Antigenic properties of California encephalitis viruses

The kinetics of inactivation by antiserum of several strains of California encephalitis virus were examined by the technique of plaque reduction. The antisera employed were obtained from rabbits following multiple injections of virus strains propagated in the brain tissues of suckling mice. The viruses used in the plaque reduction studies were propagated in cultures of baby hamster kidney cells. This was done to maximize the likelihood that the interaction between virus and antibody would involve virus-specific determinants and minimize the possibility of interference by antibodies directed against host cell components. An antiserum dilution was predetermined for each homologous virus-antiserum combination so that similar rates of inactivation resulted. That is, when the results of each of the homologous neutralization reactions of several viruses were compared graphically, the curves were found to coincide. When attempts were made to neutralize heterologous virus strains with each of these sera at their unique dilutions, three types of reaction were apparent. In some, the rates of inactivation of certain heterologous viruses were similar to that of the homologous virus, that is, inactivation was immediate and without a lag following admixture. In the second type of cross-reaction, a significant lag preceded neutralization, while in the third type of reaction no indication of cross-reaction could be detected. The demonstration of a kinetic lag preceding neutralization supported the concept of multihit inactivation kinetics rather than single hit kinetics. The differences between the two types of heterologous cross-reactivity suggested the significance of the configuration of the antigenic determinants in determining the ability of a cross-reacting antibody to recognize" a portion of the determinant. In the course of preliminary experiments it became apparent that phenotypic differences arose in genotypically identical viruses when different types of host cells were used to propagate the viruses. A significant persistent fraction of surviving virus was noted when neutralization experiments were conducted with viruses propagated in suckling mice. This fraction of non-neutralizable virus could not be reduced by any treatment of the virus suspension or the antiserum or by using different assay cells. Attempts to modify the surfaces of mouse brain-derived viruses by enzymatic digestion with trypsin, phospholipase C and neuraminidase did result in the detection of differences in the surfaces of viruses grown in these cells and in baby hamster kidney cells. However, treatment of mouse brain-propagated viruses with these enzymes did not enhance their neutralizability. As a part of this study, experiments were also conducted in which variables influencing plaque formation by California encephalitis viruses were investigated. The plaquability of the viruses was found to be extremely sensitive to the pH of the agarose overlay. In contrast to other viruses that exhibit pH sensitivities, California encephalitis viruses did not produce plaques when the initial pH of the overlay was greater than 7.5. Optimum conditions for plaque formation by most of the viruses were noted when the initial agarose overlay pH was between 7.0 and 7.2.

ANTIGENIC PROPERTIES OF CALIFORNIA ENCEPHALITIS VIRUSES by Joseph Anthony Crookston A dissertation submitted to the faculty of the University of Utah in partial fulfillment of the requirem9nts for the degree of Doctor of Philosophy Department of Microbiology University of Utah June I 1975 eo 1\1 1\} 1 T TEE 1\ J) I' R 0 V A L SUP E R V I S OR y of a dissertation submitted by Joseph Anthony Crookston �! I h;ln� read this (lisscriatiolJ d�cto:,t de�ree. _J/:Z' Date 7� ________. . I 11:1\,C read this dissertation ;)f1<! cIoct oral degree. ----" f/, · el, ! /I 7, .; -- -�' -'� ';." -":..." . --Date' I . l'Iicmbcr., Super __ i .",.·:; COfllmittee I kn'c re:ld this dis:ilTt:Jt:(JIJ and han� found it to he' of satisfactory quality fnr dlxtoral degree. . )!2{! >�. __ Date . ______ __ _ .. � P' '/,1 :\ J/' -- ..-----.----.�---- - Pa1..l1 S. - --- --._-- .----.- ... -----�--- ---_.-._- . 'Ln1iibardi, Ph. D. ]vfcmOCT, SUP("[vjs:)r/ COlnrnittcc I !J;!\,c read this diss!'rl:l�io" d()( t ral degree . � . --q-:lJrL�----�--.--.' 7 Date J..1cmb{'!·, Supervisory COfllll:illec I have re:1d this diss"rlatin:; and In\'l� fmlnd it tn h� of sa;isfac[Qry quality [or docloral degree. ('/ /� / '" _ __ � Date ___ £J . L.L __ __ _________-- - Paul S. �icholcs. Ph.D. a .. Mt�:nber, Supe�vi,{)t·y· Committee dissnt:lliolt :tnd h;\\'l' f(lund it to hi' of �,lIl't,,( tOIY qll:llity fo,' 0 ---fJ1({,'�U<;�h".k '.,. .. (;rC��-�l . ' ,� � '.A l'!"rVl., ]..>\(' . .", "... I" -'T , �i(;Jnh('r. SIIP'T\'i';(Il), COJl!lIlit 1('1' '1) �y Ii-/ . I) . . ;1 UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of the University of Utah: Joseph Anthony CrookstOE I have read the dissertation of in its final form and have found that (1) changes suggested by the Supervisory Committee have been completed in the manuscript; (2) reference citations and bibliography are consistent and in an acceptable form; (3) all illustrative materials including figures, tables, and charts are in place; and (4) the final manuscript is satisfactory and ready for submission to the Graduate School. � Paul S. ------- Lombardi, Ph.D. Member, Supervisory Committee Approved for the Major Department �eP CF'.�---� Lowell A. Glasgow, M.D. Chairman /Dean Approved for the Graduate Council '- -,, ' I //( \ (( i (if'" ,/;.;' ;» S�terling M�/ McMurrin, braduate Dean )) .- , i pH.D. • _ ,- t�_._/ ACKNOWLEDGMENTS The author wishes to express his appreciation to Dr. Douglas W. Hill for the contribution he made to th2 author's education. While the effort he extended to pro- vide guidance and counsel were of inestimable worth in conducting this investigation, the opportunity to observe and share his philosophy toward the pursuit of knowledge and one's education was invaluable in the evolution of the author's academic attitudes. A feeling of indebtedness is also held for the suggestions and efforts of Drs. L. P. Gebhardt, S. Marcus, P. S. Nicholes, and M. Rogolsky. The time, advice and stimulating discussions offered by Dr. P. S. Lombardi are especially acknowledged. The author would also like to express his gratitude for and acknowledge the support and assistance offered by his wife, Janice. Finally, the financial assistance received from the National Science Foundation in the form of a predoctoral traineeship is gratefully acknowledged. TABLE OF CONTENTS PAGE .............. LIST OF FIGUHES ·... ·. AB8TRACT • • • .. ·.. .. INTRODUCTION . . . . . ·.·... REVIEW . . . . . . . . . ·..... LI ST O}' TABLES . • • LITE~ATunE I. I I. HISTORY OF THE CALIFORNIA ENCEPHALITIS ARTHROPOD-BonNE VIRUSES • • • • • • • •• IV. VI~USES • • • • • • • • • • VIII. ·... 6 18 CELL CULTUllES • VII. 3 PLAQUE FOJ1HATION BY CALIFORNIA ENCET-'HALITIS VIRUSES • • • . II. VI. 3 13 VIHUS STRAINS . V. 1 REDUCTION (NEUT~ALIZATION) AS A SEROLOG I CAL TECHNI r~UE • • • • • • • • • • I . IV. xii PLA~UE MATERIALS AND METHODS III. x A:t'YEIGEIJIC PROPERTIES A!\T]) RELATIONSHIPS AJ,!ONG CALIFO~,ijIA ENCEPHALITIS ARTHROPOD·- BORNE III. vii ~IEDIA ·...... ·...... ·... ... AND SOLUTIONS . ·.. ASSAY OF VIRAL INFECTIVITY ANIMAL IMMUNIZATIONS ·..... ·... VIRUS PURIFICATION . . . . . . . . . . . PROTEIN DETEHHINATION •• ·..... VIRUS NEUTRALIZATIO:1 STUDIES 20 20 24 28 29 30 31 32 PAnE 34 PLAQUE FORMATlnN 31 PURIFICATION OF CALIFORNIA E:rCEPIIALITIS VIRUSES • • • • • • • • • • • • • • • 60 III. VIRUS-A~rTIBODY IHTEnACTION STUDIES 66 IV. NEUTRALI~ATION STUDIES 92 I. II. DISCUSSION • RE}i'EnE:~CES • ..... . .. .. .... ... . ... .. . 116 131 142 VIrrA • • • • • vi LIST OF FIGURES PAGE FIGURE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. California encephalitis virus plaque nurruJers as a function of sample dilution . 40 ion of BFS-283(LP) virus in primary embryo fibroblasts . . 42 Replication of BFS-283 (Sp) v ir ..ls in primary chick embryo fibroblasts . 42 Repl ion of San Angelo virus in prlmary chick embryo fibroblasts . . . . 43 R9plicat of Tahyna virus in prlmary chick embryo fibroblasts . 43 single step growth rates of Comparison litis virus (30521) in California e Melnick's and 's minimum essential med ia .. . 45 Replication of BFS-283(LP) virus in a continuous line of baby hamster kidney cells 55 Replication of LaCrosse virus in a continuous line of baby hamster kidney cells . 56 Distrib'..ltion of infectivity following clarification of suckling mouse brain-propagated Trivittatus virus by sed tation through 25% sucrose onto a 65% sucrose cushion . 62 Distribution of sucrose cushion-clarified, mouse brain-derived v virus infectivity following sedimentation to equilibrium in a 25-65 percent sucrose gradient 64 Partial purification of mouse brain-propagated San Angelo virus by centri on of a crude suspension through 25-45 percent sucrose (A) gradients and rccentri tion to equilibrium (8) of a peak fraction (#18) through 25-65 percent sucrose 65 FIGURE 12. 13. Influence of the host cell used to propagate San Angelo virus on neutralization by rabbit anti-San Angelo virus serum . 68 Neutralization of BFS-283(LP) virus by homologous "early immune rabbit serum collected 7 days after intramuscular injection of virus . 73 Neutralization of San Angelo virus added to previously incubated San Angelo virus-homologous antibody mixtures 75 Neutralization of chick cell-propagated San Angelo virus by antiserum pre-incubated with heat inactivated mouse brain-derived San Angelo virus or normal mouse brain material 77 Comparison of neutralization of mouse-brain and chick cell-derived San Angelo viruses by rabbit anti-San Angelo virus serum, assayed on primary chick embryo and modified baby hamster kidney cells 82 Neutralization of mouse brain-derived LaCrosse virus by rabbit anti-LaCrosse virus serum. 84 The effect of initial virus concentration on the rate of neutralization of chick cellpropagated San Angelo virus by homologous hyperimmune rabbit antiserum. 93 Neutralization of baby hamster kidney and chick embryo cell-derived San Angelo viruses by rabbit anti-San Angelo virus serum . . 95 Neutralization of California encephalitis viruses by mouse anti-BFS-283 virus ascitic fluid . . • . 99 Neutralization of California encephalitis viruses by mouse anti-BFS-283 virus ascitic fluid . 99 Neutralization of California encephalitis viruses by antisera collected from two rabbits submitted to similar schedules of immunization with BFS-283 virus . 100 Influence of antiserum concentration on the kinetics of homologous and heterologous Ca fornia encephalitis virus n(~utralization 102 II 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. viii FIGURE 24. 25. 26. 27. 28. 29. Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainde~ived BFS-283(LP) virus. . . • . . 105 Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainderived LaCrosse virus ..•.. 107 Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainderived Tahyna virus . • . . • . • . . • . . 109 Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainderived Jerry Slough virus .••. . •• III Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainderived Trivittatus virus . • . . . . .• 112 Neutralization of California encephalitis viruses by rabbit antiserum to mouse brainderi ved San Angelo virus •..• .•.. 114 ix LIST OF 'IIABLES TABLE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Viruses of the Californ encephalitis group included in the study • . 21 Inoculation schedule for preparation of lIearly immune I', lIimmune" and IIhyperimmune " rabbit antisera to members of the CEV group 30 Influence of temperature during virus sample dilution and adsorption to cells on California encephal is (30521) virus plaque numbers . 37 Time required for adsorption of California encephal is virus (30521) to primary chick embryo cells at 25 C 38 California encephalitis virus plaque size and quality as a function of sodium bicarbonate concentration . • . 46 Plaque formation by California encephalitis viruses on primary chick embryo cells . 49 Effect of init 1 overlay pH on the morphol::Jgy of plaques formed on pri1nary chick embcyo cells by Californ encephalitis group viruses 51 Plaque formation by California encephalitis viruses on primary chick embryo cells . 52 Comparison of plaque-forming unit titers of Ca lifornia encephali tis gr::Jup virus sllspen:3io!1s when assayed on chick embryo fibroblasts and baby hamster kidney cells . 57 Neutralization of San Angelo virus propagated in mouse brain and chick embryo cells by rab-bit antisera to normal chick cells and chick cell-propagated San Angelo virus 69 Neutralization oE mouse brain- and chick embryo cell-derived San Angelo and BFS-283 viruses by rabbit anti BFS-283 virus serum . 71 TABLE 12. 13. 14. 15. 16. 17. 18. 19. PAGE Influence of antiserum dilution on the neutralization of mouse brain-propagated San Angelo virus by rabbit anti-San Angelo virus serum . . 78 Enhancement of neutralization of monse bra propagated San Angelo virus by goat antirabbit globulin serum following incubation with rabbit anti-S3n A~gelo virus serum 80 Effect of trypsin treatment on the infectivity of partially purified LaCrosse virus propa ted in BRK cells and the suckling mouse . 88 Effect of phospholipase C treatment on the infectivity of partially purified mouse brainderived LaCrosse virus 90 Residual antiviral antibody activity of rabbit sera after 100-fold dilution beyond pre-determined optimal dilution for neutralization of homologous virus 101 Effect of extended time on the neutralization of selected California encephalitis viruses by rabbit anti-mouse brain-derived LaCrosse virus serum . 108 Neutralization of California encephalitis viruses by rabbit anti-Tahyna virus serum 110 Influence of extended time and increased concentration of antiserum on the neutralization of California encephalitis viruses by rabbit antiserum to Jerry Slough virus propagated in suckling mice . 113 xi ABSTRACT The kinetics of inactivation by antiserum of several strains of Cal ornia encephalitis virus were examined by the technique of plaque reduction. The antisera employed were ob-tained from rabbi ts following multiple injections of virus strains pro?agated in the brain tissues of suckling mice. The viruses used in the plaque reduction studies were propagated in cultures of baby hamster kidney cells. This was done to maximize the likelihood that the interaction between virus and antibody would involve virusspecific determinants and minimize the possibility of interference by antibodies directed against host cell components. An antiserum dilution was predetermined for each homologous virus-antiserum combination so that similar rates of inactivation resulted. That is, when the results of each of the homologous neutralization reactions of several viruses were compared graphically, the curves were found to coincide. When attempts were made to neutralize heterologous virus strains with each of these sera at their unique dilutions, three types of reaction were apparent. In some, the rates of inactivation of certain heterologous viruses were similar to that of the homologous virus, that is, inactivation was immediate u.nd without a lag following u.dmixture. In the second type of cross-reaction, a significant lag preceded neutralization, while in the third type of reaction no indication of cross-reaction could be detected. The demonstration of a kinetic lag preceding neutralization supported the concept of multihit inactivation kinetics rather than single hit kinetics. The differences between the two types of heterologous cross-reactivity suggested the significance of the configuration of the antigenic determinants in determining the ability of a cross-reacting antibody to "recognize" a portion of the determinant. In the course of preliminary experiments it became apparent that phenotypic differences arose in genotypically identical viruses when different types of host cells were used to propagate the viruses. A significant persistent fraction of surviving virus was noted when neutralization experiments were conducted with viruses propagated in suckling mice. This fraction of non-neutralizable virus could not be reduced by any treatment of the virus suspension or the antiserum or by using different assay cells. Attempts to modify the surfaces of mouse brain-derived viruses by enzymatic digestion with trypsin, phospholipase C and neuraminidase did result in the detection of differences in the surfaces of viruses grown in these cells and in baby hamster kidney cells. However, treatment of mouse brain- propagated viruses with these enzymesilid not enhance their neutralizability. xiii As a part of this study, experiments were also conducted in which variables influencing plaque formation by California encephalitis viruses were investigated. The plaquability of the viruses was found to be extremely sensitive to the pH of the agarose overlay. In contrast to other viruses that exhibit pH sensitivit s, California encephalitis viruses did not produce plaques when the initial pH of the overlay was greater than 7.5. Optimum conditions for plaque formation by most of the viruses Were noted when the initial agarose overlay pH was between 7.0 and 7.2. xiv INTRODUCTION The data obtained from the techniques previously appl d to the examination of antigenic differences and similar ies between viruses of the Ca group of arboviruses have been fornia encephalitis rpr~ted as an indication that only minor differences exist between many of the viruses. Considerable antigenic overlap was apparent when immunodiffusion and immunoelectrophorectic techniques were util to examine and compare members of the group: Several virus strains Were indistinguishable by these methods. Complement fixation comparisons were more discriminatory, although heavy reliance had to be placed upon the reproducibility of single two-fold serum dilution titer differences. Other attempts desi d to further the understanding of the antigenic relationships of these viruses have relied upon modification or refinement of the abovementioned techniques instead of the application of methods in which crossreactivities are minimized. attempt to gain further This situation prompted an ight into the nature of the viruses' intragroup antigenic relationships by studying the e ct of specific homologous and heterologous antiviral antibodies on the ability of different virus strains to success fully infect cells. AeJdi tionally , conditions which 2 affected neutralization of the viruses as well as factors that influenced the production of plaques by the viruses on different cells were studied. LITERATURE REVIEW I. HISTORY OF THE CALIFORNIA ENCEPHALITIS ARTHROPOD-BORNE VIRUSES -The first lation of a virus (Bakersfield Field Specimen #91, BFS-91) belonging to what has subsequently been named the Ca fornia encephalit (CE) group of arthropod-borne viruses was made in 1943 from mosquitoes during an encephalitis epidemic in southern California involving Western equine encephalitis and St. Louis encephalitis viruses (Hammon et al., 1945). virus in North America was unique The recovery of this that no known disease could be associated with the agent at that time, although the sera from several animal species, including man, possessed antibodies which neutralized the ctivity of the virus. In 1944 two additional isolations of the same virus (BFS-283 and BFS-395) were made from mosquitoes in the same locality (Hammon and Reeves, 1945). Early serological surveys implicated small mammals (ground squirrels and rabbits) as reservoirs of infection in nature (Hammon & Reeves, 1947) . Infections produced experimentally in these animals were inappareni, although a demonstrable viremia occurred (Hammon & Reeves, 1952b). These results, as well as subsequent observations, indicated thilt, as a parasite, 4 the virus was well adapted to its hosts (Sudia et al., 1971). Serological stud s of the seril from patients with undiagnosed encephalitides from areas outside California (Washington, Colorado, North Dakota, Kansas, and Japan) did not provide evidence for the presence of this virus in these areas, so the agent was assumed to have a very limited geographic distribution (Hammon & Reeves, 1952a). Corrobora- tive findings from the study of a number of domestic and wild animal spec s supported this conclusion. In North Dakota in 1948 an antigenically related but distinct virus was recovered from Aedes trivitattus mosquitoes (Hammon et al., 1952b). This virus was given a designation reflecting the species of mosquito from which it had been isolated; most other California viruses have names associated with their geographical origin. Evidence of more extensive distribution of these viruses or viruses exhibi ng close serological relationships was presented by the isolation of a strain (Melao) from mosquitoes in Trinidad in 1955 (Spence et al., 1962) and again in Brazil in 1957 (Causey et al., 1961). Similarly, repeated isolations of another virus were made from mosquitoes in Mozambique in 1959 and 1960 (Kokernot ct al., 1962). tion Lumbo virus. This strain received the designa- The presence of CE virus in Europe was ascertained in 1958 when Tahyna virus was isolated from mosquitoes in Czechoslovakia (Bardos and Danielova, 1959). SerologicCll surveys of humans in different Europe::1n areas 5 suggested an endemic status of a California virus throughout central Europe (Bardos, 1960: Bardos & Sefcovicova, 1961). In the United States, no isolations were made after Trivitattus virus until 1958, when San Angelo virus was found in mosquitoes in Texas (Grimes et al., 1962). California encephalitis viruses were detected only in mosquitoes until 1959, when the Snowshoe hare strain was isolated from the blood of an animal described as an emaciated, rather sluggish snowshoe hare . • . dorfer et al., 1961). II 11 • • (Burg- The following year, isolations of Tahyna virus were made from the sera of two apparently healthy humans (Likar & Casals, 1963). These were originally defined as new isolates and referred to as Trojica viruses, although subsequently they were found to be identical to Tahyna virus (Taylor, 1967). In the United States, the first human isolate (LaCrosse) was recovered from the brain t sues of a 4-year-old child who died of meningo-encephalitis in 1960 (Thompson et al., 1965). Additional mosquito isola- tions have given rise to other partially characterized California viruses, e.g., Jamestown Canyon in Colorado (1961; Taylor, 1967), Jerry Slough in Southern California (1963; Taylor, 1967), and Keystone in Florida (1964; Bond et al., 1966). Of interest is the observation that no CE viruses Were isolated in California during the 20 year period following the initial