Zbl. Bakt. 277, 188-192 (1992) © Gustav Fischer Verlag, StuttgartlNew York
Indirect Haemagglutination Test for the Serodiagnosis of Brucellosis using Stabilised Sheep Red Blood Cells
GANGA PRASAD RAI, GAURI SHANKER AGARWAL, and KODUMUDI SUBBARA YAN VENKATESWARAN Division of Microbiology, Defence R&D Establishment, Gwalior-474002, India
Received October 5, 1991 . Revision received January 7,1992 . Accepted February 12, 1992
Summary Sheep red blood cells (SRBCs) stabilised with formaldehyde, glutaraldehyde, pyruvic aldehyde and double aldehyde were evaluated, in the indirect haemagglutination (IHA) test, for their suitability for the serodiagnosis of brucellosis in bovines. Serum samples from 23 standard tube agglutination test (STAT)-positive and 12 clinically suspected but STATnegative cows were titrated by the IHA test. Double aldehyde and glutaraldehyde-stabilised SRBCs were found to be more sensitive in the IHA than the formaldehyde or pyruvic aldehyde-stabilised SRBCs. Double aldehyde and glutaraldehyde treatments enabled the IHA to clearly differentiate between the normal and the diseased cows. Double aldehydestabilised cells have an edge over the glutaraldehyde-stabilised SRBCs because the former treatment makes the SRBCs readily available for antigen coating without tannic acid activation.
Zusammenfassung Mit Formaldehyd, Glutaraldehyd, Pyruvataldehyd und Doppelaldehyd stabilisierte Schaferythrozyten wurden im indirekten Hamagglutinationstest auf ihre Eignung zur Brucellose-Serodiagnostik bei Rindern gepruft, Serumproben von 23 Kiihen, die im Standard-Rohrchen-Agglutinationstest positiv waren, und von 12 klinisch verdachtigen, aber in diesem Test negativen Kuhen wurden im indirekten Hamagglutinationstest untersucht. Es wurde festgestellt, daf im indirekten Harnagglutintionstest mittels Doppelaldehyd und Glutaraldehyd stabiliserte empfindlicher als mit Formaldeyhd oder Pyruvataldehyd stabilisierte Schaferythrozyten waren. Behandlung mit Doppelaldehyd und Glutaraldehyd errnoglichte eine klare Differenzierung von gesunden und erkrankten Kiihen im indirekten Hamagglutinationstest. Mit Doppelaldehyd stabilisierte Zellen sind vorteilhafter, weil die erstgenannte Behandlung die Schaferythrozyten ohne Gerbsaureaktivierung fur eine Antigenbeschichtung vorbereitet.
Serodiagnosis of Brucellosis by IHA
189
Introduction Brucellosi s, a bacterial disea se of farm cattle and human beings is prevalent all over the world. This disease is diagnosed either by isolation of the cau sative organism or by serological techniques. Isolation of Bruc ella abortus, the microbe responsible for brucellosis, from clinical materials is no doubt the method of choice but it takes a long time, even two to three week s, to establish the disease. M any serological tests like standard tube agglutination test [STAT] (1 ), indirect haemagglutinat ion test [IHA] (5), complement fixation test [CFT ] (2), immunofluorescence test (10 ) and enzyme-linked immunoso rbent assay [ELISA] (1) have been emplo yed to detect ant ibodies produced in response to Brucella abo rtus infectio n of cattle. A number of assays ut ilising sheep red blood cells [SRBCs] as markers hav e been described (4). Red blood cells stabilised either by formaldehyde, glutaraldehyde or pyruvic aldeh yde are routinely used in the IHA test (8). The present study was car ried out to evaluate the use of SRBCs stabil ised by treatment with formaldehyde, gluta raldehyde, pyruvic aldehyde and double aldehyde (SRBCs stabilised by a combinati on of pyruvic aldehyde and glut ar aldehyde separated by a tanning procedure) in th e IHA test to find out the levels of antibody in the sera of cow s infected with B. ab ortus.
