A study of the specificities of sequential antisera to variola and monkeypox viruses by radioimmunoassay*

Bulletin of the World Health Organization, 58 (1): 131-138 (1980) A study of the specificities of sequential antisera to variola and monkeypox viruses by radioimmunoassay* HARRIET H. WALLS,1 DONALD W. ZIEGLER, 2 & JAMES H. NAKANO 3 The specificities of antisera during development of the humoral antibody response to poxvirus antigens were examined in monkeys injected with chimp-9 whitepox virus or monkeypox virus. Sera were obtained from 3 African green (vervet) monkeys inoculated with chimp-9 whitepox virus, 1 rhesus monkey inoculated with monkeypox virus, and 2 rhesus monkeys inoculated with soluble monkeypox viral antigen. Thesequentially obtained serafrom each animal were adsorbed with uninfected chicken chorioallantoic membranes (CAM) or vaccinia virus-infected CAM. The adsorbed sera were tested by radioimmuno- assay to determine thespecificity ofthe residualantibodies to vaccinia, variola, andmonkey- pox viruses. The adsorbedsera at differentstages ofthe immune responseshowed increasing specificity with time after inoculation. Generally, antibodies in sera collected earlier than 21-27 days after immunization could not be identified after adsorption, but late sera could be identified unequivocally. Recent progress in the worldwide smallpox eradi- cation programme has increased the need for detecting and evaluating putative smallpox cases after the acute phase of the disease has passed. Because viral iso- lations will not always be possible, effective surveil- lance requires reliable serological procedures for differentiating smallpox from other closely related poxvirus infections. Monkeypox, variola, and vaccinia viruses, although closely related, have been shown to contain unique, type-specific antigens (1, 2). Related viruses have been successfully differentiated by identification of component antigens; however, the complexity of the methods has precluded purification of sufficient quantities of viral antigens for extensive use in sero- logical procedures. Several methods have been de- scribed for differentiating poxvirus antisera in which selective and exhaustive adsorptions of sera with crude viral suspensions are used. The usefulness of these procedures for preparing type-specific antisera is diminished because of the tedious separation tech- niques required to remove the adsorbent antigens. Previously we described a radioimmunoassay (RIA) procedure that effectively identified specific rabbit antisera after a single adsorption with crude antigen preparations (3). In this simplified procedure, the * From the US Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control (CDC), Atlanta, GA 30333, USA. The use of trade names is for identification purposes only and does not constitute endorsement by the Public Health Service of the US Department of Health, Education, and Welfare. I Research Microbiologist, Viral Immunology Branch, CDC. 2 Chief, Viral Immunology Branch, CDC. 3Chief, Viral Exanthems Branch, CDC. antigen-antibody complexes were not separated; the residual unadsorbed specific antibodies were measured directly, without further preparative steps. Antisera to monkeypox, variola, and vaccinia viruses could be differentiated by comparing the relative con- centrations of residual antibodies to each virus. Inasmuch as serum specimens are obtained from patients at various intervals after acute infection or vaccination, the relationship between strain specificity of the sera and the time since the antigenic stimulus must be considered. In this study, the specificities of sequential antibodies appearing after poxvirus infec- tion of primates were investigated. The adsorption RIA procedure was used to measure the specificities of antibodies in sera resulting from infection of monkeys with whitepox virus (chimp-9 strain) and monkeypox virus (Utrecht strain). In addition to the specificities