter, and a portion of each sample was again tested for HA activity. Only the eluate at 0.1 M NaCl was HA positive. The filtrates were 45 44 Table 2. Absorbtion Spectra at 260 and 280 mu and HA Activity of Fractions Eluted from IBV Infected Allantoic Fluid by DEAE-cellulose Column Chromatography Absorbancy Absorbancy M Tube HA M Tube HA NaCl No 260 280 Titer NaCl No. 260 280 Titer 0.02 1 0.10 0.11 0 0.55 56 0.50 0.47 0 2 0.12 0.15 0 57 0.28 0.58 0 5 0.10 0.11 0 58 0.40 0.49 0 4 0.05 0.06 0 59 0.29 0.41 0 5 0.02 0.04 0 40 0.51 0.40 0 0.05 6 0.07 0.09 0 0.40 41 0.50 0.57 0 7 0.10 0.12 0 42 0.22 0.29 0 8 0.12 0.15 0 45 0.21 0.27 0 9 0.19 0.24 4 44 0.20 0.21 0 10 0.17 0.18 8 45 0.12 0.15 0 0.10 11 0.11 0.16 8 0.45 46 0.12 0.14 0 12 0.28 0.50 16 47 0.15 0.17 0 15 0.98 1.10 128 48 0.18 0.20 0 14 1.20 1.60 256 49 0.28 0.48 0 15 1.10 1.40 128 50 0.48 0.49 0 0.15 16 0.85 0.90 52 0.50 51 0.50 0.51 0 17 0.67 0.81 8 52 0.57 0.59 0 18 0.70 0.72 8 55 2.4 1.8 0 19 0.10 0.12 0 54 2.2 2.1 0 20 0.11 0.15 0 55 0.20 0.22 0 0.20 21 0.10 0.11 0 0.55 56 0.26 0.27 0 22 0.08 0.08 0 57 0.28 0.28 0 25 0.08 0.09 0 58 0.50 0.51 0 24 0.09 0.09 0 59 0.21 0.22 0 25 0.08 0.09 0 60 0.22 0.25 0 0.25 26 0.10 0.11 0 0.60 61 0.21 0.21 0 27 0.08 0.09 0 62 0.11 0.11 0 28 0.10 0.12 0 65 0.12 0.15 0 29 0.11 0.14 0 64 0.12 0.14 0 50 0.14 0.17 0 65 0.11 0.12 0 0.50 51 0.19 0.51 0 0.65 66 0.09 0.1 0 52 0.29 0.67 0 67 0.09 0.11 0 55 1.6 2.6 0 68 0.08 0.1 0 54 1.8 2.8 0 69 0.085 0.09 0 55 0.50 0.50 0 70 0.08 0.09 0 The absorbancy of the fractions up within the range of 0.05 to 0.09. did not hemagglutinate. to 1 M NaCl was These fractions 45 18:11 EB Ofi Om ONH Oma OON Odm 0mm .mnmmumoumfiouso CESHOU mmoadaamoamdmo >9 pedam Ufloucmaam Umpommcfl H¢I>mH Eonm Umusam mGOHuomum mo >ua>fluum «E van .18 0mm can com um muuummm coeumuomnm Homz NO 558302 mm.o v.0 mm.o m.o mw.o 4.0 mm.o m.o mm.o m.o mfi.o a.o no.0 No.0 6 h e e 4 1.151211. quEDZ coauomum Hfi 00 H0 ON .a .mHm u HmUHB dm.lllu :8 0mm um mochHOmnd..l.. :8 Com um wochHOmnm ...... _ 1 4 _ u Aoueqzosqv 46 serially diluted 10-fold and tested for infectivity in chicken embryos. Only the eluate at 0.5 M NaCl was infective (Table 5). The allantoic fluid from the embryos inoculated with the undiluted eluates was treated with trypsin for HA activity. Again, only the eluate at 0.5 M NaCl was positive (Table 5). Table 5. Infectivity of the Pooled Samples Collected at 0.1, 0.5 and 0.5 M NaCl from DEAE-cellulose Column Chromatography. HA Titer of Allantoic M Fluid from Infected NaCl EIDSO Embryos 0.1 0 0 0.5 0 0 0.5 103-5 520 Chemical composition of IBV hemagglutinin. The hemagglutinin is a lipoprotein, containing traces of carbohydrate and RNA without any detectable amount of DNA (Table 4). Approximately 550 mg of freeze-dried hemagglu- tinin can be collected from a liter of IBV-41 infected allantoic fluid. Table 4. Chemical Composition of IBV Hemagglutinin Constituents Per Cent Dry Weight Protein 51.2 Lipid 54.2 Carbohydrate 0.11 RNA 0.01 DNA 0.0 47 Ultrafiltration. The hemagglutinin passed through filters with pore diameters of 100 mu or greater, but did not pass through a 50 mu diameter pore filter (Table 5). Table 5. Filterability of the IBV Hemagglutinin and its Approximate Size Pore Diameter of the HA Titer of the Filter in mu Filtrate 450 1024 500 1024 100 1024 50 0 Agglutinability of different species of erythrogytes. Only chicken erythrocytes were agglutinable at 4 C, 25 C, and 57 C (Table 6). Sometimes prozone phenomenon was observed at 25 and 57 C. Table 6. Agglutinability of Different Species of Erythro- cytes by IBV Hemagglutinin HA Titer at Species of Erhthrocytes 4 C 25 C 57 C Chicken 1024 1024 1024 Cow, dog, horse, sheep, rabbit and man 0 0 0 48 Thermostability of the IBV hemagglutinin. The freeze-dried hemagglutinin in sealed ampules was stable for at least 5 months at -70 C, -20 C, and 4 C without any decrease of HA activity when the sample was reconstituted to its original volume. When 2000 i_100 HA units per ml of HA buffer was incubated for varying periods, the hemagglutinin was stable for 5 to 4 days at 4 C, 1 to 2 days at 25 C, 4 to 8 hours at 57 C, and 5 to 4 hours at 56 C (Table 7). Table 7. Effect of Temperature and Time on IBV Hemagglutinin Temperature in C Time 4 25 57 56 0 2048 2048 2048 2048 5 min * 2048 1024 2048 10 min * * 2048 1024 15 min 2048 2048 1024 1024 50 min * * 1024 1024 45 min * * 1024 1024 1 hr 2048 2048 512 1024 2 hr 2048 1024 1024 1024 5 hr * 1024 1024 1024 4 hr * 1024 512 128 8 hr 2048 1024 512 < 4 12 hr 2048 1024 < 4 < 4 24 hr 2048 1024 < 4 * 48 hr 2048 256 * * 72 hr 1024 16 * * 96 hr 16 < 4 * * * Not done During the experiments it was observed that the hemagglu- tinin sometimes crystallizes at 4 C in 24 to 48 hours. When incubated at 56 C for 50 to 45 minutes, the inhibition of 49 hemagglutination in the initial dilution tubes (prozone phenomenon) did not occur. Effect of pH on IBV hemagglutinin. In phosphate buffered saline solution, at 0.07 ionic strength but at different pH levels, hemolysis occurred at pH 4.5 or less. The hemagglutinin precipitated at pH 4 and sometimes at pH 4.5. From pH 5.0 to 7.5 there was no Signifi~ cant effect of hydrogen ion concentration on the hemagglutinin. At pH 8.0 and above, there was marked progressive decrease in HA activity (Table 8). Table 8. Effect of pH on IBV Hemagglutinin "U m E r-3 '4. r1. (D H ”- O I‘P U'I * UIOUIOUIOUICNOUTOU‘IOO P O [\3 HS. OOCOCOCDODQNNOEO'DUIUICN I—‘P *- Hemolysis 50 Hemagglutination inhibition test. The HI titer of the normal rabbit serum was 40. The titer of the anti-hemagglutinin rabbit serum increased to a h and 10th days (Table 9), after maximum of 520 on the 8t which the titer progressively decreased to the original on the 18th day. The titer of the anti—IBV rabbit serum in~ creased to a maximum of 160 on the 10th day and on the 12th day (Table 8), and decreased to 80 to continue through the th 20 day (Fig. 2). Table 9. The Development of Hemagglutination Inhibitors Against IBV—Hemagglutinin and IBV in Rabbit HI Titer of Time in Anti-hemagglutinin Anti-IBV Days Rabbit Serum Rabbit Serum 0 40 40 2 20 40 4 80 40 6 160 80 8 520 80 10 520 160 12 160 160 14 80 80 16 80 80 18 40 80 20 40 80 Effect of IBV-hemagglutinin on the receptors on chicken erythrocype. Receptors on the chicken erythrocytes for PR-8 strain of influenza virus and NDV were not affected, and the HA titer was the same as that with the normal erythrocytes. Hl Titer 51 560 — 520 — 280 —' **———‘ 240 — 200 — 160 ’ 120 T ' 40}.— O /O————O/ -——O———O I I 1 1 I 1 J 1 1 l 2 4 6 8 10 12 14 16 18 20 Time in Days Fig. 2. Development of hemagglutination inhibitors against IBV hemagglutinin and IBV in rabbit. 