Materials and Methods Antigen. B. abortus 5-99 obtained from the Indian Veterinary Research Institute, Izatnagar, India was grown on Brucella agar medium (Difco, USA) and sonicated antigen from this culture was prepared (1). The protein content in the soluble sonicated extr act was measured using the method described by Lowry et a1. (6). The aliquots were stored frozen at - 20 °C till further use. Sera. Twenty-three standard tube agglutination test positive (STAT-positive) and twelve clinically suspected but standard tube agglutination negative (STAT-negative) serum samples were collected from the Central Military Veterinary Laboratory, Meerut, India. Serum samples from twenty normal healthy animals with neither a previous history of brucellosis nor vaccination against B. abortus in the recent past were included in the study as normal controls. Before perform ing the IHA test, sera were inactivated at 56 °C for 30 min and heterophilie agglutinins were removed from each serum sample by adsorption with unsensitised SRBCs and also with whole cells of Brucella melitensis, Brucella suis, Escherichia coli, Vibrio cholerae and Yersinia enterocolitica 0 : 9. Stabilisation of sheep red blood cells. Sheep blood was collected into Alsever's solution and cells were washed thrice in phosphate-buffered saline (pH 7.4). Cells were treated with different stabilising agents (8) such as formaldehyde, glutarald ehyde, pyruvic aldehyde and double.aldehyde. IHA test. SRBCs stabilised with formaldehyde, glutaraldehyde or pyruvic aldehyde were treated with tannic acid [1 in 20 000 (w/v) dilution] before being sensitised with antigen. With double aldehyde-stabilised cells, the tanning step was omitted because the cells had been tanned during stabilisation itself. The optimal sensitising dose of antigen was determined for each type of sta bilised cells by checker-board titration in the IHA test (9). The titre of the serum was determined by two-fold serial dilution of serum samples diluted ten times. Statistical analysis. The median anti body titre was calculated from the individual titre values by IHA using four different stabilising agents. Dat a were analysed by the Wilcoxon Rank Sum Test to compar e the median antibody titre in different treatment s of SRBCs for sta bilisation. Two-tailed P values of < 0.05 were considered significant.
190
G. P. Rai, G. S.Agarwal, and K.S.Venkateswaran
Results The median antibody titres (MATs) of the STAT-positive, STAT-negative and normal controls are given in Table 1. The MATs of both STAT-positive and STATnegative sera were significantly (P < 0.05) higher than the normal control animals in all the four different stabilisation treatments of SRBCs. The MATs of the STATpositive animals were two to three times higher than the median ± 2 standard deviation of the control animals. Distributions of the individual titres are summarised in Table 2. It is obvious from this table that a titre value of:s 80 was the titre seen in the normal controls by IHA and a titre of 1 in 160 could be considered as doubtful or a borderline case whereas titres of :::::: 320 clearly indicated the presence of diagnostic antibody levels in brucellosis by IHA. Double aldehyde and glutaraldehyde-stabilised SRBCs could clearly differentiate between the normal and the diseased (STAT-positive as well as animals suspected by clinical symptoms) animals. Pyruvic aldehyde and
Table 1. Median Antibody Titres" of the sera of cows by IHA using different stabilising agents Stabilising Agent
Normal Control STAT+ (n = 20) (n = 23)
STAT-
Formaldehyde Glutaraldehyde Pyruvic aldehyde Double aldehyde
2.0±0.05 2.5 ±0.60 1.5±0.76 3.0 ± 0.44
4.0±0.57 4.5 ±0.98 4.0±0.72 5.0± 1.11
(n = 12)
5.0±0.78 6.0± 1.03 4.0±0.72 8.0± 1.29
* Expressed as lOx Logj,
Table 2. Distribution of IHA Titres of sera of STAT-positive and doubtful (STAT-negative) cows of B. abortus infection Stabiliser for SRBC
Titre range
Normal Control
STAT+
STAT-
Formaldehyde
0-80 160 2: 320
20 0 0
0 3 20
5 7 0
Glutaraldehyde
0-80 160 :::::: 320
20 0 0
0 0 23
3 3 6
Pyruvic aldehyde
0-80 160 2: 320
20 0 0
0 12
11
4 6 2
0-80 160 :::::: 320
20 0 0
0 0 23
1 2 9
Double aldehyde
Serodiagnosis of Brucellosis by IHA
191
formaldehyde treatments of red blood cells were not sensitive enough to demarcate the uninfeeted from the infected animal and also produced a few doubtful IHA titre value s of 1 in 160.