of these sera, the specificities of sequential sera from monkeys immunized with soluble monkeypox (Copen- hagen) viral antigen were examined. MATERIALS AND METHODS Virus strains Poxvirus strains were provided by the Viral Exan- thems Branch, Center for Disease Control (CDC), Atlanta, Georgia. Except for vaccinia, the viruses were originally isolated from crust materials of infected patients or infected nonhuman primates, and all isolates were passaged in embryonated chicken chorioallantoic membranes (CAM). The following virus strains were used in this study: 3932 - 131 - H. H. WALLS ET AL. Wyeth strain of vaccinia (Vac) virus was obtained from Wyeth Laboratories, Philadelphia, PA.; African smallpox (ASP) virus, a variola (Var) major strain (CDC No. V68-I-258), was isolated from a patient in Sierra Leone in 1968 and was described by Lourie et al. (4); Utrecht monkey monkeypox (MMP) virus was isolated from an orang-utan during apoxlike epizootic among nonhuman primates in a zoological garden in Rotterdam (5); Copenhagen MMP virus was isolated from cynomolgus monkeys in 1958 and was described by von Magnus et al. (6); and chimp-9 (Ch-9) virus, a variola-like whitepox virus strain (CDC No. V71-I-224), was isolated in Moscow from the kidneys of an apparently healthy chimpanzee caught in Zaire in 1971 and was described by Marennikova et al. (7). Antigen preparation A soluble MMP viral antigen was used for inocu- lation of rhesus monkeys. The soluble antigen from CopenhagenMMP virus was prepared as described by Olsen et al. (8). Antigens for RIA and antiserum adsorption were prepared in chicken CAM as described previously (3). The antigens used for RIA included uninfected CAM and Vac-, Var-, and MMP-infected CAM. The anti- gens for adsorption procedures included only un- infected CAM and Vac-infected CAM. Antisera preparation Antisera to Ch-9 virus were prepared in 3 African green (vervet) monkeys by subcutaneous (SC) inocu- lation of Ch-9 virus prepared in CAM and having a titre of 106.8 pock-forming units (PoFU) per 0.1 ml. An inoculum of 0.1 ml was given at each of 6 sites on the animals' shaved backs. Blood samples were col- lected at least weekly by femoral venous puncture during the first 6 weeks and at varying intervals thereafter for up to 4 months. Antisera to infectious Utrecht MMP virus and Copenhagen MMP soluble antigen were produced in rhesus monkeys. One animal (Rh-3-76) was inoculated with 10565 and 107.4 PoFU of infectious Utrecht MMP virus at 0 and 6 days, respectively, and was boosted with infectious Utrecht MMP virus (107.4 PoFU) at 33 days. For comparison with antisera produced to infec- tious MMP virus, 2 other animals (Rh-1-76 and Rh-2-76) were immunized with Copenhagen MMP soluble antigen at 0, 7, 17, and 24 days. At 33, 39, and 66 days after the original inoculation with soluble MMP antigen, both monkeys were challenged with 105.5, 107.4, and 1074 PoFU of infectious Utrecht MMP virus, respectively. All the animals were bled from the femoral veins at 3- to 7-day intervals for up to 2 months. Samples of the sera were prepared and stored at -70 °C until all sera were collected. All the sera were then tested by RIA in a single series of experiments. Adsorption procedure In a previous investigation we described an RIA adsorption procedure in which rabbit hyperimmune poxvirus antibodies displayed characteristic binding patterns when reacted with orthopoxviruses after adsorption with selected heterologous poxvirus anti- gens (3). The investigation revealed that anti-Var (anti-ASP, anti-Harvey [Har], or anti-Ch-9), anti- MMP (or anti-human monkeypox [HMP]), and anti- Vac sera are distinguishable when they are adsorbed with Vac virus and reacted in RIA with Vac, Var, and MMP viruses. In the procedure each antiserum was also adsorbed with uninfected CAM to remove non- specific antibodies that react with egg material. To determine specificity, each antiserum was adsorbed with Vac virus and tested