World
Journal
of Microbiology
and Biotechnology,
9, W-96
An ELBA-disc procedure for antibodies to Pseudomonas pseudoma//ei: application for a serological study of melioidosis in an endemic area N. Embi,* D. Devarajoo, R. Mohamed and G. lsmail The optimization and development of an ELISA-disc procedure for the detection of antibodies to whole cell surface antigens and purified exotoxin of pseudomonas pseudomallei is described. Comparison of the serum agglutination test (SAT), the serum based enzyme-linked immunosorbent assay (ELISA) and the ELBA-disc procedures used on goat and human sera demonstrated a high correlation in their ability to detect antibodies specific for P. pseudomallei antigens. A serological survey using the ELISA-disc method was carried out on a normal human population in Sabah, Malaysia, an area known to be endemic for melioidosis. The prevalances of antibodies towards cell surface antigens and exotoxin of P. pseudomallei were 28% and S%, respectively. As a procedure, the ELISA-disc technique reported here is technically simple and provides savings in costs and is thus deemed suitable for seroepidemiological surveillance of melioidosis in remote areas of South-East Asia. Key wards: ELISA, exotoxin,
immunodiagnosis,
melioidiosis,
Melioidosis offers a great challenge to clinicians, epidemiologists and laboratory technicians serving in endemic areas of South-East Asia (Howe et al. 1971; Punyagupta 1983; So 1985; Thisyakorn 1986; Smith ef al. 1987) and tropical Australia (Rode & Webling 1981). Information on the prevalance and incidence of melioidosis in Malaysia is much needed to know precisely the extent of morbidity and mortality due to Pseudomonas pseudomallei infections. Various assays have been developed for the serological surveillance of P. pseudomallei infections in these endemic areas, including indirect haemagglutination and complement fixation tests utilizing heat-killed or sonicated bacterial suspensions (Nigg 1963; Nigg & Johnson 1961; Strauss et al. 1969; Alexander ef al. 1970; Khupulsup & Petchclai 1986). These procedures, however, hardly provided an acceptable level of specificity and sensitivity because non-specific reactions occurred with P. aeruginosa and other diverse antibacterial sera. The ELISA-based procedure developed for the detection of antibody to P.
N,. Embi. S.D. Devarajoo, R. Mohamed of Life Science, Universiti Kebangsaan aysia. *Corresponding author. @ 1993 Rapid
Communicafions
and G. Ismail are with the Faculty Malaysia, Bangi, Selangor, Mal-
of Oxford
Pset.tdomanas psetrdomallei, serum agglutination.
pseudomallei exotoxin (Ismail et al., 1987b) is very sensitive and specific but still suffers from some drawbacks: it requires collection tubes, serum processing and serum samples which are hard to handle and transport. These factors limit the usefulness of serum-based ELISA to mainly urban hospitals, precluding its use in interior areas of South-East Asia where an epidemiological study of melioidosis is most crucial and urgently needed. The collection of blood on filter paper for combined use in the ELISA procedure has been described for several infectious diseases, including hepatitis B (Zhuang et al. 1982), brucellosis (McLean 81 Hulbink 1989), leprosy (Douglas & Worth 1984)‘ schistosomiasis (Evengard & Hagi 1988) and human irnmunodeficiency virus infection (Farzadegan et al. 1987). This ELISA-disc method provides a simple, cheap and accurate screening test that does not require elaborate confirmatory procedures and can be carried out on a mass scale in a central laboratory. Its minimal range of testing variability over time further renders this approach especially suitable for sero-epidemiological studies in remote areas where transport of blood samples is a complicating factor. This paper describes an investigation of an ELISA-disc method in comparison with a serum-based ELISA procedure
Lfd W&i
]oumal of Mxrobdogy
and Biotechnology. Vol 9, 1993
91
N. Embi et al. for
the
detection
and purified
Materials
exotoxin
of antibodies of
against
whole
P. pseudomallei in humans
cell
antigens and goats.