Discussion Bovine bru cellosis caused by B. abo rtus is an economically important disease associated with abort ions and infertility. Though officially recommended conventional tests like STAT, CFT and rose bengal test are still being used for routine serodiagnosis of brucello sis, none of the se indi vidual tests can identify all the cases of bovine brucellosis. The IHA test has been found to be sensitive and specific allow ing timely diagnosis and ident ification of animals with brucellosis at an earlier time with better accuracy than any other conventional test (5). Th e major limitation with IHA test lies in the use of fresh red blood cells as indicator. Because fresh cells are fragile, pr one to haemolysis and bacterial contamination, they can be stored only for a brief period. This limitation can be overcome by stabilising the red blood cells with aldeh ydes and these cells can be used for up to eight months with out the loss of much antigen-coating efficiency (7). Sensitivity of the IHA test wa s found to depend on the meth od of stabilisation in the diagnosis of malaria (3). In the present investigation, it wa s observed th at the use of double aldehyde-sta bilised cells greatl y increased the sensitivity of the IHA test and the sta bilisation by glutar aldehyde also gave satisfacto ry result s comp ared to formaldehyde and pyru vic aldehyde sta bilisatio n. Our observat ions are in confo rmity with the other reported result s on stabilisation of SRBCs in IHA for the detection of other diseases (3, 8). Therefor e, double aldehyde-stabilised SRBCs can be emplo yed for routine serodiagnosis of bru cellosis by IHA. Acknowledgement . Authors are grateful to Brig. K. M. Rao (Retd.), former Director, Defence R &D Establishment for his interest in this study and constant encouragement. Thanks are also due to Commandant , Central Military Veterinary Research Laboratory, Meerut for providing the serum samples.
References 1. Batra, H. V., P. Chand, L. Ganju, R. Mukerjee, and j. R. Sardana: Dotenzyme linked immunosorbent assay for the detection of antibodies in bovine brucellosis. Res. Vet. Sci. 46 (1989) 143-146 2. Chand, P., j. R. Sardana, H. V. Batra, and R. S. Chauhan: Comparison of the dotimmunobinding assay wih the complement fixation test for the detection of Brucella antibodies in sheep. Vet. Microbiol. 20 (1989) 282-287 3. Farshy, D. C. and I. G. Kagan: Use of stab le sensitised cells in indirect microhaemagglutination test for malaria. Am. J. Trop. Med. Hyg. 21 (1972) 868-872 4. Gupta, R. K., C. N. Mishra, V. K. Mehtha, V. K. Gupta , L. N. Rao Bhau, G. Singh, D. Gowal, and S. N. Saxena: Indirect haemaggIutination test for the assay of antibodies against Japanese encephalitis viral antigen in human and animal sera. Ind. J. Med. Res. 91 (1990) 315-320 5. Khairov, S. G. and O. Yu. Yusupov : Experiences with indirect haemagglutinat ion test in brucellosis. Veterinariya 1 (1989) 60-62 6. Lowry, O. H ., N. j. Rosebrough, A. L. Farr, and R. j. Randall: Protein measurement with the folin phenol reagent. J. BioI. Chern. 193 (1951) 265- 275
192
G. P. Rai, G. S.Agarwal, and K. S.Venkateswaran
7. Nagarkatti, P. S., M. Nagarkatti, and K. M. Rao: Development of a kit for the assay of haemagglutination inhibition antibodies to flaviviruses using formalinised goose erythrocytes. Trans. Roy. Soc. Trop. Med. Hyg. 74 (1980) 22-25 8. Parija, S. C. and N. Ananthakrishnan: Evaluation of stabilised cells in the indirect haemagglutination test for echinococcosis. ]. Med. Microbiol. 19 (1985) 95-98 9. Rai, G. P., K. Zachariah, and S. Shriuastaua: Comparative efficacy of indirect haemagglutination test, indirect fluorescent antibody test and enzyme linked immunosorbent assay in serodiagnosis of typhoid fever. ]. Trop, Med. Hyg. 92 (1989) 431-434 10. Rodriquez, A. N., J. Pachon Diaz, R. T. Satigo, J. A. C. Contreras, P. V. Fernandez, L. L. Cortes, J. c. de la Fuente, and A. M. Marin: Indirect immunofluorescence and rose bengal test for the diagnosis of brucellosis. Rev. Clin. Esp. 175 (1984) 27-32