by RIA for residual activity to Vac, MMP, and Var viruses. With hyper- immune sera (anti-Vac, anti-Var and anti-MMP) adsorption with Vac virus removed virtually all the Vac virus antibody activity. After Vac virus adsorp- tion the residual antibody activity upon reaction with MMP virus was negligible for anti-Vac, negligible or low for anti-Ch-9 and anti-Var, and high for anti- MMP and anti-HMP. In a similar procedure, when these same Vac virus-adsorbed antisera were reacted in RIA with Var virus, the residual antibody activity was low for anti-Vac, low for anti-MMP and anti- HMP, and moderate for anti-Var and anti-Ch-9. In the light of these results, sequentially obtained antisera from monkeys were adsorbed with only two antigens, uninfected CAM or Vac virus-infected CAM. Each serum was diluted either twofold or three- fold in phosphate-buffered saline (PBS) containing 200 ml of fetal calf serum (FCS) and 0.4 g of sodium azide per litre. Aliquots of each dilution were trans- ferred to separate tubes, and an equal volume of uninfected CAM or Vac virus-infected CAM was added to each dilution. As controls, uninfected or Vac virus-infected CAM was added to tubes containing FCS and sodium azide in PBS. The serum dilution- adsorption mixtures were incubated for 2 h at 35 °C, held at 4 °C for 16-18 h, and then incubated at 35 °C for 30 min. Aliquots (0.025 ml) of the uncentrifuged adsorption mixture were then assayed for residual antibody activity by using the RIA procedure. In pre- liminary trials, adsorbed sera were also tested with and without centrifugation to remove the adsorbent anti- gen material. The results suggested that centrifugation did not facilitate differentiation of specificity. Therefore, the centrifugation step was not incorpor- ated in the routine RIA procedure. 132 ANTIBODY RESPONSE TO POXVIRUS ANTIGENS RIA procedure The RIA procedure was identical with the assay de- scribed by Hutchinson et al. (3). In the procedure, the antiviral antibodies were reacted with virus-infected CAM, which were affixed to the wells of microti- tration plates. Antipoxvirus antibodies were quan- tified by adding an optimal amount of 125I-rabbit antimonkey globulin (9) and measuring the radio- activity affixed to the antigen-antibody complexes in the microtitration wells. The antibody activity to Vac, Var, and MMP viruses after Vac virus adsorption was measured and compared to the antibody activity to Vac virus after CAM adsorption. The relative amount of antibody remaining after Vac virus adsorption was expressed as the percentage of the antibody reactive with Vac virus after CAM adsorption and was calcu- lated by the following formula: Percentage residual activity = Antibody titre (Vac, Var, or MMP) after Vac adsorption x 100 Antibody titre (Vac) after CAM adsorption 3000 2000 1000 800 600 500 400 300 200 100 80 60 50 40 30 20 10 RESULTS Specificity of sequential antibodies resulting from infection with Ch-9 virus Serum specimens were obtained sequentially from three African green (vervet) monkeys that had been inoculated with a crude suspension of Ch-9 virus. The animals were inoculated on day 0 and received no additional injections. The animals were bled at least weekly during the first 6 weeks and at varying intervals thereafter for up to 4 months. Preliminary serum titrations by RIA to determine the sequential appearance of poxvirus antibodies (Fig. 1) revealed no apparent antibodies in sera earlier than 9 days after inoculation. At 9 days each animal showed an antibody response, with precipitous increase to maximal titre between 11 and 16 days. Sub- sequently, the RIA titres decreased slightly but re- 3000 2000 1000 800 600 500 400 300 200 a) P- 00~ 0 a) 100 80 60 50 40 30 20 l10 0 10 20 30 40 50 60 70 80 90 100 110 0 10 20 30 40 50 60 70 80 90 100 110 Days After Primary Immunization Fig. 1. Sequential radioimmunoassay antibody titres reac- tive with vaccinia virus after inoculation with chimp-9 virus on