52 The receptors for the trypsin-modified IBV (T-IBV), ether- treated IBV (E-IBV), and the IBV-hemagglutinin (HeIBV) itself were not destroyed and the erythrocytes were not reagglutin~ able (Table 10). Table 10. Effect of IBV Hemagglutinin on the Receptors on Chicken Erythrocytes HA Titer with Normal Erythrocytes Treated Viruses Erythrocytes with H-IBV PR-8 2048 2048 NDV 512 512 T-IBV 1024 < 4 E-IBV 2048 < 4 H-IBV 1024 < 4 Effect of neuraminidase on the receptors on chicken eryphrogytes. Chicken erythrocytes treated with neuraminidase from .y. cholerae did not have any receptors available for H~IBV, T-IBV, E-IBV, and PR-8 strain of influenza virus (Table 11). Table 11. Effect of Neuraminidase on the Receptors on Chicken Erythrocytes HA Titer Normal Neuraminidase treat- Hemagglutinin Erythrocyte ed Erythrocytes H-IBV 1024 0 T-IBV 1024 0 E-IBV 2048 0 PR-8 2048 0 55 Effect of neuraminidase on the infectivity of IBV—41. When 0.008 or more units of neuraminidase were used inhibition of infection was 90 per cent or more. With 0.002 unit, there was a 50 per cent inhibition of infection (Fig. 5). Neuraminidase at a concentration of 0.001 unit or less did not inhibit growth of IBV-41 in chicken embryos (Table 12). Table 12. Effect of Neuraminidase on the Growth of IBV-41 in Chicken Embryos Units of Neuraminidase HA Titer of Per Cent Inhi- per Embryo Allantoic Fluid bition of HA 0 1280 0 0.0001 1280 0 0.0005 1280 0 0.001 1280 0 0.002 640 50 0.004 160 87.5 0.006 520 75.0 0.008 40 96.87 0.01 80 95.75 0.02 40 96.87 0.04 40 96.87 0.50 40 96.87 Effect of IBV hemagglutinin on the electro~ kinetic charge on chicken erythrocytes. The electrophoretic mobility of chicken erythrocytes, 1.14, was reduced 51.2 per cent to 0.784 u sec‘l volt‘l -1 cm , after treatment with IBV hemagglutinin; 46 per cent, to 0.616 u sec‘l, after treatment with PRm8 influenza virus; Percent Inhibition 100 90 80 70 60 50 40 50 20 10 54 l 1 I 1 1 1 1 1 1 1 0.002 0.006 0.01 0.014 0.018 Unit of Neuraminidase Fig. 3. Inhibitory effect of neuraminidase (y, cholerae) on IBV-41 in chicken embryo. 55 and 66 per cent to 0.588 u sec-l, after treatment with neuraminidase from y, cholerae (Table 15). Table 15. Electrophoretic Mobility of Erythrocytes Treated with IBV Hemagglutinin, PR-8 Strain of Influenza Virus and Neuraminidase, at pH 6.4 and F/2,0.0726 Electrophoretic Per Cent Mobility Reduction of Reagents u sec‘l volt”l cm'“l Mobility None 1.14 0 H-IBV 0.784 51.2 PR-8 0.616 46.0 Neuraminidase 0.588 66.0 The surface charge density on the surface of erythrOM cyte was calculated according to Abramson §£1§l° (1942). v=§<§+ri>, where V = the electrophoretic mobility in u sec‘l volt“l cmwl, n = the viscosity in poise, 6 = the surface charge density in electrostatic units (esu) per cma, ri = the mean radius of the counter ion (sodium ion), and k = the DebyewHuckel function in cm'l. In order to use this equation it was assumed that (1) the erythrocyte acts like a particle of very large radius of curvature so that ka >> 1 (Abramson egflal., 1942) a being the radius of the erythrocyte and (2) the equation is applicable at ionic strength of 0.0726. 56 The electrophoretic mobilities were corrected to the viscosity of water at 25 C by multiplying 1.28, the rela~ tive viscosity of the electrophoresis buffer. The relative viscosity of the electrophoresis buffer was measured by an Ostwald's viscometer at 25 C. At 25 C in water, k = 0.527 x 108 JF72 in cm‘l, where F/2 = the ionic strength of the buffer. The radius of the sodium ion, r1, was assumed to be 2.67 A (Hunter, 1960). A typical example to calculate 6, the surface charge density: V U (l/k + ri) (j : At 0.0726 ionic strength, (1/k + ri) = 14.05 R. Vn = 1.46 x 10‘”2 u sec"l volt‘l cm“; X poise, and 6 was computed as 5120 esu cm‘a. The total electric charge per erythrocyte was calcum lated by multiplying the charge density by the area of the erythrocyte and with the knowledge that one unit of electronic charge, the charge of an electron, is equal to 4.80 x 10‘‘10 esu. The area of chicken erythrocyte is 225 pa according to Eylar §;_g;. (1962). The surface charge density on the normal chicken erythrocyte was 5120 esu cm‘g. This was reduced, with a consequent loss of electrons from each of the erythrocyte, due to the action of IBV hemagglutinin. PR-8 strain of influenza virus, and neuraminidase (Table 14). 57 Table 14. Surface Charge Density and Electrons Lost from Chicken Erythrocyte Due to IBV Hemagglutinin, PR—8 Strain of Influenza Virus, and Neuraminidase ————__.-—_--a Number of Number of Charge Electrons Electrons Density per Lost per Reagents esu cm'2 Erythrocyte Erythrocyte None 3120 1.46 x 107 ~- H-IBV 2146 1.01 x 107 4.50 x .106 PR-8 1686 7.90 x 106 6.70 x 106 Neuraminidase 1062 4.98 x 106 9.62 X 106 Release of Sialic acid from erythrocytes by IBV-hemagglutinin. The hemagglutinin of IBV released 25.4 ug of NANA from 8.9 x 109 erythrocytes in comparison to 55.58 ug by PR-8 strain of influenza virus and 59.60 ug by neuraminidase (Table 15). Assuming that the sialic acid released due to these agents to be free NANA of molecular weight 510, the number of NANA molecules released was maximum with eruthro- cytes treated with neuraminidase and minimum with erythro~ cytes treated with IBV hemagglutinin. If it is assumed that the net negative charge on chicken erythrocyte is mainly due to NANA, then theoretically, each electron lost should equate with the carboxyl (COO') group of a molecule of NANA released. The ratio of the 58 Table 15. Amount of NANA Released Due to IBV Hemagglutinin, PR-B Strain of Influenza, and Neuraminidase Total NANA, ug, Molecules of Released per NANA Released per Reagents 8.9 x 109 Erythrocytes Erythrocyte H-IBV 25.4 5.12 x 106 PR—8 55.56 1.17 x 107 Neuraminidase 59.60 1.50 x 107 experimental and theoretical yields of NANA was greatest in the case of PR-8, and lowest in the case of IBV hemagglutinin (Table 16). Table 16. Comparison of the Physical and Chemical Changes on the Chicken Erythrocyte Treated with Various Reagents Number of Number of NANA Ratio of the Electrons Molecules Rew Experimental Lost per leased per and Theoretical Reagents Erythrocyte Erythrocyte Yield of NANA H-IBV 425 x 108 5.12 x 108 1.14 PR~8 6.7 x 10‘ 1.17 x 107 1.75 Neuraminidase 9.62 x 106 1.50 x 107 1.55 59 Characteristics of IBV neuraminidase. The neuraminidase activity associated with the IBV~ hemagglutinin was further established by measuring the amount of NANA liberated from NANA-L. When incubated with 2000 HA units of hemagglutinin per ml of 0.1 M phosphate buffer, pH 6.5, 18 ug of NANA was liberated from 1 mg of NANA-L in 8 hours (Table 17). During the first 50 minutes the reaction appeared to be linear and 17 ug of NANA was liberated in the first 4 hours of incu- bation at 57 C (Fig. 4). Table 17. Hydrolysis of NANA-L as a Function of Time. One mg of NANA-L was Incubated with IBV Hemagglu- tinin for Various Times at 57 C Time NANA Liberated (ug) 50 min 5 1 hr 10 2 hr 15 4 hr 17 8 hr 18 The optimum pH for liberation of NANA from NANAuL due to neuraminidase activity associated with IBV hemagglutinin in phosphate buffer was in the range of 6 to 6.5 (Fig. 5). Heat at 56 C for 45 minutes inactivated neuraminidase associated with the IBV hemagglutinin (Table 18). 