isolation, although a large nUI~er of mosquitoes Were trapped and examined each year 6 (Gresikova et al., 1964). Since 1965, California group viruses have been found in at least 26 additional states and the number of isolations have continued to increase each year (Sudia et al., 1971). Coincidental with the increasing frequency of virus isolations has been the number of human infections detected, d gnoseo and reported. Hammon and Reeves (1952a) described three clinical cases presumably attributable to these viruses in 1943-44; the next human case was not recorded until 1963, a significant span of 20 years (Quick et al., 1965). Available data suggest that the incidence of human involvement was, at best, limited before 1964. Beginning in that year, epidemics were reported from Indiana (Marshall, 1964) and Ohio (Spencer, 1965) and subsequently from Wisconsin, Minnesota and Iowa (Sudia et al.,1971). II. ANTIGENIC PROPERTIES AND RELATIONSHIPS AMONG CALIFORNIA ENCEPHALITIS ARTHROPOD-BORNE VIRUSES When the increasing prevalence of Cal ornia enceph- alitis viruses became apparent, attempts were made to clarify the antigenic relationships of these viruses among themselves and other arthropod-borne viruses. The earliest efforts Were restricted by the techniques then available and by a limitation shown by some of the viruses: Despite repeutc:c1 efforts, hemagglutination could not be de1110nstratcd for some virus strains. Thus, none of the MelLio 7 strains (Spence et al., 1962: Whitman and Shope, 1962), the Lumbo (Kokernot et al., 1962) or Snowshoe Hare (Burgdorfcr et al., 1961) viruses were initially thought to "produce" hemagglutinins. Conversely, Tahyna (Bardos et al., 1961), LaCrosse (Thompson et al., 1965) and BFS-283 (Whitman and Shope, 1962) viruses were shown to exhibit hemagglutinating activity, although the antigens were characteristically low-titred or irregularly extractable. More recently, techniques have been described which have resulted in the isolation of hemagglutinating preparations from all California viruses studied. These methods include trypsin treatment of sucrose-acetone-prepared antigens (Hannoun, 1968) and sonication followed by trypsin treatment (Ardoin et al., 1969). Less vigorous procedures have been required to obtain satisfactory complement-fixing antigens (Hammon and Sather, 1966). This property proved advantageous when com- plement-fixation (CF) tests were observed to be more discriminatory than neutraliza on (assayed by inoculation of virus-serum mixtures into suckling or weanling mice) in the differentiation of strains of CE virus. For example, anti- serum to LaCrosse virus neutralized 2.9 logs of the homologous virus, 2.8 logs of BFS-283, and 2.9 logs of the Snowshoe Hare virus. When assayed by CF, the titer of the homologous virus-serum mixture exceeded those of the heterologous mixtures by a factor of 4 (Thompson et al., 1965). Other studies have corroborated and extended this observation 8 (Hi:.lmmon et ul., 1952b; Newhouse et a 1., 1964; Bond et ill., 1966) . The in vivo measurement of neutralization is not without advantage, however. Although the greater apparent specificity of CF resulted in reliance upon that test for estimation of the antigenic relationships among CE viruses, the relatively short time period over Which antibod spe s fic for CF antigens can be detected in sera limits the testis utility. Hammon and Reeves (1952a) found 32 of 294 human sera capable of neutralizing BFS-283 virus but none had demonstrable complement-fixing activity. Similarly, 44 of 118 human sera examined by Gresikova et al. (1964) exhibited significant neutralizing capacity While none was able to fix complement. Recent clinical studies have pro- vided evidence indicating that antibodies capable of interacting with hemagglutinating antigens are short-lived, increasing rapidly to maximum titer and then decreasing in titer to undetectable levels after a short time (Quick et al., 1965). Antibodies to complement-fixing CE virus antigens increase in activity more slowly but are also short-lived (Bond et al., 1966). In contrast, the neutral- izing activity of sera increases rapidly and is maintained at detectable levels for a much longer tim;:'! (Chun et al., (1968). The unique antigenic character of the original CE isolate wi t.h respect to other then known arboviruses was 9 first indicated by the experiments of Hammon, Reeves and Gillindo (1945); no cross-reactivity was noted with any other virus. Subsequently, Espana and Hammon (.l948) demonstrated a subtle relationship between BFS-283 and St. Louis and Japanese encephalitis viruses. In the CF (but n,:)t neutralization) test, antiserum t() the California virus reacted with undiluted, benzene-extracted an~igens of the heterologous viruses; no relationship could be detected in the recipr8cal tests. The L'. un'ba strain fro~ South America was foun:] to be neutralized by Bwamba fever virus antisera (Kokernot et al., 1962). The latter, isolated from humans in Afri2a, is the prototype of a small but distinct serolog group (Smithburn et al., 1941). The re detectable by neutralization and was An antigenic similarity 1 arbovirus ionsh was only unidirecti~nal. CE group viruses and members of the B"dnyam\vGra virus l)"roup \Vas su!]ges ted 'tlhen G'Jaroa virus was Sh·:)Wll to exh ibi t and Shope, 1962). llle::n~"J2~ Th features of both groups (Whi tman virus has been classified as of the Bunyam\v[=cd group be:::;aus:~ d of cr::-' cross-reactions with other representatives of the same group. Neither Guaroa virus nor its antiserum reacted with Bunyamwera virus antisera or the respective viruses in hemagglutination inhib ion or neutralization tests. relationsh Conversely, the between Guaroa virus and the California viruses could not be detected by CF studies (Sather and Harrunon, 10 1967); cross-reactions between the viruses and heterologous sera were apparent only in neutralization and hemagglutination inhibition studies (Whitman and Shope, 1962). Within the California group of arthropod-borne viruses the antigenic relationships appear extremely complex when examined by the CF technique (Sather and Hammon, 1967). The high degree of cross-reactivity has been interpreted as an indication of relatedness attributable to "antigenic instability" (Hammon and Sather, 1966). In several instances virus isolates have been designated as distinct strains, although their specific antisera reacted to equal titer in the CF test with some heterologous virus antigens or d tinguished the homologous virus from some of the heterologous viruses by only a two-fold dilution difference (Sather and Ham~on, 1967). These d iculties notwithstand- ing, the California virus group was tentatively organized into 11 subgroups, eight representing the viruses alluded to above that had been isolated in the United States. Murphy and Coleman (1967) applied the Ouchterlony technique of immunodiffusion to the serological study of these viruses, using infected suckling mouse brain tissues as the source of antigen. Their results were agreement with the complement fixation studies of Sather and Hammon (1967). Subsequently, 'Wel1ings et ale increase the speci (1970) were able to city of the immunodiffusion technique by e l.iminat.ing (through adsorptio:1) non-viral antigen-antibody 11 reactions and by using antisera obtained from inoculated animals within two weeks of the primary injection. In 6 of 9 instances these sera were strain ("type") specific while sera collected after subsequent injections possessed broader group specificit s. These sera were also used to examine the electrophoretic properties of the viral antigens and the same specificities were noted (Wellings et al., 1971) . Sera obta d after a series of virus injections gave rise to multiple arcs of precipitation folloifJing electrophoresis of the homologous or heterologous viruses. Preliminary adsorption of these sera with a heterolog0~s, cross-reacting virus did not necessarily eliminate arcs present in the homologous virus-antibody reaction. Based on the results of these experiments, it was suggested that the classif tion of the California group be revised to contain three subgroups, represented by BFS-283, Trivittatus and Keystone viruses, the remaining members to be considered as strains of BFS-283. Guaroa virus reportedly shared "minor antige:1ic determinants" with all members tested except Trivittatus, with which two "antigens" were found In common. Calisher and Maness (1970) screened imm'.lne sera from several host sources representing a variety of imnunization schedules by complement fixation, hemagglutination inhibi tions, and neutralizatio:1 tests. Selecting sera [or titer, specificity, availability and other features, the cross- 12 reacti v i tics of group members were co:npared by the d :::rJ.b Ie diffusion-in-gel tech~ique. the density of the precipitin arc, reaction, w~s Reactivity determined by "0" indicClting no "4" indicating "maximum intens i ty and a narro''''' precipitin arc". S~owshoe In these experiments Hlre virus could n0t be distinguished from seven other viruses while BFS-283 and LaCrosse viruses were only marginally different from each other and five oth'2r viruses. The precipitin lin·2s 0f th2 hO:1101090US react ions of these v iruses were interpreted as 114+" arcs whereas the pre pitation intencaS~2 s it ies of the six heterolog'')us v lruses in each reported as "3+". V/!2re Mela::::> and Trivittatus viruses exhibited absolute specificity by lacking cross-reactivity with other group me;:nbers and were sugg-ested as sub·-grol.lp prototy?es. Three o:'her suh9:LOUPS 'Nere proposed: These consisted of B23-283, SnJwshJ'2 Hare and LaCrosse viruses; Jamestown Canyon and Jerry Slough viruses; and rrahyna and Lum:::>o viruses. San Ange 10 v lrus "vas unique in pro\! ld ing an "an-tigenic bcidge" between the lJroups containing BfS-283 anrJ Jamestown Canyon v iruse:3. P:cesently, this group of viruses hilS be:::;!) separated into three complexes accor-d ing ·to the Sl1bcolTI!,l:l. t Irrunl1nolo9ical R,~lationships on Among CataloguGc] Arth_~:-t)pod-borne Viruses (citecl in Sudia et al. includ(~ te(~ I 1971). Th(~se c,)jnple:(c-~s Tri\l.ittatus, Melao and BFS-283 as prob::>types ,vh3_1<.::: the rema. i-,lin(] (characl:eriz(~d) members are regc:u,"'(]ed as !3ubtypC:3 13 of BPS -283. Lllmbo ancl Jer ry S 10ugh virusc~s are cons ide! Led to be var iants of Tahyna and lJamesto\Vl1 Canyon viruses, respectively. III. PL1\Q:JE REDUC'f ION (NEU'fIV\.LIZAT ION) AS A SEROLOGICAL TECHNIQUE Casals (1957) demonstrated that in the study of arbo- viruses, compleroont fixation occupies an intermediate position between hemagglutination inhibition and neutralization (assayed in mice) in specificity, i.e., lack of cross-reactivity. Although each of these methods has as a feature a quantal nature, neutralizatio~ was considered to be the m8st specific or least cross-reactive. This increased specificity is, in part, dependent on the route by which the assay animals, usually mice, are challenged. Numerous studies have also indicated the greater sensitivity of these in. ~ivQ. neutralization tests, especially When mice are inoculated intracranially rather than intraperitoneally (Burgdorfer et al., 1961; Porterf 1964). when C:l Id, 1962a; Kunz et al., Further, although no differences might be expected virus-antiserum mixture is assayed in vivQ. or in vitro (Porterfield, 1962a), the latter has generally been reported to be more sensitive as well as specific for detection of antigenic similar (Porterf ies and d ferences between viruses Id, 1962b; Daniels, et al., 1961). Utilizing the quantitative techniques of plaque formation, Dulbecco ct al. (1956) demonstrated that the 14 ncutrillization of an infectious virus suspension could be followed kinetically and, hence, the rate of inactivat.ion could be ascertained. McBride (1959) presented evidence to indicate that the rate of inactivation of a poliovirus serotype by its homologous antiserum was always greater than that observed when a heterologous serum was used. observation has been co~firmed This by a number of other investi- gators employing a variety of viruses and antisera (Ashe & Scherp, 1963; Ozaki & Tabeyi, 1967). Inherent to this approach is the idea that the degree of antigenic relatedness will be reflected in the slope of the plotted neutralization curve, the curve of the ho~~logous virus-serum reaction (as a function of time) being characterized by a greater slope than any heterologous reaction and the curves of heterologous combinations reflecting the relative antigenic similarities among different viruses. Because the slope of a curve can be described mathematically, attempts have been made to summarize the antigenic relationships of viruses in quantitative terms (McBride, 1959). Viral neutralization is presumed to be a bimolecular event (or series of bimolecular events) in which one or more molecules o~ antibody interact with an infectious virus particle in some manner to bring about inactivation. Because the serum concentration is usually great enough that the chungQ in the number of antibody molecules is negligible, ncutrQlization hilS generally been described in terms of 15 first order (monomolecular), rather than second order, chemical kinetics. dV - dt Thus, (1) = kV That is, the change (decrease) in virus concentration with respect to time -~~ is proportional to the concentration of the virus (V) or equal to the product of the virus concentration and a velocity constant, k. When this relation- ship is integra ted between two virus concentra tions V and o Vt at zero minutes and t minutes, respectively, k can be calculated from the following: (2 ) Dulbecco et ale (1956) utilized equation (2) in modi- fied form to include the influence of the antiserum: K(t-t o ) In Vo = D Here, K represents the the serum dilution. (3 ) Vt m~dified velocity constant and D, Graphically, this modified first order equat.ion resembles its parent form, the plot being linear and negative in slope (in the presence of excess antibody). Whereas true first order reactions are characterized graphica lly by linear exponential decline, m~st virus- antibody systems depart radically from this by approaching a horj.zontal asymptote as the reaction proceeds (Lewenton- 16 Kriss and Mandel, 1972). This departure from first order kinetics has been described as non-neutralizable virus (Mandel, 1961), a persistent fraction (Dulbecco et al., 1956) a protected fraction (Lafferty, 1963a) and simple equilibrium between dissociated and undissociated virus and antibody (Burnet et al., 1937). The size of the per- sistent fraction varies between antiserum preparations (Lafferty, 1963b) as well as within single antiserum pre·paratio:1s collected at different times during the immunization process (Lewenton-Kriss and Mandel, 1972); between virus preparations (Lewenton-Kriss and Mandel, 1972); and between cell species up:Jn which neutralizatio:1 is being assayed (Kjellen and Schlesinger, 1959). While the explanation of the persistent fractio:1 has not been unequivocally ascertained, its existence is predicted by most models regarding virus-antibJdy interactio:1 (Lafferty, 1963a; Lewenton-Kriss and Mandel, 1972). Neu- tralization is considered to be a tWo-step reaction 1n which the virus particle first becomes reversibly sensitized by interactiO:1 with one or more molecules of antibody and then beco~es stabilized (neutralized) or continues to exist as an infectious virus-antibody complex. Stabilization has been attributed to binding of the second combining site of an antibody molecule to a second antigenic determinant on the same virus particle (Lafferty, 1963b); to the attachm~nt of an ilntiboc1y ffi:Jlccule t_o anyone of several "critical" 17 sites present on the virus surface (Rappaport, 1970); or to the attachment of a final antibody molecule to a "critical area" (composed of "groups of antigen-antibody complexes" but otherwise similar to any other area of the virus surface; Westaway, 1965b). The existence of infectious com- plexes of virus and antibody has been demonstrated both in vivo (Notkins et al., 1966) and in vitro (Ashe & Notkins, 1966). That antibody is associated with non-neutralized virus is indicated by the secondary neutralization resulting from the addition of specific antiglobulin. A second point of departure from linearity, manifested by an initial lag phase, has also been described but for fewer systems. Lafferty (1963a) described a "shoulder" which preceded the linear, exponential loss of infectivity following mixture of virus and antibody. Although this lag was only apparent at temperatures below 37 C, the same phenomenon has been reported when antibody was present at low concentration (Burnet et al., 1937; Kalmanson et al., 1942) and when 195 antibodies were employed (Philipson, 1966). The demonstration of a lag in the kinetic study of virusantibody interaction suggests "multiple hit" neutralization kinetics, that is, the requirement for more than one molecule of antibody to neutralize a virus particle (Westaway, 1965b). Experimental evidence for this has been presented by Gard (1957), Philipson (1966), and Wallis and Melnick (1970). The hypothesis that neutralization is 18 normally effected by a single molecule of antibody (single hit kinetics) hus, however, received more popular support (Rubin and Franklin, 1957; Lafferty, 1963b; Lewenton-Kriss and Mandel, 1972). IV. PLAQUE FORMATION BY CALIFORNIA ENCEPfffiLITIS VIRUSES Several investigators have described conditio~s sup- porting formation of plaques by members of the California encephalitis group of viruses. and continu~~s cell lines have Primary cultures of cells bea~ studied b~t limitatio~s have been reported for most of these. Some of the diffi- culties include length of incubation time required before plaques large eno~gh to detect are formed, poor efficiency of plaquing and inability to extend the use of a particular cell to all CE strains. Among primary cell cultures, thQse of chick embryo cells have been used to plaque Tahyna virus (Porterfield, 1960) and BFS-283 (Crookston et al., 1968): Duck embryo cells also supported plaque formation of these viruses as well as Snowshoe Hare and a New Jersey isolate designated as South River virus (Henderson and Coleman, 1971). Plaq~es were not produced on duck embryo cells by Trivittatus '=>r Melao while LaCrosse and Keystone viruses gave rise to faint, unco'-lntable plaques. Henderson and Coleman (1971) also exa~in'2d plaq'-1c forination on primary ham3ter embryo, h3.'nster kidney und rh2sus monkey cells: With each of these at least one 19 of the above-mentioned viruses did not produce plaques while the plaques of some of the strains that did form were too faint to be counted. Hamster embryo cells were most suscep- tible to plaque formation by these viruses (except Keystone), however, the plaque titers observed were at least one log lO'wer than those determined by suckling mouse intracranial inoculation. Continuous cell cultures have also been shown to support plaque form3tion by CE viruses. Berghold and Maz~ali (1968) were able to plaque Melao virus on baby hamster kidney cells (BHK-21), green monkey kidney cells (VERO), and rhesus monkey kidney cells: no other members of the group were studied. Stirn (1969) used VERO and rhesus monkey kidney (LLC-MK 2 ) cells and was able to demonstrate plaques for every CE virus that he studied. The tim:; necessary before plaques became visible varied between 2 and 10 days, while maximum plaque diameters reached between 16 and 19 days was 1-15 mm. Henderson and Coleman (1971) studied plaq',18 formatio:1 on BHK-21, porcine kidney (PK-13), HeLa, green m~nkey kidney, KB and human diploid lung cells. Only Blffi-21 cells supported plaque formation of all the viruses tested. In each case suckling mouse inoculation indicated higher titers than were obtained by plaque quuntitation. MATERIALS AND METHODS I. V IRlJS S'fR,\ INS The strains of California encephalitis virus used throughout these stud sources. s were obtained from several The virus designation, passage history at time of receipt and the individual and laboratory from whom each virus was obtained are indicated in Table 1. The virus designated "3052111 was isolated on June 22, 1965, from Aedes dorsalis mosquitoes at Blue Lake, Utah. Hemagglutination inhibition, complement fixation and neutralization tests, the latter in mice, were used to identify the virus as being closely related to BFS-283 (Crane et al., 1970). Upon receipt, each strain of CE virus was inmediately inoculated intracerebrally into suckling mice. When the animals became moribund they were frozen at -70 C. After thawing, the brains were removed and mixed with phosphatebuffered saline (PBS) to obtain a 10 percent (vol/vol) suspension. Dilutions of this material were inoculated onto mono- layers of chick embryo cells and assayed for plaque formation. Plates containing fewer than 10 plaques were used to obtain plugs from the agarose above characterist.ic plaques: 21 TABLE 1 VIRUSES OF THE CALIFORNIA ENCEPr~LITIS GROUP INCLUDED IN THE STUDY: VIRUS DESIGNATION, PASSAGE HISTORY AND SOURCE. Passage Virus BFS-283 MBP-8, HP-l* 30521 MBP-3 " LaCrosse (LaX) MBP-8 Rocky Mounta Lab Hamilton, MJntana (L.A. Thomas) Tahyna (stra 92, TAH) MBP-20 II Snowshoe Hare (SNH) MBP-14 II Jerry S18ugh (BFS-44 70, JS) MBP-5 " San Angelo (20230, SA) ? .. Tr i vi-t ta t:1S (7941, TVT) MBP-3 " Jamesto#D Canyon (JC) MBP-5 Publi2 Health Service Lab., Greeley, Colorado (J.S. Lazuick) Melao (Tr9375, MEL) MBP-3 Center for Disease Control, Atlanta, Georgia (G.E. Sather) Keystone (KEY) ? * MBP mouse brain pass; HP Dugway Prov Grour}rJs Utah (K.L. Smart) Walter Reed Army Institute of Research Washington, D.C. (J.M. Dalrymple) = hamster pass. I 22 these were mixed with PBS and inoculated into litters of suckling mice. This plaque purification procedure was repeated two additional times; after the third passage into suckling mice, stocks were prepared by inoculating cranially several litters of the same animals. int~a­ These stock pools consisted of 10 percent (vol/vol) infected mouse brain suspensions in PBSi cellular debris was removed by low speed centrifugation. at -70 C. The preparations were dispensed and stored Viruses propagated in different kinds of cells (chick embryo or baby hamster kidney) were obtained by infecting cell cultures of each of these with a mouse brain virus preparation and collecting the supernatant medium after infection. These virus-containing fluids were clarified by low speed centrifugation, dispensed into one ml aliquots and stored at -70 C. II. CELL CULTURES The techniques described by Dolana (1968) were followed in the preparation of primary chick embryo cells. Ten day old chick embryos were aseptically removed from eggs and placed in a sterile Petri dish containing cold phosphatebuffered saline without magnesium or calcium salts belo\v). (PD, see After removing the head, limbs and viscera from each eniliryo, the remaining material was washed 4 times in cold PD, drained and minced with scalpels. These minced tissues were repeatedly washed with cold PD until red blood cells 23 could not be detected in the supernatant fluids. The washed, minced tissues were digested by repeated 5 minute exposures to 0.25 percent (wt/vol) trypsin dissolved in PD. After each incubation with the enzyme the supernatant, containing dispersed cells, was filtered through 6 layers of cotton gauze and collected in chilled Melnick's growth medium (see below) containing 20 percent (vol/vol) calf serum. The latter was relied upon to prevent continued trypsinization. When sufficient numbers of cells had been collected, the suspension was centrifuged, the supernatant decanted and the pellet of cells resuspended in fresh growth medium. The cell concentration was determined by hemocyto- meter count and enough growth medium was added to dilute the cells to a final concentration of 1 x 10 6 cells per mI. Five ml of this suspension were added to each 15 x 60 mm plastic Petri dish (Falcon Plastics, Oxnard, California). After 40-48 hour's incubation the cells were confluent and were used to assay viral infectivity. A mutant strain of the continuous line of baby hamster kidney cells (BHK-2l/C13) was also employed. This cell, designated BHK, was provided by J. C. Taylor and was obtained by treatment of the parent clone with nitrosoguanidine (N-methyl, N'nitro-N-nitrosoguanidine, Aldrich Chemical Co., Milwaukee, Wisconsin) and subaequent selection procedures involving dependency upon added thymidine, hypoxanthine and folic acid. The mlltant clone of cells was characterized 24 by ,lluch slower acidification of growth med iUIll than was th::! wild type cell. Cells used for infectivity assays or propagation of pools of virus were cultured in 250 ml plastic tissued culture flasks California). (Falcon Plastics, Oxnard, Confluent monolayers were remaved from the plastic surface and dispersed by a trypsin-EDTA solution (ATv, See below) and then suspended in Eagle's minim.llll essential mediu:l1 (MEJi1, see below) supplemented with ').2 p2rcent Bactotryptose (Difco, Detroit, Michigan) percent (vol/vol) calf serum. a~d 10 The cell concentratio~ was adjusted to contain 1.5 x 105 cells per mI. Five ml of the cell suspension were added to each large flask. After 4)- 48 hO;..1rs of inc'ubation, cells had reached confluency. III. MEDIA AND Melnick vatio~ I S S'JLU'rIO~S complete gcovvth ;l1ediulTI of chick embryo cells, while E3.g tial medium (MEJv1) was '..1sed for kid:1ey cells. ~..lsed VJ3S IS propagatio~ for cl.llti- minim'..1:l1 essenof baby ha~;3tei:' Th2 latter mea iulU was reconst itLlteeJ from ':h2 ?owelcred [orin (Grand Island Biolo9ical Co., Gra:1d Island, Ne'w York) with rjistilled 'dater a~d was supplemented '/Jith 10 percent (vol/vol) calf serum, 0.2 percent (wt/vol) tryptose (Difco, Detroit, Michigan) and 0.35 gil sodium bicarbonate. Th,::! complete medium tlll-OiJ(J~1 \-JdS steri 1 ized by Seitz filtratiO'1 type ST ster i 1 iz ing f i 1 ters tion, S21n Francisco, California). (Here ule s Fi 1 tor Corporu- 25 Each Ii tor of Melnick s growth medium was prepared as I follows: lOX Hank1s balanced salt solution. 