THE JOURNAL OF INFECTIOUS DISEASES • VOL. 131, NO.1· © 1975 by the University of Chicago. All rights reserved.
JANUARY 1975
NOTES Counterimmunoelectrophoresis with Influenza Antigens. I. Use of Avian Plague Virus to Detect Type-Specific Antibodies to Influenza A in Human Sera J. c. Russi-Cahill, M. c. Mogdasy, R. E. Somma-Moreira, and M. H. de Peluffo
From the Department of Bacteriology and Virology, Institute of Hygiene, School of Medicine, Montevideo, Uruguay
Avian plague virus was used as antigen in a counterimmunoelectrophoresis technique. This virus was selected because it detects only type-specific influenza A antibodies in human sera, avoiding the possible interference of other antigens with anodic migration. The results with reference sera, as well as the correlation of positive sera found by counterimmunoelectrophoresis and complement fixation with the proposed antigen, in the absence of other types of antibodies to fowl plague virus antigen, support the conclusion that the counterimmunoelectrophoresis technique reveals type-specific antibodies. The test is more sensitive than immunodiffusion but less sensitive than complement fixation. Its sensitivity, simplicity, and rapidity make it suitable for serologic surveys of human influenza A.
Counterimmunoelectrophoresis (CIE ) has been successfully used in various fields of biology. Berlin and Pirojiboot reported the application of the technique for diagnosis of influenza infection, with human strains as antigens [1]. The present work describes a modification of their technique for evaluation of the possible detection of exclusively type-specific antibodies to influenza in human sera, by the introduction of avian plague virus (FPV) as antigen.
times and centrifuged at 400 g for 10 min. The supernatant was dialyzed against distilled water and concentrated with polyethylene glycol about five times, yielding an antigen with a CF titer of 1:256. As a control, a similar membrane preparation was obtained with uninoculated eggs. A ribonucleoprotein antigen of influenza type A prepared at the Center for Disease Control (CDC) was also used in CF tests. Sera. Samples of human serum were obtained from 100 adult donors. Reference antisera to ribonucleoprotein of influenza A and B prepared in guinea pigs according to Lief et al. [2] were furnished by the CDC. Pooled normal guinea pig sera were used as a control. CF was performed according to the standards set by the CDC [3]. Human sera were tested by this technique with soluble (ribonucleoprotein) A2/Hong Kong/8/68 (H3N2) antigen. For tests of neuraminidase inhibition, the method of Webster and Pereira was used [4]. The HAl test was performed according to standards set by the CDC [5], with formalinized 1% chicken red blood cells, and an allantoid fluid antigen prepared from embryonated hen's eggs inoculated with FPV. Counterimmunoelectrophoresis was performed on glass plates (5 X 25 em) covered with a 2mm layer of 0.9% Noble agar (Difco, Detroit, Mich.) in veronal buffer (pH 8.6; ionic strength
Materials and Methods
A ntigen. Chorioallantoic membranes of embryonated hen's eggs (11 days old) were inoculated with 100 50% infectious doses of A/FPV/ Dutch/27 (Hav1Neq1). The eggs were incubated for 36 hr at 37 C, and the membranes were harvested and ground in a Waring Blender (20,000 rpm). The membranes were freeze-thawed three
Received for publication January 21, 1974, and in revised form June 21, 1974. We thank Professor I. Conti-Diaz for helpful advice in counterirnmunoelectrophoresis technique, Dr. V. Vila for providing the sera, and Mrs. Ana M. Serpa for technical assistance. Please address requests for reprints to Dr. J. C. RussiCahill, Instituto de Higiene, A. Navarro 3051, Montevideo, Uruguay.