60 20‘— ug NANA J i J l l IV) 4 6 8 Time in Hours Fig. 4. Hydrolysis of NANA-L as a function of time. One mg of NANA-L was incubated with IBV hemagglutinin for various times at 57 C. ug NANA 61 Fig. 5. Effect of pH for liberation of NANA from NANA-L due to neuraminidase activity associated with IBV hemagglutinin in 0.1M phosphate buffer adjusted to appropriate pH. 62 Table 18. Effect of Heat on Neuraminidase Associated with IBV Hemagglutinin at 56 C. NANA-L was Used as a Substrate Heat at 56 C NANA Liberated (09) Time in 4 Hours at 57 C 0 17 15 min 10 50 min 5 45 min 0.5 60 min 0.01 Hemadsorption due to IBV-41. Chicken erythrocytes adsorb to IBV-41 infected CEKC, 50 hours after infection (Fig. 6), but not to normal CEKC or IBV-42 infected CEKC. Antigenic relationship of IBV hemagglutinin to IBV-41 in allantoic fluid, and NAF. When tested against anti-IBV hemagglutinin‘rabbit serum, both IBV hemagglutinin and IBV-41 give one precipitin line compared to no precipitin line for NAF in agar gel double diffusion (Fig. 7), and immunoelectrophoresis (Fig. 9). When tested against anti-IBV rabbit serum, IBV-41 produces two precipitin line, IBV hemagglutinin produces rather a sharp precipitin line and NAF produces a diffuse line (Fig. 8). 65 Fig. 6. Adsorption of chicken erythrocytes onto the chicken embryo kidney cells infected with IBV-41. Fig. 7. Fig. 8. 64 Diffusion of IBV hemagglutinin and IBV infected allantoic fluid against anti-IBV hemagglutinin rabbit serum. 1 = anti-IBV hemagglutinin rabbit serum, 2 = IBV hemagglutinin, 5 = IBV-41 infected allantoic fluid, 4 = NAF. ———-———— 0.0\// Diffusion of IBV hemagglutinin, IBV infected allantoic fluid and NAF against anti-IBV rabbit serum. 1 = Anti-IBV rabbit serum, 2 = IBV hemagglutinin, 5 = IBV-41 infected fluid, 4 = NAF. Fig. 9. 65 ._+ O \_/ ———,.- [ J __). O ' + ——)- o —'> r 4 fl 0 Antigenic relation of IBV hemagglutinin, IBV infected allantoic fluid and NAF as ‘demonstrated by immunoelectrophoresis in agar gel. Anti-IBV hemagglutinin-rabbit serum, IBV hemagglutinin, IBV infected allantoic fluid, NAF FPUJNP DISCUSSION The hemagglutinin of IBV is ordinarily masked. Trypsin exfoliates the surface of the virus, either by acting di~ rectly on the protein coat of the virus or by destroying inhibitor(s) present in the allantoic fluid, so that the trypsin-modified virus can attach to a substrate on chicken and turkey erythrocytes to cause agglutination (Corbo and Cunningham, 1959; and Muldoon, 1960). Ether treatment of certain lipid containing viruses has resulted in the disruption of the virus particle with consequent separation of the hemagglutinin from the infectious nucleoprotein. Ether treatment of II“! in allantoic fluid can also induce hemagglutination. Anion exchange DEAE-cellulose column chromatography has been successfully used for the separation of the IBV hemagglutinin. Since surface potentials of the virus and the ionogenic functional groups of the adsorbent are involved, the process usually purifies a virus preparation by selec- tively eluting impurities or other components of the virus having differentially charged groups. Elution depends on the ionic strength of the buffer solution used (Kabat and Mayer, 1961). Sodium chloride, 0.1M in 0.0067 phosphate buffer elutes the hemagglutinin of IBV; 0.5 M elutes 66 67 proteinaceous materials without any detectable activity re~ lated to IBV; and 0.5 M NaCl elutes the infectious portion of IBV. The hemagglutinin of IBV is a lipoprotein, containing 54 per cent lipid and 51 per cent protein with traces of carbohydrate and RNA. The hemagglutinin of myxoviruses (Hoyle §£_al., 1961; Waterson, 1964; Rott, 1964) and vaccinia virus (Stone, 1946; Katesfln 1962; Neff gt_a_l_., 1965) are also lipoproteins. Myxoviruses contain about 20 to 40 per cent lipid (Frommhagen §£_§l,, 1959; and Kates §£_al., 1962), whereas vaccinia virus contains about 4 to 9 per cent lipid (Kates §£_al,, 1962; and Zwartouw, 1965). Ether disrupts the myxovirus particle, and hemagglutinating units are separable from the infectious nucleoprotein. Vaccinia virus hemagglutinin can be separated from the infectious portion of the virus easily by DEAE-cellulose column chromatography (McCrea and O'Laughlin, 1959). The origin of the lipoprotein associated with the HA activity of the viruses is still controversial. Knight (1946), Stone (1946), and Frommhagen et al. (1959), have shown chemically and serologically that the lipoprotein of influenza virus is not very distinct from that of the normal host cells, whereas Franklin EE_§l- (1957), concluded that a lipid is an integral part of NDV since lecithinase hydro- lyzes the lipid envelope and virus is inactivated. Anti- genically NDV hemagglutinin is related to the virus but not 68 to the host material (Rott, 1964). The lipoprotein of vaccinia virus may be a hybrid of the cellular and viral in- duced components in the vaccinial infected HeLa cells and CAM (Neff et al., 1965). Anti-IBV hemagglutinin-rabbit serum specifically inhibits the hemagglutinin (HA activity) and produces a single precipitin line in agar gel, indicating its antigenic relation to the virus. To explain the mechanism of hemagglutination by IBV, two parameters may be taken into consideration: (1) the surface of IBV which must be exfoliated, either by trypsin, ether, or anion exchange chromatography, (2) the substrate on the erythrocyte to which the exposed surface of the virus (hemagglutinin) attaches. That the substrates of trypsin or ether-modified IBV and the hemagglutinin are the same is indicated by the finding that the hemagglutinin removes the substrates for all from chicken erythrocytes. This also indicates that the removal of the substrates may be enzymatic. The hemagglutinin of IBV is associated with neuramini- dase—like activity. Structurally, the hemagglutinin resides in the lipoprotein complex of NDV and PR-8 strain of influ— enza virus, which can be disintegrated by lipid solvents (Laver, 1965) into small spheres with "spikes" on their sur- face, or to elongated structures with "bottle brush" appearance (Waterson, 1964). It is also known that the hemagglutinin and neuraminidase of certain myxoviruses are two different entities (Mayron et al., 1961; Howe gt al., 69 1961) residing side by side (Bayer, 1964; Noll §£_al,, 1962). Spikes are also present on the surface of IBV (Berry et al., 1964), but untreated virus does not cause hemagglutination. According to one report (Berry et al., 1964), trypsin does not remove the spikes of the virus. It is possible that trypsin removes or destroys inhibitor(s) present around the surface projections of the virus or a structural reorgani- zation of the protein coat enables the virus to hemagglutinate. If that is the case, then the mechanism of anion exchange chromatography to separate the hemagglutinin from the in- fectious portion of the virus becomes apparent. Depending upon a gradient of electric potentials around their surface, it may be possible that the hemagglutinin of IBV is selectively separated from both the inhibitor around it and the infectious portion within it. Electron microscopic study reveals that the hemagglu— tinin in the eluate from DEAE at 0.1 M NaCl is about 60 mu (Nazerian, Personal communication) which is supported by the finding that hemagglutinin can pass through a 100 mu milli- pore filter but not through a 50 mu pore filter. The agglutinability of the erythrocytes by IBV hamagglu— tinin can now be explained better on a physicochemical basis. It has long been recognized that the carboxyl (COO‘) groups of sialic acids are the dominant ionogenic groups mainly responsible for net negative charge on the erythrocytes (Hunter, 1960; Cook et al., 1961, and Eylar et al., 1962). 