Dr. G. P. Rai, Division of Microbiology, Defence Rand D Establishment, Tansen Road, Gwalior-474 002, India
Indirect Haemagglutination Test for the Serodiagnosis of Brucellosis using Stabilised Sheep Red Blood Cells
GANGA PRASAD RAI, GAURI SHANKER AGARWAL, and KODUMUDI SUBBARA YAN VENKATESWARAN Division of Microbiology, Defence R&D Establishment, Gwalior-474002, India
Received October 5, 1991 . Revision received January 7,1992 . Accepted February 12, 1992
Summary Sheep red blood cells (SRBCs) stabilised with formaldehyde, glutaraldehyde, pyruvic aldehyde and double aldehyde were evaluated, in the indirect haemagglutination (IHA) test, for their suitability for the serodiagnosis of brucellosis in bovines. Serum samples from 23 standard tube agglutination test (STAT)-positive and 12 clinically suspected but STATnegative cows were titrated by the IHA test. Double aldehyde and glutaraldehyde-stabilised SRBCs were found to be more sensitive in the IHA than the formaldehyde or pyruvic aldehyde-stabilised SRBCs. Double aldehyde and glutaraldehyde treatments enabled the IHA to clearly differentiate between the normal and the diseased cows. Double aldehydestabilised cells have an edge over the glutaraldehyde-stabilised SRBCs because the former treatment makes the SRBCs readily available for antigen coating without tannic acid activation.
Zusammenfassung Mit Formaldehyd, Glutaraldehyd, Pyruvataldehyd und Doppelaldehyd stabilisierte Schaferythrozyten wurden im indirekten Hamagglutinationstest auf ihre Eignung zur Brucellose-Serodiagnostik bei Rindern gepruft, Serumproben von 23 Kiihen, die im Standard-Rohrchen-Agglutinationstest positiv waren, und von 12 klinisch verdachtigen, aber in diesem Test negativen Kuhen wurden im indirekten Hamagglutinationstest untersucht. Es wurde festgestellt, daf im indirekten Harnagglutintionstest mittels Doppelaldehyd und Glutaraldehyd stabiliserte empfindlicher als mit Formaldeyhd oder Pyruvataldehyd stabilisierte Schaferythrozyten waren. Behandlung mit Doppelaldehyd und Glutaraldehyd errnoglichte eine klare Differenzierung von gesunden und erkrankten Kiihen im indirekten Hamagglutinationstest. Mit Doppelaldehyd stabilisierte Zellen sind vorteilhafter, weil die erstgenannte Behandlung die Schaferythrozyten ohne Gerbsaureaktivierung fur eine Antigenbeschichtung vorbereitet.
Serodiagnosis of Brucellosis by IHA
189
Introduction Brucellosi s, a bacterial disea se of farm cattle and human beings is prevalent all over the world. This disease is diagnosed either by isolation of the cau sative organism or by serological techniques. Isolation of Bruc ella abortus, the microbe responsible for brucellosis, from clinical materials is no doubt the method of choice but it takes a long time, even two to three week s, to establish the disease. M any serological tests like standard tube agglutination test [STAT] (1 ), indirect haemagglutinat ion test [IHA] (5), complement fixation test [CFT ] (2), immunofluorescence test (10 ) and enzyme-linked immunoso rbent assay [ELISA] (1) have been emplo yed to detect ant ibodies produced in response to Brucella abo rtus infectio n of cattle. A number of assays ut ilising sheep red blood cells [SRBCs] as markers hav e been described (4). Red blood cells stabilised either by formaldehyde, glutaraldehyde or pyruvic aldeh yde are routinely used in the IHA test (8). The present study was car ried out to evaluate the use of SRBCs stabil ised by treatment with formaldehyde, gluta raldehyde, pyruvic aldehyde and double aldehyde (SRBCs stabilised by a combinati on of pyruvic aldehyde and glut ar aldehyde separated by a tanning procedure) in th e IHA test to find out the levels of antibody in the sera of cow s infected with B. ab ortus.