and Methods
Bacteria Pseudomonas pseudomallei was obtained from the Microbiology Department, Faculty of Medicine, National University of Malaysia, Kuala Lumpur. The strain was originally isolated from a male patient who died of melioidosis. The organisms were propagated on Brain/Heart Infusion Agar (Sigma Chemical Co., St. Louis, MO, USA) slants and stored at 4’C. Antiserum Production The P. pseudomallei whole-cell antigen was prepared as previously described (Ismail et al. 1987~) and the exotoxin was purified by the method employed by Mohamed et al. (1989). Rabbit sera specific for whole cell antigens and exotoxin were used as controls in the SAT and ELISA. Rabbits were intravenously inoculated with 1.0 ml of heat-killed or sonicated whole-cell antigens of P. pseudomallei, (lo6 cells/ml) in an equal volume of Freund’s incomplete adjuvant and sampled at weekly intervals. For antitoxin sera, rabbits were immunized with IOO pgg/ml of purified exotoxin. The antigen for the SAT was prepared from a 7-day broth culture of P. pseudomallei that has been autoclaved and centrifuged. The supematant fluid was removed, preserved with phenol (0.5% by vol) and stored at 4’C before use. Blood Samples Human blood samples from normal individuals were collected directly onto filter paper discs by the finger-prick method. A total of 418 samples was collected from the Kota Kinabalu, Tuaran, Tambunan id Penampang Districts of Sabah, Malaysia. Human cord blood samples of newborns were obtained from the maternity ward of Queen Elizabeth Hospital, Kota Kinabalu, Sabah, Malaysia. Goat sera that were positive for P. psedomallei infection utilizing the SAT were obtained from the Veterinary Department, Sabah, Malaysia. ELlSA Procedures The detection of specific antibody to P. pseudodlei exotoxin was performed according to method previously described (Ismail et al. 1987b). Polystyrene, flat-bottomed, 96-well microtitration plates were used for the assay (NUNC Immunoplate II, Gibco Laboratories, Grand Island, NY, USA). Wells were coated with 100 ~1 of purified exotoxin (0.5 $/ml) in carbonate-bicarbonate binding buffer, 0.1 M, pH 9.6. After overnight incubation at 4’C, the plates were washed five times with phosphate-buffered saline containing 0.05% (w/v) Tween 80 (PBST) and stored at 4’C. A 100-p sample of antiserum diluted in phosphate-buffered saline (PBS) was added to each appropriate well and the plates were incubated, with shaking, at room temperature for 30 min. After incubation, the plates were washed five times with PBST, and 100 ~1 of peroxidase-labelled rabbit anti-mouse immunoglobulin G (Sigma Chemical Co.) diluted I in 200 in PBST was added. The plates were incubated for 20 min with shaking at room temperature, washed five times with PBST and then 100 ~1 of substrate solution was added. The substrate solution was 0.012% H,O1 in citrate buffer (pH 5.0) containing 6 mg 2-2’-azinobis (3-ethylbenz-thiazoline) sulphonic acid (ABTS)/ml. Reaction mixtures were incubated with shaking at room temperature for 10 min. Control wells were treated identically, except that PBS
92
World ]ouml
of Microbiology and Biotechnology. Vol 9, 1993
replaced the serum. The turbidity was read at 405 nm in a MicroELISA Reader (Dynatech). The ELISA-disc method was a modification of the method of McLean & Hulbink (1989). Discs were made out of filter paper for all blood collections in the study. Wells of flat-bottomed microtitre plates were coated with 100 ~1 of purified exotoxin (5 hg/ml) or sonicated heat-killed cells (5 pg/ml) in carbonate/bicarbonate binding buffer, 0.1 M, pH 9.6. After an overnight incubation at 4’C, plates were washed three times with PBST. To minimize non-specific adsorption, 100 ~1 of 2% (v/v) skimmed milk was used to coat the wells and incubated at 37’C for 1 h. Plates were stored at 4°C until used. Filter paper discs (diameter 5 mm) saturated with the blood samples obtained by the finger-prick method were put into each well. A volume of 100 ~1 PBST was added to each well and incubated on an orbital shaker apparatus at room temperature for 1 h. After incubation, discs were removed and plates were washed three times with PBST before incubation with peroxidase-labelled anti-IgG antibody (Sigma). The assay conditions for the ELISA-disc method and several physical parameters were compared. Discs made out of Whatman No. 4 and Whatman No. 5 filter paper were compared with the discs originally used in the study, made out of Schleicher & Schull filter paper. The effect of detergent (Tween 20) in the elution step of blood-soaked discs was also investigated. Other physical parameters studied included: (i) sample volume effect: two volumes of samples (5 ~1 and 10 ~1) were compared; (ii) sample storage temperature: samples were stored at room temperature, 4’C and - 20°C prior to testing and (iii) sample storage-duration: samples were kept for 1, 2, 5, 10 and 15 days before use in the assay.