70 Sialic acids provide a substrate for neuraminidase and most of the myxoviruses. After the initial attachment to the substrate on the erythrocyte surface, under optimum conm ditions, neuraminidase hydrolyses the ketosidic linkage of the sialic acid. Release of the sialic acid is also responsi- ble for reduction of the electrophoretic mobility of the erythrocyte because of the loss of electrons associated with this COO_ group of the sialic acid. Theoretically, each electron lost should have equated with the C00_ group of a molecule of sialic acid released. The ratio of the experi- mental and the theoretical yields is more than one in each case. This is more so in the case of influenza virus. This ratio is more than 2 in the case of human erythrocytes (Cook et al., 1961) and about 1.5 when chicken erythrocytes were used with neuraminidase (Eylar et al., 1962). Many reasons can be accounted for this discrepancy. The equation used to calculate the surface charge density assumes a broad, smooth surface of the erythrocyte. It does not take into account the surface irregularities, microcrenations on the erythrocyte where the distribution of sialic acid may not significantly contribute to the net electric charge but may still be accessible to the enzyme action. That the hemagglutinin of IBV is associated with neuraminidase activity, is further established by the direct measurement of NANA released from sialolactose. In the initial 50 minutes the reaction seems to be linear, and 71 17 ug of NANA were released in the first 4 hours at 57 C from 1 mg of NANA-L by 1 mg of hemagglutinin that contained 2000 HA units. After 4 hours the activity was considerably reduced. The optimum pH for IBV neuraminidase is 6.5, similar to that of Asian strain of influenza virus (Mayron et al., 1961). Ada et al. (1965) have shown that antiserum against LEE virus neuraminidase inhibited almost completely homologous action on fetuin and ovine submaxillary mucoid but only partially on NANA-L. It is possible that NANA-L is not a suitable substrate for IBV neuraminidase or that it is also possible that neuraminidase concentration in 1 mg hemagglutinin was low. In either event it is significant that IBV hemagglutinin hydrolyses sialolactose. The thermostability of the IBV hemagglutinin and neuraminidase are interrelated. The HA activity is reduced after 4 to 8 hours at 57 C, and the neuraminidase activity also begins to decrease after 4 hours at 57 C. At 56 C, however, neuraminidase is inactivated in 50 to 45 minutes, whereas the hemagglutinin is stable for at least 5 hours. Trypsin—modified IBV is more thermostable for 9 hours at 56 C (Muldoon, 1950). Since the IBV hemagglutinin is a separated lipoprotein of relatively pure form, thermostability may be much less than the trypsin-modified IBV, and it is possible that egg white trypsin inhibitor added to the trypsin modified IBV, serves as a protective colloid in the mixture. No biochemical assay could be made with trypsin— modified IBV. 72 Neuraminidase destroys the substrates for T—IBV, E-IBV and IBV hemagglutinin, PR-8 strain of influenza, and NDV on chicken erythrocytes. Neuraminidase from y. cholerae in- hibited the growth of IBV-41 in chicken embryos. To obtain the maximal inhibitory effect, 0.008 units of the enzyme per embryo was necessary. Measurement of the HA activity of the allantoic fluid from embryos inoculated with the virus was the criterion to judge the 'protective power' or the "inhibitory effect" of neuraminidase. However, it does indicate that neuraminidase destroys the receptors for IBV-41 in chicken embryos. Under similar conditions, 0.0004 units of neuraminidase per embryo provides maximal protection against Lee-B virus, and 0.006 unit of the enzyme was re- quired for protection against PR-8 strain of influenza virus (Johnson et al., 1964). The present finding also supports the proposition that sialic acid-containing substrates are necessary for IBV-41 infection of chickens. Hemadsorption is a phenomenon related to hemagglutin- ation. Chicken erythrocytes adsorb to the surface of IBV infected CEKC. Since CEKC are the only cells known to cause hemadsorption due to IBV.it is imperative that this cellular surface dynamics be studied further on established cell lines for a better understanding of the host-cell and virus interaction. Since the hemagglutinin of IBV displays such signifi- cant biological activities as to agglutinate chicken 75 erythrocytes and liberate sialic acids from suitable sub- strates, it may be well asked, what is the surface of the IBV like, and what is the nature of its substrate? This is perhaps the only virus known, in which its ionogenic surface has to be exposed, either by trypsin modification, ether treatment, or by anion exchange chromatography, to provide a substrate on the chicken erythrocyte. The surface of IBV has not been clearly defined and it is not known where the environment of the inhibitors, if there were any, ended, or where the virus surface thus ex- posed, began. Electron microscopists are satisfied with the micrographs of the "spikes" around the virus particle, and it is tempting to borrow ideas and interpret results similar to those for NDV or influenza viruses, where the hemagglutinin is also associated with "spikes," "rosettes,' or even ”bottle brush" like structures. At the molecular level the picture is more obscure. One can appreciate only, that there are ionogenic groups on a complex macromolecular surface composed mainly of lipid and protein, which under proper orientation will bind to a suitable susceptible receptor. These molecular compon- ents are-associated with two biological activities, the hemagglutinating and neuraminidase activities. It is possible to separate these two entities in myxoviruses (Mayron et al., 1961; Laver, 1965). The chemical identities, perhaps associated with the lipoprotein complex, of these 74 two biological entities so closely related with each other, are yet not fully known. Their origin may be partly regu- lated by the viral genome, the RNA, which must have code words for their construction and assembly in the envelope of the virus. The receptors for IBV on erythrocytes is interesting. N-acetylneuraminic acid was released from chicken erythro- cytes by nauraminidase. It is known that human erythrocytes contain sialic acids similar to those found on the chicken erythrocytes (Klenk and Uhlenbruck, 1958). If that is the case, human erythrocytes also should have been agglutinable by IBV hemagglutinin. This discrepancy may possibly be explained by the current concept about the Specific receptor mechanism (Koshland, 1965). Belyavin (1965) hypothesized that more "points” (Friess et al., 1962) in the receptor element makes a more "specific" receptor, and