day 0: VM-1-75 0-* ; VM-2-75 A-A ; VM-3-75 v-v All antisera obtained 9 days or later after inoculation were adsorbed with uninfected chorioallantoic membranes before being tested by RIA. Days After Primary Immunization Fig. 2. Sequential radioimmunoassay antibody titres to poxviruses after inoculation on day 0 with chimp-9 virus (VM-2-75). Sera adsorbed with uninfected chorioallantoic membranes and reacted with vaccinia virus *- ; ad- sorbed with vaccinia virus and reacted with vaccinia virus 0; adsorbed with vaccinia virus and reacted with variola virus A- ; and adsorbed with vaccinia virus and reacted with Utrecht monkeypox virus VV. 133 134 H. H. WALLS ET AL. mained high throughout the test period (70-100 days). Sera obtained at 9 days or later were tested by adsorption RIA to determine the relationship between the poxvirus strain specificity and the time after immunization. The specificity was ascertained by adsorbing the sera with poxvirus antigens and measur- ing residual antibodies. The specificities of the sequen- tial specimens were similar for each of the 3 animals; therefore, results of the adsorption studies for only 1 animal infected with Ch-9 virus (VM-2-75) are shown in Fig. 2. Adsorption of early sera (9 or 11 days after immunization) with Vac virus yielded sera with re- sidual antibody consistently reactive with Vac virus but randomly reactive with Var and MMP viruses (Table 1). The random distribution of residual anti- Table 1. Characterization of antibodies in sequential sera of monkeys immunized with chimp-9 virus Vac adsorbed b CAM Vac Var MMP Serum Vac Identifi- number titrea Titre % c Titre % Titre % cation d VM-1-75 (11)e 2869 422 15 138 5 270 9 7 (18) 1548 207 13 358 23 128 8 Var± (25) 1982 165 8 268 14 196 10 Var+ VM-2-75 (9) 325 109 34 .: 10 - 3 ( 10 4 3 ? (16) 2395 438 18 115 5 163 7 7 (27) 1654 289 18 185 11 428 26 ? (65) 2377 195 8 483 20 155 6 Var (105) 1447 52 4 585 40 129 9 Var VM-3-75 (9) 955 150 16 87 9 .10 1 7 (13) 519 448 86 280 54 ( 10 ( 2 7 (15) 2113 797 38 1018 48 524 25 Var± (31) 1541 123 8 384 25 149 10 Var (71) 1264 99 8 873 69 148 12 Var a Titres of CAM adsorbed antisera reacted with Vac virus antigen. b Titres of Vac virus adsorbed antisera reacted with antigen as indicated. c Percentage residual antibody activity = Vac virus adsorbed titre CAM adsorbed titre x 100 d Interpretive identification. e Days after primary immunization. body activity to Vac, Var, and MMP viruses persisted through about 27 days. Later in the immune response (>31 days), however, the residual antibodies reactive with Vac and MMP viruses declined, and those reac- tive with Var virus increased. Hence, only those sera collected at 31 days or later showed adsorption pat- terns characteristic of anti-Ch-9 (or -Var) sera and could be identified correctly. Specificity of sequential antibodies resulting from injection with MMP virus A rhesus monkey (Rh-3-76) was inoculated with infectious UtrechtMMP virus on day 0 and day 6. The animal was bled three times during the first month after primary inoculation. Poxvirus antibody activity in serum collected 9 days after inoculation was not sufficient for determination of specificity. MMP viral titres were nearly maximal 16 days after inoculation and resembled the humoral antibody responses of the monkeys inoculated with Ch-9 virus (Fig. 3). How- ever, unlike the sera of monkeys inoculated with Ch-9 virus, the adsorption pattern of the 16-day serum was characteristic of MMP antiserum (Table 2). In both the 16-day and 29-day sera, the residual antibodies reactive with Vac virus were somewhat elevated, but 3000 2000 lo8oo ./ A 11000 800 I 600 / 500 I 400 300 < 200 pn110 loLo _ 80 IV60 I 50_ 40 30 20 / ~10 Di 0 10 20 30 40 50 60 70 80 90 100 110 A 'A A la laA la A A ~~ADays After Primary Immunization Fig. 3. Sequential radioimmunoassay antibody titres reac- tive with vaccinia virus after inoculation with (A) soluble Copenhagen monkeypox viral antigen on