64
65
CIE with Influenza Antigens
0.06). Five to six rows of wells (3 mm in diameter) were cut in the agar leaving 6 mm between wells. Under these conditions it was possible to perform 80-96 reactions per plate. The electrophoresis was performed during 1 hr at 5 rnA (approx. 3V/cm), and the results were read with oblique light against a dark background. For double gel-diffusion tests, 0.9% ion agar (Oxoid Ltd., London, England) in 0.15 M NaCI was used. One central and six equally spaced peripheral wells (3 mm in diameter) were cut in a 2-mm agar layer. The distance between the central and peripheral wells was 3 mm. The plates were incubated at 24 hr in a humid chamber at 24 C. Results
The results obtained with reference antisera in CIE are shown in table 1. Anodic precipitin lines were formed only with antisera to ribonucleoprotein of type A, and with the FPV antigen prepared in this laboratory; no reaction was observed between the same antigen and other sera. An antigen prepared with uninoculated chorioallantoic membranes did not react with any of the sera included in the reaction. Of 100 human sera tested in CIE with FPV antigen, 54% yielded positive reactions. Meanwhile, the same sera did not give any precipitin line when tested with the normal chorioallantoic membrane antigen. CIE-positive sera were processed by HAl and neuraminidase inhibition Table 1. Counterimmunoelectrophoresis with reference antisera. Antigens Sera*
FPVt
Antisera to RNP of influenza A (CDC, lot 6) Antisera to RNP of influenza A (CD~ ~t 7) Antisera to RNP of influenza B (CDC, lot 2) Nonnal guinea pig sera
=
*
CIE with FPV appears to be adequate for the detection of type-specific antibodies to influenza A. This is supported by the results obtained with the reference antisera to ribonucleoprotein of influenza A, as well as by the correlation between CIE and CF. In addition, no antibodies to other viral or cellular components were present in the human sera. Nevertheless, the contribution of ribonucleoprotein and the internal protein to the precipitin lines have not yet been established. Reactions of the ribonucleoprotein and the internal protein might be similar. Staining of the precipitate with acridine orange did not furnish further information. Table 2. Comparative sensitivity of counterimmunoelectrophoresis (CIE) , gel diffusion (G D) , and CF techniques. Sera
CIE
1 2
2
3 4 5
6 7 8 9 10
+
11 12
=
=
Discussion
+
NOTE. ( +) anodic precipitin line; (-) no precipitin line. ribonucleoprotein; CDC Center for Dis* RNP ease Control. t FPV avian plague virus-infected membrane antigen. CAM = normal chorioallantoic membrane antigen.
= =
against FPV, with negative results; so the possibility that the precipitin lines described might be due to those viral components was discarded. The sensitivities of gel diffusion, CF, and CIE techniques were compared in a selected group of sera (table 2). These results indicate that CIE is more sensitive than gel diffusion, although less than CF, because more positives were found by this last technique, mainly at higher dilutions.
13 14 15 16 17
18 19 NOTE.
GD
CF 40 10
+ + + + + +
~ ~
W W W W
2
40 10 10 10 20 10 20 20 160 80 40
+ +
+ + 8 8 2
2 2
Results are expressed as reciprocal titers. (+) undiluted; (-) no precipitin line.
= serum positive when tested
=
66
Schild demonstrated that ribonucleoprotein and the internal protein are both type specific [6], but it has not been determined whether the antibodies to the internal protein are as short-lived as those to ribonucleoprotein. Therefore, the time course of these precipitating antibodies shown by CIE must be established before the ability to detect recent infection by influenza A can be attributed to the technique. Studies on decay of these antibodies are now in progress; preliminary data indicate that the antibodies revealed by the CIE technique correlated with the curve of antibodies detected by CF with ribonucleoprotein antigen. Comparative analysis of the results by the described serologic techniques indicates that crn has a reasonable sensitivity, although CF is better. In addition, CIE is a technique that combines specificity, speed, and simplicity, so it appears adequate for large serologic surveys, as well as for epidemiologic surveillance of influenza A.
Russi-Cahill et al.
References
1. Berlin, B. S., Pirojiboot, N. A rapid method for demonstration of precipitating antibody against influenza virus by counterimmunoelectrophoresis. J. Infect. Dis. 126:345-347, 1972. 2. Lief, F. S., Fabiyi, A., Henle, W. Antigenic analysis of influenza viruses by complement fixation. I. The production of antibodies to the soluble antigen in guinea pigs. J. Immunol. 80:53-65, 1958. 3. Casey, H. L. Standardized diagnostic complement fixation method and adaptation to microtest. Washington, D.C., U.S. Government Printing Office. 1965. 34 p. 4. Webster, R. G., Pereira, H. G. A common surface antigen in influenza viruses from human and avian sources. J. Gen. Virol, 3:201-208, 1968. 5. Methods for the Laboratory Diagnosis of Respiratory Viruses. U.S. Department of Health, Education and Welfare. Atlanta, Georgia, Communicable Disease Center. 53 p. 6. Schild, G. C. Evidence for a new type-specific structural antigen of the influenza virus particle. J. Gen. Virol. 15:99-103, 1972.