it is the configuration and spatial distribution of such "points“ that confer "specificity" upon a chemical compound. Accordingly it is possible that ”points" necessary for IBV hemagglutinin "binding“ are suitably oriented (configurated) on chicken erythrocytes but not on human erythrocytes, even though their structure remains the same. A receptor gradient (Burnet gt_al,, 1946) was proposed for the graded ability of different myxoviruses to destroy receptor sites on erythrocytes. Multiplicities of enzyme and/or substrates have been proposed to explain the 75 mechanism underlying the concept of the receptor gradient. Recently, Ackermann (1964) has attempted to explain receptor gradient in terms of a mosaic theory which propounds that overlapping, circumscribed areas on erythrocytes, for each virus, may be structurally different, but still containing sialic acid linked in the critical terminal positions for neuraminidase action. These areas are recognized by a particular hemagglutinin which in turn is associated with the enzyme concerned. These hypotheses, however, do not take into account the difference (e.g., thermostability) of the 'neuraminidase“ of different viruses and the susceptibility of the receptors containing sialic acids on different species of erythrocytes. This discrepancy can be taken into account if one considers that (1) the hemagglutinin and neuraminidase molecules of different viruses are differently arranged, partly because of their genetic make-up, so that their ionogenic surfaces are different, and more importantly, (2) the configuration of the structure of the sialic acid contained on the muco- protein complex that are known to be the substrates for the hemagglutinin or neuraminidase are dissimilar in different species of erythrocytes. This should Specifically explain the situation where mutual stereospecificities of the mole- cules or ionic groups only, can attract and bind each other, and concomittantly reduce the electronic potential. 76 Experimentally this would involve a formidable chemi— cal operation, not only to dissect and identify the molecules or structures involved, but also to determine their configur- ation and general spatial distribution on a complex and complicated surface such as those on the virus and, of course, those on the erythrocytes. SUMMARY 1. A hemagglutinin was isolated from IBV—41 infected allantoic fluid by anion exchange diethylaminoethyl—cellulose column chromatography. In a sodium chloride gradient in 0.0067 M phosphate buffer pH 7.4, the hemagglutinin eluted at 0.1 M and the infectious portion of the virus eluted at 0.5 M NaCl. Diethyl ether was also used to isolate the hemagglu— tinin of IBV. 2. Only chicken erythrocytes were agglutinable by IBV. Human, cow, horse, sheep, dog, and rabbit erythrocytes were not agglutinable. 5. The hemagglutinin was identified as a lipoprotein, containing 54 per cent lipid and 51 per cent protein, with traces of carbohydrate and RNA. 4. The size of the hemagglutinin was estimated to be about 60 mu in diameter. The hemagglutinin was inactivated in 4 to 8 hours at 57 C or in 5 hours at 56 C. A pH range of 5 to 7.5 did not significantly influence the HA activity- Freeze-dried hemagglutinin was preserved for at least 5 months at 4 C. 5. Neuraminidase-like activity was also associated with the hemagglutinin. In phosphate buffer, pH 6.5, it 77 78 liberated N-acetylneuraminic acid from N-acetylneuraminic acid—lactose, the maximal activity being at 57 C. Neuramini- dase like activity of IBV was inactivated in 45 minutes at 56 C. 6. Receptors on chicken erythrocytes for trypsin- modified IBV, ether-treated IBV and the hemagglutinin itself were destroyed by the hemagglutinin. The receptors for PR~8 influenza virus and NDV were not affected under similar conditions. 7. The hemagglutinin liberated 5.12 x 106 molecules of NANA and removed 4-5 x 106 electrons from each erythrocyte, in comparison to 1.5 x 107 molecules of NANA and 9.62 x 108 electrons removed by neuraminidase from V, ghgleraex 8. Neuraminidase of Vibrio cholerae removed the sub- strates for trypsin-modified IBV, ether-treated IBV and IBV- hemagglutinin, and PRe8 strain of influenza virus. 9. A concentration of 0.008 units of neuraminidase per chicken embryo was necessary for maximal inhibition of infection by IBV-41. 10. Hemadsorption occurred with IBV-41 infected CEKC. 11. Maximal inhibition of hemagglutination occurred with anti-IBV-hemagglutinin rabbit serum collected on the 8th and 10th day post—inoculation of the hemagglutinin. This serum produced a single precipitin line against the hemagglutinin in agar gel. EIBLIOGRAPHY Abramson, H. A., L. S. Moyer, and M. H. Gorin, 1942. Electrophoresis of Proteins and Chemistry of Cell Surfaces. New York: Reinhold Publishing Corp. Ackermann, W. W., 1964. Cell surface phenomena of Newcastle disease virus (pp. 155~165) in Newcastle Disease Virus. Ed. by R. P. Hanson. Madison, Wisconsin. The Uni— versity of Wisconsin Press. Ada, G. L. and E. L. French. 1959. Purification of bacterial neuraminidase (receptor destroying enzyme). Nature, Lond., 185: 1740-1741. Ada, G. L. and J. D. Stone. 1950. Electrophoretic studies of virus-red cell interaction: .Mobility gradient of cells treated with the viruses of influenza group and the receptor—destroying enzyme of V. ghglerag. Brit. J. Exptl. Path., 51:265-274. Ada, G. L., P. E. Lind and W. G. Laver, 1965. An immuno- logical study of avian, viral and bacterial (Vibrig ‘ghglerae) neuraminidase based on specific inhibition of enzyme by antibody. J. Gen. Microbiol.,_§gfi 225-255. Akers, T. G. 1965. Some cytological studies of the multiplication of infectious bronchitis virus in chicken embryo kidney cells. Ph. D. Thesis, Michigan State University, East Lansing, Michigan. Anderson, S. G., 1959. Hemagglutination by animal viruses, Animal Viruses (pp. 21—50) in the Viruses. Vol. 5, Ed. by F. M. Burnet and W. M. Stanley. Academic Press, NO Y. Andrews, C. 1964. Viruses of Vertebrates. The Williams and Wilkins Co., Baltimore. Bangham, A. D., B. A. Pethica, and G. C. F. Seaman, 1958. The charged groups at the interface of some blood cells. Biochem. J., 69; 12-19. Barron, A. L., Z. Zakay-Roness, and H. Bernkoff. 1965. Hemagglutination of chicken erythrocytes by the agent of psittacosis. Bact. Proc. (1965) pp. 117. 79 80 Bateman, J. B., A. Zeliner, M. S. Davis, and P. A. M. Caffery. 1956. The electrophoretic properties of red blood cells after reaction with influenza virus hemagglutinin. Arch. Biochem. Biophys., 60; 584~587. Bayer, M. E., 1964. An electron microscope examination of urinary mucoprotein and its interaction with influenza virus. J. Cell Biol. 21; 265-274. Beaudette, F. R. and C. B. Hudson, 1957. Cultivation of the virus of infectious bronchitis. J. Am. Vet. Med. Assoc. 90; 51-60. Belyavin, G. 1965. The mechanism of influenza virus hemagglutination as a model for specific receptor adsorption. J. Gen. Microbiol. 52; 15‘17. Berlin, B. S., J. S. McQueen, E. Minuse and F. M. Davenport. 