100 ml 50 ml Calf serum Lactalbumin hydrolysate (Nutritional Biochemical Corp., Cleveland, Ohio) 5.0 g NaHC0 3 (0.035 g/ml stock solution) 10 ml CaC12.2H20 (1.325 g/ml stock solution) 1.4 ml Penicillin 105 units Streptomycin Distilled H20 . 850 ml Calf serum was collected at a lo~al slaughterhouse and processed or was purchased from Grand Island Biological Co. Each lot of serum was tested for cytotoxic and antiviral activity before use. The completed medium was sterilized by Seitz filtration. Concentrated stocks of Hank1s balanced salt solution contained the following chemicals in each liter of lOX solution: NaCl 80.0 9 KCl 4.0 9 MgS0 4 ·7H 2 0 . 2.0 9 KH 2 P0 4 • 0.6 9 Na 2 HP 04 0.48 9 Glucose 10.0 9 Phcno 1 Reel • 0.2 9 26 For prop~gation gI":)VlLh tn:l(li!.Hn V·F1S of large pools of virus u~:;ed Mel~ick's after slight modification: 'llhe calf serum concentration was reduced to 2 percent (vol/vol) and Tryptos~3 phenol red was not included. (2 gil) was added when the med ium \,.,as to be used wi th baby hamster kidney ce lIs . . The overlay used during the assay of viral infectivity consisted of MSM, 5 percent (vol/vol) calf serum, 0.2 percent (wt/vol) tryptose, 0.02 M ·tris (tris aminornethane I Ohio» (hydro~{ymeth:!l) Nutri tional Bioclv::;mical Corp. I Cle-"eland I and 0.4 percent (wt/vol) agarose (Van Waters R0gers Scientific, San Francisco, California). a~d The additi8n of agarose contributed to acidification of the overlay; thi3 was anticipa.ted \vhi1e init lly adjusting the pH of the tris SOl 11tion. Ph:::;,sphate-buEfered saline (PBS, pH ).2) was used to wash cells and, after addition of 0.2 percent (wt/vol) b:::;,vine serum albumin (BSA, Fraction V, Calbiochem, Los A'1tj'21es, California) was also USGd as tht~ virus d i 11ent. the following (per liter): NaCl . 80.0 9 KCl 2.0 g N . a 21~-IP') -.4 KH 2 P04 11.4 '3 . 2.0 9 27 fro prepare PBS, the following were added to 100 ml of the above lOX solution: MgC12·6H2° (0.1 g/ml stock solution) 1.0 ml CaC12·2H20 (0.13 g/ml stock solution) 1.0 ml Penicillin Streptomycin . Distilled H2 O · . . . . · . . . . . . . . . . . . . . . · 105 units 105 ug 900 ml These solutions were also sterilized by Seitz filtration. Phosphate b~ffer (PD, pH 7.0) was prepared as above, except that calcium and magnesium salts were not incorporated. This solution was used during trypsinization procecLJ.res. Tris-b'.J.ffered saline (TBS) was used in 2urificatio:1 procedures involving sucrose gradient centrifugation and consisted of 0.01 M tris, 0.001 M EDTA (ethyle:::1,::=(diamin,::=) tetraacetic acid, J.T. Baker Chemical Co., Phillipsburg, New Jersey) and 5.85 gil NaCl. The desired pH was estab- lished by the addition of 12 N HCl. Chick e~)ryJ cells were freed from tissues and dispersed from monolayers with a solution of 0.25 percent (wt/vol) trypsin (Difco) in PDi the pH 'lias adjusted to 7.4 with 2 N NaQn. Baby ha~ster kidney cells (BHK) were dispersed by a solution containing trypsin and versene (ATV). Each liter conta in2d th'2 following: Ni.lCl 8.0 g KCl . 0.4 g 28 Glucose . 1.0 9 ... NaHC03 . Trypsin 0.58 9 0.5 9 ... EDTA 0.2 9 The presence of virus plaques on monolayers was detected by the addition of a neutral red s01ution. Each liter of this solution contained 0.2 9 neutral red and 8.5 9 NaCl. IV. ASSAY OF VIR~L INFECTIVITY Two techniques were used to quantitate th3 infectivity preparatio~s: of virus assay. Suckling mouse titration and plaque The assay in mice was limited to a determination th'3 di luti:Jll 1")£ ~f a virus sample that was letha.l fo::- 50 ?'2r- cent of the inoculated mice (LDSO). The LDSO was determined by the method of Reed and Muench (1938). For each titration litters containing at least 6 suckling mice were inoculated intracranially with serial tenfold dilutions of the sample. The plaque assay was the second m3th~d titating infectivity of virus samples. conditions that would pro.n~te used for quan- Determination of the formation of plaqu3s by CE viruses constituted a significant portion of the work to be presented; details of the procedures employed are discussej in the Results section. layers .~£ Briefly, 4J-48 hCHJT old mon t ) - chick embryo or baby hamster kidney cells were washed with PBS to remove cell debris and bicarbonatebuffered growth .1\(~dium and th.3n appropr to d i lu':ioos OC 29 each virus sample were inoculated onto the washed cells 1n 0.2 ml amounts. Two, three or four monolayer cultures were used for each dilution assayed. Virus 'IJas allowed to adsorb to the cells for 60 minutes at room temperature or 37 C; five ml of molten overlay were then pipetted into each 50 mm Petri plate containing infected cells. The overlay medium 'Nas prepared by mixing double concentration solutions of buffered MEM and molten agarose solution. were equilibrated to 42 C before ~ixture B:Jth of th3se anj this temperature was maintained during addition of the overlay. When the medium had solidified, the plates were incubated at 37 C for 60 0:- 96 hours, at which time 2 ml of ne;J.tral red solutio::1 Were added to each plate. After 2 hours at 37 C the stain was rem07ed and th3 cultures were left in a dark room Eo= 12 hours before plaques Were counted. V. ANIMAL IMMUNIZATIONS In early trials, antisera were prepared by variej 'nea,':1S in rabbi ts a!ld mice. descriptio~s W'.lf3~e approp= iate, more d etai led are provided in the Results section. Table 2 outlines the schedule follo'lled to obtain speci fie antisera against 8 virus strains. This schedule was folloNed in the hope that sera representing different specificities wO;J.ld be available. Blood samples were removed follO'tJinl] cardiac [J'.1ncture of the anesthetized animals. After the blood had clotted at room temperature for 1-2 hours, it was 30 refrigerated an additional hour to facilitate separation and removal of the serum. Sera Were collected and clarified by low speed centrifugation and then stored at -20 C in 'J.5 :nl aliquots. TABLE 2 INOClJL7\T ION S':;HEDULE FOR PREPAR..l\T ION OF "E..I\RLY IMMUNE", "IMMUNE" AND "HYPERIMMUNE" RABBIT ANTISERA TO MEMBERS OF THE CEV GROUP. Day Trea tml=!1t o Normal serum sample removed :) First injection, 1 ml, each hind flank 7 First bleeding, 8 Second injection, as above intram~scular, "EARLY IMMUNE" serum 15 Second bleedin3', "IM1'11JNE" 43 Third injectio~, as above 52 Exsanguination, VI. S'2~um "HYPERIMMUNE II serum VIR0S NEUTR1\LIZA.TION STUDIES The techniques utilized were essentially those detailed by Habel (1969) for polio,.,lrus. The virus in()clllum '."'3S diluted in PBS-BSA to contain 1-5 x 10 6 plaque-forming units (PFU) per mI. The viruses were propagated in the brain tissues 'Jf sucklin9 ;nice and in cultures of chick ernbryo and baby hamster kidney (BHK) cells. 56 C for 30 minutes before use. Sera were heated at An at tempt was ilkld~ to 31 determine a dilution for each antiserum that would result in similar amounts and rates of neutralization by each antiserum for the homologous mixtures (see Results). Aliquots of serum and virus were equilibrated at 37 C for 20 minutes It was necessary to obtain zero before mixture. tim!~ sa:nples from mix·tures of normal serum and virus because of the rapidity of the virus-antibody interaction. At designated tim3s, 0.1 ml s3'11£)les -were removed from -':.h8 incubatin3 mixtures and diluted immediately into 9.9 ml of chilled PBSBSA; these were maintained in an ice bath until all sam?les had b2en taken. At this time additional dilu·tions, if necessary, were made and then these were assayed for surviving infectivity. VII. VIRUS PURIFICATIO~ Viruses of th9 CE group ware partially purified by sedimentation through sucrose gradients. Pools of virus propagated in suckling mice or in cultures of chick embryo or baby hamster kid~9y cells were Itered throug~ pore diam'3ter membranes (Millipore Filter Corp. I 1.2 un Watertown, Massachusetts) and th'3n clarified by sedimentation::::>nto sucrose cush ions. Th'3se were prepared by layering 10 ml of 20 or 25 percent (wt/vol) sucrose (SchwarZ/Mann, Orangeburg, New Jersey) onto 5 ml of: a high d'2ns i ty "pad" composed ::::>f 65 percent (wt/vol) sucrose. Ten to 12 ml of a filtered virus sample were layered onto the lower density solution 32 of sucrose and these were centrifuged for 90 minutes at 20,OJJ RPM in a SW 25.1 rotor in a Spinco Model L preparative ultracentrifuge. Fractions constituting the sucrose-sucrose interface Were obtained by drop collection fro:n the gradient tube bottom. twofold with tris b~ffered This material was diluted saline and layered onto gradients containing 25-65 percent sucrose. These Were prepared by sequentially layering 3 ml of 5 percent increments of sucrose solutions, beginning with th3 65 percent solution. Th latter gradients ware centrifuged for 360 minutes under the same conditions as above. Fractions were collected by min-3ral oil drop'JI!ise dis?lacement of sucrose drops; th")se containing the highest levels of infectivity were used in subsequent experimental VIII. proced~res. Pi{')TEIN D:2::TERMINATION The :nethod of Lowry et al. (1951) was used after some modification to estimate the amount of protein present in various fluids. Cupric s~..llfate and SGdiu;ll ?:Jtassium tar- trate were prepared separately as one and 2 percent solutions (wt/vol), res?2ctively. Prior to use equal volumes of these were combined and one ml of the resulting mixture was added to 50 ml of a 2 percent (wt/vol) sodium carbJnate in s,")dium hydrox.i..d:::; solution. One ml of the ~.l N second~ixture was pipctted into each test tube containing a 200 microliter prated n sample and rapio ly mixed. After 10 :ninute3 , 0.1 tnl 33 of Folin reag0nt (Fisher Scientific Co., Fairlawn, New Jersey) was added to each sample with imillediate, vig8rous mixing. This reagent hild previously been diluted with an equal volume of d tilled water. After 30 minutes' incubation at room temperature the absorbance of each sample was determin2d at 660 :lm in a Beckman DU spectroph8tom'3ter. A standard curve correlating absorbance to protein concentration was constructed every time pro:.ein concentration det.erminations were made. Bovine serum albumin diluted in tris- buffered saline was utilized as the reference pro:.ein. RESULTS I. .FOR.!~~T ION PL.~QUE Initial attempts toward the development of a plaque assay applicable to every member of the California enceph·alitis (CE) grou.p of viruses involved the use of primary cultures of chick embryo cells (CEF). tion on CE~ While plaque forma- has been reported for Tahyna virus (Porterf 1960) using an agar-co~taining .?laqu~s 'tJere observed 'Nh~n were used as a control. (1967) overlay, Bales et ale were unable to plaque BFS-283 virus on these cells, Id, althaug~ green monkey kidney cells (VERO) Using this second virus, regarded as the group prototyp2, and Western equin,= encel.:)~alitis (WEE) virus as a control, an effort was made to demonstrate plaque formation on CEF monolayers. When the overlay contained Melnick's 'jcowt.h medium, calf serum, s''Jdium bicarbonate and purified 0.9 percent (wt/vol) agar (Difco, Detroi t, Michigan), plaqu'8s were consisten t ly obs-9rved 'Ni th WE:~ virus 'Nhi le no evidence of cell destruction was apparent in the presence of the CE virus. Agar, which is a mixture of polysaccharides, inhibits plaque fo:-mat ion:JE a variety ,")f envc loped (Colo:! et al., 1965) and non-enveloped (Takemoto and Liebhaber, 1961) viruses. Inhibition has been attributed to the neg~tively 35 charged sulfate and carboxyl groups associated with agaropectin (Ventura, 1968) and can be largely prevented by the use of agarose, a neutral galactose polymer component of agar (Borden et al., 1972). e Alternatively, the adverse of these polyanionic components can be lessened by including polycationic substances in the overlay medium (Val'le, 1971). diethy Experimentally, neither the inclusion of noethyl dextran (DEAE-D, Sigma Chemical Co., St. Louis, Missouri) nor protamine sulfate (Sigma Chemical Co.) at concentrations of 20-1000 and 20-400 micrograms, respectively, faci tated the formation of CE virus plaques under Subst ion of agar by 0.5 percent (wt/vol) agarose resulted in areas within the monolayers, although the appearance was not that of a typical plaque. The "plaques" were characterized by irregular size, diffuse edges and an overall faint appearance that prevented quantitation. By minimum essent ing Melnick's medium with Eagle's 1 medium (MEM) the number of "plaques ll increased and the overall appearance (size, contrast, definition of edges) i Further refinement was achieved by decreasing the concentration of agarose to 0.4 percent. Plaques produced by a 1 mosquito isolate (#30521) under these conditions were uniform in size (2mm) and well-defined with respect to contrast and cd BFS-283 virus stock sllspensions exhibited two (] istinct plaque size populations, 36 one larg'2 (4mm) and the other small (1 mm). The incubation time required to attain these sizes was 60 hours. Having determined the corditions that would permit plaque formation and, hence, quantitation of at least two CE viruses, the plaquing efficiency was compared to assays performed by intracranial inoculation of suckling mice. Litters of mice 36 to 43 hours old ~ere employed and each .mouse was inoculated with 0.025 ml of one of several dilutio:1s ofa stock preparation of the desert isolate. sa~l= At th2 time, 0.2 inl of the sam2 dilutions were inoculated onto washed replicate CEF monolayers and assayed fa . . . plaque fo:r-mation. assay i~ In se~Je.ral suckli~g the animals (LD-..)\O: comparisons the titer e;3tina:'ed by ~e as the dos2 lethal for 50 perceDt of R~ed and Muench, 1938) was characteris- tically higher than that determined by the plaque assay, although thi2 difference seld''Jm exceeded J.5 1095. Similar results were noted when the titers of BFS-283 and Tahyna were compared by these methods of assay. Several characteristics of the CE virus-CEP cell interaction were examined. These included the influence of temperature and time on adsorption of virus, susceptibility of the cells to infection by the viruses, a~d the pattern of CE virus replication as a function of time. '11he effect of temperature on th2 infectivi ty 0:: OIV:~ CE vjrus (3J521) during dilution and adsorption to the assuy cells was fjrst studied. Separate samples of a 37 virus pool were equilibrated to 4 C or 25 C, diluted to contain a countable number of plaque-forming units and then maintained at these temperatures for 60 minutes. At this time aliquots Were pipetted onto washed monolayers7 half of the inoculated plates were incubated at 37 C for 60 minutes while the remaining mon81ayers were maintained at 25 C. At the end ~f this adsorption period all cultures were overlaid with agarose-containing medium and incubated for the time required for plaque formation. The results (Table 3) indicated that the virus CQuid be manipulated at room temperature without untoward effects and that the differences in adsorption temperatures (25 C vs. 37 C) were not sufficient to influence the number of plaques formed. TABLE 3 INFLUEN2E OF TEM?E~z\TURE DURINI:; 'JIRU..3 SAM.PLE DILUTIoN AND XDS')Ri?TION ro CELLS 0N CALIFORNIA ENCEPHALITIS (30521) VIRUS PLAQUE NUMBERS.a -----.--- -- _._._._---- -.-~.----- --- -. -~~~_.- -_._._-- - Adsorption Temperature (C) Dilution Temperature (C) 25 37 4 48± 7.0 b 482: 11 25 61± 6.7 60± 5.0 a. Virus samples held at the dilution and adsorption temperature for 60 ~inutes each. b. Average number of plaques and standard deviation of 7 replicate plates. --~ 38 rl1he time required for adsorption of the desert isolate (30521) to eEF cells was determined (Table 4). At the indicated times after addition of the virus inoculum to replicate monolayer cultures, unadsorbed virus was removed and the plates washed 4 times before addition of molten overlay. Maximum adsorption, determined by numbers of plaques counted, was achieved 1 minute after pipetting virus onto the cells. TABLE 4 TINE REQUIRED FOR ADSORPTION OF CALIFORNIA ENCEPHALI'rIS VIRUS (30521) TO PRIt1ARY CHICK EMBRYO CELLS AT 25C. Time (minutes) Titer (PFU x 10 7 ) 0 2.4± 0.6* 1 4.9± 0.9 2 3.8± 0.8 5 4.7± 0.8 10 6.0± 0.4 20 5.4:!: 0.2 40 5.1± 1.0 60 5.1± 0.8 Inoculum by previous titration 5.0± 0.2 * Standard deviation of four plate counts. 39 The susceptibility to infection of the cells which com~Jrise a monolayer of chick embryo "fibroblasts" prepurQd in the manner described (cf. Materials and Methods section) was ascertained by two techniques. One approach, regarded as a standard criterion to b '2 employed when evaluating the susceptibility of a cell spec s to plaque formation by a virus (Cooper, 1961), was to determine tha relationship between incremental dilutions of a virus sample and the number of plaques observed when assayed on th9 cells b'2in9 studied. In Fig~-1re 1 th3 nu·Th")er of plaques observed has been plotted as a function of sample dilution between 10- 5 and 10- 6 . The results indicated a linear relationship betwe<2n number of plaque-forming uni ts and virus concentration. The second m9thod of determining the susceptibility of chi~k cells to mon~layers in£ecti~n by these viruses 4as to in~ect cell at a virus-to-cell ratio that would guarantee the infection of every cell and to subsequently assay each for infection by the added v ~ell (infective center assay). This second step was accomplished by trypsinizing monolayers of infected cells after adsorption at 37 C, determining the number of cells present after dispersion and then plating dilutions of these onto uninfected monolayers, similar to th9 steps follow·ad for a plaque assay. The results indicated that at a mUltiplicity of infection of 5 all cells were infected. According to th9 40 80 . N 0 60 ~ (j) Pi W +' .r-l ~ :::> 1~0 till ~ .r-l S ~ 0 ct-; I (j) '~ i 20 ::5 ry ill r-l P.. o v F 283 ( PLE DJ CONFI 1. CALIFORNIA » s dilution EPHALITIS VIRUS ( PLAQUE J>rUlVIBERS A FUNCrr I ON OF ION. REPRESENT 95 PERC E LIlVIITS. 41 Poisson equation, at least 99.3 percent of the cells would, theoretically, have been infected. Similarly, at a virus ty of 0.005, less than 2 percent of the cells were multipl infected while, theoretically, less than one percent should have been. The replication of some members of the CE virus group In CEF cells was investigated. Cultures contained in plastic tissue culture flasks were inoculated with enough virus to ensure infection of every cell. Simultaneous in- fection of these cells permitted the study of a single cyc of virus replication. Following adsorption at rODm tempt.::ra- ture for 60 minutes, unadsorbed virus was removed before adding fresh medium by washing monolayer cultures 3 times with cold PBS. At va ous times aliquots of the growth medium covering the infected cells were removed for ass of infectivity and frozen until all samples had been collected. The tures of replication are indicated for four CE iso- lates in Figures 2 through 5. gro·.vth curves ·:Jf the S3.ffi'2 In para lIe 1 experim'2nts the viruses in CEF cells pretreated with actinomycin D (AcD, Calbiochem, LaJolla, California) were determined. Cells were incubated In the presence of 2.5 ug AcD per culture (2 x 10 7 cells) for 90 minutes prior to infection. These results are included in each graph. The churacteristics of replication were simi each of th(~ r for viruses examined (San Angelo, Tahynu and large (LP) ilnd smull pluque (SP) variants of BFS-283). Virus, 42 9 AcD-treated cells 8 7 6 5 4,________ ________ ________ ______ 6 12 18 24 ~ Time ~ ~ ~ ter infection (hours) FIGURE 2. REPLICATION OF BF 283( IMARY C EMBRYO FIBROBLASTS. ) VIRUS IN 9Untreated c Ui r4 r4 Q) o o r4 Of) o H 4, __________~____----~--------~---------~ Time after infection (hours) F' I GUHE PRIIVlAHY 3. REP IJ I CA~ll ION 0 F S - 2 f) J ( S p) V I RUSIN CHICK EMBRYO FIBROBLASTS. 