days 0, 7, 17, and 24 and infectious monkeypox virus on days 33, 39, and 66: Rh-1-76 0- , Rh-2-76 A--A or (B) infectious monkey- pox virus on days 0, 6, and 33: Rh-3-76 +-* . All antisera obtained 9 or more days after inoculation were adsorbed with uninfected chorioallantoic membrane before testing by radioimmunoassay. Symbols on the abscissa indicate days of inoculation for each group of animals. ANTIBODY RESPONSE TO POXVIRUS ANTIGENS 135 Table 2. Characterization of antibodies in sequential sera of monkeys immunized with monkeypox virus Vac adsorbed b CAM Vac Var MMP Serum Vac Identifi- number titrea Titre % c Titre % Titre % cation d Rh-1-76 (17)e 70 42 60 18 26 13 19 ? (21) 329 289 88 58 18 342 104 MP ? (28) 324 262 81 95 29 486 150 MP (36) 306 84 28 66 22 382 125 MP (39) 780 103 13 121 16 349 45 MP (62) 759 154 20 119 16 1138 150 MP (75) 2714 204 8 290 11 1411 52 MP Rh-2-76 (21) 101 39 39 (10 . 10 128 127 MP (28) 359 46 13 43 12 427 119 MP (42) 2071 149 7 75 4 3044 147 MP (49) 1481 327 22 172 12 1585 107 MP (62) 854 149 17 152 18 1358 159 MP (75) 2407 168 7 147 6 1622 67 MP Rh-3-76 (9) 19 . 10 - 53 < 10 . 53 . 10 . 53 7 (16) 1200 244 20 118 10 1046 87 MP (29) 1059 402 38 162 15 1250 118 MP (42) 2165 69 3 128 6 3745 173 MP a Titres of CAM adsorbed antisera reacted with Vac virus antigen. b Titres of Vac virus adsorbed antisera reacted with antigen as indicated. c Percentage residual antibody activity = Vac virus adsorbed titre CAM adsorbed titre x 100 d Interpretive identification. e Days after primary immunization. Monkeys, Rh-1-76 and Rh-2-76, were immunized on days 0, 7, 17, and 24 with Copenhagen MMP soluble antigen. They were challenged with infectious MMP virus at 33, 39, and 66 days. Monkey Rh-3-76 was immunized with infectious MMP virus (1055 and 1074 PoFU) at 0 and 6 days, respectively, and boosted with infectious MMP virus (107.4 PoFU) at 33 days. the high reactivity with MMP viral antigen was characteristic of MMP antiserum specificity as estab- lished in a previous study (3). Thirty-three days after the primary inoculation, the monkey was given another inoculation with infectious MMP virus and 9 days later (42 days after primary inoculation) the CAM-adsorbed titre had increased approximately twofold. At this time, after adsorption with Vac virus, the residual antibodies reactive with Vac and Var viruses were low (<lO1Vo), and the anti- body activity remaining for MMP virus was very high. Specificity of sequential antibodies resulting from immunization with soluble MMP viral antigen Two rhesus monkeys (Rh-1-76 and Rh-2-76) were injected with soluble Copenhagen MMP antigen and received booster injections 7, 17, and 24 days after the primary inoculation. At 33, 39, and 66 days after the primary injection the animals were injected with infec- tious Utrecht MMP virus and were bled at least once per week for up to 21 months. As with the animal inoculated with infectious virus, the early sera from these animals (Fig. 3) had little or no poxvirus anti- body until 9 days after inoculation. Nine days after inoculation, antibodies were detected in sera from both animals. In contrast to the animals inoculated with either of the infectious poxviruses, the monkeys immunized with soluble antigen (Rh-1-76 and Rh-2-76) did not attain maximum titres until 21-28 days after the primary immunization (Fig. 3) and produced lower titres than animals that received infec- tious virus. Upon subsequent multiple inoculations of the monkeys with infectious virus, the titres increased precipitously, equalling the high titres attained in the animals inoculated with infectious Ch-9 virus. 