Materials and Methods Antigen. B. abortus 5-99 obtained from the Indian Veterinary Research Institute, Izatnagar, India was grown on Brucella agar medium (Difco, USA) and sonicated antigen from this culture was prepared (1). The protein content in the soluble sonicated extr act was measured using the method described by Lowry et a1. (6). The aliquots were stored frozen at - 20 °C till further use. Sera. Twenty-three standard tube agglutination test positive (STAT-positive) and twelve clinically suspected but standard tube agglutination negative (STAT-negative) serum samples were collected from the Central Military Veterinary Laboratory, Meerut, India. Serum samples from twenty normal healthy animals with neither a previous history of brucellosis nor vaccination against B. abortus in the recent past were included in the study as normal controls. Before perform ing the IHA test, sera were inactivated at 56 °C for 30 min and heterophilie agglutinins were removed from each serum sample by adsorption with unsensitised SRBCs and also with whole cells of Brucella melitensis, Brucella suis, Escherichia coli, Vibrio cholerae and Yersinia enterocolitica 0 : 9. Stabilisation of sheep red blood cells. Sheep blood was collected into Alsever's solution and cells were washed thrice in phosphate-buffered saline (pH 7.4). Cells were treated with different stabilising agents (8) such as formaldehyde, glutarald ehyde, pyruvic aldehyde and double.aldehyde. IHA test. SRBCs stabilised with formaldehyde, glutaraldehyde or pyruvic aldehyde were treated with tannic acid [1 in 20 000 (w/v) dilution] before being sensitised with antigen. With double aldehyde-stabilised cells, the tanning step was omitted because the cells had been tanned during stabilisation itself. The optimal sensitising dose of antigen was determined for each type of sta bilised cells by checker-board titration in the IHA test (9). The titre of the serum was determined by two-fold serial dilution of serum samples diluted ten times. Statistical analysis. The median anti body titre was calculated from the individual titre values by IHA using four different stabilising agents. Dat a were analysed by the Wilcoxon Rank Sum Test to compar e the median antibody titre in different treatment s of SRBCs for sta bilisation. Two-tailed P values of < 0.05 were considered significant.
190
G. P. Rai, G. S.Agarwal, and K.S.Venkateswaran
Results The median antibody titres (MATs) of the STAT-positive, STAT-negative and normal controls are given in Table 1. The MATs of both STAT-positive and STATnegative sera were significantly (P < 0.05) higher than the normal control animals in all the four different stabilisation treatments of SRBCs. The MATs of the STATpositive animals were two to three times higher than the median ± 2 standard deviation of the control animals. Distributions of the individual titres are summarised in Table 2. It is obvious from this table that a titre value of:s 80 was the titre seen in the normal controls by IHA and a titre of 1 in 160 could be considered as doubtful or a borderline case whereas titres of :::::: 320 clearly indicated the presence of diagnostic antibody levels in brucellosis by IHA. Double aldehyde and glutaraldehyde-stabilised SRBCs could clearly differentiate between the normal and the diseased (STAT-positive as well as animals suspected by clinical symptoms) animals. Pyruvic aldehyde and
Table 1. Median Antibody Titres" of the sera of cows by IHA using different stabilising agents Stabilising Agent
Normal Control STAT+ (n = 20) (n = 23)
STAT-
Formaldehyde Glutaraldehyde Pyruvic aldehyde Double aldehyde
2.0±0.05 2.5 ±0.60 1.5±0.76 3.0 ± 0.44
4.0±0.57 4.5 ±0.98 4.0±0.72 5.0± 1.11
(n = 12)
5.0±0.78 6.0± 1.03 4.0±0.72 8.0± 1.29
* Expressed as lOx Logj,
Table 2. Distribution of IHA Titres of sera of STAT-positive and doubtful (STAT-negative) cows of B. abortus infection Stabiliser for SRBC
Titre range
Normal Control
STAT+
STAT-
Formaldehyde
0-80 160 2: 320
20 0 0
0 3 20
5 7 0
Glutaraldehyde
0-80 160 :::::: 320
20 0 0
0 0 23
3 3 6
Pyruvic aldehyde
0-80 160 2: 320
20 0 0
0 12
11
4 6 2
0-80 160 :::::: 320
20 0 0
0 0 23
1 2 9
Double aldehyde
Serodiagnosis of Brucellosis by IHA
191
formaldehyde treatments of red blood cells were not sensitive enough to demarcate the uninfeeted from the infected animal and also produced a few doubtful IHA titre value s of 1 in 160.