JANUARY 1975
NOTES Counterimmunoelectrophoresis with Influenza Antigens. I. Use of Avian Plague Virus to Detect Type-Specific Antibodies to Influenza A in Human Sera J. c. Russi-Cahill, M. c. Mogdasy, R. E. Somma-Moreira, and M. H. de Peluffo
From the Department of Bacteriology and Virology, Institute of Hygiene, School of Medicine, Montevideo, Uruguay
Avian plague virus was used as antigen in a counterimmunoelectrophoresis technique. This virus was selected because it detects only type-specific influenza A antibodies in human sera, avoiding the possible interference of other antigens with anodic migration. The results with reference sera, as well as the correlation of positive sera found by counterimmunoelectrophoresis and complement fixation with the proposed antigen, in the absence of other types of antibodies to fowl plague virus antigen, support the conclusion that the counterimmunoelectrophoresis technique reveals type-specific antibodies. The test is more sensitive than immunodiffusion but less sensitive than complement fixation. Its sensitivity, simplicity, and rapidity make it suitable for serologic surveys of human influenza A.
Counterimmunoelectrophoresis (CIE ) has been successfully used in various fields of biology. Berlin and Pirojiboot reported the application of the technique for diagnosis of influenza infection, with human strains as antigens [1]. The present work describes a modification of their technique for evaluation of the possible detection of exclusively type-specific antibodies to influenza in human sera, by the introduction of avian plague virus (FPV) as antigen.
times and centrifuged at 400 g for 10 min. The supernatant was dialyzed against distilled water and concentrated with polyethylene glycol about five times, yielding an antigen with a CF titer of 1:256. As a control, a similar membrane preparation was obtained with uninoculated eggs. A ribonucleoprotein antigen of influenza type A prepared at the Center for Disease Control (CDC) was also used in CF tests. Sera. Samples of human serum were obtained from 100 adult donors. Reference antisera to ribonucleoprotein of influenza A and B prepared in guinea pigs according to Lief et al. [2] were furnished by the CDC. Pooled normal guinea pig sera were used as a control. CF was performed according to the standards set by the CDC [3]. Human sera were tested by this technique with soluble (ribonucleoprotein) A2/Hong Kong/8/68 (H3N2) antigen. For tests of neuraminidase inhibition, the method of Webster and Pereira was used [4]. The HAl test was performed according to standards set by the CDC [5], with formalinized 1% chicken red blood cells, and an allantoid fluid antigen prepared from embryonated hen's eggs inoculated with FPV. Counterimmunoelectrophoresis was performed on glass plates (5 X 25 em) covered with a 2mm layer of 0.9% Noble agar (Difco, Detroit, Mich.) in veronal buffer (pH 8.6; ionic strength
Materials and Methods
A ntigen. Chorioallantoic membranes of embryonated hen's eggs (11 days old) were inoculated with 100 50% infectious doses of A/FPV/ Dutch/27 (Hav1Neq1). The eggs were incubated for 36 hr at 37 C, and the membranes were harvested and ground in a Waring Blender (20,000 rpm). The membranes were freeze-thawed three
Received for publication January 21, 1974, and in revised form June 21, 1974. We thank Professor I. Conti-Diaz for helpful advice in counterirnmunoelectrophoresis technique, Dr. V. Vila for providing the sera, and Mrs. Ana M. Serpa for technical assistance. Please address requests for reprints to Dr. J. C. RussiCahill, Instituto de Higiene, A. Navarro 3051, Montevideo, Uruguay.