43 9 -Untreated cells rn ----------.---------8 r-f rl Q} o 8 7 6 5 4 6 Time 12 18 24 ter infection (hours) FIGURE 4. REPL]CATION OF SAN ANGELO VIRUS PRIMARY CHICK FIBROBLASTS. 9 ed cells rn ill o 8 treated cells C"- o rl :x! (\J 7 H OJ ~ H 6 Q) ..p ·ri ..p o rl hD 5 o ~-l 4 6 12 18 Time after infection (hours) FIGURE • REPLJCfiJ'ION OF MAHY CHICK YO 1~AIIYNA FIBROBLASrrS. VIRUS IN PRI- 44 presumably progl.:ny, was detected in the supernatant fluids within 3 hours after infection. Exponential release of virus was evident through the twelfth hour; further increase in titer after this time was negligible. did not beco~e Cytopathic changes h~urs. apparent until 30-36 Cells pre-treated with AcD 3upported replication of all CE strains studied. The activity of the AcD preparation was tested by its ability to inhibit the replicati8fl of vaccinia, a DNA-containing virus, under conditions similar to those Replication was mJre than 90 percent in- described above. hibited. Thi~se res~Jlts s'Jg3ested that DNA was nC)t involved in the replication of CE viruses. In support of this, McLerran and Arlingshaus (1973) have dete::;ted RNA in purified preparations of LaCrosse virus. As noted above, the growth medium used in the plaque assay determined I in part, plaqu9 :n::>rpholo9Y an::] nu:nbers. The replication C)f the desert isolate (30521) in fluid cultures, i.e., in the presence of Melnick's or Ea31e's MEM growth ned ia lacking a-J.:l r 0::::- a]arose ,was determined. The data of Figure 6 indicate that no significant differences in virus propag.3.tion could be detected with either mediuiTI. Having defined the ::;onditions necessary for plaque formation by two CE virus group members and inves 1. igated som3 basic aspects of the virus-cell interaction, the a?plicahility of the assay to other viruses in the group was examined. It was found that the several viruses studied 45 9 -;--m ill rl rl 8 OJ 0 C'0 rl 7 >< N H OJ PJ 6 H OJ ...p .r-I 4.::> 5 0 rl oD 0 H 4 6 18 12 24 Time after infection (hours) FIGURE 6. COMPARISON OF IA (G,) AND IV.EDIA. s~rEP S (#30521) MIN RATES IN IAL 46 vari.ed considcrubly in plnque morphology. Besides size variation, the contrast between the plaques and uninfectcd areas made detection of the plaques of some of the viruses difficult. It had been noted that contrast could be improved by doubling the bicarbonate concentration, so an experim2nt was desig~ed to determine the concentration of bicarba~ate that .would be optimal for the detection of Tahyna and "3052111 virus plaques (Table 5). The concentration of the buffer, no.Lioully p:-esent in thf3 overlay at 0.35 g/1, was varied between one-fo'Jrth and four times that amount. This range was accompanied by differences in pH of from 6.9 to 7.9 apparent at the time th·3 o',erlay was added to the infected cultures. During incubation, the infected, overlaid cells Were maintain.3d in an atmosphe!:"e of apPY'oximately 5 percent (vol/vol) carbon dioxide (C0 2 ). TABLE 5 CALIFORNIA ENCEPfffiLITIS VIRUS PLAQUE SIZE AND QUALI'ry AS A FUNC'rrON OF 80DIU1"1 BICAR30UAT2 ·:OH·CEN'rR}\.TION. Virus BicarbJnate Concentration (g/l) .0875 Tahyl1i1 (l)a,b #30521 (1-2) a. P1acIUC d iam(~ter b. ParcntlE~ses plaque's. 1.4-] .175 .35 .70 1.05 (1 ) ( 1) 1-2 2-3 (2-3 ) 2 2-3 2-3 2-3 2-3 (mm) indicate poor qua1i ty, d iff icul t-i-.o-score 47 Both Tahyna and the desert isolate appeared equally sensitive to the conditions produced by low bicarbonate concentrations (pH 6.9). At the highest bicarbonate concentration the plaques of Tahyna virus were diffuse, poorly contrasted and difficult to count, thereby restricting the range over which good quality plaques were observed. The mQrphology of 30521 virus plaques did not change above the lowest concentration of bicarbonate. In addition, the number of plaques formed by the desert isolate did not change as the concentration increased above 0.35 gil; for Tahyna virus plaque numbers were highest at 0.7 and 1.05 gil. Because the pH and bicarbonate concentration increased concomitantly, it was not possible to attribute the changes in morphol09Y and numbers of plaques observed during this experiment to a direct influence of the bicarbonate ion. The acidity observed in growth media as the metabolism of cultured cells proceeds has been attributed, in part, to the accumulation of non-volatile 3-carbon acids (Cooper, 1961). Atmospheric carbon dioxide (C0 2 ) also contrib~tes to acidification by combining with water to form carbonic acid that subsequently dissociates into hydrogen and bicarbonate ions: + An increase or decrease in C02 tension is accompanied by a corresponding pH change. Experimentally this becam9 appitrent when bicarbonate-buffered overlay media were added 48 to infected cultures and left at roo(n temperatur.e dur:-i(}(] :;:Jlidification of the overlay: pres~_FlI~e The r.elatively IIJ\II p:::trt.ial of carbon dioxide prescnt in the atmJspherc caused an increase in alkalinity. Because the pH fluctuated with atmospht2ric changes between experiments, the initial pH was difficult to control. An awareness of the difficulty associated with the use of bicarbonate in cell o7erlays led Porterfield (lQ60) to replace this buffer with tris Hel (TRIS amino:nethane). was a CE ~irus (hyjroxy~TI(~thyl) One of the viruses studied by Porterfield strain, Tahyna, and the cells used were from chick emb::yos. Hi.s results prompted an effort to determine the applicability and use of this co:npound to plaque oth~r CE v.irus8s. For these studies plates contain- ing infected cells Were incubated in air-tight boxes (Labline Instruments, Inc., Melrose Park, Illinois). the concentration of the CO 2 to ~hat This liu~ted amount present when the containers were initially sealed and also to that produced by the respiration of th3 cells. Tris-Hel was used at concentra tions betw '2(?n (). 01 M and 'J. 05 M; the pH of each overlay was adjusted initially to 7.2 with H-::::l. By vis 1al 1 comparison rJur in9 incubation, the increase in acid i ty of th'3 tris-buffered overlay was slower than similar plates overlaid with the bicarbonate-blJEEered in a 5 percent CO 2 atmosphere. ~U(:~d ium and incubated The plaques produced under the tris-buffered overlay were similar to those noted [or the bicarbol1ztte-buEEered media in all respects, e.g., diameter, 49 contrust tlnu sharply-defined cdges. It was concluded that the usa of tris buffer was acceptable under the conditions described. Pltlque formation by BFS-283 and 30521 viruses was studied using overlays containing both buffering substances, i_e., tris Bel a!1t1 bicarbonate (Table 6). In addition to TABLE 6 PLAQUE FORMATION BY CALIFORNIA ENCEPHALITIS VIRUSES ON PRIMARY CfIIC~ EMI3.:.~YO CE~LS: IN5'LUEI'JCE ~)g CAl{gON DIOXIDE 1 S0';)IU."1 BICARBONATE AND TRIS (HYDROXU1ETHYL) AMINOMETHANE a -~------ pH~J HCO] --_. (g/l) ._--"- --- --------""- - - -----.- -"- #30521 _._- CO 2 No CO 2 B?S-233 --- -~- --.-- ---.---- .. CO NJ C°2 2 _____ w_'_' _______ 0.0975 6.9D 95 c 61 TNCd 31 0.175 7.15 84 74 83 13 0.35 7.25 74 66 55 0 0.70 7.70 79 74 15 0 1.05 7.80 68 51 0 0 1.40 7.90 63 19 0 0 tris alone 7.30 84 85 TNC 59 ------'-~---- ---- a. OJerlny consisted oE 0.4 percent agaros~, Eagle's M~M! 0.02 M tris adjDsted to pH 7.3, and indicated amJunts of sodiun bicarbonate. b. Determined at the time of overlay. c. Plaqu(; forinin'~ units pj~ r O. 2 ml. too numerous to coun t. 50 bicarbonate, tris-Hel was present in all overlay media at 0.02 M concentration and pH 7.3. The amount of bicarbonate present increased in doubling concentrations gil to 1.40 gil. fro~ 0.0875 This resulted in a range of pH's. Dupli- cate sets of plates were prepared for each virus so that one could be incubated in an atm':)sphere of 5 percent CO other without added CO . 2 Significant differences between the pH :ind/or bicarbonate sensitivities of the were noticed. and the 2 virus:~s As the bicarbonate concentration increased in the presence or absence of added CO , the 2 nu~)ers of plaque=;; dec-cea3ed with both viruses, altholJr;3'h the sensitivity of B2'5-283 CO , the 2 formc~d -"'dS markedly greater. bicarbQ~ate In the presence of added concentration range over which BFS-233 plaques was ext.ended. For the desert isolate, plaques Were app3rent over the entire range, but the numbers did increase when CO 2 was prese7.lt. Th::=se results SU99Gst.ed a d 2£inite inf luence of the small pH differences that eXLsted l at th::= tim::= the infected ffiJnolayers were overlaid but because bicarbonate was present in relatively large am~unts a definitive conclusion could not. be reached without first eliminating th2 added bicarbonate. 'rris-HCl was previously sho'",n to support th'2 forma tion of plaqu'2s by members of th8 CE virus 0:: added bicarbonate and ov~rlilYs g.COUp carbon dioxide. in thf:! abs;3nce Tris-buffered of decreasing pH were prepared by adding increasing amollnts of 6 N hy,Jrochloric acid to each oE severa 1 l.iquificd 51 overlays. Because the tris concentration was constant while the initiol pH of each o'J'crlay was unique, the assumptiol1 was made that any effect on plaque formation would be attributable to the pH. Tables 7 and 8 summarize the experi- Over the pH range studied (7.0 to 8.0) mental findings. variation was evident in plaque morphol:Jgy (Table 7) and TABLE 7 EF.'BEC'r 01" INITI}\L OVERLAY pH ON THE MJRPHOLOSY OF PLAQTJES FORMED ON PRIMARY CHICK EMBRYO CELLS BY CALIF'OR~ IA ENCEPH.\LIT IS G~~OUP ~J IR~-;3ES. a OVerlay pH Virus 7.7 7.5 7.3 7.0 ---- 8.0 7.9 ----~.--.-"-------------------~----- ------.-.-- #30521 ++++b ++++ ++++ ++++ ++++ ++++ LaCrosse ++ +++ +++ ++ ++ a B:"S-233 +tt ++ 0 0 a a San ++ t- t- + 0 0 0 0 Jerry Slou'3h + + 0 0 0 0 Sn:::)'Nshoe Hare + + 0 0 0 a Trivittatus + 0 0 a 0 0 Melao 0 0 0 a a 0 Ang'~lo a. Th'2 overlay contained 0.4 percent agarose, Eagle IS MEM and 0.02 M tris (hydroxym~thjl) d:ninometha:.le and initial pH was adjusted by addition of 6 N Hel. b. +t-t-t-, greater than 2 m~ diam., well-defined edges, good contrast; +++, 1.5-2 1011 diam., well-defjned edg~:.3, gi,Jod contrast; +t-, 1-2 mm diam., diffuse ed;:;cs, fair contrast: +, cytopathic changes of irregular shape, uncountable; 0, no plaques. 52 rrhc rna j<::>r i ty of the v lruses used in this numbers (Tub Ie 8). experiment formL:d plaques only over a narrow range of pH values. An exception was the desert isolate (#30521) which produced distinct, 3-4 mm pH. r]iamc~ter plaques at each overlay While the data are not presented, it was noted that the ability to form plaques was greatly reduced when the ini t ~a lover lay pH was bE.~low 7. O. The resul ts in Tab le 8 indicate the countable plaques at different overlay pH's of four of the viruses used in this experiment and suggest the "recognition" by the viruses of an optimum pH fo:: plaq:ue formation. In a parallel experiment, the effects of pH on TABLE 8 FORMATIO~ BY CALIFORNIA ENCE!?IL\LITIS VIRD3ES ON ?~ I MARY CHICK E.MBR-{O C£l..J.JS ~ INFLuENcE O? pH :)N NUMBERS OF PLAQUES.a PLAQUE Overlay pH Virus 7.0 b 7.3 7.5 7.7 7.9 8.0 75 57 52 52 33 #30521 30 L~Cross:~ 22 54 31 16 10 0 B:?S-233 71 34 a a 0 0 San Angl.=lo 43 a a a 0 0 ,--~.-~,-- -- ----- a. Oilerlay consisted of 0.4 percent agarose, Eagle's MEI1 and 0.02 M tris (hydroxymethyl) aminomethane adjusted to indicated pH values by addition of 6 N Hel. b. Avcr a :-F:> n 'J mb e r 0 f P la '1.11 e s for s ~? V t:! n rep 1 i Cd t e p 1 ate:3 . 53 plaqu~ formation by four Group A Arboviruscs (Sindbis, Western, Eastern and Venezuelan equine encephalitis viruses) w(~re stur]ied: NJ dif£erenc(:;s could be detected in plaqu.2 numbers, size or contrast over the same pH range. The plaques produced on chick embryo cells by Jerry SloUljh, Trivittatl1s, S:1oyJ'shoc Hare, M'2lao or Jamesto\'Jn Canyon viruses could not be improved by overlay pH adjustments: Keystone and Lumbo 'viruses 'flere not studied. Wh"~n Tahyna virus 'flas inoClllated onto chick emb.:::-yo cell cultures aYld th'~n overlaid with the tris-buffered agarose medium at a pH that pro'vided optimal conditions foe plaque fO::-iOation I incorporation of 50-80 u9 Dill\E-Dextran into the overlay increased the plaque size from 1-2 mm to 4-5 mm: Inclusion of this p8lycation did not affect th9 :::lumbers of plaques. cells. Concentrations above 100 ug/ml were toxic for the The plaques of Trivittatus, Jamestown Canyon, Jerry Slough and Melao viruses were :::lot improved by the addition 8£ DRA...E-D. It was of some interest to determine the effect of pH on the replicatio:1 of the viruses When the cells had n:)t been covered with d:'1 a9d rose--containinC] ov2.r lay tn(~d i umw For this experiment San Angelo virus, Which formed plaques at pH 7.0 but not at pH 7.3, was selected. The conditions 9rowt::.h (~I]rV~? :;tu(1ics in terHlS of mUltiplicity of infection, ad- sorption and wilshings. The m9d ium (Melnick IS) was b;..1ffored 54 by tris with no added bicarbonate. 0f the culture 7.3 or 7.6. fecti~n, mediu~ The pH of throe uliquots was initially adjusted to pH 7.0, During th'2 24 hours of incubation after 1n-- the pH did not change in any culture flask by wore than 0.1 pH ~nit. N~ differences were noted in the infec- tivity titers achieved between pH 7.0 and 7.6. The data indicuted that, while pH affected plaque forioati:>J1 of San A'19t~lo virus, it did not necessarily influence replicatio::1. The applicability of a second cell for plaquing of CE viruses was also investigated. Tnis was prompted by the d if Eictll ties exper ienced wh'2n at tempts 'tlere made to demo::1strate plaques for some of the viruses. The cell (BHK) was a strain:>f the continous line of bally hamster k cells (BH.<-21/C13) isolated by Stoker & lji1(~Y McPherson (1964). The origin of this clone is described in the Materials a':1d Methads section. One desireable property o~ the cell, compared to the wild type, was its slower rate of acidification of culture media than the wild type. ments that had b(;(~n [:H.?r formed with chick emb::-yo ce Ils repeated, using the BHK line. co~ducted Several experi- For exa~ple, '.th~re studies Were to determine the characteristics of replication (): the large plaque variant o~ BH{ cells (Figurffi 7 and 8). BFS-233 and LaCrosse viruses in While the general features of replication resembled th'Jse using chick embryo cells, thl2 pL:.iquc~ fOrHl.ln(} t i tors CY::i1(:riJ lly h igll(~r. dt~tected when Ba:< Ct)lls \rJere As was noted with chick om~ryo US'~:l cells (Nere I 55 Untreated cells W rl rl '-......... 9 - ------~G1 (I) o 8 AcD-treated celIE 0....-- 7 6 o r-; oD o 5 ~ 4 6 12 24 18 Time after infection (hours) FIGURE 7. C INUOUS CATION OF BFS-28J(1P) VIRUS IN A BABY HA1VlSTER KI C 11S. 56 Untreated cells rn rl rl 9 (!) o C"O 8 cells rl X N o ~ o 5 ~ 4 6 12 18 Time after infection (hours) FIGUHE 8. REPLICA1 ION OF LACROSSE VIRUS IN A CONT:! S IIJ BABY HAMSTER C 1 57 actino~~cin D had no de~~nstrablc effect on th2 replication of th,? viruses. A ~oiTIparison was also clade b'2tween the plaquing effi- ciencies of chick embryo cells and BHK cells: San Angelo, Tahyna and LaCrosse vIrus stock suspensions were used in the compa.ris·'Jn (Table 9). Whi Ie the number of San Angelo virus plaques was higher on chick cells than o~ h3~ster baby TABLE 9 COMPARIS')0J 87 Ph"!\QUE-POR:qIN3 UNIT TITERS OF CALIFORNIA ENCEPHALITIS GROUP VIRUS SUSPENSIONS WHEN ASSAYED ON CRIer< EI13RYO ?I8,::<J)3~l\STS AND 3AB'{ H}\MSTER KIDNE{ CELLS.a Assay cell Virus BHK CEF 7 San Ang'210 2.1-3.1 x 10 7b 3.0-4.4 x 10 Ta"hy!la 1.8-2.5 x 10 8 2.0-2.6 x 10 8 LaCrosse 1.4-2.1 x 10 7 1.2-2.0 x 10 7 0.4 percent agarose, HEM and 0.02 M tris (hydroxym3thyl) amin'Jmethane; the pH of th(~ oV2rlay l..lsed -to plaqu!~ San An}2lr:::> and LaCrosse viruses was 7.1, and Tahyna was 7.4. a. O'ler lay cons isted of b. Range of PFO/ml at thf.::! 95 percent confid'2nc(~ limits. kid~ey cells, the data were not regarded as significant On01.19h to discontinue use of th,:; cells. Generally 1 the than the corresponding virus plaques on chick embryo cells. 58 The effect of virus sample dilution on numbers of plaques observed on BHK cells was studied, using BFS-283 (large plaque variant) propagated in suckling mice and in BBK cells. The results were similar to those noted when chick embryo cells were used as the assay cells: As dilu·· t.iol1 of the YJirus-containing sample increased, the of plaques appearing decreased. the ability of single, n~.lmbers Again, the data suggested infectious virus particles to infect Infective center assays were not performed. cells. The variation in numbers and morphol03ical features of the plaques of several CE viruses as a functi~n oE initial pH differences in the overlays added to BHK cells was examined as each virus strain was received. The stability of the plaque characteristics of each vIrus on BK{ cells was found to be influenced by the initial pH of the overlay media. In contrast to chick embryo cells, plaques were fO:lood on the baby hamster kidney cells by Trivittatus, M'21ao, Ja:nestowi.1 Canyon and Snowshoe Hare viruses. The optimum pH for plaque format. ion was between pH 7.0 Canyo~ a~d 7.2. byth(~:3 Plaque formation by e vIrus s Ja~estown and Melao viruses was observed to be greatly influenced by slight differences in overlay pH: observed When the overlay pH The demonstratio~ virllses for pl<tqlll~ 'Ill Plaques 'Nere only as initially adjusted to 7.1. of an apparent pH dependence of CE format.ion would hav\:.; add'2d an unrJ'2:3ire-- c1blc var l.able requiring mani.pulation and intcrpretat_ion if 59 different overlay media had had to be used for each virus stud ies were per focmor]. decided to s tandarJ i one overlay pH: Zt:; the plaque assay of these v iruses to The value selected was pH 7.1. allowed formation of plaqll(~S Tahyna virus, Which exhib th~ but. S numbers of A'1 add it i')na 1 va r This by every member stlHl:lec]. d the highest pH optimum, duced smaller plaques at th,~ It was pro- lower pH (1-2 Inm 'J'erSU3 3-4 !TIm), \'Jer:e similar. Ie Which influenced the m~rphology of the CE virus plaques was the calf serum lsed in the QVerlay met]ium. In the course of these studi9s it was neces to use several different lots and preparations of calf sera and with each ~f th'3se the mo.:::-phology of th9 plaques pr-J- duced by each of the viruses varied. The most apparent differences were associated wi th the d iameters As noted abov2, sera inhibitory for o~ the 'J~ the plaque:31 isola~es were not used. To summarize: of Primary chick embryo cells and a strain hamster kidney cells both the Californ d replication of encephalitis group viruses studied. Plaque format on both types of cell was markedly dependent on the init 1 pH of the overlay. The BHK cells supported formation by a greater number of the CE viruses than did the chick embryo cells: Every member of the group stue] ied formed cas i ly detectab Ie plaques on the hamster cell line. 60 II. PURIFICATION OF CALIFORNIA ENCEPHALITIS VIRUSES The need for preparations of CE viruses from which con-taminating non-viral matter had been removed was anticipated and resulted in an attempt to devise a procedure that would increase the specific activity, i.e., PFU: protein ratio, of each virus preparation. Precipitation of the infective virus present in crude mouse brain or chick cell suspensions by buffered, sat~rated solutions of ammonium sulfate was attempted but the recovery was low, seldom exceeding 30 percent. Because the level of infec- tivity remaining in the supernatant fluid did not total the amount of unprecipitated virus, it was assumed that the virus had been inactivated. The concentration of infective particles that could be precipitated was not increased by different pH values (6.3 to 8.1), by different levels of salt saturation (35 to 65 percent), by the concentration of protein present (3 to 11 percent), or by the method of addition or mixture of the salt to the virus preparation. Aggregation did not contribute sign icantly to the high percentage of virus lost during precipitation: Filtration of the resuspended precipitate by Millipore membranes with pore diameters 2-4 times greater than the diameter of the virus particles (Murphy et al., 1968) did not reduce the levels of infectivity. Although precipitation of virus with ammonium sulfate did decrease the level of contaminatinq protein by at least 80 percent, the accompanying loss 61 of infectivity precluded its usc. In several instances virus preparations prccjpitated with ammonium sulfate were further purified by centrifugation through 15-34 percent sucrose gradients. In every experiment complete recovery of the input virus was noted. Furthermore, the distribution of detectable protein in these gradients was limited to the topmost fractions, indicating that the nonviral material was unable to sediment into the higher sucrose concentrations. This suggested the appl bility of sucrose gradients to the purification of CE viruses. Gradients could be composed of a low density solution of sucrose capable of preventing sedimentation of nonviral substances but permitting the sedimentation of infectious particles. A high density sucrose solution could serve as a cushion on which the virus particles would collect. In practice the cushion consisted of 65 percent sucrose while the low density sucrose solution contained 20 or 25 percent. When 25 percent. sucrose was used, a plot of the distribution of infectivity indicated the presence of significant concentrations of infectious particles throughout the low density solution of sucrose (Figure 9). Although the amount of contaminating protein excluded was greater when this concentration was used, replacement by 20 percent sucrose resulted in greLlter recovery of infectivity in the frQctions constituting the sucrose-sucrose interface and so the latter was employed throughout purification 62 9 4000 rl ~ 7 rl ~ " ~ ~1 ~ 3000 " ~ .~ ill () I rl ~ 0 < 0 .