1965. A method for increasing the sensitivity of the hemagglutination inhibition test with equine influenza virus. Virology. '21; 665-666. Berry, D. M., J. G. Cruickshank, H- P. Chu, and R. J. H. Wells, 1964. The structure of infectious bronchitis virus. Virology, 25; 405-407. Biswal, N. 1965. Electrokinetic studies on erythrocytes treated with modified infectious bronchitis virus. M. S. Thesis. Michigan State University, East Lansing, Michigan. Bogoch, S. and J. Kaufman, 1965. Relation of state and concentration of virus to production of sialo- responsin in vitro. Bact. Proc. (1965) pp. 101. Bogoch, S., R. F. Gilfillan and P. E. Vans. 1962. Sialoresponsin. Neuraminic acid containing substance which accumulates in chorioallantOic fluid during the 1st minutes of virus infections. Nature. 196; 649-651. Brown, W. E., S. C. Schmittle and J. W. Foster, 1962. A tannic acid modified test for infectious bronchitis of chickens. Avian Dis. 6; 99-106. Buckland, F. E. and D. A. J. Tyrell. 1965. A comparative study of virus hemagglutinins. The stability of hemagglutinin and red cell receptors to certain physical and chemical treatments. J. Gen. Microbiol. .53: 241-245. 81 Buckley, S. M. 1959. Propogation, cytopathogenicity, and hemagglutination-hemadsorption of some arthropodeborne viruses in tissue culture. Ann. N. Y. Acad. Sci. 2;; 127-187. Burnet, F. M. 1942. The affinity of Newcastle disease virus to the influenza virus group. Aust. J. Exp. Biol. Med. Sci. 29; 81-88. Burnet, F. M. and J. D. Stone. 1947. The receptor destroy” ing enzyme of y. gholeggg. Aust. J. Exp. Biol. Med. Sci. 22; 227-228. Burnet, F. M., J. F. McCrea, and J. D. Stone, 1946. Modifi- cation of human red cells by virus action. I- The receptor gradient of virus action in human red cells. Brit. J. Exp. Path., 21; 228—256. Bussell, R. H., D. T. Karzon, and F. T. Hall, 1962. Hemagglutination and hemagglutination inhibition studies with ECHO viruses. J. Immunol., 88; 58‘46. Buthala, D. A., 1956. Some properties of the avian bronw chitis virus. Ph. D. Thesis. Iowa State College. Buzzell, A. and M. Hanig, 1958. The mechanism of hemaggluw tination by influenza virus. Ad. virus Res. 2: 289-546. Cairus, H. J. F., and P. J. Mason, 1955. Production of influenza A virus in the cells of the allantoic- J. Immunol. 2;; 58-40. Cassel, A. 1957. Multiplication of vaccinia virus in the Ehrlich Ascites Carcinoma. Virology. 5: 514~524- Choppin, P- W. and I. Tamm. 1960. Studies of two kinds of virus particles which comprise as virus strains. J. Exp. Med. 112: 895-920. Cook, G. M. W., D. H. Heard, and G. V. F. Seaman, 1960. A sialomucopeptide liberated by trypSin from human erythrocyte. Nature, 182; 1011m1012. Cook, G. M. W., D. H. Heard, and G. V- F. Seaman, 1961. Sialic acids and Electrokinetic charge of the human erythrocyte. Nature, 221; 44~47. Corbo, L. J. and C. H. Cunningham, 1959. Hemagglutination by trypsinmmodified infectious bronchitis virus. Am. J. Vet- Res., 20; 876n885. 82 Cunningham, C. H., 1957. Symposium on immunization against infectious bronchitis virus. I. Some basic properties of infectious bronchitis virus, Am. J. Vet. Res. :18: 648=654. Cunninghfim, C. H., 1963. A Laboratory Guide in Virology. 5 ed. Burgess Pub. Co., Minneapolis, Minnesota. Cunningham, C. H-, and H. 0. Stuart, 1947. The pH stability of the virus of infectious bronchitis of chickens, Cornell Vet., 31. 99‘103. Cunningham, C. H. and M- P. Spring, 1965. Some studies of infectious bronchitis virus in cell culture, AVian Dis- 9. 182~192, Curtain, C. G., E. L. French, and J. Pye- 1953. The preparation and properties of an inhibitor of influenza virus hemagglutination. Aust. J. Exp. Biol, Med. Sci.. 31; 315-322. Darrell, R. W. and C. Howe, 1964- The neuraminidase of Parainfluenza Virus (Type 2). Proc. Soc. Exp. B101. and Med. 116: 1091-1094. Eylar, E. H.. M. A. Madoff. O. V. Brody and J. L. Oncley. 1962. The contribution of Sialic acid to surface charge of erythrocyte. J. Biol- Chem. 231; 1992n2000. Fazekas de St. Groth, S- 1948- Destruction of influenza virus receptors in the mouse lungs after artificial destruction. Aust. J. Exp. Biol. Med, Sci., 26; 271m285. Ferguson, G. H., 1958. Growth of infectious bronchitis virus in suspended chorioallantoic membranes. M, S. Thesis. Michigan State University. East LanSing, Michigan. Folch, J- S-, Arsove, and J. A. Meath. 1951. Isolation of brain strandin, a new type of large molecule tissue component. J. Biol. Chem., 191; 819 831. Francis, T. 1947. Dissociation of hemagglutinating and antibody measuring capaCities of influenza Virus. J. Exp. Med. g5; 1-7. Franklin, R. M., H. Rubin and C. A. DaVis, 1957. The pron duction,purification and properties of Newcastle disease virus labeled with radiOphosphcrus. Virology. é: 96‘114. 85 Friess, S. W., B. Witkop, R. C. Durant, L. J. Reber, and W. C. Thommesen. 1962. FUrther aspects of stereo~ specificity in interaction of polyfunctional amino derivatives with biological receptors. Arch. Biochem. and Biophys. lgéz 158~165. Frommhagen, L. H. and C. A. Knight. 1959- Column purifie cation of influenza virus- Virology.~ 8; 1989208. Frommhagen, L. H., C. A. Knight and N. K. Freeman. 1959. The ribonucleic acid, lipid, and polysaccharide constituents of influenza virus preparations- Virology 8: 176~197. Furchgott, R. E. and E. Ponder, 1941- Electrophoretic studies on human red blood cells- J. Gen. Physiol. .gg: 447w457. Gillespie, J. H. 1962. The virus of canine distemper. Ann. N. Y. Acad. Sci. .191: 540-547. Goldfield, M. S., Srihongse and J. P. Fox. 1957. Hemagglutinins associated with certain human enteric viruses. Proc. Soc. Exp. Biol. Med- 98; 788w791. Gommatos, P. J. and I. Tamm. 1965. The secondary structure of reovirus RNA. Proc- Nat. Acad. Sci. 49; 707r714. Gottschalk, A. 1954. The influenza virus enzyme and its mucoprotein substrate. Yale J. Biol. Med. fig; 552-364. Gottschalk, A. 1957. Neuraminidase, the specific enzyme of influenza virus and Vib£l©.¢h§l§£§§~ Biochem- et Biophys. Acta. §§:-6459648. Gottschalk, A. 1958. The influenza virus neuraminidase. Nature. London. 1813 377-578- Gottschalk. A. 1960. The chemistry and biology of sialic acids. London, Cambridge Univ. press. Halperan, S., H. J. Eggers, and I- Tamm. 1964. Complete and coreless hemagglutinating particles produced in ECHO 12 virus infected cells. Virology, 35; 81-89. Hanig, M. 1948. Electrokinetic changes in the human erythrocyte during adsorption and elution of PR“8 influenza virus. Proc. Soc. Exp. Biol. Med., §§g 585~592. 84 Heard, D. H. and G. V. F. Seaman. 1960. The influence ofgfli and ionic strength on electrokinetic stability of human erythrocyte membrane. J. Gen. Physiol. 45; 655-654. Heberling, R. L. and F. S. Cheever. 1965. Simian entero- virus SV 2 hemagglutination studies. I. Relationship of infectivity to hemagglutination. Proc. Soc. Exp. Biol. Med. 118: 151-154. Heimer, R. and K. Mayer. 1956. Studies on sialic aCid of submaxillary mucoid. Proc. Nat. Acad. Sci- Wash. 42: 728-754. Hiller, H. and J. F. Hoffman. 1955. Ultrastructure of plasma membrane as determined by the electronmicro- scope. J. Cellular. Compar. Physiol. 4;; 205-248. Hirschfeld, J. 1960. Immunoelectrophoresis procedure and application to the study of group specific vari- ations in sera. Sci. Tools. 