Discussion Bovine bru cellosis caused by B. abo rtus is an economically important disease associated with abort ions and infertility. Though officially recommended conventional tests like STAT, CFT and rose bengal test are still being used for routine serodiagnosis of brucello sis, none of the se indi vidual tests can identify all the cases of bovine brucellosis. The IHA test has been found to be sensitive and specific allow ing timely diagnosis and ident ification of animals with brucellosis at an earlier time with better accuracy than any other conventional test (5). Th e major limitation with IHA test lies in the use of fresh red blood cells as indicator. Because fresh cells are fragile, pr one to haemolysis and bacterial contamination, they can be stored only for a brief period. This limitation can be overcome by stabilising the red blood cells with aldeh ydes and these cells can be used for up to eight months with out the loss of much antigen-coating efficiency (7). Sensitivity of the IHA test wa s found to depend on the meth od of stabilisation in the diagnosis of malaria (3). In the present investigation, it wa s observed th at the use of double aldehyde-sta bilised cells greatl y increased the sensitivity of the IHA test and the sta bilisation by glutar aldehyde also gave satisfacto ry result s comp ared to formaldehyde and pyru vic aldehyde sta bilisatio n. Our observat ions are in confo rmity with the other reported result s on stabilisation of SRBCs in IHA for the detection of other diseases (3, 8). Therefor e, double aldehyde-stabilised SRBCs can be emplo yed for routine serodiagnosis of bru cellosis by IHA. Acknowledgement . Authors are grateful to Brig. K. M. Rao (Retd.), former Director, Defence R &D Establishment for his interest in this study and constant encouragement. Thanks are also due to Commandant , Central Military Veterinary Research Laboratory, Meerut for providing the serum samples.
References 1. Batra, H. V., P. Chand, L. Ganju, R. Mukerjee, and j. R. Sardana: Dotenzyme linked immunosorbent assay for the detection of antibodies in bovine brucellosis. Res. Vet. Sci. 46 (1989) 143-146 2. Chand, P., j. R. Sardana, H. V. Batra, and R. S. Chauhan: Comparison of the dotimmunobinding assay wih the complement fixation test for the detection of Brucella antibodies in sheep. Vet. Microbiol. 20 (1989) 282-287 3. Farshy, D. C. and I. G. Kagan: Use of stab le sensitised cells in indirect microhaemagglutination test for malaria. Am. J. Trop. Med. Hyg. 21 (1972) 868-872 4. Gupta, R. K., C. N. Mishra, V. K. Mehtha, V. K. Gupta , L. N. Rao Bhau, G. Singh, D. Gowal, and S. N. Saxena: Indirect haemaggIutination test for the assay of antibodies against Japanese encephalitis viral antigen in human and animal sera. Ind. J. Med. Res. 91 (1990) 315-320 5. Khairov, S. G. and O. Yu. Yusupov : Experiences with indirect haemagglutinat ion test in brucellosis. Veterinariya 1 (1989) 60-62 6. Lowry, O. H ., N. j. Rosebrough, A. L. Farr, and R. j. Randall: Protein measurement with the folin phenol reagent. J. BioI. Chern. 193 (1951) 265- 275
192
G. P. Rai, G. S.Agarwal, and K. S.Venkateswaran
7. Nagarkatti, P. S., M. Nagarkatti, and K. M. Rao: Development of a kit for the assay of haemagglutination inhibition antibodies to flaviviruses using formalinised goose erythrocytes. Trans. Roy. Soc. Trop. Med. Hyg. 74 (1980) 22-25 8. Parija, S. C. and N. Ananthakrishnan: Evaluation of stabilised cells in the indirect haemagglutination test for echinococcosis. ]. Med. Microbiol. 19 (1985) 95-98 9. Rai, G. P., K. Zachariah, and S. Shriuastaua: Comparative efficacy of indirect haemagglutination test, indirect fluorescent antibody test and enzyme linked immunosorbent assay in serodiagnosis of typhoid fever. ]. Trop, Med. Hyg. 92 (1989) 431-434 10. Rodriquez, A. N., J. Pachon Diaz, R. T. Satigo, J. A. C. Contreras, P. V. Fernandez, L. L. Cortes, J. c. de la Fuente, and A. M. Marin: Indirect immunofluorescence and rose bengal test for the diagnosis of brucellosis. Rev. Clin. Esp. 175 (1984) 27-32
Dr. G. P. Rai, Division of Microbiology, Defence Rand D Establishment, Tansen Road, Gwalior-474 002, India