~ I ~ H ~ ~ 5 ~ 1000 Fraction No. FIGURE 9. DISTRIBUT ON VITY OLLOWING C IFICAT ON OF SUCKLING ~OUSE -PROPAGATED TRIV US VIRUS BY SEDIION 25% SUCRO ON~O A 65% I 63 expc~r imcnt s . Fractions containing the highest conccntra- tions of virus particles were collected, diluted 1:2 with tris-buffered saline and layered onto gradients containing 25-65 percent sucrose. lengths of time: These were centrifuged for varying Isopycnic banding was achieved by 3 houris centrifugation. When viruses were centrifuged onto a cushion and then sedimented to density equilibrium under these conditions, the highest concentrations of infectivity, i.e., the peak, was spread over several fractions (Figure 10). In contrast, when virus prepared by sedimentation into 25-45 percent gradients of sucrose (Figure llA) and then resedimented - the second time to equilibrium - in gradients of 25-65 percent sucrose (Figure lIB), the peak of infectivity was much narrower, that is, was distributed over fewer fractions. These results were interpreted as an indication of particle size or density heterogeneity. By either method the degree of purification was such that protein could not be detected (less than 2 ug protein/ml) by the Lowry procedure after the virus had been "partially purified", that is clarified and sedimented to equilibrium in sucrose gradients. To summarize: California encephali t.is viruses Were purified without substantial loss of infectivity by a twostep procedure. Virus particles were first sedimented onto a sucrose cushion; this resulted in removal of substantial amounts of nonviral material. Virus particles thnL had 64 8 rl 8 "'" n ( ~ fI--t r.:y 0 rl QO 6 0 H 5 10 20 30 Fraction no. FIGURE 10. DISTRIBUTION OF SUCROSE CUSHIOf\-,CLAB.IFIED, MOGSE BRAINDERIVED TRIVITTATUS VIRUS INFECTIVITY FOLLOWING SEDIMENTATION TO EQUILIBRIUNi IN A 25-65 PERCEN'I' SUCROSE GRADIENT. 65 8 7 6 $.-1 Q) 5 P.J H Q) ..p ·rl -tJ 7 B 0 r1 QO 0 r-=! 6 5 10 20 30 Fraction No. FIGURE . PARTI PUR IC ION MOUSE N-PROPAGATED SAN ANGELC VIRUS BY ION OF A CRUDE SUSPENSION 25 PERCENT SUCRD (A) GRADIENTS AND 1FUGA11ION 110 EQU LIER UM (B) A PEAK FRACTION (II ) THROUGH 25-65 SUCHOSE. 66 sedimcntcll onto t~h(:! sucrose-sucrose interface were collected, layered onto sucrose gradients consisting of 25-65 percent sucrose in tris buffer, and then sedimented to density equilibrium. Fractions of each gradient containing the highest amount of infectivity were collected and used as "partially purified" virus. III. VIRUS-ANTIBODY INTERACTION STUDIES The conditions permitting plaque formation by members of the California encephalitis virus group were determined in order that plaque techniques could be applied to the serological investigation of the viruses' antigenic properties and intragroup relationships. Before this was attempted, however, some basic features of the CE virus-antibody interaction Were studied. In early immunization trials, mice and rabbits were inoculated intraperitoneally and subcutaneously, respectively, at 7 day intervals for 35 days with suspensions of BFS-283 and "30521 11 viruses propagated in suckling mice. Ten days after the sixth injection, sera were collected following exsanguination by cardiac puncture. The neutral- izing capacity of each serum was determined by making serial tenfold dilutions of the sera and viruses and incubating mixtures of each combination for 60 minutes at 37 C. Assay for virus surviving the interaction with antibody indicated that in no i.nstance did the decrease in 67 infectivity by neutralization exceed one log of the inoculum present at zero time. Moreover, cross-reactions in the heterologous combinations were complete, i.e., heterologous virub was neutralized to the same extent as homologous virus. Immune sera specific for these viruses were also obtained from K. L. Smart at Dugway Proving Grounds, Utah and studied. Similar results Were noted. To examine the possibility that the animals had responded poorly to the antigenic stimulus provided by the mouse brain-derived viruses, two rabbits were injected with chick cell-propagated San Angelo virus emulsified in complete Freund's adjuvant. Both were inoculated in each foot pad and hind quarter (intramuscular); three weeks later subcutaneous injections were administered at three positions along the back in addition to intramuscular injections in both hind quarters. Two weeks following the second injec- tion blood samples were removed and the sera collected. The neutralizing capacity of each serum was tested with homologous mouse brain-derived (MBdSA) and chick cellderived (CCdSA) San Angelo viruses (Figure l2). As was previously observed for BFS-283 and the local isolate, the neutralization of MBdSA did not exceed one log during 30 minutes' incubation. In contrast, the neutralization of CCdSA was rapid and exceeded three logs of inactivation within 10 minutes. These results suggested differences betweon viruses propagated in the two types of cells, 68 w 100 Mouse brain-derived virus ----------- ::5 $-~ ·M :> on ~ 10 \ \ 'M • :> :> 'M H ::5 w \ \ 1 \ Ct--! 0 \ +' (J) 0 H . \ ~ 0.1 (]) Chick cell-derived virus "- tl. 0 10 20 Time (minutes) FIGURE 12. INFLUENCE OF THE HO USED TO PROPAGATE VIRUS ON NEUTRALI ION BY ANTI-SAN ANGELO VIRUS (C DERIVED) SERUM. 30 69 although the possibility could not be discounted that antibodies to host cell substances had either interfered with the inactivation of MBdV (in the earl r studies) or had enhanced the neutralization of CCdV in this experiment. To examine the possibility that anticellular antibodies had influenced viral inactivation, two experiments were initiated. In the first experiment, dilutions of antiserum obtained from a rabbi t in jected wi th ho'mogenized normal chick embryo cells were mixed with CCdV and incubated; fresh guinea pig serum was included in one reaction set TABLE 10 NEUTRALIZATION OF SAN ANGELO VIRUS PROPAGATED IN MOUSE BRAIN (MBdSA) AND CHICK EMBRYO CELLS (CCdSA) BY RABBIT ANTISERA TO NORMAL CHICK CELLS (Ra-ANTI CC) AND CHICK CELL-PROPAGATED SAN ANGELO VIRUS (Ra-ANTI CCdSA).a Serum ---------- Present MBdSA CCdSA Absent MBdSA CCdSA Normal rabbit 6 1.4 x 106b 1.6 x 10 1.5 x 10 6 1.7 x 10 6 Ra-anti CCdSA 4.9 x 105 5.0 x 10 2 3.1 x 105 5.0 x 10 2 Ra-anti CC 1.6 x 10 6 2.4 x 10 4 1.8 x 10 6 1.9 x 10 6 -------a. Assay on chick embryo cells. b. PF'U/ml c. Fresh unheated or heated guinea pig serum present at 5 percent (vol/vol). 70 while heated guinea pig serum was present in another. 'rhe data sumnorized in Table 10 indicate that anti-chick cell antibodies did cause inactivation of virus but only when unheated guinea pig serum was included. The effect WaS specific for the chick cell-derived agent; MBdV was not inactivated. Similar experiments with anti-normal mo~se brain cell sera and mouse brain-derived viruses indicated no neutralizing effects of anticellular sera. Moreover, pre-incubation of either type virus with the corresponding anticellular serum, followed by incubation with specific antiviral serum, did not decrease the amount of virus neutralized. In the second experiment, antiserum obtained after immunization of a rabbit with mouse brain-derived BFS-283 virus was examined for its ability to neutralize the homo- logous mouse brain- and chick cell-derived viruses. The data presented in Table 11 indicate that this antiserum did not inactivate MBdV to the extent detected with CCdV. This indicated that the host source of the virus used in the immunization procedure did not affect the ability of the antiserum to neutralize virus. In contrast, the host source of the virus used in the neutralization test was important in determining the amount of virus inactivation. Numerous viruses, enveloped and non-enveloped, have been shown to maintain substant 1 amounts of infectivity following incubation with neutralizing antibodies. The 71 TABLE 11 NEu'rRALIZATION OF MOUSE BRAIN- AND CHICK EMBHYO CELL- DERIVED SAN ANGELO AND RFS-283 VIRUSES BY RABBIT AN11IBFS-283 VIRUS (MOUSE BRAIN-DERIVED) SERUM. San Angelo 283-LP Antiserum CCdV MBdva Ra-anti MBdBFS-283 RNS 8.7 x 106b 3.4 x 10 4 3.0 x 10 7 2.6 x 10 6 MBdV CCdV 6.0 x 10 6 4 x 10 3 5.8 x 10 6 4 x 10 6 a. MBdV mouse brain-derived virus; CCdV derived virus. b. PFU/ml chick cell- virus resisting neutralization has been referred to as the persistent fraction, protected fraction or simply nonneutral~zable causes. virus and has been attributed to a variety of The discovery of two viruses - presumably of identical genotype but distinct phenotype - that exhibited such marked differences in neutralization by specific antisera, presented an opportunity to examine the nature and, perhaps, a cause of the persistent fraction. Besides the contribution of the virus to the persistent fraction, the possible influence of other components involved in the plaque reduction technique was also stud d. These included the assay cells and factors associated with serum such as spccj fic antibodies, cornplcmc.:nt and other "accessory fact.ors I: 72 Components of complement can contribute to the neutrulization of most viruses, especially when early, IgM-containing sera are used (Daniels et al., 1970: Yoshino and Taniguchi, 1965). In addition, cytomegalovirus neutraliza- tion by hyperimmune, IgG-containing sera has also been shown to be enhanced when complement is present (Andersen, 1972). Other serum components have also been identified which increase the amount of virus that can be neutralized by antiserum. These "accessory factors'll are distinct from complement, although they may constitute a complex of factors in which one or more components of complement participate (Way and Garwes, 1970). Earlier, data were presented which indicated that the addition of fresh guinea pig serum did not increase the amount of mouse brain-derived virus neutralized (Table 10). In other experiments with different CE viruses from the same cell source, the size of the surviving fraction was not altered by the addition of fresh guinea pig sera. Similarly, the addition of unheated normal rabbit serum to mixtures of antiserum and viruses prepared from infected mouse brains had no affect on the neutralization of the viruses. As a comparison and an indication of the activity of the guinea pig serum preparation, the latter was included in a mixture of BFS-283 and homologous early immune serum (Figure 13). The rate of jnactivation of this virus was considerably greater in the presence of unheated guinea pig serum. It was concluded 73 10 U) ::i ~I .,-j ? ttO s:::: .,-j ? .,-j ? H ::i U) ct-J 0 ...j-.:> s:::: Q) 0 H 0.1 • Q) P-t 0 o 20 Time (minutes FIGURE 13. NEUTRALIZATION BFS283 (LARGE PLAQUE VARIANT) VIRUS BY HOMOJ-lOGOUS "EARLY II'v1IV1UNE HABBIT SERUM COLLECTED 7 DAYS INTRAMUSCULA.R I I OF VIRUS: EFFECT OF FRESH (8) (0) GU G SERUM. SENT 95 PERC C E S. ff 74 that heat labile components of sera as well as other "accessory factors" influenced the inactivation of CE viruses only when sera that might be expected to contain 19 S, IgM-type antibodies Were present. The reduction or depletion of antibody concentration was investigated as a possible cause of the persistent fraction by Lafferty (1963a) and Lewenton-Kriss and Mandel (1972), although neither were able to demonstrate sufficie~t depletion to account for the surviving infective virus. This possibility was also entertained with the CE viruses and their antisera (Figure 14). Mouse brain- and chick embryo cell-derived San Angelo viruses were both used. time zero o~e At virus type was mixed with an equal volume of homologous antiserum; when the surviving virus was limited to that present in the persistent fraction (pre-determined by examination of data from neutralization curves) enough fresh virus (MBdV or CCdV) was added to return the level of infectivity to the concentration present at zero time. This resulted in dilution of the antiserum by a factor of two. When MBdV was added to previously incubated mixtures containing virus propagated in either cell type the neutralization rate was similar to that of the initial mixture of MBdV and antiserum. This was also found to be the case when chick cell-derived virus was added to a previously incubQtcd mixture containing virus from this sume cell source. Because of the extremely high persistent frilction 75 100 1 0.1 m :::::i N A 0.01 .r-l :> bJ) ~ .r-l :> :> .r-l 100 10 N :::::i m 1 ct-I 0.,1 0 B 0.01 ----.-I 0 N Q) flt ~. 100 10 \ I I 1 I 0.1 __..J 0.01 8 Time (minu 24 16 s) FIGURE 14. NEUTRALIZATION OF SAN ANGELO VIRUS TO PREVIOUSLY INCUBATED SAN VIRUS-HOMOLOGOUS ANT MIXT SEC ADDIT ON OF VIRUS WAS MADE AT 12 MINUTES. 0, MOUSE BRAIN IVED VIRUS; ~t CHICK C IVED VIRUS. 76 of MBC]V, it was not possible to follow the inactivation of CCdV added secondarily to a mixture of MBdV and antiserum. Instead, heat-inactivated MBdV was incliliated with the antiserum before CCdV was added. The pattern of neutralization (Figure 15) was similar to that observed in Figure 14B. Because the effect of heating on the viruses' antigenic structures was not known, the mouse brain-derived virus used was heat-inactivated at 37 C and 56 C. As a control, normal mouse brain material was treated the same as mouse brain-derived virus, inactivated at 37 C, and then incubated with antiserum prior to the second add ion of virus. Again, the capacity of the antiserum to neutralize virus added secondarily was not significantly diminished. In precipitin reactions the presence of excess antibody gives se to a "soluble" complex characterized by markedly reduced precipitation. An analogous state has been described for some viruses (Lafferty, 1963b). High antiserum concentrations reduce the amount of virus neutralized, i.e., increase the size of the protected fraction. By mixing MBc]V with increasing dilutions of antiserum, it could be shown that the concentration of antiserum employed in the experiments descr d to this point did not prevent neutralization of MBdV (Table 12). Moreover, a 1:800 dilution of rabbit anti-San Angelo virus (chick cell-derived) serum 'NdS found to contain enough antibody to neutralize all dctecLLlble infectivity of the homologous chick cell- 77 w ::> 100 $..4 • .-1 :> :> 10 -.-I :> $..4 :'J w tt-; 1 0 -J-:> ~ OJ 0 $..4 OJ ~ 0.1 0 3 6 9 Time (minutes) F 15RALIZATION OF CHICK C PROPAGATED SAN ANGELO VIRUS BY ANTISERUM PRE-INCUBATED W HEAT INACTIVATED MOU 3RAIN-DEVED SAN ANGELO VIRUS OR MOUSE BRAIN MATERI . At 37 C HEATED MBdSA; s, CHEATED MBdSA; 37 C NMB. .t 78 TABLE 12 INFLUENCE OF ANTISERUM DILUTION ON THE NEUTRALIZATIO':'J OF MO'JSE BRAIN-PROPAGArr'ED SAN ANGELO VIRUS BY RABBIT ANTI-SAN ANGELO VIRUS (CHICK CELL-DERIVED) SERUM. Expt. No. Rec procal of Antiserum Dilution Incubat time (minutes) 30 0 1 0 1.7 x 106a NDb 60 2.0 x 10 6 BOO ND 3.0 x 105 2.4 x 105 1,200 ND 3. B x 105 2.4 x 105 1,600 ND 3.B x 105 2.B x 105 2,OOJ ND 4.4 x 105 2.B x 105 2 0 1.5 x 10 6 ND 1.7 x 10 6 2,00:) ND 1.6 x 105 1.4 x 105 4,OOJ ND 2.4 x 105 1.2 x 105 B,OOJ ND 7.5 x 105 2.B x 105 16,000 ND 1.2 x 10 6 7.0 x 105 -------------------a. PFU/ml b. Not determined 79 derived virus (Figure 12), although only one log of the corresponding mouse brain-derived virus was neutralized at this dilution. Further dilution of the antiserum did not decrease the persistent fraction of the latter virus. These results suggested that the concentration of antiserum utilized did not contribute significantly to the quantitative level of the persistent fraction. Mandel (1958) achieved secondary neutralization in a poliovirus-rabbit antiserum mixture by the addition of guinea pig anti-rabbit globulin serum. The increased inactivation was attributed to antiglobulin-induced stabiliz3tion of primary virus-antibody b~nds. Inactivation of virus after antiglobulin serum has been added suggests that viruses can survive the interaction with antiviral antibodies, even when the latter are bound to the virus particle. An attempt was made to reduce the level of the p2rsistent fraction associated with mouse brain-derived viruses by adding goat-anti-rabbit globulin serum to the previously incubated mixture of virus and rabbit antiviral serum. Although the experimsnt was performed twice with different commercial preparations of goat-antiglobulin serum and included a wide range of dilutions of the goat antiglobulin and rabbit antiviral sera, secondary neutralization could not be detected (Table 13). The activities of the anti- globul in sera 'Nere not tested 7 however I the first lot was biologically active against rabbit globulins in a 80 'rABLE 13 ENHANCEMENT OF NEUTRALIZAT ION OF MOUSE BRAIN-PROPAGATED SAN ANGELO VIRUS BY GOArr ANrr I-RABBIT GLOBULIN SERUM FOLLOWING INCUBATION WITH RABBIT ANrI-SAN ANGELO VIRuS SEK;JM. ------~----,--------.------------ Initial incubation of virus wi: .l6.Q"I Subsequent incubation-L6QII) w/: 3.6 x 10 NRS* . GARS IRS Titer (PFU/ml) of surviving virus 7 1.2 x 10 7 Diluent 3.0 x 10 7 (1:20) (1:8:J) 2.5 x 10 6 . Diluent 1.2 x 10 6 " IRS 1.2 x 10 6 II GARS (1:20) 1.4 x 10 6 II GARS (1:100) 1.5 x 10 6 II GARS (1:200) 1.3 x 10 6 5.8 x 10 6 IRS (1:8:J0) II Diluent 3.0 x 10 6 II IRS 1.8 x 10 6 II II .. * NRS normal rabbit serum; San Angelo) seru~. GARS serUl1. GARS (1:20) 3.0 x 10 6 GARS (1:100) 3.0 x 10 0 GARS (1: 200 ) 3.0 x 10 6 t:: IRS immune rabbit (antigoat anti-rabbit glob~lin 81 radioimmlnoprecipitin assay with polyoma virus (Dr. P. Lombard i, personal communicat ion) • Attention was next directed to the possible influence of the cells employed to assay virus surviving neutralization. Kjellen and Schlesinger (1959) found that "neutral- ized" vesicular stomat lOOO~fold is virus was characterized by a high2r persistent fraction when assayed on primary chick embryo cells than on a continuous line b::>ne marroN cells. similar a~~unts leukemic Similarly, Lafferty (1963b) reported of neutralization of rabbitpJx virus whan assayed In chorioallantoic ffit-=mbranes and in suckling mice o~ th2 the mixture was 3ssayed in rabbit skin. He but substantially lower am::>unts of inactivation virus Wh211 concluded th3t one of the causes of the resistant fraction was th2 inability of certain species of cells to recognize virus as being neutralized, even though illolecules of "neutralizing" antib::>dy were firmly bound to the virus particle. When chick cells and baby h3:l1ster kidney cells -were used to assay neutralized mouse brain-derived San Angelo virus, significant differences in the level of surviving virus could not be detected (Figure 16). When these same cells Were used to assay neutralized chick cell-derived virus, a detectable difference was observed, i.e., the degree of neutralization assayed on BHK cells. ~f CCdSA appeared greater wh2n 82 UJ ::> 1 ~ -"? A QO ~ -"? -"? 0.1 ----.-J 3 6 9 ~ ::> UJ Y--l 0 +' ~ (l) 0 ~ (l) Ili 10 1 a.1 L _ _ _ _ _ .--'--_ _ _ _ _-'-·-_ _ _ ·_···_ _ _ -' 3 6 9 Time (minutes) FIGURE 16. COMPARISON OF NEUTRALIZATIO~ OF MOUSE-BRAIN (A)-AND CHICK CELL(B)-DERIVED SAN ANGELO VIRUSES BY RABBIT ANTI-SAN ANGELO VIRUS ( DR N-DERIVED) ASSAYED ON PRIMARY CHICK (~) AND MODI BABY HA[V1S reER KIDNEY (.) CELLS. BARS INDICATE 95 PERCENT CONFIDEr'{CE LIMIrfS. t 83 The contribution of the virus component to the high surviving fraction, as well as the material with which it was mixed as a pool, was also studied. The experiments included a determination of the presence of aggregates within m~use brain-derived virus samples and the existence of non-infectious virus-specific antigens that might have possessed serum blocking power. Both have previously advanced as underlying causes of the persiste~t bee~ fraction (Wallis and Melnick, 1967; Cardiff et al., 1971). Wallis and Melnick (1967) were able to eliminate the persistent fraction of several viruses by preliminary filtration of samples through me~branes with porosities not exceeding twice the reported diam2ter of the individual virus particles. Their results suggested that virus sequestered within aggregates did not come into contact with antibody and, hence, escaped neutralization. Similar experiments were applied to mouse brain-derived CE groi.lp viruses, using 220 nm Millipore m2:nbrane filters. An attempt was made to differentiate between viral aggregates present before interaction with antibody and immJne a:Jgregates that might be present after reaction with antibody. However, filtration did not alter the neutralization rate significantly or decrease the extent of the persistent fraction (Figure 17). An Qlternate approach was attempted which provided information concerning the possible interference of 84 100~----~- _________________ Q 10 ~ bD ~ A .,j ~ .,j 1 ~ H ::> UJ 100 '. " '0, ct-t 0 "0 - _ _ Q_ +' ~ (l) 0 (J\ 10 ~'0- _ ----- H (l) 0... ....... B 1 $ -. ~ _ _--'--_ _ _I'-----_ _L -_____ J 40 80 Time (minutes) FIGURE 17. NEUTRALIZ RIVED LACRO VIRUS RABB V S ( BRAIN-DERIVED) REMOVAL VIRAL(A) AND FI ION THR 11 220 nm AFT 60 MIN AT LINES: REACT OF V WITH DASHED LINES: REACTION OF V SERUM. 0 85 neutralization by non-infectious, virus-specific, antigenic comp~nents. Cardiff et ale (1971) were able to separate two mouse brain-derived Dengue virus hemagglut by sedimentation through a sucrose grad ting ant nt. The "rapidly sedimcnting h2mag'j"lutinin" consisted c>f whole, infectioi.l:3 virions, while the "slowly sedim2nting hemagglutinin" Was regarded as incomplete virions containing unique anti cO!1formations and protein strllctures. The presence of the latter in a neutralization mixture was suffic inactivat bodies. of the ic to block ctious particles by specific anti- An effort was made to separate infectious CE virus particles from nO!1-infect viral components that might have served to block neutralization. Mouse brain-derived viruses were fuged thro'..1gh solutions of 20 percent sucrose onto 65 percent sucrose cushions and then sedim3nted into 25-65 percent gradients of sucrose; a fractio~ of each gradient containing a part of the infectivity peak (F h~mc>lo- ure 11) was mixed w gous antiserum and incubated. In every case th'3 frac t ion ''Jf virus surviving neutralization was similar to that measured when crude infected mouse brain susp'2nsions were used. The experiments described above ~ere each performed in an attempt to identify conditions that might contrib~te information relevant to the existence and nature of th(~ fraction of CE virus resistant to neutralization by specific antisera. None were fruitful in terms of a definition ~f Ei6 the factors that wO~lld gi ve rise to this stilte. p." 