1; 18-25. Hirst, G. K. 1941. The agglutination of red cells by allantoic fluid of chick embryos infected with in- fluenza. Sci. 24; 22-25. Hirst, G. K. 1942. Adsorption of influenza hemagglutinins and virus by red blood cells. J. Exp. Med. 16; 195-209. Hirst, G. K. 1959. Virus—host cell reaction in Viral and Rickettsial diseases of man (pp. 96-114). 3rd. ed. Ed. by T. M. Rivers and F. L. Horsfall. Philadelphia, J. B. Lippincott Co. Hotchin, J. E., R. Deibel and L. M. Benson, 1960. Location of nonpathogenic myxovirus plaques by hemadsorption. Virology. 19: 275-280. Howe, C. 1951. The influenza virus receptor and blood group antigens of human erythrocyte stroma. J. Immunol. 66; 9-55. Howe, C., H. M. Rose, and L. Schneider, 1957. Enzymatic action of influenza virus on human erythrocyte stromacomponents. Emoc. Soc. Exp. Biol. Med., 96: 89-94. Howe, C., H. M. Rose, and L. T. Lee. 1961. Observation on the relationship between hemagglutinin and neuramini- dase of influenza viruses. Proc. Soc. Exp. Biol. Med., 105: 420-425. 85 Howe, C., L. T. Lee, and H. M. Rose. 1961. Collocalia mucoid. A substrate for myxovirus neuraminidase. Arch. Biochem. Biophys. 95: 512-520. Hoyle, L. 1950. The multiplication of influenza viruses in the fertile egg. J. Hyg. 48: 277-297. Hoyle, L. 1952. Structure of influenza virus. The relation between biological activity and chemical structure of virus fractions. J. Hyg. 59: 229-245. Hoyle, L., R. W. Horne and A. P. Waterson, 1961. The structure and composition of the myxoviruses. 11. Components released from the influenza virus particle by ether. Virology, 15; 448-459. Hunter, R. J. 1960. Constancy of the surface charge density of human erythrocytes at different ionic strength. Arch. Biochem. Biophys. 88: 508-512. Johnson, C. A., D. J. Pekas, and R. J. Winzler. 1964. Neuraminidases and influenza virus infection in embryonated eggs. Sci. 145: 1051—1052. Kabat, E. A. and M. M. Meyer, 1961. Experimental Immuno- chemistry. Thomas Books, Springfield, Illinois. Karzon, D. T. 1962. Measles virus. Ann. N. Y. Acad. Sci. 101: 527-559. Kasel, J. A., W. P. Rowe, and J. L. Nemes, 1960. Modifi- cation of erythrocyte receptors by a factor in adenovirus suspensions, Virology, 19: 588-591. Kates, M., A. C. Allison, D. A. J. Tyrell, A. T. James. 1962. Origin of lipids in influenza virus. Cold Spring Harbor Symp. Quant. Biol. 21: 292-501. Keck, K. 1956. An ultramicro technique for the determin- ation of deoxy pentose nucleic acid. Arch. Biochem. Bi0phys. 65; 446—451. Klenk, E. and G. Uhlenbruck, 1958. Uber ein neuraminsaure- haltiges mucoproteid ans Rindererythrocytenstroma. Hoppe-Seyl. z. 511: 227 cited in "The Chemistry and Biology of Sialic Acids," pp. 55. A. Gottschalk, London: Camb. Univ. Press. Knight, C. A., 1946. Precipitin reactions of highly purified influenza viruses and related materials. J. Exp. Med. ‘85: 281-294. 86 Koshland, D. E. Jr., 1965. Correlation of structure and function in enzyme action. Sci. 142: 1555-1541. Laver, W. G., 1965. The structure of influenza viruses. 5. Disruption of the virus particle and separation of neuraminidase activity. Virology. 29: 251-262. Lefkowitz, S. S. and M. M. Sigel, 1965. Passive hemagglu- tination reaction with avenovirus antibody. Bact. Proc. 1965. pp. 125. Lennette, E. H., 1959. Serologic reactions in viral and rickettsial infections of man. 5rd. ed. pp. 250-250. Ed. by T. M. Rivers and F. L. Horsfall. Philadelphia: J. B. Lippinscott Co. Lief, F. S. and W. Henle, 1956. Studies on the soluble antigen of influenza virus. I. The release of S antigen from elementary bodies by ether. Virology. 2; 755-711. Lowry, O. H., N. J. Rosenbrough, A. L. Farr and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 195: 265-275. Macrae, A. D., 1962. Reovirus of man. Ann. N. Y. Acad. Sci. 101: 455-460. Madoff, M. A. and E. H. Eylar, 1961. The contribution of sialic acid to red blood cell mobility. Fed. Proc. 20: 62. Mallmann, V. H. 1960. The demonstration of a cell-attach- ment and growth factor and comparative studies of infectious bronchitis virus before and after separa- tion from a factor. Ph. D. Thesis. Michigan State University, East Lansing, Michigan. ' Marcus, P. I. 1962. Dynamics of cell surface modification in myxoviruses infected cells. Cold Spring Harbor Symp. Quant. Biol. 21: 551-565. Marcus, P. I., and J. M. Slab. 1965. Myxovirus-specific receptors on isolated HeLa nuclei. Bact. Proc. (1965) pp. 100. Mayron, L. W., B. Robert, R. J. Winzler, and M. E. Rafelson, Jr. 1961. Studies on the neuraminidase of influenza virus. I. Separation and some properties of the enzyme from Asian and PR-8 strains. Arch. Biochem. Biophys. 92: 475-485. 87 McClelland, L. and R. Hare. 1941. The adsorption of influenza virus by red cells and a new in-vitro method of measuring antibodies for influenza virus. Canad. P. H. J. 55; 550-558. McCrea, J. F. 1948. Mucins and mucoids in relation to influenza virus action. II. Isolation and characterization of serum mucoid inhibitor of heated influenza virus. Aust. J. Exp. Biol. Med. Sci. 55; 555-570. McCrea, J. F. 1955a. Studies on influenza virus receptor substance and receptor-substance analogues. I. Prep- aration and properties of a homologous mucoid from the salivary gland of sheep. Biochem. J. 55: 152-158. McCrea, J. F. 1955b. Studies on influenza virus receptor substance and receptor substance analogues. II. Isolation and purification of a mucoprotein receptor substance from human erythrocyte stroma treated with pentane. Yale J. Biol. Med., 25; 191-210. McCrea, J. F., and J. O'Loughlin, 1959. Separation of vaccinia hemagglutinin from infectious virus particle by chromatography on DEAE columns. Virology, 5: 127-129. Mizutani, H., T. Beals, A. V. Hennesy, and F. M. Davenport. 1962. A single procedure for purification of viral hemagglutinin. Virology, 11; 210-212. Mohanty, S. B., H. M. Devolt, and J. B. Faber. 1964. A fluorescent antibody study of infectious bronchitis virus. Poultry Sci. 45; 179-182. Morris, D. L., 1948. Quantitative determination of carbo- hydrates with Dreywood's Anthrone reagent. Sci. 191; 254-255. Moskowitz, M. and M. Calvin. 1952. On components and structure of the human red cell membrane. Exp. Cell. Res. 5; 55-46. Muldoon, R. L. 1960. Some characteristics of the hemagglu- tinating activity of infectious bronchitis virus. Ph. D. Thesis. Michigan State University, East Lansing. Mussagay, M. and R. Rott. 1964. Studies on the structure of a hemagglutinating component of a group A Arbor virus (Sindbis). Virology, 25: 575-581. 88 Nazerian, K. 1960. Electron microscopy studies of the virus of infectious bronchitis and erythrocytes agglutinated by trypsin-modified virus. M. S. Thesis, Michigan State University, East Lansing, Michigan. Neff, B. J., W. W. Ackerman, and R. E. Preston, 1965. Studies of vaccinia hemagglutinin obtained from various vaccinia infected tissues. Density gradient centrifugation and electron microscopy. Proc. Soc. Exp. Biol. Med. ‘115: 664-671. Newlin, S. C. and A. P. McKee, 1965. Effects of proteolytic enzymes on reovirus and on erythrocyte receptors to reoviruses. Bact. Proc. (1965). pp. 101. Noll, H., and T. Aoyagi, J. Orlando. 