3000 2000 1000 800 600 I 500/ 400 / a) 300 Al < 200 olOO 1\ 20 2oloo~~~~~~~~~~~ 410 0 10 20 30 40 50 60 70 80 90 100 110 Days After Primary Immunization Fig. 4. Sequential radioimmunoassay antibody titres to pox- viruses after inoculation with soluble Copenhagen monkey- pox viral antigen on days 0, 7, 17, and 24 and with infectious monkeypox virus on days 33, 39, and 66 (Rh-1-76). Sera adsorbed with uninfected chorioallantoic membranes and reacted with vaccinia virus 44; adsorbed with vaccinia virus and reacted with vaccinia virus *- ; adsorbed with vaccinia virus and reacted with variola virus - and adsorbed with vaccinia virus and reacted with Utrecht monkeypox virus v-v. H. H. WALLS ET AL. Marked increases in titre were observed 6-9 days after each booster injection (Table 2). Sequentially collected sera from animals inoculated with soluble MMP antigen were appropriately ad- sorbed, and residual activity was tested against Vac, Var, and MMP antigens. The RIA adsorption titres of a representative animal (Rh-1-76) are shown in Fig. 4. The titre of the earliest CAM-adsorbed serum (17 days) tested against Vac virus was low, and its reactivity with Vac virus in RIA was reduced only slightly by adsorption with Vac virus. On the other hand, adsorption with Vac virus virtually eliminated anti-Var and anti-MMP activity. Thus, the pattern of activities of the earliest serum, of relatively low activity (< 100), was not characteristic of hyper- immune anti-MMP sera and suggested lack of speci- ficity. Sera obtained 21 and 28 days after inoculation, unlike the 17-day serum, had higher titres, and after a single Vac virus adsorption the residual antibodies were reactive with both homologous MMP virus anti- gen and with Vac virus. The homologous MMP titres after Vac virus adsorption for the 21- and 28-day sera were 10407o and 150%o relative to the CAM-adsorbed Vac virus titres, respectively. The Var virus titres after Vac virus adsorption were lower (18% and 29%7o, respectively) than theMMP or Vac virus titres. In sera collected later than 28 days after inoculation, the Vac virus-adsorbed homologous titres (MMP virus) re- mained high, whereas the heterologous titres (Vac and Var viruses) decreased with values ranging from 28%o to less than 10%o relative to the CAM-adsorbed Vac virus titres. Thus, through 75 days, the reaction of adsorbed serum with homologous antigen (MMP virus) remained high, and the reaction with hetero- logous Vac and Var viruses declined. Therefore, the later sera were readily identifiable as anti-MMP sera. Sera from the other animal (Rh-2-76) inoculated with soluble MMP antigen showed adsorption pat- terns similar to those of sera from the first animal. The earliest serum tested (21 days) had a low titre and moderately high residual antibody reactive with Vac virus. Nevertheless, the percentage of residual anti- body reactive with Vac virus was lower than that of the previous animal, and reactivity was so high with MMP virus that it was easily distinguished as an MMP anti- serum. All subsequent adsorptions readily permitted identification of the sera as MMP antisera. These results suggest that specific antibodies for monkeypox virus can be elicited by soluble monkey- pox virus antigen as well as by infectious monkeypox virus (Table 2). This specificity was evident at 21-28 days after inoculation with soluble monkeypox viral antigen, which was before the animals were challenged with infectious monkeypox virus. Furthermore, the specificity was comparable to that observed in the 16-day serum obtained from the animal inoculated with only infectious monkeypox virus. For most of the Vac virus-adsorbed MMP sera, the residual antibody reactive with MMP virus exceeded 100%o. This is explained by the fact that the specific MMP viral antibody titres of Vac virus-adsorbed sera are compared with Vac virus titres of CAM-adsorbed sera. Similar observations were reported in a previous investigation of the poxvirus specificities of rabbit hyperimmune sera (3). DISCUSSION Antibodies to variola, vaccinia, and monkeypox viruses are cross-reactive in most serological pro- cedures because of antibodies arising to common anti- genic determinants (10, 11). However, hyperimmune antisera specific for monkeypox, variola, and vaccinia viruses can be