'xami na- tion of th2 observation regarding the inactivatin' genetically identical viruses propagated in diffet It cell species suggested that phenotypic differences could account California for the distinct patterns of neutralization. encephalitis viruses presumably mature when an envelope is acquired as a result of passage through the hast cell membrane. Differences in me:nbrane compositions between cell species would be reflected in the surface properties of the viruses. gate the Three experiments were designed to investisibility that viral envelope differences had determined the reactivity of virus with antibody and, the size of the persistent fraction of CE viruses. each Jf these, the viruses were prepared by h~nce, In sedimentatio~ onto sucrose cushions and then centrifuged into 25-65 percent sucrose gradients. Purified preparations were mixed with enzymes that might be expected to alter the antigenic properties of the viruses - if their specific substrates were present at the virion surface. For each enzyme, a range of concentrations was studied; at. different tim::s the enzymatic activity was arrested and homolo:l:)'Js viral antiserum was added to the mixture. each enzyme on viral infectivity W3S The effect of determined as well as its influence on the' ability of specific antisr3rum to neutralize the treated virus. rrho first enzyme studied was trypsin. Hannoun (19G8) reported that trypsin trcatm3nt of sucrose-acetone 87 preparations of CE virus hemagglutinating antigens caused a significant increase in the hemagglutination ti tors. is contrasted by enzymatic treatment of Gro~p rfhis B Arbovirus antigen preparations, which exhibit loss of hemagglutinating activity after trypsin treatm3nt (Cheng, 1958). was made between the trypsin m~use particles of s~nsitivities A comparison of the infective brain- and baby hamster kidney cell propagated LaCrosse viruses (Table 14). Trypsin (Nutritional Biochemical Corp., Cleveland, Ohio) was prepared as a 400 ug/ml solution in tris-b~ffered sa line, pH 7.5, and was sterili zed by fi 1 trat ion thrOi.lrJh a 220 nm Millipore membrane before use. Tenfold dilutions of the stock solution were IMde prior to each experiment. Equal volu~es preparatio~s of 37 C equilibrated trypsin and virus were mixed and incubated. At designated times the activity of the trypsin was arrested by adding 0.1 m1 heat-treated calf seru~ to each ~l of the mixture. neutralization of trypsin-treated viruses appropriate mixtures. seru~ ~as When stujied, dilutions were added to each of these At the lowest trypsin concentration added (2 ug/ml), 9J percent of the BHK-derived virus 'Nas inactivated. In co~trast, at the highest trypsin concentration used (20,] ug/ml), less than 10 percent of the mouse bra in-propagated virus was inactivated. Trypsin treatment of the latter virus had no effect on the level of non-nQ~tral­ izec] virus following reaction with antiserum. ThGS(~ TABLE 14 EFFECT OF TRYPSIN TREATMENT O~ THE INFECTIVITY OF PARTIALLY PURIFIED LACROSSE VIRUS PROPAGATED IN BHK CELLS AND THE SUCKLING MO~SE.a Time (minutes) Trypsin CO!1cn. MBdLaXb (ug/ml) 0 BHKdLaXb 10 20 ND ND 40 0 1.OxlO 6 1.2xlO 6 10 20 40 ND ND 1.6xlO 6 0 S.Oxl0 5c 2 ND 1.2xlO 6 1.lxl0 6 1.2xl0 6 ND 1.lx10 6 7.2xl0 5 1.6x10 5 20 ND 1.3xlO 6 1.2xl0 6 1.2xlO 6 ND S.Oxl0 4 4.0xl0 4 7.0xlO 3 200 ND 1.lxl0 6 1.Ox10 6 9.0xl0 5 ND 3 3 3.0xlO3 3.0xlO 3.0xlO a. Virus pools clarified by sedimentation onto sucrose cus.hiJn, purification by centrifugation through a sucrose gradient. b. M3dLaX virus. c. PFU/m1. d. Not determined. mouse brain-derived La.Crosse; BHKdLaX = f~llovled by BHK cell-derived LaCrosse 00 00 89 findin~fS W'2rc also noted when Trivittatlls virus was used. The results indiCdted that susceptible peptide bonds associated with BHK-propagated virus were the mOUSe n~t exp0sed on brain-derived virus. LaCrosse virus propagated in the brain tissues of suckling mice Was treated with a preparation of phospholy remove phospholipid lipase C in an attem?t to select compone.:.1ts of the viral envelope that could contrib:.r':e to steric hindrance of antibody kn~::r.rJn to cleave the rn~lecules. ph~sph~diester phatidic acid and the nitrog(~nous (Mahler and Cordes, 1971). The enzyme is linka h3twee.:.1 ph'~s­ base of phospholipids Friedman and Pastan (1969) were able to remove 60 p'::!rcen t of th'2 phosph'Jlipid of Semliki Forest virus withaut causing loss of infectivity. In con- trast, the infectivity of influenza virus (WSN strain) was greatly decreased a a short exp0su~e to the enzyme (Simpson and Hauser, 1965). LaCrosse virus was mixed with increasing phosph'Jlipase C (Worthingto:) cO.:.1centratio~s Biochemical Corporation, Freehold, New Jersey) and the preequilibrated (37 C) mixtures were incubated for variou3 tim~s sal (Table 15). The dilu'2nt resembled tr b'.lt contained .OJI M CaC12 and n'~ -b'..1£fered EDTA. Because th2 enzym2 requires calcium ions (Friedman and Pastan, 1969), its activity could be arrested by adding 0.2 ml of 0.01 M EDTA to each ~l of th8 mixture. expcrilIv~ots Subsequ'2nt neutra 1 iZdt. io() 'w'2re per for me.: c1 in a manner similar to the 90 tryps in studies. Exp0sure of the virus to 10'J u'J of phos- ph()lipase C resulted in an increase in the infectivity titer of the virus preparation 'tlhich "was rnaintuined for at least In the presence of 1 mt] of 15 :ninutes. th(~ enzyme increasec1 infectivity was noted after 5 minutes but by 15 minutes a considerable loss was apparent. The size of this persistent fraction of th0 phospholipase C-treated virus susl?ension was similar to that of the control suspension that had not been incubated with the enzyme. TABLE 15 EFFECT OF PHJSPHOLIPASE C TRE..l\'TMENT ON THE INFECrIVITY OF PAR"fIALLY P'JRIFIED M'J~JSE B!l..l\IN-DERIV£.iJ LACROSSE VIRJS. PhJspholipase C Con2entrat on Time (minutes) -- ____.__________ .LlJ9.LIIl!:.t:_________ ~ _____ _ -.-----.- o 10 0 4.2 x 107b 5 ND 3.8 x 10 15 108 7 ND c ND ND 4.8 x 10 7 1.0 x 10 8 9.9 x 10 7 7 8.6 x 10 7 2.2 x: 1J 7 4.0 x 10 Cl. Worthin3ton Biochemical Corp., 1.5 b. PFU/m1. c. Not determined. Expcri~ents 100J -_._-,------_.- --.--.-- --~.-~------ u~its/mg. were also conducted with a preparati~n J£ ne',JraminidZlSC (Sigma ChemicQ1 Company, St. Louis, MissQuri). rrho rat ionale behi nd t.h:! usc of this enzylw:: was simi lar to that of phJsp!l,)lipase C: RCITl:Jval of negatively-churgcd 91 molecules of neuraminic acid present on viral glycolipids and glycoproteins (Klenk and Choppin, 1970) might alter the surface of the virus in such a way that molecules of 'rhe enzyine antibody would no longer be sterically hindered. was used at a concentration of 0.2 units/ml~ was inhibited by the addition of 0.01 M its activity N-acetylneu~aminic acid (0.2 ml per ml of the incubating mixture). hours of incubation the mouse brain-der~ved After 3 LaCrosse virus titer decreased from 2.2 x 10 7 PFO/ml to 1.2 x 10 7 PFO/ml. When specific antiserum '/'Jas added 30, 60 or 120 min'ltes after neuraminidase had been, the amount of virus neutralized wa.s similar to that observed for untreated controls. To summarize: The use of California encephalitis viruses (propaga ted in th 3 brains of suc::kling :nice) in l plaque reduction studies resulted in the presence of an extremely high level of non-neutralizeable virus. the source of the used to proj~ce antiseru~ nor the source of the virus the antiserum influenced th9 surviving virus. Neither In contrast, the h~st amou~t of cell in which the virus used in the neutralization mixture was propagated determined th2 degree of virus inact.ivation, and h'2nce, the level of the persistent mouse brain-df~Livcd specific, non iTI,)use brain fracti~n. Neutralization of virus was not blocked bi virus- nfectious material present in infected s~..lspensions and filtration to remove viral ag'3rcgatcs did not inccease th·; amount of neutral iZdt ion. 92 The add ion of antiglobulin t.O virus-antibody mixtures did not result in increased loss of infectivity. Indirect evidence was obtained sllggesting that surface differences between viruses propagated in the two cell types lTIay have been responsible for the marked differences in degree of neutralization. IV. NEIJTRl1.LIZl\TION STUDIES Comparis0n of the antigenic relationshi several California enc among litis viruses was attempted by applying the plaque reduction techniques desccibed by Dulbecco at ale (1956), and used by M=Br study of antigen lly related viruses. e B2 (1959) in the the riments W(2re performed, the influence of several potential variables was investigated. The first variable considered was that of initial virus concentration and its effect on the resulting rate of viral inactivati:Jn by a!1t.isex-um .. It was found that varying the concentrati8n of virus present in the neutralization 'nix"tul-e at time zero over a hundred fold d ilut.iol1. ran'3t:; did not affect the rate of neutralization permitted at gure 18). This ast a tenEold dilutian away from the undiluted v pools. Subsequt=ntly, virus stock prepara- tiol1S were diluted at the time Df each experiment to contCl in c:. 1-5 x 10° P';:"U/ml. 93 100 en :::; ~l .r-j :> QD s:::! .r-j :> :> .r-j H :5 en 4-l 0 1 ...j-:> s:::! a o (J) 0 ... H (J) P-t 0.1 0 3 9 Time (minutes) FIGURE 18. EFFECT OF JNITIAL V TRATION RATE NEUTRALIZATI CELL-PROPAGATED SAN ANGELO V S HYPERlMMU RABB ISERUM. ~, 10 · e, 10 ·5 PFU/ml; A, 105.5 PFU/ml. CONC CHICK OUS PFU/ml; 94 The nurnber of virus particles in a virus-antiserum mixture inastivated by neutralizing antibodies have been shown to be influenced by the type of cells on which the mixture is assayed (Kjellen and Schlesinger, 1959) and als'J by the h·")st cells in which the test virus has been propa3ated (Lewenton-Kriss and Mandel, 1972). An atte:npt was made to detect any contr ib'.1tion ei th'er of these might make to the CE virus-rabbit antiserum system ·-1nder study . . Chick e:nbryo and baby hamster kidney cells were used to propagate as well as to assay the neutralization of San Angelo virus. The results of this experiment are pres r3nted in Figure 19. Chick cell- or BHK cell-derived San Angelo virus \!vas mixed wi th an equa 1 volume of appropr ia te ly diluted rabbit antiserum to virus. m~)llSe br.-ain-deri vee] San Angel,") Each component of thf3 mixt.ure '.-las equilibrated to ~3ed 37 C before admixture and both virus preparati8ns were at similar concentrations. Infec ti vi ty sample:3 '·rlere removed from -:'h '3 incubatin9 m=.x·tures at on·3 miOtlte intervals between ze r-o tim(:;! and 6 minu tes. Previously, it had been nJted that the curves d:cawn to represe.::1t th r3 k Lnet CCdV neutralization frequen·tly exhibited an in it that preceded exponential loss of infectivity. of 1 la3 This lag (sh0ulder) was evident in the experimental results presented in Figure 19Jn both typ:=s of assa.y cells. Th2 deviation [rom linea.rity of the plot of B:-iKdSA in3.ctivc1ti')n W~1S not as prot1:Junced ()n ei thor tYP'2 of co 11. Th0 rcsl11ts 95 100 10 1 A o•l _________ L o 2 _ _ _ _--.-L_ _ _ 1 4 6 100 10 1 B 0.1 t o 2 Time (minutes) FIGURE 19. NEUTRALIZATION OF BABY HAIVlSrrER KIDNEY (,,) -AND CHICK EMBRYO CELL (A)-DERIVED SAN ANGELO VIRUSES BY RABBIT AN~I(MOUSE BRAIN-DERIVED) SAN ANGELO VIRUS SERUM: ASSAY ON PRIMARY CHICK ElVlBRYO (A) AND MODIFIED BAP), HAIVi~;ITER KIDNEY (B) CELLS. BARS INDICATE 95 PERCENT CONFIDENCE LIMrrS. 96 also indicated that the two cell types Were similar in their discrimination between neutralized and infectious viruses. c neutralization rate constants have been Whi le used by several inve~tigators to indicate the ant lC relatedness of viruses (McBride, 1959; Ozaki and th·~ 1967), th,~ir ob3erved deviations from linearity precluded appl ication to the campa rison of CE v l.rLlses . Simi larly, oth';::r met.hods that have been us in quantitat . lpS terms, the antigenic relat viruses ware nat used taken into to compare acco~nt beca~se, ~~ans an a c 0_ generally, they have nat the possible ex tence of an initial lag in neutralization or have overlooked the neutralized virus. f sen::::::e of nO(1- ·th~ This necess i tated deve lopmen t of of presenting a meanin 1 comparison of the antigenic celationships of the CE viruses. Num~rous of investiga.to:-s have pointed out lo\ving th.;:: interaction of v thl~ import.an~e cles and 3.nti- sera kinetically, particularly When se'leral viruses are studied that demonstrate a high de Bes 2S nl~utralization tests, kinet d to complement fixation campa cross-rea~tivity. e method3 have been sons of viruses (Hatgi and Sweet, 1971) and hemagglutination inhibiti0n stul]ies of related viru3 isolates (Casa.ls rate th~n any heterologous reilcti~n. I Th~jS, 19(4). kin.~t_ In each ic techn iqUC:3 97 pennit easy identification of homolo90us viruses and in some instances can be used to indicate relative relationships among heterologous viruses. Conversely, Hashi~oto and Prince (1963) were able to detect virus-antibody dissociation by followin9 the neutralization of Japanese encephalitis virus kinetically. With these observations in mind, an effort was made to determine a dilution for each antiserum that would result in a similar rate of inactivation of the respective hom~)logous viruses. Within the limitations inheren·t to the plaque assay, comparis:)l1s of the neutral iza·tion of any virus by different antisera would then be possible. The assumption was made that antisera giving rise to similar virus inactivation rates with their respective h()lllOlo90US viruses would contain similar numbe::::-s of neutra l.izing antibody molecules. It was found I experimen·- tally, that dilutions of each serum could be determined that would leave 20-25 percent, 1.7-5.0 percent, a~d 0.3-1.0 :percent of the oJ.:-iginal infectivity at 3, 6 and 9 minute:3, respectively. W:.lS The semi-logarithmic plot of these values a straight line. the different s~ra In practice the fina 1 dlletti ')ns for tested varied between 1:80 and 1:1400. Tne va 1 id ity of the assu;nptions mentioned above was tested by comparing the neutralization of BFS-233 and th~ee heterologous viruses by two different preparations of antiserum obtainQc] by inoculation of mice with BPS-23J Cl.!1d foll~)\v'ed by collecti:)l1 of the immune ascitic fluids 98 (Figures 20 and 21). These sera were prepared at the National Institutes of Health in 1963 and 1956 and distribu.ted as reference antisera. '/J1;!re Th,= optimum d i llltions Were determined to be 1:1100 ior the earlier preparation and 1:408 for the fluid prepared in 1966. Experimentally, the reactions of the heterologous viruses with each an-tiserum were unique but were similar between antiserum pre ra tio!1s. Reprodu::::ibi li ty between sera rabbits was also noted. pared in It was observed that the rabbit sera t=xhibited a degree of variation that might be expected between similar animals In every instance the gure 22). patterns ''Jbserved b'2tween th'2 pa o~ rabbit sera Were heterologo~s similar with respect to th9 neutralization of viruse:3. The possibility that neutralizing antibodies continued to inactivate virus comprising the surviving infect ity sa~ples fra~tion after had been removed from the incuDatin9 mixtures was also considered. In neutralization experi- ments, samples removed for assay of residual infectivity were routinely diluted ol1e hundredfold in chilled buffer as quickly as ~ossible to minimize further neutralization. An indication of the neutralizing activity contained in sera diluted this far beyond the described optimum dilution was ascertained by incubating sera at the h and hOl11o viruses for 3 hours (Table 16). dilutions In no instance di(l the amount of hOlTIt)logolls virus inactivatccl exceed on(~ ll')g. 99 100 \ 10 \ \ 1 San Angelo 0.1 ----1.------1._-.1-- ---o-e 100 ,, 10 '0 1 Trivittatus LaCrosse 0.1 --L_--,-I_ - - ' - _ - - ' a 6 J 60 9 a J 6 60 9 Time (minutes) FIGURE 20. NEUTRAIIZATION OF CALIFORNIA ALITIS VIRUSES BY MOUSE I-BFSASCITIC FLUID (NIH REFERENCE ANTI PARED 6-66.) ANTISERUM DILUTED 1:400. 100 rn ::5 ~ or-! ~ 10 \ :> bD s::: 1 \ 0 or-! :> :> 0.1 ::5 rn 100 or-! ~ ct-1 0 .p ----4t-----.--------, 10 s::: (J) 0 ~ -~~~ • " '0 1 O.l Trivittatus LaCrosse Q) P-i ,, I 0 I 6 ') J I 9 60 0 0 Time (minutes) FIGURE'; 21. NEU llRALIZ ION OF CALIFORNIA EPHSES MOUSE ANTI S-28J VIRUS ALITIS V ASCIr:l'IC 1D (NIH REFERENCE ISERUUl, PAHED 3-26 3.) AN~iISERUM DILU1 1: 1100. f 1 100 -'- -- ....• -.---.-.- ....... -.--.---- ~-'. ---0 100 ----4~ '- .... '- " .. 10 \ \ \ \ 1 \ A ---lI _ _ _ _-' 0.1 _ - J . ._ _ _----L_ _ _ _ --l-.---JI"-~---Ar- - - _ _• 100 '- 10 1 B 0.1 o ----~------~I------~----~ ) 6 9 60 Time (minutes) FIGURE 22. TRALIZATI OF CALIFORNIA EPHALrrIS VIRU BY ANTISERA COLI,ECTED FROM TWO RAEB S SUB~ITTED TO S ILAR SCHEDULES OF I ION WI1'H BFS-28) VIRUS. IT A SERUM DILUT 1:600; RABBIT B SERUM DILUTED 1: 00. e, TRIVITTATUS VIRUS; A, JERRY SLOUGH VIRUS; 0, BFS-2o) VIRUS. 101 TABLE 16 RESIDUAL ANTIVIRAL ANTIBODY ACTIVITY OF RABBIT SERA AFTER lOO-FOLD DILUTION BEYOND PRE-DE'rERM[N8D OPT I'1AfJ DILUTION FOR NEUTRALIZF\TION OF HOMOLOGOUS VIRUS. Rabbit hyperimmune, antiviral serum 9irected against: Time (minutes) 0 60 180 13:-l a 6 Jerry Slough 5 1.9 x 106c 1.4 x 10 6 8.5 x 10 1.4 x 10 Trivittatus 1.4 x 10 6 1.4 x 111 6 1.4 x 10 6 1.2 x 10 San Angelo 2.6 x 10 6 2.1 x 10 6 1.6 x 10 6 2.2 x 10 6 San Angelob 2.3 x 10 6 2.3 x 10 6 1.6 x 10 6 2.0 x 10 6 Tahyna 2.4 x 10 6 2.6 x 10 6 2.3 x 10 6 2.8 x 10 6 B8'S -283 2.0 x 10 6 1.6 x 10 6 1.5 x 10 6 1.4 x 10 6 L.:.Cross8 8.8 x 105 5.5 x 10 5 3.0 x 105 ND d a. Normal rabbit serum controls. b. Rabbit anti-S:::.n .l\ngelo virus (chick cell-propagated) serum. All other sera are rabbit anti(mouse brain propagated) virus sera. c. PFtJ/ml d. Not determined. 6 Another variable considered was the influence of antiserum concentration on the neutralization of heterologous vIruses. tow~ The two viruses selected, Jerry Slough and James- Canyon, have been classified as variants of one virus (Sudia et al., 1971). When mixed with the optimal dilution of LaCrosse antis(.jrum (Figure 23B), inactivation of lJamestown Cunyon v irus could not: be detected. However, when 102 rn ::::5 1 ~ .r-! :> A b.O ~ .r-! :> :> 0.1 .r-! ~ ::5 rn ct--l 0 100 -j-J ~ Q) 0 H Q) Pi 10 1 0.1 ~-------~----~----__~ o 3 6 9 lJ.lime (minutes) FIGURE 23. INFLUENCE OF ANTISERUM CONCENTRAlJ. ION ON THE KINE'I'ICS OF' HOMOLOGOUS AND HETEROLOGOUS CALIFORNIA ENCEPHALITIS VIRLS NEUTRALIZATION. At RABB ANTISERUM TO OCOUSE BRAIN-DERIVED LACROSSE VIRUS DILUTED 1:100; B, 1:]00 ("OF:[1IMAL") e, LACROSSE VIRUS; A, JERRY SLOUGH VIRUS; 6, lTAlVlESTOWN CANYON VIRUS. 1 103 the antiserum concentration was increased threefold (Figure 23A), neutralization was apparent. Interestingly, the neutralization of Jerry Slough virus was not affected by these antiserum changes . . It was concluded that the use of rabbit antisera at the dilutions effecting the described rate of neutralization would provide an acceptable basis for the comparison of CE viruses by homologous and hetero19gous antisera. Having considered the influence of assdY cell, virus source and concentration, antiserum ~oncentration and biological differences between irnmllnized animals, several experiments were performed to determine the degree of cross-reactivity detectable when heterologous viruses "'Iere mi.xed with each of several different antisera. For these experimen~BFS-283 (large plaque), LaCrosse, San Angelo, Trivittatus, Tahyna, Jerry Slough and Jamestown Canyon viruses were each inoculated into +:.wo rabbits which wece then immunized according to the schedule outlined in Table 2. sera of these rabbits were stud d, When the hyperimmune was noted that each serum except those specific rqr Jerry Slough and Jamestown Canyon viruses could be used in neutralization at dilutions greater than 1:100. experime~ts Jerry Slough antiserum was used at a 1:80 dilution; neither serum frOILl 1:..11e rabbits injected with Jamestown Canyon virus was able to neutralize homologous virus at dilutions as low as 1:40 and so they Were not used. The rabbits Were inoculated with viruses 104 propagated in mouse brain tissue whih.~ the vi.rI.U:)(~~; used in plaque reduction studies were propagated in BHK cells: This maximized the likelihood that antibody would react with virus ific antigenic components and that anti-host cell antibod ization. s would not interfere with or contribute to neutralInfectivity samples were removed from the incubating mixtures at 3 minute intervals for 9 minutes; the next . samples were rem:)ved at 60 and 180 minutes. Normal serum was used in zero time and 18J minute control tubes: 3 hours' incubation, s()me loss of After ctivity was noted, although this seldom exceeded 0.5 log (32 percent). each of the In llo\"ing F l gures the neut La 1 i za ti on of the homologous viruses is plotted in the lefthand graph. The neutralization of viruses that did not cross-react to a significant degree, i.e., lose more than 50 percent of the vi ty by 60 rninutes, has bec?n reported zero ti me infec separately as the reduction of infectivity at 180 minutes. The advanta of kinetic campa when the results of inactivation of sons became evident se~/eral CE viruses by BFS-283 antiserum (E'igure 24) Were compared with data obtained by the immunodiffusion studies of Calisher and Maness (1970). In the neutralization stud s significant sh;')ulders were evident in the plots in Which no virus hao been inactivated. These were apparent for each the heterologous viruses batween zero and 6 or 9 minutes. The homologous virus was eusily distin9uished from all o:'her .--. --.------. 100 l \ 10 \ \ 1 :J H •.-1 \ San 3-283 W . . -. \ LaCrosse • 10 0.1 :> , '6 100 :> .r-! :> '\. 10 '0 H ::::> w +..:> ......... 1 Jerry Slough Trivittatus Janie s town Canyon 'H o ~" 0.1 I" "" " I c C> u H 100 ~-----w Q I) ,. " 10 I) "Q ~ \ f \ 1 Keys \ Tahyna o 3 I 6 \ t determined [ Me~ao o 3 loooV 9v Q 60 (minutes) URE 24. NElJTRA1IZATION CALIFORNIA MOUSE BRAIN IVED BFS1:800. IS VIRUSES BY VIRUS. 1r-' o U1 106 viruses studied; the heterologous viruses could be grouped according to their patterns of inactivation. Thus, San Angelo and Tahyna viruses were easily distinguished from Jamestown Canyon, Trivittatus and Keystone viruses. La- Crosse virus was distinct in that it did not cross-react with anti-BFS-283 serum during 60 minutes' incubation. After 180 minutes less than one log (0.7 log) of this virus had been neutralized. In the experiments of Calisher and Maness (l970), LaCrosse, San Angelo and Jamestovln Canyon viruses exhibited "3+" precipitin reactions against BFS-283 antiserum while the homologous virus was characterized by a "4+" reaction. The results of experiments on neutralization of several California encephalitis viruses by laCrosse hyper-immune serum have been summarized in the graphs of Figure 25. The time course of neutralization of Jerry Slough virus was similar to that of the homologous LaCrosse virus, although an initial lag was evident and after 9 minutes the rate of neutralization decreased considerably. As has be~n noted previously, Jamestown Canyon and Jerry Slough viruses are regarded by some investigators as variants of one virus (Sudia et al., 1971). It was inter- esting to note the unique patterns of neutralization of these viruses when mixed with anti-LaCrosse serum. For the other viruses the degree of cross-reactivity was limitcc]. The extent of neutralization of Trivittatus, 100 0 • i,"'" i 10 L " i 1 W :::5 H .r-j O• 1 " L II l 0, "~ La.,C"Y'o("'s . ~ c e ~I !_ I,.., U D. D 0 v I ~ [ .. II sa~ An~elO roo 100 .. .. - , , '0 283 IOV>,A,I I 100"\A,1 0--0 \ > H :::5 tn 10 \ I \ 1 LTr\Vi tU~"AA.1 <+-1 o o H C) P-t r ~ Q ~ --.... () I 10 L 1 ,.---.y I Y--y II 1 [Me~~o Tahyna 0.1 Jerry S10u.gh '.It. tt1a 0.1 100 ,, 0 o o TimE~ TO (minu 3 1 6 1--.1 9 60 S) ZATICN CALIFORK ENCEPHALITIS VIRU SE BRAIN-DERIVED LACROSSE VIRUS. ANTI- 1:300. t--' o -..J 108 Melao and San Angelo viruses after 180 minut.es in Table 17. lS in(lif~at!:!ll The neutralization of each of these viruses did not exceed one log after 3 hours. TABLE 17 EFFECT OF E>ITEND2:D TIME ON THE NETJrRALIZATION OF SELECrrED CALIFO::<!.JIA E~\IC£PHALITIS VIRUSES BY RABBIT ANT 1MOUSE BRAIN-DERIVED LACROSSE VIRUS SERUM. a Time (minutes) Virus Strain -~--,-~-- Ob 180b 180 60 TVT ,.. 