64
65
CIE with Influenza Antigens
0.06). Five to six rows of wells (3 mm in diameter) were cut in the agar leaving 6 mm between wells. Under these conditions it was possible to perform 80-96 reactions per plate. The electrophoresis was performed during 1 hr at 5 rnA (approx. 3V/cm), and the results were read with oblique light against a dark background. For double gel-diffusion tests, 0.9% ion agar (Oxoid Ltd., London, England) in 0.15 M NaCI was used. One central and six equally spaced peripheral wells (3 mm in diameter) were cut in a 2-mm agar layer. The distance between the central and peripheral wells was 3 mm. The plates were incubated at 24 hr in a humid chamber at 24 C. Results
The results obtained with reference antisera in CIE are shown in table 1. Anodic precipitin lines were formed only with antisera to ribonucleoprotein of type A, and with the FPV antigen prepared in this laboratory; no reaction was observed between the same antigen and other sera. An antigen prepared with uninoculated chorioallantoic membranes did not react with any of the sera included in the reaction. Of 100 human sera tested in CIE with FPV antigen, 54% yielded positive reactions. Meanwhile, the same sera did not give any precipitin line when tested with the normal chorioallantoic membrane antigen. CIE-positive sera were processed by HAl and neuraminidase inhibition Table 1. Counterimmunoelectrophoresis with reference antisera. Antigens Sera*
FPVt
Antisera to RNP of influenza A (CDC, lot 6) Antisera to RNP of influenza A (CD~ ~t 7) Antisera to RNP of influenza B (CDC, lot 2) Nonnal guinea pig sera
=
*
CIE with FPV appears to be adequate for the detection of type-specific antibodies to influenza A. This is supported by the results obtained with the reference antisera to ribonucleoprotein of influenza A, as well as by the correlation between CIE and CF. In addition, no antibodies to other viral or cellular components were present in the human sera. Nevertheless, the contribution of ribonucleoprotein and the internal protein to the precipitin lines have not yet been established. Reactions of the ribonucleoprotein and the internal protein might be similar. Staining of the precipitate with acridine orange did not furnish further information. Table 2. Comparative sensitivity of counterimmunoelectrophoresis (CIE) , gel diffusion (G D) , and CF techniques. Sera
CIE
1 2
2
3 4 5
6 7 8 9 10
+
11 12
=
=
Discussion
+
NOTE. ( +) anodic precipitin line; (-) no precipitin line. ribonucleoprotein; CDC Center for Dis* RNP ease Control. t FPV avian plague virus-infected membrane antigen. CAM = normal chorioallantoic membrane antigen.
= =
against FPV, with negative results; so the possibility that the precipitin lines described might be due to those viral components was discarded. The sensitivities of gel diffusion, CF, and CIE techniques were compared in a selected group of sera (table 2). These results indicate that CIE is more sensitive than gel diffusion, although less than CF, because more positives were found by this last technique, mainly at higher dilutions.
13 14 15 16 17
18 19 NOTE.
GD
CF 40 10
+ + + + + +
~ ~
W W W W
2
40 10 10 10 20 10 20 20 160 80 40
+ +
+ + 8 8 2
2 2
Results are expressed as reciprocal titers. (+) undiluted; (-) no precipitin line.
= serum positive when tested
=
66
Schild demonstrated that ribonucleoprotein and the internal protein are both type specific [6], but it has not been determined whether the antibodies to the internal protein are as short-lived as those to ribonucleoprotein. Therefore, the time course of these precipitating antibodies shown by CIE must be established before the ability to detect recent infection by influenza A can be attributed to the technique. Studies on decay of these antibodies are now in progress; preliminary data indicate that the antibodies revealed by the CIE technique correlated with the curve of antibodies detected by CF with ribonucleoprotein antigen. Comparative analysis of the results by the described serologic techniques indicates that crn has a reasonable sensitivity, although CF is better. In addition, CIE is a technique that combines specificity, speed, and simplicity, so it appears adequate for large serologic surveys, as well as for epidemiologic surveillance of influenza A.
Russi-Cahill et al.
References
1. Berlin, B. S., Pirojiboot, N. A rapid method for demonstration of precipitating antibody against influenza virus by counterimmunoelectrophoresis. J. Infect. Dis. 126:345-347, 1972. 2. Lief, F. S., Fabiyi, A., Henle, W. Antigenic analysis of influenza viruses by complement fixation. I. The production of antibodies to the soluble antigen in guinea pigs. J. Immunol. 80:53-65, 1958. 3. Casey, H. L. Standardized diagnostic complement fixation method and adaptation to microtest. Washington, D.C., U.S. Government Printing Office. 1965. 34 p. 4. Webster, R. G., Pereira, H. G. A common surface antigen in influenza viruses from human and avian sources. J. Gen. Virol, 3:201-208, 1968. 5. Methods for the Laboratory Diagnosis of Respiratory Viruses. U.S. Department of Health, Education and Welfare. Atlanta, Georgia, Communicable Disease Center. 53 p. 6. Schild, G. C. Evidence for a new type-specific structural antigen of the influenza virus particle. J. Gen. Virol. 15:99-103, 1972.