prepared by adsorption with appro- priate antigens (10, 12). Because of tedious method- ology, these procedures are not applicable for sero- logical characterization of patients' sera. The RIA procedure, which requires no exhaustive adsorption and centrifugation steps (3), offers a method for studying the specificity of antibodies that appear sequentially after natural infection or vaccination. Therefore, we used the adsorption RIA procedure to assess the specificity ofantibodies appearing with time after animals were immunized with poxviruses. Our results showed that after adsorption, the residual antibodies in sera collected earlier than 21-27 days after poxvirus infection were often more difficult to identify than those in sera collected later. Regard- less of the infectious virus (chimp-9 or monkeypox) used for immunization, the temporal changes of the adsorbed antisera were similar. Furthermore, multiple injections with infectious virus or soluble antigen did not alter the pattern of increased specificity with time. The mechanism of temporal changes in specificity as measured by adsorption studies is not understood. One possible explanation is the heterogeneity of the early antibody population. Several investigators have shown that in animals immunized with haptens, proteins, or viruses antibody reactivity varies with time after injection (13-17). Generally, low-affinity antibodies appear early in the immune response, and high-affinity antibodies appear later (13, 14). Cross- reactivity of the antibodies is reported to increase simultaneously with the appearance of antibodies of high affinity (14). If low-affinity antibodies are the dominant species occurring early in the immune re- sponse, as Webster (13, 14) suggested, antibodies reacting with the adsorption antigen may enter into a state of equilibrium with the antigen affixed to the microtitre plates. If high-affinity antibodies pre- dominate late in the immune response (13), the equi- 136 ANTIBODY RESPONSE TO POXVIRUS ANTIGENS 137 librium would favour the formation of an antigen- antibody complex; and the adsorbing antigen should more consistently remove the cross-reactive anti- bodies. Antibodies of different immunoglobulin classes may also contribute to the different adsorption pat- terns. Although IgM antibodies in early sera may bind with the adsorbent antigen, unreacted antigen- binding sites may still be available for further inter- action with the antigen affixed to the microtitration plates in the RIA because of the pentameric structure of the IgM molecule. Under our test conditions dif- ferent proportions of IgM antibodies to poxviruses in early sera could contribute to the variable specificity. The effects of each of these factors could modulate the binding reaction and contribute to the apparent non-specificity of early antibodies as determined in the adsorption studies. Regardless of the cause of the variable specificity of early serum specimens, our observations indicate that the poxvirus strain speci- ficity is more consistent in sera obtained later than 30 days after infection or immunization. Much work remains to be done before the RIA adsorption pro- cedure can be applied to human sera, but these results suggest that interpretation of the poxvirus antibody specificity will require an accurate history defining the time of onset of infection. ACKNOWLEDGEMENTS The authors thank David E. Wells and Brian P. Holloway for excellent technical assistance in this study. RtSUMt tTUDE PAR TITRAGE RADIO-IMMUNOLOGIQUE DE LA SPECIFICITE D'ANTISERUMS PRELEVES SUCCESSIVEMENT A L'EGARD DES VIRUS DE LA VARIOLE ET DU MONKEYPOX La specificite des antiserums, temoignant de l'apparition et du developpement de la reponse en anticorps circulants dirig6e contre les antigenes des divers poxvirus, a ete ana- lysee chez des singes a qui avaient ete inocules le virus white- pox chimp-9 ou le virus monkeypox (souche d'Utrecht). Les scrums ont ete recueillis