10° 1.4 x 7.0 x 10 5 5.0 x 10 5 1.0 x 10 6 MEL 1.8 x 10 6 9.0 x 10 5 1.6 x 10 5 1.0 x 10 6 SA 1.7 x 106c 1.1 x 10 6 2.0 x 10 5 1.6 x 10 6 a. Antiserum diluted 1:300. b. Normal rabbit serum controls. c. PFU/ml. When Tahyna virus antiserum was used to inactivate CE viruses only Su.:! Angelo and BPS -283 'were neutralized to a significant extent (Figur2 26). As was noted with the other antisera, inactivation of these viruses was preceded by a lag. LaCrosse, Trivittatus, Jerry Slough, Keystone and M21ao viruses were indis tral nguishabl'2 tion curves Were compared. whC:.:tl their neu- Extension of the reac- tion time of four of the isolates to 180 minlltes die] not Eaei 1. i tat(~ (TaLle IFn. their serological separation by th is antiserum 100 (Z 10 II II I,,(I~ .-.. -. III \ \ \ \ 1 [{;) :::s ~ .r-l () o . 1 LTahyna .____--'___ I \ IZI San Angelc ~or.r.1 I ~~\J 1 I" r. LaCrosse I A > 100 43---<bl!.~---_ > > 10 c75 1 .,-1 6 ~ ....... A, "D H o +-" C (]) o H o•1 I I ('\vo"c\v t o _'-11---------,...-.. " Q) P-i Jerry Slough Trivittatus Jamestown Canyon 'H '. 10 ~ "y -Q \ \ 1 \ Keystone BFS-28J \ o Ir. I v 9 v6 o r. "" o 0 J 6 v I I o Melao I J I 6 I" r... ' " I 9 v~60 Time (minutes) DI 26. NEUTRALIZATION OF CALIFORNIA ENCEPHALITIS VIRUSES BY ANTISERUM ~O MOUSE BRAIN-DERIVED TAHYNA VIRUS. ANTISERUM l:JOO. I--' o \.0 110 TABLE 18 Nf:UTRALIZATION OF CALIFORNIA ENCEPHALITIS VIRUSES BY RABBIT ANTI-TAHYNA (MOUSE BRAIN-DERIVED) VIRUS SERUM. a Time (minutes) Virus Strain Ob 60 JS 5.0 x 10 5 8.0 x 10 LaX b 180 180 4 2.0 x 10 4 4.8 x 10 5 1.7 x 10 6 7.0 x 10 5 1.1 x 10 5 1.7 x 10 6 TVT 1.1 x 10 6 2.0 x 10 5 1.0 x 10 5 1.0 x 10 6 KEY 5.0 x 10 5 1.0 x 10 5 4.0 x 10 4 4.0 x 10 5 a. Antiserum diluted 1:300. b. Normal rabbit serum controls. The antiserum s ic for Jerry Slough virus was the most specific of the sera used: Fewer cross-reactions were noted with this antiserum than with any (Figure 27). After 180 minutes the amount of each virus that had been of neutralizat f·J1d the ot.heJ:-s i·1.crea.s(~ vated had not increased from the degree noted at 60 minutes (Table 19). in the antiserum concentrat significantly increase the amount of A four- did not virus inact (Table 19). Keystone, Melao, Tahyna and San Angelo viruses cxhib significant cross-reacti\'it.if~s virus antiserum (Figure 28). when mixed with Trivittutus The neutralization of the other viruses was not eXi:lmined at extended times. d I 100 AI >-, l:r~, W :3 H .~ o•1 :-> ~Je Slough l I I 10 0 I ') q a a -6 I San Angelo h ' "I :-> t 10 termined +' Q) 1- 0.1 ~ o H Trivittatus Jamestown Canyon Q) 100 P-t BFS-283 ---.-. I 1 ct-l o lI I, , , 1 0 0- -e L H :3 tf) O--EJ - .....0 vvv o :-> 0 I I 100 .~ [], La~ros~e ~ I Ip oe~ r ~, • "'- 10 I 11 [Keylstone 0.1 I 6 0 3 r () () ()- -() I- I Tahyna 10v 9 ~ v0 v I 60 0 V, "- • Melao I j 3 6 10 " 0 I 9 v v 60 Time (minutes) FIGURE 27. NEUTRALIZATION OF CALIFORNIA EPHALITIS VIRUSES BY RA5B ANTISERUM TO MOU BRAIN IV JERRY SLOUGH VIRUS. 1SE? DI 1:80. t-' I-' I-' i !~ ill ::::s H .r-! \ \ San Angelo Trivi ttatus I ! I Ls n 10 A "'> I ""- i \ l~I ~O 0.1 0 GJ 100 ~, • If) v f) v D I r t BFS;-28 3 , IL ,~,J > .r-! > > H .r-! ::::s ill lOC! ~ A ........ 10 1 o•1 '00 ° Jerry Slough LaCrosse Jamestown Canyon !f-i o .A A: ........ 6- -6 10 v " <po I'. I 0 "0' >:: OJ u H OJ P-t 100~ ~ \ 10 L I 1 \ Tahyna Keystone 0.1 o " \ " '~ o 3 6 \ Lr" r.. A'rt 9 V6 o V v '" lao o 3 6 --'...c,"-~ vv 9 60 Time (minutes) FIGURE 28. NEUTRALIZAT RA3B ANTISERUM SERUM DI 1:900. OF CALIFOR~IA EPHALITIS VIRUSES BY BRAIN-DERIVED TRIVITTATUS VIRUS. ANTIf-' t-' tv 113 TABLE 19 INFLUENCE OF Kl(TENDED T IMg AND [I..J"~~REASED CONCENTRATION OF' ANfrISERrJM 0::1 Ti-IE NEUTRi\LIZATION OF CALIFORNIA ENCEPa~LITIS VIRUSES BY RABBIT ANTISERUM TO JERRY SLOUGH -vIRUS PROPAGATED IN SUCKLING MICE. Time (minutes) Virus Strain Oc 180 a 180 ND ND b 180 c JS 1.4 x 10 6d rrAH 1.4 x 10 6 1.7 x 10 6 1.4 x. 10 6 1.6 x 10 LaX 1.2 x 10 6 1.2 x 10 6 7.7 x 105 1.2 x 10 6 SA 2.2 x 10 6 1.8 x 10 6 1.6 x 10 6 1.9 x 10° Tv'r 1.6 x 10 6 1.2 x 10 6 7.7 x 105 2.2 x 10() 283 1.8 x 10 6 1.8 x 10 6 1.2 x 10 6 1.6 x 10 6 KEY 2.0 x 10 6 1.5 x lOS 4.2 x 10 4 1.2 x 10 6 MEL 5.0 x 105 1.4 x 10 4 3.0 x 10 3 5.0 x 105 a. Antiserum diluted 1:80. b. Antiserum diluted 1:20. c. Normal rabbit serum controls. d. PFU/ml. e. Not determined. 1.3 x 10 6 6 r r When San Angelo virus antiserum was used to neutralize other CE viruses, cross-reactivities were noted with B?S-283, Jamestown Canyon, Trivittatus, Jerry Slough, Keystone and Tahyna viruses; only LaCrosse virus was not neutralized by the serum (Figure 29). Melao virus was not studied. San Angelo virus, significantly, could not be distinguished 100~ -. - '1II~ 10 1_ ~ I 1 L ::s H 'M o•1 \ San w \ LaCrosse 10 I_ _ L... __1. I ov"',.p,) 5-283 \ b., 1"\,0.,,1'\. , U 0 ,,b :> cD >=: 'M :> 'M :> 100i~ 10 L , " H ::s m ....... 1 o .... 0 Trivittatus ct-1 ' ... Jerry Slough 0.1 +> >=: (!) o H 100 \iii ';I n , (!) P-i 10 \ \ 1 Keystone 0.1 ~ Not , , '~ o l----kv-V'VJ 3 6 \ Tahyna 9 60 o 3 \ lov'vl'.,f) 6 9 60 determined I Melao 1. o 3 I I ...... v v0 v I 6 9 60 'Time (minutes) FIGURE 29. NEUTRALIZATION RP..B3IT ANTISERUM TO MOUSE SERUM LUTED 1: O. IS VIRUSES ANGELO V • ANTIf-' f-' ~ 115 from Trivittatus virus by the anti-San Angelo virus serum during the first 9 minutes of reaction. Significant, immediate inactivation was also apparent with the neutralization of BFS-283; however, the rate was lower than that of the homologous reaction. DISCUSSION The experiments described herein Were conducted in an attempt to further the understanding of the complex antigenic relationships which characteri~e encephalitis group of viruses. co~pare the California While the intention was to the viruses by plaque reduction techniques, the studies found necessary before this could be achieved prised a significant proportion of the experiments. c~~­ For purposes of discussion the data can be separated into four topics: pH and plaque formation; the persistent fraction in neutralization reactions; the kinetic lag noted in homologous and heterologous neutralization reactions; and the cross-reactivities observed between members of the CE virus group. The behavior of the two types of cells used to investigate plaque formation by CE viruses was similar in most mensurable respects. One obvious distinction was the ability of the line of hamster cells to support formation of plaques by certain members of the virus group that did not form countable plaques on chick embryo cells, at least not under the conditioI)s studied. St~veral experilOont.3 indicated an extraordinary sensitivity of most of the viruses to the conditions prevailing during incubation of 117 infected cells under overlay media containing agarosc. That this gelling agent may have influenced plaque formation Was suggested by several experiments. For example, the size and number of plaques Viere influenced by the medium employed in the overlay, while the patterns of replication - in fluid cultures - in terms of the kinetics of appearance of viruses and yield of particles - were not affected by the medium used (Figure 6). overlay ~9dium at the ti~e The pH of the it was added to infected cells also influenced plaque formation. This affect was also influenced by the presence of agarose: A virus that pro- duced plaques at pH 7.0 but not at pH 7.3 replicated in fluid cultures equally well at pH 7.0, 7.3 or 7.6. The observation that plaque sizes were increased when DEAEDextran was included in the overlay medium suggested the presence of contaminating ani~nic substances i , the agarose preparations that were capable of inhibi ting th,e virusGs. The noted increase in plaque sizes and numbers when the concentration of the agar-)sc; ',Jas decreased also seemed to indicate the presence of inhibitory substances, although the decrease in concentration that effected this improvcmc~nt was not great. Indeed, the slight change in concentration could have had a greater affect on the diffusibility of nutrients or metabolic products through the gf~l. 118 Plaque formation of several viruses is known to be affected by the conditions generated by the use of solidifying substances (Ventura, 1968; Colter et al., 1964). The improvement in plaque sizes and numbers observed when agar is replaced by agarose has been interpreted to mean that the presence of contaminating sulfated polysaccharides is responsible for virus inhibi 1961a). on (Takemoto and Liebhaber, Furthermore, when polycations are added to agar- containing overlay med I plaque sizes and, sometimes, numbers increase; this has also been interpreted as an indication that polyanions interfere with infection of cells by viruses (Liebhaber and Takemoto, 1961b). Restriction of diffusion has also been advanced as the explanation of virus inhibition by gelling substances. Wallis and Melnick (1968) noted that polio and herpes viruses can be inhibited by dextran sulfate, a sulfated polyglucose molecule, while natural agar polysaccharides have no effect. In contrast, Campbell and Colter (1965) reported that the plaques of mengo-virus were enhanced by dextran sulfate. These results suggested to Wallis and Melnick (1968) that polyanions present in the gels interfered with the diffusion of viruses through semi-solid med per to the particles. but did not bind, These investigators also indicated that the inhibitory affect could be eliminated by decreusing the gel concentration used. 119 Reports of viruses sensitive to the overlay medium pH have indicated that, where sensitivities can be demonstrated, they are due to prevailing acidic cond Vogt et ale ions. (1957) showed that the tld" mutants of polio- virus produced significantly fewer plaques when the pH, established by changing the bicarbonate concentration, was below 7.1. Similarly, small plaque-producing strains of rubella virus were inhibited by acid pH overlay conditions (Fogel and Plotkins, 1969). The plaques of foot-and- mouth disease virus variants were also inhibited under agar media when the initial pH was 7.0-7.2 (Martinsen, 1970). In each of these examples, the viruses produced plaques at higher pH values, even at pH 8.6. The observations reported for the CE viruses were markedly different: The majority of the variants studied did not produce plaques above pH 7.5 and produced plaques of greatest size, number and clarity between pH 7.0 and 7.2. Results of the experiments involving plaque formation by Californ encephalitis viruses d not favor either of the above-mentioned proposed mechanisms of inhibition. The addition of a polycation such as DEAE-Dextran would have contributed to an increase in the net positive charge of the agarose gel. This might have been expected to enhance the ability of the virus particles to diffuse through the overlay meclium beCCluse of charge differences or becuusc anioni.c inhibitors had been sequestered by cationic 120 components. Similarly, the increased plaque sizes which accompanied the decrease in agarose concentration could have been due to enhancement of diffusibility by reducing the viscosity or density of the gel or by reducing the concentration of the inhibitor. Wallis and Melnick (1968) suggested that reports of enhanced plaque formation under agarose Were probably not attrib~table to the absence of sulfated polysaccharides but, instead, to the higher pH which characterizes preparations of agarose. of the CE virus studies do not support this The results explanati~n. Still to be explained is the extraordinary affect of small initial pH jifferences on the ability of CE virus2s to produce plaques on the two cell species studi9d. The determination of condi t.ions tha t ""QuId pr J!fiot'::! formation of plaques by CE viruses permitted application of plaque re~uction techniques to the determinati~n of the i':ltragrollp antigenic serol~gical methods. I:1 relationshi~)s preli~inary sugges ted by other expe meni:s was noted that the host cell in which the virus to be used in ne~li::..ra 1 izd. ti ~n experiments \vas prepared de-!:::.erminedt.h(~ outcome of virus-dntibody interactions, while the virus us~~d to immll1,ize animals had no influence. This was represented gr3phically by the presence of a significant persistent fraction of surviving virus following neutralization when the virus used had been propagated in suckling mouse brain tissues. When viruses propagated in chick 121 embryo cells or baby hamster kidney cells were studied, the surviving fractions were much smaller or could not be demonstrated. Other investigators have also noted diffi- culties associated with the use of mouse brain-derived viruses. Hashimoto and Pr ince (1963) concl u(l,:;:!:l I:hat neu- tralization kinetics were not applicable to the study of Japanese encephalitis virus: The virus used was derived from preparations of infected mouse brains. Cardiff et al. (1971) demonstrated the presence of virus-specific, noninfectious substances in mouse brain preparations of virus that were able to interfere with the neutralization of infectious particles. Removal of these by centrifugation reduced the persistent fraction considerably but did not eliminate it. The significance of the virus source to be used in neutralizati·-:>n stud others .. Le~vent;:>n~-K.r-iss s has ,:llso been suggest:.ed by and Mandel (1972) found that the persistent fraction of poliovirus propagated in HeLa cells was hi0her than the surviving fraction of the same virus prepared in monkey kidney cells. These differences suggest the involvement of host cell-determined variations which could also be used to explain the above-mentioned differences between CE viruses of one strain propagated in dissimilar host cells. phen~typic Cells are known to contribute to differences between otherwise similar viruses. For example, a single passage of Venezuelan equine encephulitis virus into a host or host cell unrelated to 122 that in which the virus was first propagated, resulted in detectable differences in the lipid composition of the viral envelope (Heydrick et al., 1966). Similarly, the lipid composition of SVS virus varied with the host cell in which it was propagated and resembled the lipid composition of the plasma m3mb.r.anes been demonstrated when one v has been propagated Stenback and Durand (lg63) showed tha t N'2wcastle disease v gruwn was characterized by lower dens g~ted ~lsa Density diffecences haV8 (Klenk and Choppin, 1969). different host cells. the respective cells in mammalian cells. i11 ce lIs of avian Qt- i ~ s than the same virus In an experiment that has not been presented as a part of this thesis, the rate of sedimentation of mouse bra -derived virus was found to be Sll)vlei:- than that of the same virus propagated in baby hamster kidney cells. This suggested different densit s of the two virus types and prompted the attempts to demonstrate dissimilarit s their envelopes. When purified preparations of each type of virus were treated with trypsin, distinct d became evident. rences between their properties The sensitivity of hamster cell-derived viruses at low concentrations of the enzyme was contrasted sharply with the resistance of mouse brain-de even at high enzyme concentrations. ved viruses, Trypsin is an cnclopcptidase Which exhibits specificity for pept contuining the carboxyl groups of arginine or lysine bonds 123 (Mahler and Cordes, 1971). The susceptibility of the one virus to this enzyme indicated that the determinants associated with infectivity contained at least one of these amino acids. Conversely, the resistance of the other virus suggested that enzyme, determinants had been protected from the ther by proteins that were not affected by trypsin or by non-prote substances. That the latter may was suggested by the results of an have been invo ri- ment in which mouse brain-propagated virus was treated with phospholipase C. This enzyme has been used by others to release phospholipids and cholesterol from the envelopes of viruses (Friedman and Pastan, 1969; Cartwright et al., 1969). As has been mentioned, the effect of phospholipase C on infectivity var In the strain used, an increase In While it would provide support stance of the CE virus ct ity was observed. evidence to interpret this finding as an indication that the enzyme cleaved material from the virus in such a way as to render inactive particles infectious by exposing cr ical sites, the data could also be interpreted to indicate that the enzyme had dispersed aggregates of virus. The ability antibodies to inactivate mouse brain-der neutralizing viruses was not enhanced by treatment of the viruses with either trypsin or phospholipase C. The apfBrent interference with neutralization by surface st.ructures associated with viruses grown mouse 124 brain ssues seemed to suggest that, while antibod ios may have been able to attach to these viruses, they may not have been able to assume stable configurations or bonds that might have resulted in inactivation. Lafferty (1963b) hypothesized that the initial interaction between an infectious v ion and a molecule of antibody is reversible and involves one viral determinant and only one of the two combining sites of the antibody molecule. This inter- action, which could involve more than one determinantcombining site pa , was referred to as "sensitization ll and by definition preceded neutralization of the particle. Neutralization sensitized particles would follow the reaction of a second combining site with a second viral determinant on the same virus particle. When th stabili- zation does not occur the sensitized viruses remain infectious and give rise to the persistent fraction. Mandel (1958) has shown that some preparations of poliovirus retained infectivity in the presence of neutralizing antibod s which were associated with the virus particles: Addition of anti-globulin serum to a "neutralized"mixture decreased the amount of surviving virus. The persistent fraction associated with CE viruses prepared from infected moUSe brain tissues can be explained in terms of Lafferty's hypothes The resistance of mouse brain-derived viruses to trypsin, contrasted by the sensitivity of hamster cell-derived viruses, suggested the 125 presence of non-protein muterial which, in effect, protected the antigenic determinants from the action of the enzyme. This masking e could also have prevented antibody molecules from assuming stable configurations with antigenic determinant pa either by ste lly hindering the anti- body molecule or by making the effective distance between available determinants too When the results of the neutralization of chick cellderived CE v s were plotted semi-logarithm second departure from linear was detected. lly, a For a limited time after virus and hyperimmune serum had been mixed, the rate of inact ion was slower than was noted at subsequent times (Figure 19). This lag was also evident when early immune sera were studied (Figure 13). The application of specific rate constants to the comparison of neutra ion of related viruses depends, in part, on the assumption that a single molecule of antibody is sufficient to inactivate a virus icle. If the additional assumption is made that the change in the concentration of antibody molecules after reaction is negligible, the equations of first order or monomolecular reaction kinetics may be applied. A semi- logarithmic plot representing this kind of reaction is a straight line, while departure from linearity, i.e., a lag, indicates a reaction greater complexity (Svehag, 1968). The demonstration of a lag preceding the exponential loss of CE virus infectivity in at least two situations seemed l26 to preclude the use of these mathematical techniques. In- stead, a method was developecJ which permitted visualization of antigenic similarities and dissimilarities of the viruses by determining and plotting the kinetics of inactivation of each strain. similar: Experimentally, the homologous reactions Were This led to the assumption that, at the dilutions used, the concentration of neutralizing antibody molecules in each Serum was approximately the same. In the heterolo- gous combinations, three types of cross-reaction were apparent: There was either no detectable reaction, cross- reactivity in which the curves of inactiva on were charac- terized by significant lags, or reactions in which no lag was demonstrable and the rate of inactivation resembled the rate of neutra zation of the homologous virus. These distinct patterns provided the basis of an attempt to explain the cross-reactions and, hence, relationships within the CE virus group. The presence of a kinetic lag in neutralization experiments suggests that more than one event must transpire or that the critical event leading to inactivation proceeds slowly. The rapidity of primary antigen-antibody reactions (Haurowitz, 1968) favors the first possibility. If only one antibody molecule is required for neutralization, a lag would not be expected, even in heterologous reactions. Rather, a linear neutralization curve of decreased rate would be noted. If, on the other hand, more than one 127 antibody molecule is necessary to bring about hct{~r.