Results Correlation with Serum-based ELISA and SAT The conventional serum-based ELISA, ELISA-disc and SAT procedures were compared with respect to the sensitivity of each assay in detecting antibodies specific to P. psecdomallei antigens (Table I). Comparisons were made on eight serum samples obtained from goats, four of them being positive for P. pseudomallei infection as indicated by SAT. Generally, the absorbance values obtained from the ELISA-disc procedure were consistently higher than the values from the conventional ELISA. All SAT-positive sera gave high absorbance readings in both the ELISA and ELISA-disc methods. For SAT-negative serum samples, the absorbance values were lower but well above the cut-off level value of 0.002 for negative controls. The relative sensitivity of the ELISA-disc compared with the conventional ELISA was studied using 71 goat sera that were positive for P. pseudomallei infection as indicated by SAT (Table 2). Both ELISA and ELISA-disc procedures employed in the study showed similar sensitivity in detecting the presence of antibodies against P. psetrdomallei antigens. Data indicate that 54% of the serum samples were reactive towards cell surface antigens whereas only 26% were positive for exotoxin antibody. Further, not all the serum samples that were positive for exotoxin antibody showed reactivity towards cell surface antigens.
EllSA
Table 1. Comparlson of serum-based ELISA, ELISAdisc procedures in the detection of specific antlbodies pseudomallel whole cell antigens in goat sera. Sample
Absorbance
at 405 nm
ELlSA
ELISA-disc 0.101 0.102
+ +
0.060 0.081 0.058 0.044
0.084 0.129 0.065 0.036
+ + -
0.085 0.063
0.091 0.064
-
of serum-based ELISA and detection of antibodies against of P. pseudomalleiin goat sera.
No. of sera tested
Sera Whole
ELISA ELISA-disc
Table 3. ELISA-disc
71 71
Comparison procedure
16 (25.8%) 16 (25.8%)
Mean
absorbance at 405 mn*
0.012 0.009
+ 0.004 * 0.005
Whatman
No.
0.006
+ 0.004
NS-not
analysis significantly
values obtalned types of filter paper.
in
SignificanceT
N.S
represents
means
was
using
done
Exotoxln
38 (53.5%)
& Schull No. 5
values
to:
38 (53.5%)
Schleicher Whatman
4
positive
cell antigens
of absorbance using three different
Type of filter paper
l Absorbance samples. TStatistical
SAT
0.099 0.094
Table 2. Relative sensltivities ELISAdisc procedures in the whole cell antigens and exotoxin Method
and SAT to P.
of the
21
goat
serum
Student’s
t-test.
different.
Table 4. The effect of the presence of detergent Tween-20 the buffer solution used in the ELISA-disc procedure. Buffer
Mean
absorbance at 405 nm’
Phosphate-buffered saline (PBS)
0.003
_+ 0.001
Phosphate-buffered saline with
0.005
f
Tween-20 ‘Absorbance samples. T Statistical
Significancet
P = 0.01 0.001
(PBST) values analysis
represent was
done
means using
the
of
21
Student’s
goat t-test.
serum
in
procedures for melioidosis
Blood samples from umbilical cords of newborn babies gave negative results when tested for the presence of antibodies against P. pseudomallei whole cell antigens or purified exotoxin. Similar results were obtained when processed sera from cord blood instead of whole blood samples were used in both assay procedures (data not shown). Comparison of Physical Parameters Results from the comparison of absorbance values obtained in the ELISA-disc method using three different types of filter paper are given in Table 3. Macroscopical examination was done to assess the soaking capacity of sampling discs and all the three types of filter paper exhibited even blood spreading and absorbance. Filter paper discs from Schleicher & Schull possessed the optimum soaking capacity followed by the Whatman No. 5 and Whatman No. 4 filter paper. The former were therefore used routinely. The effect of Tween-20 in the buffer solution used to elute the samples from the discs was investigated. Samples eluted with buffer containing detergent gave higher absorbance values than those washed with buffer alone (Table 4). indicating the consistently-improved yield of sampled antibody in the presence of detergent. Two volumes of samples (5 and 10 ~1) were tested on the filter discs to determine whether different sample volumes adsorbed to the discs can affect the absorbance values of the assay. For the two different volumes tested, there was no observable difference in the absorbance readings, indicating that the total amount of antibodies extractable from the discs, even at the lower 5 ~1 volume, had reached the optimal practical level detectable by the ELISA-disc procedure (data not shown). Figure I shows that blood samples adsorbed to filter paper discs could be kept up to a maximum of 10 days at a suitable temperature before use in the ELISA-disc assay. The absorbance values however tended to decrease after storage for more than 10 days. Further, samples on discs could be stored at room temperature and 4°C for up to 10 days without showing a decline in the absorbance readings. The absorbance values however showed a downward trend when samples were kept at - 20°C before use in the assay. Application for Sero-epidemiological Study Figure 2 gives the results of serological surveys carried out on human population from four districts in Sabah, Malaysia in relation to their age groups. In general, the data show that 28% (117/418) of the normal population tested were positive for the presence of antibodies towards the cell surface antigens but only 8% (341418) were positive for anti-exotoxin antibodies. The incidence of antibody to P. pseudomallei antigens was highest in the age groups ranging from 21-35 years.