chez 3 singes verts africains pour le premier virus et I singe rhesus pour le second; 2 singes rhesus a qui avait ete injecte un antigene soluble du virus monkey- pox (Copenhagen) ont fait l'objet d'un prelevement analogue. Les echantillons de strum recueillis successive- ment sur chaque animal ont ete adsorbts avec membrane de chorio-allantoide de poulet non infectte (CAM) ou avec CAM infect&e par le virus de la vaccine. Les serums adsorbes ont ete soumis A des epreuves radio-immunologiques pour verifier la sp6cificit6 des anticorps residuels A 1'egard des virus de la vaccine, de la variole et de l'infection A monkey- pox. La specificite des serums adsorbes aux differentes etapes presumees du d6veloppement de la r6ponse immuni- taire a augmente avec le temps ecoule entre l'inoculation et le pr6levement. D'une maniere generale, l'identification des anticorps dans les serums recueillis avant un delai de 21-27 jours apres l'inoculation et adsorbes n'a pas e possible, alors qu'elle a et6 faite sans equivoque dans les serums preleves ulterieurement. REFERENCES 1. EsposITO, J. J. ET AL. The virion and soluble antigen proteins of variola, monkeypox, and vaccinia viruses. Journal ofmedical virology, 1: 95-110 (1977). 2. ESPOSITO, J. J. ET AL. Orthopoxvirus DNA: Strain differentiation by electrophoresis of restriction endo- nuclease fragmented virion DNA. Virology, 89: 53-66 (1978). 3. HUTCHINSON, H. D. ET AL. Differentiation of variola, monkeypox, and vaccinia antisera by radioimmuno- assay. Bulletin of the World Health Organization, 55: 613-623 (1977). 4. LOURIE, B. ET AL. Human infection with monkeypox virus: laboratory investigation of six cases in West Africa. Bulletin ofthe World Health Organization, 46: 633-639 (1972). 5. PETERS, J. C. Eine "Monkeypox" Enzootic im Affen- haus des Tiergartens "Blijdorp." Die Kleintier-Praxis, 11(3): 65-70 (1966). 6. VON MAGNUS, P. ET AL. A pox-like disease in cyno- molgus monkeys. Acta pathologica et microbiologica scandinavica, 46: 156-177 (1959). 7. MARENNIKOVA, S. S. ET AL. Poxviruses isolated from clinically ill and asymptomatically infected monkeys and a chimpanzee. Bulletin of the World Health Organization, 46: 613-620 (1972). 8. OLSEN, R. G. ET AL. Preparation and evaluation of a non-infectious monkeypox virus vaccine. Journal of clinical microbiology, 6: 50-54 (1977). 138 H. H. WALLS ET AL. 9. HUTCHINSON, H. D. & ZIEGLER, D. W. Criteria for preparing, evaluating, and standardizing iodinated globulins for radioimmunoassay procedures. Applied microbiology, 28: 935-942 (1974). 10. EsposITO, J. J. ET AL. Serologic relatedness of monkey- pox, variola, and vaccinia viruses. Journal of medical virology, 1: 35-47 (1977). 11. GISPEN, R. ET AL. Monkeypox-specific antibodies in humans and simian sera from Central Africa. Bulletin of the World Health Organization, 53: 355-360 (1976). 12. GISPEN, R. & BRAND-SAATHOF, B. Three specific anti- gens produced in vaccinia, variola, and monkeypox infections. Journal of infectious diseases, 129: 289-295 (1974). 13. WEBSTER, R. G. The immune response to influenza virus. II. Effect of the route and schedule of vaccination on the quantity and avidity of antibodies. Immunology, 14: 29-37 (1968). 14. WEBSTER, R. G. The immune response to influenza virus. III. Changes in the avidity and specificity of early IgM and IgG antibodies. Immunology, 14: 39-52 (1968). 15. KIM, Y. T. & SISKIND, G. W. Studies on the control of antibody synthesis. VI. Effect of antigen dose and time after immunization on antibody affinity and hetero- geneity in the mouse. Clinical and experimental immu- nology, 17: 329-338 (1974). 16. URBAIN, J. ET AL. Increase and decrease in binding affinity of antibodies during the immune response. Immunochemistry, 9: 121-136 (1972). 17. WERBLIN, T. P. ET AL. Studies on the control of anti- body synthesis. III. Changes in heterogeneity of anti- body affinity during the course of the immune response. Immunology, 24: 477-492 (1973).