\~)lol}()uS virus neutralization, a lag should be demonstrable. Further- more, the extent of the lag might be expected to be influenced by the specificity, avidity and heterogeneity of the antibod s part ipating in the neutralization reaction. In this context, spe ficity is defined as the degree of cross-reactivity an antiserum exhibits while avidity is viewed as the tendency of antigen-antibody complexes to dis soc spontaneously or upon dilution. Webster (1968) measured the avidity of antisera reacting with influenza viruses by determining the equilibrium constants of the reactions. His data indicated that antisera of low avidity reacted only with the homologous antigens and were, therefore, highly spe fie. Similarly, antisera of high avidity were found capable of reacting with heterologous antigens; these were considered to be low specificity. These observations can be extended to the interpretation of the CE virus neutralization results but do not account for all the observations. For these, other explanations must be advanced. That little is known about the mechanism of viral crossreactions is evidenced by the controversy between proponents of the single determinant and mosaic hypotheses regarding the antigens involved in neutralization (Rappaport, lr)~·)L). A virus exemplifying the single determinant hypothesis contains repeating units of a sing antigen: a cross-reacting 128 virus would be characterized by a structurally simi antigen present as a repeating unit. Adherents of the mosaic hypothesis propose the presence of more than one antigen, each capable of interacting with neutralizing antibodies. A cross-reactive virus would possess at least one of these anti ant and it would be structurally identical to the present on the other virus. Support for either hypothesis is limited and in most instances indirect. Serologically, a lag would not be expected if the relationships between antigenically similar viruses were attribuIe to common, identical antigenic determinants and if only one molecule of antibody was necessary to inactivate the virus. Among hemagglutinating luenza viruses, the peptide maps of gens isolated fro~ closely related strains (determined by cross-reactions) have been found to be very similar. In contrast, similar antigens isolated from viruses before and after a major antigenic shift were observed to be s 1972). ificantly different (Webster and Laver, These observations suggest ions in single determinants rather than the presence of mUltiple discrete antigens. Evidence from the study of other enveloped viruses also s sts the presence single determinants. For example, Kennedy and Burke (1972) isolated a single envelope prate from vesicular stomatitis virus. Neutra1- izing antibody specific for this polypeptide inact.ivated the homologous virus only (Wagner et al., 1969). Whi 129 several investigators have reported the presence of single polypeptides in the envelopes of arboviruses (Strauss et al., 1969; Shapiro et al., 1971), Sehlesinger et ale (1972) identified a second glycoprotein in Sindbis virus that was found to co-migrate in polyacrylamide gels with the previously identified envelope glycoprotein. Three LaCrosse virus polypeptides have been separated by polyacrylamide gel electrophoresis, although their biologIcal function or position in the virus was not reported (McLerran and Ar 1 , 1973). Two d tinct patterns of heterologous CE virus neutral- ization were apparent in the studies of CE virus crossreactions. In the one, neutralization was signif lag, and in some a rate of inactivation simi reaction, , lacked a nces was characte zed by to that of the homologous sting a high degree of structural simi between the antigenic determinants of the viruses (cf. Figures 25 and 29). rity volved That these viruses were not identical, however, was indicated by the unidirectional nature of their identities and by the dissimilarit s in their reactions with immune sera corresponding to other CE viruses. The second type of heterologous virus neutralizat was characterized by an absence of detectable inactivation, lasting varying periods of time following admixture. The lengthy lags might be (]scribec1 to the presence - in hi9h 130 concentrations - of antibodies of low specificity as well as low avidity. Initial interaction between virus particles and these antibodies would prevent reaction with antibodies of greater avidity. However, dissociation of the form8r would give the more avid antibody molecules the opportunity to react with the virions. As the number of these inter- actions increased neutralization would become measureably evident. Inherent to this interpretation is the assumption that neutralization requires more than o.ne molecule of antibody to be affected. The detailed comparisons presented herein for selected CE virus isolates and their specific hyper immune sera supported, in some cases, the classificatory proposals advanced by other investigators using less sensitive as well as less specific serological techniques. In other instances, however, significant differences were noted. With these observations in mind the value of extensive studies with each CE virus isolate seemed questionable without providing parallel data using previously applied techniques such as complement fixation and hemagglutination inhibition and without examining larger numbers of immune sera representing different immunization routes, schedules, etc. The most significant observation made, with respect to the antigenic relationships of the CE viruses, was that the employment of kinetic techniques permit much greater discrimination in determining the nature of the relationships of CE viruses that could not be differentiated by other serological techniques. REFERENCES Andersen, J. K. 1972. The influence of complement on cytomegalovirus neutralization by antibodies. Arch Fur Die Gesamte VirusforschunCj. 36:133-140. Ardoin, P., D. H. Clarke, and C Hannoun. 1969. The preparation of arbovirus hemagglutinins by sonication and trypsin treatment. Amer. J. Trop. Med. and Hyg. 18: 592-598. Ashe, W. K., and A. L. Notkins. 1966. Neutralization of an infectious herpes simplex virus-antibody complex by anti-gammaglobulin. Proceedings of the Ndtional Academy of Sciences, u.S. 56:447-451. Ashe, W. K., and H. W. Scherp. 1963. Antigenic analysis of herpes simplex virus by neutralization kinetics. J. Immun. 91:658-665. Bales, G. L., L. L. Salomon, and R. W. Tew. 1967. Techniques for rapid detection of Californ encephalitis virus. Technical Report, Dugway Proving Ground, T 67116. Bardos, W. 1960. Immunological study of antibodies neutralizing Tahyna v in the sera of inhabitants of Czechoslovak J. Hyg., Epidem., Microbiol., and Immunol. 4:54-60. Bardos, W., and V. Dan lova. 1959. The Tahyna virus, a virus isolated from mosquitoes in Czechoslovakia. J. Hyg., Epidemiol., Microbiol., and Immunol. 3:264-271. Bardos, W., and L. Sefcovicova. 1961. The presence of antibod s neutralizing Tahyna virus in the sera of inhabitants of some European, Asian, African, and Australian countries. J. Hyg., dem., Microbiol., and I~nunol. 5:501-504. BergolC), G. H., and R. Mazzali. 1968. P formation by arboviruses. J. Gen. Virology. 2:273-284. Bond, J.~., W. McD. Hammon, A. L. Lewis, G. E. Sather, and D. J. Taylor. 1966. Californ 9rouP Qrboviruscs in F da and Report of a neW strQin, Keystone virus. Public Health Reports. 81:607-613. 132 Borden, E. C., G. W. Gary, anc1 F. A. Murphy. 1970. Compar on of agar and agarose preparations for Mcngovirus plaque formation. Applied Microbiology 20:289291. Burgdorfer, W., V. F. Newhouse, and L. A. Thomas. 1961. Isolation of California encephalitis virus from the blood of a Snowshoe Hare (Lepus Americanus) in Western Montana. Amer. J. of Hyg. 73:344-349. Burnet, F. M., E. V. Keogh, and D. Lush. 1937. The immunological reactions of filterable viruses. Australian Journal of Experimental Biology and Medical Science 15:227-247. Calisher, C. H., and K. S. C. Maness .. 1970. Arbovirus identification by an agar-gel diffusion technique. Applied Microbiology 19:557-564. Campbell, J. B., and J. S. Colter. 1965. Studies on three variants of Mengo encephalomyelitis virus. III. Effect of overlay and polyanions on plaque size. Virology 25:608-619. Cardiff, R. D., W. E. Brandt, T. G. McCloud, D. Shapiro, and P. K. Russell. 1971. Immunological and biophysical separation of Dengue-2 antigens. J. of Virology 7:15-23. Casals, J. 1957. Viruses: the versatile parasites. I. The arthropod-borne group of animal viruses. Transactions of the New York Academy of Sciences 19:219-235. Casals, J. 1961. Procedures for identification of arthropod-borne viruses. Bulletin of the World Health Organization 24:723-734. Casals, J. 1964. Antigenic variants of Eastern Equine encephalitis virus. Amer. J. Trop. Med. and Hyg. 119:547-565. Cartwright, B., C. J. Smale, and F. Brown. 1969. Surface structure of vesicular stomatitis virus. Journal of General Virology 5:1-10. Causey, O. R., C. E. Causey, O. M. Maroja, and D. G. Macedo. 1961. The isolation of arthropod-borne viruses, including members of tv-/O hitherto underscribed serological groups, in the Amazon region of Brazil. Amer. J. Trop. Med. and Hyg. 10:227-249. 133 Cheng, P-Y. 1958. The inactivation of group B arthropoc]borne animal viruses by proteases. Virology 6:129-136. Chun, R. W. M., W. H. Thompson, J. D. Grabow, and C. G. Matthews. 1968. Californ arbovirus encephalitis In children. Neurology 18:369-375. Colon, J. I., J. B. Idoine, C. M. Brand, and R. D. Costlow. 1965. Mode of action of an inhibitor from agar on growth and hemagglutination of group A arboviruses. J. Bact. 90:172-179. Colter, J. S., M. A. Davies, and J. B. Campbell. 1964. Studies of three va ants of Mengo encephalomyelitis virus. II. Inh ion of interaction with L cells by an inhibitor and by protamine. Virology 24:578585. Cooper, P. D. 1961. The plaque assay of animal viruses. In advances in Virus Research, (K. M. Smith and M. A. Lauffer, eds.) Vol. 8, pp. 319-378. Academic Press, New York. Crane, G. T., R. E. Elbel, D. E. Klimstra, and K. L. Smart. 1970. Arbovirus collections from mosquitoes col cted in central Utah in 1967. Amer. J. Trap. Med. and Hyg. 19:540-543. Crookston, J. A., E. R. Kern, and D. W. Hill. 1969. Plaque assay and growth characterist s for California encephatis group viruses. Proceedings, Annual International Northwestern Conference on Diseases in Nature Communicable to Man 24:9-10. Daniels, C. A., T. Boros, H. J. Rapp, R. Snyderman, and A. L. Notkins. 1970. Neutralization of sensitized virus by purified components of complement. Proceedings of the National Academy of Sciences 65:528- 5. Daniels, J. B., J. J. Ratner, and S. R. Brown. 1961. P reduction, a sensit test for Eastern encephalit s antibody. Science 133:640-641. Dolana, G. H. 1968. Physical and chemical characteriza tion of purified group A arboviruses. Doctoral Thesis, Univcrsity of Utah, Department of Microbiology. Dulbccco, R., M. Vogt, and A. Str land. 1956. A study of the bLlS ic uspccts of ncut.ra lization wi th two anima 1 viruses, Western Equine encephalitis virus and Poliomyelitis virus. Virology 2:162-205. 134 Espana, C., and W. McD. Hummon. 1948. An improved bczencextracted complement fixing antigen applied to the diagnosis of the arthropod-borne virus encephul ides. J. of Immunol. 59:31-44. Fogel, A., and S. A. Plotkin. 1969. Markers of rubella virus strains in RK13 cell culture. J. of Virology 3:157-163. Fr man, R. M., and I. Pastan. 1969. Nature and function of the structural pholipid of an arbovirus. J. of Mol. Biology 40:107-115. Gard, S. 1957. Immunoinactivation of poliovirus. fur die Gesamte Virusforschung 7:449-460. Archiv Gresikova, M., W. C. Reeves, and R. P. Scrivan. 1964. California encephal virus: An evaluation continued endemic status in Kern County, Cali Amer. J. Hyg. 80:229-234. Grimes, J. E., E. H. Garza, and J. V. Irons. 1962. Is tion of California virus from Anopheles pseudopunct nnis. Paper presented at the Eleventh Annual Meetof the American Soc ty of Tropical Medicine and Hygiene, Atlanta, Georgia. Habel, L. 1969. Virus neutralization test. Fundamental Techniques in Virology (K. Habel and N. P. Salzman, eds.), pp. 288-296. Acad c Press, New York. Hammon, W. McD., and W. C. Reeves. 1945. Recent advances in the epidemiology of the arthropod-borne virus encephalitides. Amer. J. of Public Health 35:994-1004. Hammon, W. McD., and W. C. Reeves. 1947. Interepid studies on arthropod-borne rus encephalitides and poliomyelitis in Kern County, California, and the Yakima Valley, Washington, 1944. Amer. J. Hyg. 46:326335. Hammon, W. McD., and W. C. Reeves. 1952a. California litis virus, a newly described agent. I. Evidence of natural infection in man and other animals Ca Medicine 77:303-309. Hammon, W. McD., W. C. Reeves, and P. Galindo. 1945. Epidemi c stuclies of encephulitis in the San ,Jouquin Valley of California, 1943, with the isolation of viruses from mosquitoes. Amer. J. Hygiene 42:2'))306. 135 HQmmon, W. McD., W. C. Reeves, and G. E. Sather. 1952b. California ence~laliLis virus, a newly described agent. II. Isolations and attempts to identify and characterize the agent. J. of Immunol. 69:493-510. Hammon, W. McD., and G. E. Sather. 1966. History and recent reappearance of viruses in the California encephalitis group. Amer. J. Trop. Med. and Hyg. 15: 199-204. Hannoun, C. 1968. Arbovirus hemagglutinins: Differential susceptibility to trypsin. Nature 219:753-755. Hashimoto, N., and A. M. Prince. 1963. Kinetic studies on the neutralization between Japanese encephalitis virus and antiserum. Virology 19:261-272. Hatgi, J. N., and B. H. Sweet. 1971. Complement fixat1.;rl kinetic studies with Dengue viruses. Archiv fur die Gesamte Virusforschung 35:323-338. Haurowitz, F. 1968. Immunochemistry and the Biosynthesis of Antibodies, 1st ed., pp. 301. Interscience Publishers, New York. Henderson, B. E., and P. H. Coleman. 1971. The growing importance of California arboviruses in the etiology of human disease. In Progress in Medical Virology (J. L. Melnick, ed.), pp. 404-461. S. Karger. Heydrick, F. P., R. F. Wachter, and J. J. Hearn. 1966. Host influence on the characteristics of Venezuelan Equine encephalomyelitis virus. J. of Bacteriology 91:2343-2348. Kalrnanson, G. M., A. S. Hershe, and J. Bronfenbrenner. 1942. Factors influencing the rate of neutralization of bacteriophage by the antibody. J. of Immunol. 45: 1-12. Kennedy, S. I. T., and D. C. Burke. 1972. structural proteins of Semliki Forest Gen. Virology 14:87-98. Studies on the ruse J. of Kjellen, L. E., and R. W. Schlesinger. 1959. Influence of host cellon residual infectivity of neutralized vesicular stomatitis virus. Virology 7:236-238. Klenk, H-·D., and P. W. Choppin. 1969. Lipids of Plas rna membranes of monkey and hamster kidney cells and of parainfluenza virions grown in these cells. Virology 38:255-268. 136 Kokernot, R. H., B. M. McIntosh, C. Brookeworth, T. DeMorais, and M. P. Weinbren. 1962. Isolation of viruses from mosquitoes collected at Lumbo, Mozambique. I. Lumbo v , a neW virus isolated from Aedes (Skusea) Pembaensis Theobald. Amer. J. Trop. Med. and Hyg. 11: 678-682. Kenz, C., S. M. Buck , and J. Casals. 1964. Antibodies in man against Tahyna and Lumbo v uses determined by hemagglutination-inhib ion and ssue culture neutralization tests. Amer. J. Trop. Med. and Hyg. 13:738741. 1963a. The interaction between v and Lafferty, K. J. antib ,){ly. I. Kinetic studies. Virology 21:61-75. 1963b. The interaction between virus and Lafferty, K. J. antibody. II. Mechanism of the reaction. Virology 21:76-90. Laver, W. G. , and R. G. Webster. 1972b. Studies on the origin of pand c influenza. II. Peptide maps of the light and heavy polype ide chains from the hemagglutinin subun s of A2 influenza viruses isolated before and after the appearance of Hong Kong luenza. Virology 48:445-455. Lewenton-Kriss, S., and B. Mandel. 1972. Studies on the non-neutralizable fraction of poliovirus. Virology 48: 666-678. Liebhaber, H., and K. K. Takemoto. 1961. Alteration of plaque morphology of EMC virus with polycat,ions. Virology 14:502-504. Likar, M., and J. Casals. 1963. Isolation from man in Slovenia of a virus belonging to the Cali ia complex of arthropod-borne viruses. Nature 197:1131. Lowry, o. H., N. H. Rosebrough, A. L. Parr, and R. J. Randall. 1951. Protein measurements with the Folin Phenol reagent. J. of Biological Chemistry 193:265-275. McBride, W. D. 1959. Ant nic analysis of poliovirus~s by kinetic studies of serum neutralization. Virology 7: 45-58. McLerran, C. J., and R. B. Arlinghaus. 1973. Structural components of a virus of the California encephalitis complex: LaCrosse v Virology 53:247-257. 137 Mahler, H. R., and E. H. Cordes. 1966. try. Harper and Row, New York. Biological Chemis- Mandel, B. 1958. Studies on the interactions of poliomyel is virus, antibody, and host cells in a tissue cdlture system. Virology 6:424-447. Mandel, B. 1961. Reversibility of the reaction between poliovirus and neutra zing antibody of rabb origin. Virology 14:316-328. Marshall, A. L. 1964. Encephalitis - Indiana. and Mortality Weekly Reports, 13:414. Morbidity Martinsen, J. S. 1970. The effect of diethylaminoethyl dextran and agar overlay pH on plaque formation by two plaque-size variants of Foot-and-Mouth diseasG virus. Canad n Journal of Comparative Medicine 34:13-19. Murphy, F. A., and P. H. Coleman. 1967. Ca f orn ia g rOJ.] p arboviruses: Immunodiffusion studies. J. of Immunol. 99:276-284. Murphy, F. A., S. G. Whitfield, P. H. Colema~, C. H. Calisher, E. R. Robin, A. B. Jenson, J. L. Melnick, M. R. Edwards, and E. Whi tney. 1968. Ca lifornia group arbovirus;2s: Electron microscopic stud s. Experimental and Molecular Pathology 9:44-56. Newhouse, V. F., J. A. McKiel, and W. Burgdorfer. 1964. California encephalitis, Colorado tick fever and Rocky Mountain spotted fever in Eastern Canada. Canadian Journal of Public Health 55:257-261. Notkins, A. L., S. Maher, C. Scheele, iln~ J. Goffman. 1966. Infectious virus-antibody complex in the blood of chronically infected mice. Journal of Experimental Medicine 124:81-97. Ozaki, Y., and K. Tabey. 1967. Studies on the neutralization of Japanese encephalitis virus. I. Application of kinetic neutralization to the measurement of the neutralizing potency of antiserum. J. of Immunol. 98 : 12 1 B.- ~~ _.; . Philipson, L. 1966. Interaction of poliovirus and immunoglobulins. II. Basic aspects of virus-antibody interaction. Virology 28:35-46. 138 Porterfield, J. S. 1960. A simple plaque-inhibition test for the study of arthropod-borne viruses. Bulletin of the World Health Organization 22:373-380. Porterfield, J. S. 1962b. In Recent Advances in the DemJnstration of Antigen-Antibody Reactions (J. F. A~kroyd and J. L. Turks, eds.). Blackwell, London. Quick, D. T., A. G. Smith, A. L. Lewis, G. E. Sather, and W. McD. Hammon. 1965. California encephalitis virus infection: A case report. Amer. J. Trop. Med. and Hyg. 14:456-459. Rappaport, I. 1961. Analysis of the cross-reaction between two strains of tobacco mosaic virus. Nature 189:986990. Rappaport, I. 1970. An analysis of the inactivation of MS-2 bacteriophage with antiserum. J. of Gen. Virology 6:25-32. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty percent endpoints. Amer. J. of Hyg. 2 7 : 493 -49 7 . Rubin, H., and R. M. Franklin. 1957. On the mechanism of Newcastle disease virus neutralization by immune serum. Virology 3:84-95. Sather, G. E., and W. McD. Hammon. 1967. Antigenic patterns within the California encephalitis virus group. Amer. J. Trop. Med. and Hyg. 16:548-557. Schlesinger, M., S. Schlesinger, a~d B. W. Burge. 1972. Identification of a second glycoprotein in Sindbis virus. Virology 47:539-541. Shapiro, D., D. Trent, W. E. Brandt, and P. K. Russell. 1972. Comparison of the virion polypeptides of group B arboviruses. J. of Virology 6:206-211. Simpson,R. W., and R. E. Hauser. 1965. Structures associated 'with influenza virus suspensions treated with phospholipase C. Virology 27:642-646. S mi t h burn, K. C., A . F. Ma ha f f y, and J. H. Pau 1 . 1941 • Bwamba fever and its' causative virus. Amer. J. Trop. Med. 21:75-90. Spence, L., C. R. Anderson, T. H. G. Aitken, and W. G. Downs. 1962. Molao virus, a new agent isolated [rom Trinidadian mosquitoes. Amer. J. Trop. Med. and Hyg. 11: 687-690. 139 Spencer, C. 8. 1965. California encephalitis infection revealed among Ohio patients. Ohio State Medical Journal 61:768-769. Stenback, W. A., and D. P. Durand. 1963. Host influenza on the density of Newcastle Disease virus (NDV). Virology 20:545-551. Stirn, T. B. 1970. Dengue virus plaque development in Simian cell systems. I. Factors influencing virus adsorption and variables in the agar overlay medium. Applied Microbiology 19:751-755. Stirn, T. B. 1970. Dengue virus plaque development in Simian cell systems. II. Agar variables and effect of chemical additives. Applied Microbiology 19:756-759. Stoker, M., and I. MacPherson. 1964. Syrian hamster fibroblast cell line BHK-21 and its derivatives. Nature 203: 1355-1357. Strauss, J. H., B. W. Burge, and J. E. Darvell. 1969. Sindbis virus infection of chick and hamster cells: Synthesis of virus-specific proteins. Virology 37: 367-376. Sudia, W. D., V. F. Newhouse, C. H. Calisher, and R. W. Chamberlain. 1971. California group arboviruses: Isolations from mosquitoes in North America. Mosquito News 31:576-600. Svehag, S-E. 1968. Forr~tion and dissociation of virusantibody complexes with special reference to the neutralization process. Progress in Medical Virology 10: 1-63. Takemoto, K. K., and H. Liebhaber. 1961. Virus polysaccharide interactions. I. An agar polysaccharide determining plaque morphology of EMC virus. Virology 14 :456-462. Taylor, R. M. 1967. Catalogue of Arthropod-borne Virusc=;; the World, No. 1760, pp. 898. u.S. Government Printing Office, Washington, D. C. Thompson, W. H., B. Kalfayan, and R. o. Anslow. 1965. Isolation of California encephalitis group virus from a [utal humLln illness. Amer. J. of Epidemiology 81: 245-253. Valle, M. 1971. Factors affecting plaque assay of animal viruses. ACTA Patholo(]ic'l et Microbiologica ScancJinuvica Section B, Supplement No. 219, pp. 69. 140 Ventura, A. K. 1968. Comparison of starch, agarose, and agar for plaquing Venezuelan encephalitis virus. Applied Microbiology 16:808-809. Vogt, M., R. Dulbecco, and H. A. Wenner. 1957. Mutants of poliomyelitis viruses with reduced efficiency of plating in acid medium and reduced neuropathogenicity. Virology 4:141-155. Wagner, R. R., T. A. Schnaitman, and R. M. Snyder. 1969. Structural proteins of vesicular stomatitis virus. Journal of Virology 3:395-403. Wallis, C., and J. L. Melnick. 1967. Virus aggregation as the cause of the non-neutralizable persistent fraction. Journal of Virology 1:478-488. Wallis, C., and J. L. Melnick. 1968. Mechanism of enhancement of virus plaques by cationic polymers. Jo~rnal of Virology 2:267-274. Wallis, C., and J. L. Melnick. 1970. Herpes virus neutralization: Induction of the persistent fraction by insufficient antibody. Virology 42:128-137. Way, H. J., and D. J. Garwes. 1970. Serum accessory factors in the measurement of arbovirus neutralization reactions. Journal of General Virology 7:211-223. Webster, R. G. 1968. The immune response to influenza virus. III. Changes in the avidity and specificity of early 19M and IgG antibod s. Immunol. 14:39-52. Webster, R. G., and W. G. Laver. 1972a. Studies on the origin of pandemic influenza. I. Antigenic analysis of the A2 influenza viruses olated before and after the appearance of Hong Kong influe~za using antis2ra ta the isolated hemagglutinin subunits. Virology 48:433444. Wellings, F. M., G. E. Sather, and W. McD. Hammon. 1970. A type-specific immunodiffusion technique for the California encephalitis virus group. Journal of Immunology 105:1194-1200. Wellings, F. M., G. E. Sather, and W. McD. Hammon. 1971. Immunoelectrophoretic studies of the California cnceph(llit.L~:; virus group. Journal of Immunology 107: 252-259. E. G. 1967<1. The neutralization of arbovirusC's. I. Nc;utr<11ization in homologous virus-serum mixtures with two group B arbovlruses. VirOlogy 26:517-527. Wc~~tuW.~1y, 141 Westaway, E. G. 1967b. The neutralization of arboviruses. II. Neutralization in heterologous virus-serum mixtures with four group B arboviruscs. Virology 26:528537. Whitman, L., and R. E. Shope. 1962. The Californiu. .~omplex of arthropod-borne viruses and its relationship to the Bunyamwera group through Guaroa virus. Amer. J. Trop. Med. and Hyg. 11:691-696. Yoshion, K. and S. Tamguchi. 1965. Studies on the neutralization of herpes simplex virus. I. Appearance of neutralizing antibod s having different grades of complement requirement. Virology 26:44-53. t