1962. The structural relationship of sialidase to the influenza virus surface. Virology, 15; 154-157. Norrby, E. 1965. Hemagglutination by Measles virus. III. Identification of two different hemagglutinins. Virology, 15; 147-157. Oda, M. 1964. Plaque-hemadsorption technique for the detection of nonhemagglutinating vaccinia virus. Virology, 25; 452-454. Ozawa, Y. 1959. Some properties of infectious bronchitis birus propagated in isolated chorio allantoic membrane. Ph. D. Thesis. Michigan State University, East Lansing, Michigan. Padgett, B. L. and D. L. Walker, 1964. Enzymatic variants of influenza virus. III. Function of neuraminidase in the viral growth cycle. J. Bacteriol. 51: 565-569. Paucker, K., A. Birch-Anderson and P. von Magnus. 1959. Studies on the structure of influenza virus. I. Components of infectious and incomplete particles. Virology, 5; 1-20. Pereira, H. G., R. J. Huebner, H. S. Ginsberg and J. Vander- veen. 1965. A short description of the adenovirus group. Virology, 29: 615-620. Philipson, L. 1960. Separation on DEAE-cellulose of com- ponents associated with adenovirus reproduction. Virology, 19: 459-465. Plowright, W. 1962. Rinderpest virus. Ann. N. Y. Acad. Sci. 101: 548-565. 89 Ponder, E. 1951. Effect produced by trypsin on certain properties of human red cell. Blood, g; 550-556. Pye, J. 1955. Assay of inhibitors of influenza virus hemagglutination by electrophoresis. Aust. J. Exp. Biol. Med. Sci. 55; 525-524. Reagan, R. L., and A. L. Brueckner. 1952. Electron micro- scope studies of four strains of infectious bronchitis virus. Am. J. Vet. Res. 15; 417-418. Reed, L. J. and H. Muench. 1958. A simple method estimat~ ing fifty per cent end points. Am. J. Hyg. 21; 495-499. Rhim, J. S., J. I. Kato, and W. Pelon. 1965. Hemagglutin- ation by Reovirus propagated in various cell lines. Proc. Soc. Exp. Biol. Med. 118: 455-459. Rosen, L. 1964. Hemagglutination. pp. 257-276, in Techniques in Experimental Virology. Ed. by R. J. C. Harris. Acad. Press, N. Y. Rosenberg, A. 0., C. Howe, and E. Chargaff. 1956. Inhibi- tion of influenza virus hemagglutination by brain lipid fraction. Nature, London, 177: 254-255. Rott, R. 1964. Antigenicity of Newcastle disease virus. pp. 155-146, in Newcastle disease virus. Ed. by R. P. Hanson. The University of Wisconsin Press. Madison. Rowe, W. P., and J. W. Hartley, 1962. A general review of the adenovirus. Ann. No Y. Acad. Sci. 101: 466-474. Sabin, A. B. and E. L. Buescher. 1950. Unique physico~ chemical properties of Japanese B encephalitis virus hemagglutinins. Proc. Soc. Exp. Biol. Med. 14: 222-250. Schalk, A. F., and M. C. Hawn, 1951. An apparently new respiratory disease of chicks. J. Am. Vet. Med. Assn. 18; 415-422. Schleisinger, R. W., and H. V. Karr. 1956. Influenza virus and its mucoprotein substrate in chorio allantoic membrane of the chick embryo. II. Stepwise inacti- vation of substrate and its relation to the mode of viral multiplication. J. Exp. Med. .195: 555-549. 90 Schmidt, N. J., J. Dennis and E. H. Lennette. 1964. Studies on filtrates from cultures of a psychrophylic pseudomonas sp. which inactivates non-specific serum inhibitors for certain hemagglutinating viruses. J. Immunol. 95: 140-147. Schneider, W. C. 1955. Phosphorus compounds in animal tissues. J. Biol. Chem. 161: 295—505. Schramm, G., and E. Mohr. 1959. Purification of neuramini- dase from y, cholerae. Nature, Lond., 185: 1677-1678. Seaman, G. V. F., and D. H. Heard. 1960. The surface of washed human erythrocyte as a polyanion. J. Gen. Physiol., 44: 251-268. Seto, J. T. and Y. Hokama. 1964. Disc electrophoresis analysis of sialidase from influenza virus. Ann. N. Y. Acad. Sci. 121: 640-650. Shelokov, A., J. E. Vogel and L. Chi. 1958. Hemadsorption (adsorption-hemagglutination) test for viral agents in tissue culture with special reference to influenza. Proc. Soc. Exp. Biol. Med. .91: 802-809. Singh, I. P. 1960. Some properties of infectious bronchitis virus as determined by thermal and formalin inacti- vation. Ph. D. Thesis. Michigan State University. East Lansing, Michigan Smith, C. E. G. and D. Holt. 1961. Chromatography of arthropod-borne virusescx1calcium phosphate columns. Bull. World Health Organ. 24: 749-759. Smith, J. W. and A. F. Rasmussen. 1965. Neurotropin and neuraminidase activity in the myxoviruses. Bact. Proc. 1965. pp. 100. Spring, M. P. 1960. Cultivation of infectious bronchitis virus in chicken embryo kidney cells. M. S. Thesis. Michigan State University, East Lansing, Michigan. Stone, J. D. 1946. Lipid hemagglutinins. Aust. J. Exp. Biol. Med. Sci. 24: 197-202. Stone, J. D. 1948. Prevention of virus infections with enzyme of y, cholerae. I. Studies with viruses of mumps-influenza group in chick embryos. Aust. J. Exp. Biol. Med. Sci. 26; 49-64. 91 Stone, J. D. 1949. Tryptic inactivation of the receptor destroying enzyme of y, cholerae and the enzymatic activity of influenza virus. Aust. J. Exp. Biol. Med. Sci. ‘21: 229-244. Stone, J. D., and G. L. Ada. 1952. Electrophoretic studies of virus-red cell interaction. Relationship between agglutinability and electrophoretic mobility. Brit. J. Exp. Path. .55: 428-459. Stultz, W. E. 1962. Fluorescent antibody studies of infectious bronchitis virus. M. S. Thesis, Michigan State University, East Lansing, Michigan. Svennerholm, E., and L. Svennerholm. 1958. Quantitative paper partition chromatography of sialic acid. Nature. 181: 1154-1155. Takatsy, G., and K. Barb. 1959. New inhibitor and erythrocyte receptor substances for certain Asian strains of influenza virus. Nature, London, 185: 52-55. Tamm, I., and F. L. Horsfall. 1952. A mucoprotein derived from human urine which reacts with influenza, mumps and Newcastle disease viruses. J. Exp. Med. .95: 71-97. Tevethia, S. S., 1964. Studies of antigens of infectious bronchitis virus by immunodiffusion. Ph. D. Thesis. Michigan State University, East Lansing, Michigan. Warren, L. 1959. The thiobarbituric acid assay of sialic acids. J. Biol. Chem. 254: 1971-1975. Waterson, A. P., 1964. The morphology and composition of Newcastle disease virus. pp. 119-152, in Newcastle disease virus. ed. by R. P. Hanson. The University of Wisconsin Press. Madison. Wilner, B. I. 1964. A classification of the major groups of human and lower animal viruses. 2nd ed. Cutter Laboratories, Berkeley, California. Wilson, V. W., and M. E. Rafelson. 1962. Isolation of neuraminidase from influenza virus with chymotrypsin. Fed. Proc. 21; 249. Wright, B. S., and B. P. Sagik. 1958. Plaque formation in monolayers of chicken embryo kidney cells. Virology, é: 575-574. 92 Zilliken, F., G. H. Werner, R. K. Silver, and P. Gyorgy. 1957. Studies on the enzymatic properties of influ- enza viruses. I. The action of influenza B virus and RDE on hemagglutinin inhibitor of human meconium. Virology, 5: 464-474. Zwartouw, H. T. 1964. The chemical composition of vaecinia virus. J. Gen. Microbiol. 54: 115-125. ‘IICHIGQN STQTE UNIV. LIBRARIES \IHl WI II! M “I 1““ ill“ ll“ ll“ II! II“ W l\ ||l|| Will 31293200022089