N. Em&i et al.
O.OOi
t
2 5 10 15
12
51015
Incubation m
Room Temperature
1 2 5 1015
Time 1Days) qoc
-2yJc
m
Figure 1. The effect of sample storage at different temperatures for a maximum period of 15 days before use in the ELBA-disc procedure.
Discussion Infection by P. psetrdomallei may not cause any form of clinical manifestations or may result in different symptomatologies affecting various organs (Smith et al. 1987; Brown et al. 1990). Moreover, melioidosis can be further complicated by the fact that these clinical manifestations or recrudescences of chronic infections have been reported in
40
L
(6) n - o-1
1-5
6-10
m
11-15
Whole
16-20
21-25
association with a variety of stressful events, e.g. thermal injury, surgery, pneumococcal pneumonia, diabetic ketoacidosis (Jackson ef a/. 1972) and influenza A infection (Mackowiak & Smith 1978). Therefore, a sero-epidemiological investigation that can facilitate measuring the intensity and geographical distribution of l? psetrdomullei infections is required to permit optimal use of the available resources in public health programmes. The ELISA-disc method of detecting antibodies to P. pseztdomullei antigens reported here provides a useful technique for such investigation. Comparison of SAT, serum-based ELISA and ELISA-disc procedures gave a high correlation in the ability of these three serological tests to detect antibodies specific for P. pseudomullei antigens. However, not all SAT-positive sera were reactive for antibodies to whole cell antigens and exotoxin of P. pseudomullei when assayed using either conventional ELISA or ELISA-disc procedures. This may indicate that SAT is less specific for P. psetldomullei antigens and that the false serum-positive reactions observed may be attributed to immunological cross-reactions with other bacterial antigens, as previously reported (Nigg & Johnson 1961). The detection of antibodies against exotoxin is of particular relevance in serological surveillance of melioidosis because several studies have shown that the exotoxin possesses a variety of biological activities in vitro (Ismail ef al. 1987a; Mohamed ef al. 1989, 1991) and is produced in viva in naturally-infected animals (Ismail et ul. 1991), suggesting its possible role as a virulence factor in the pathogenicity of P. pseudomullei. Small discs, with a diameter of 5 mm, fit into the wells of microtitre plates, resulting in an even extraction of blood
26-30
Age group cell
antigen
31-3536-40
41-4546-56
(years) m
Exotoxin
Figure 2. Prevalence of antibodies to P. pseudomallei whole cell antigens and exotoxin in normal healthy individuals from four districts of Sabah, Malaysia, according to age group. Numbers in parentheses indicate the numbers of individuals tested in each age group.
94
World Journal of Microbiology and Biotechnology. Vol 9, I%X
ELLSA procedures for melioidosis samples. The Schleicher & Schull filter paper made the best sampling discs, probably due to its pore size, trapping the antibody molecules more efficiently. The presence of detergent helped in enhancing the absorbance values for the ELISA-disc tests by lessening the non-specific adsorption of serum components to the wells and causing lysis of interfering red blood cells during the elution process. Each finger-prick droplet of blood is approximately 25 to 50 ,ul and this appears to be far in excess of that required for optimal absorbance reading in our ELISA-disc test. Blood samples on discs can be kept at either room temperature or 4’C for up to 10 days without showing any decline in the absorbance values obtained. Thus, as a procedure, ELISA-disc possesses substantial advantages over the serum-based ELISA method. It provides savings in costs because transport of samples to the testing laboratory is greatly facilitated by simply mailing the filter paper samples without refrigeration. For sera obtained from normal healthy individuals in four different localities in Sabah, Malaysia, the prevalence rate of antibodies towards cell surface antigens of P. pseudomallei was 28% and for exotoxins, 8%. The reason(s) for this difference is not clear. The positive reaction to anti-exotoxin may indicate more recent infection or exposure to the organisms whereas the presence of anti-whole cell antibody may represent the occurrence of natural agglutinins to P. pseudomnllei in individuals living in endemic areas. The high positive rate for antibodies to either exotoxin or whole cell antigens is indicative of the high incidence of asymptomatic infection in the human population surveyed. Previous studies by Nigg (1963) and Strauss ef al. (1969) gave values of 15 to 30% for apparently healthy populations in Malaysia and Thailand. Another study, recently carried out in Malaysia and utilizing the serum-based ELISA, reported higher prevalences of antibodies in military personnel; 54% were positive for antibodies to exotoxin and 66% for whole cell antigens (Embi et al. 1992). These high rates were, however, expected since melioidosis is renowned for its high incidence in military personnel serving in endemic areas (Thin et al. 1970; Sanford & Moore 1971; Everette & Nelson, 1975). It is also noteworthy that these individuals showing the highest positive rates for antibodies against P. pseudomallei antigens in the present study were 21 to 35 years old and also formed the major work force, serving as road construction labourers, rice and vegetable farmers etc, in the four districts surveyed. These jobs expose the workers to P. pseudornallei found in the natural habitats (e.g. in soil and surface water) and place them at high risk to melioidosis infection. Newborns surveyed in our study were negative for P. pseudomallei antigens, suggesting that prior exposure to the causative organisms found in the natural environment is required before any detectable level of specific antibody to the bacteria appears in the host. In conclusion, the ELISA-disc method described here can
be used as an alternative tool for serological surveillance of P. pseudomullei infection in endemic areas. The good correlation of results from the ELISA-disc method with those from the serum-based ELISA and the technical simplicity of the ELISA-disc method make the latter particularly suitable for use in sero-epidemiological studies of melioidosis in remote endemic areas of South-East Asia.
Acknowledgements This study was supported by IRPA Grant 07-03-012 and EEC Grant TS2-0230-UK. The authors are thankful to the Superintendent of the Queen Elizabeth Hospital; the Director of Animal Disease Center, Sabah and the Flying Doctors of the Sabah Foundation for their assistance throughout this study. We also thank MS F. Sanusi for typing and the secretarial assistance rendered during preparation of the manuscript.
References Alexander, A.D., Huxsoll, D.L., Warner, A.R. Jr., Shepler, V. & Dorsey, A. 1970 Serological diagnosis of human melioidosis with indirect hemagglutination and complement fixation tests. Applied Microbiology 20, 82.5-833. Brown, A.E., Dance, D.A.B., Chaowagul, W., Webster, H.K. & White, N.J. 1990 Activation of cellular immune responses in melioidosis patients as assessed by urinary neopterin. Trunsactions of the Royal Sociefy for Tropicul Medicine and Hygiene 17, 183-191. Douglas, J.T. & Worth, R.M. 1984 Field evaluation of an ELISA to detect antibodies in leprosy patients and their contacts. International ]ournal of Leprosy 52, 2&32. Embi, N., Suhaimi, A., Mohamed, R. & Ismail, G. 1992 Prevalence of antibodies to Pseudomonas pseudomaliei exotoxin and whole cell antigens in military personnel in Sabah and Sarawak, Malaysia. n/iicrobio/ogy and Immunology (in press). Evengard, B. & Hagi, H. 1988 A filter paper technique for the detection of IgG and IgM class schistosome specific antibodies in an endemic area. Annals of Tropical Medicine and Parasitology 82, 307-309. Everette, E.D. & Nelson, R.A. 1975 Pulmonary melioidosis: Observations in 39 cases. American Review of Respiratory Diseases 112, 331-340. Farzadegan, H., Quinn, T. & Polk, B.F. 1987 Detecting antibodies to human immunodeficiency virus in dried blood on filter papers. Journal of Internal Diseases 155, 1073-1074. Howe, C., Sampath, A. & Spotnitz, M. 1971 The Pseudomallei Group: A Review. Journal of Infectious Diseases 124, 598-607. Ismail, G., Embi, M.N., Omar, O., Mohamed, R. & Razak, N. 1987a Toxigenic properties of Pseudomonas pseudomallei extracellular products. Tropical Biomedicine 4, 101-109. Ismail, G., Embi, M.N., Omar, O., Razak, N., Allen, J.C. & Smith, C.J. 1987b Enzyme immunoassay for the detection of antibody to Pseudomonas pseudomallei exotoxin in mice. FEMS Microbiology Letters 40, 27-31. Ismail, G., Embi, M.N., Omar, O., Razak, N., Allen, J.C. & Smith, C.J. 1987~ A competitive immunosorbent assay for the detection of Pseudomonas pseudomallei exotoxin. Journal of Medical Microbiology 23, 253-257.
World ~oournaiof Microbiology and Bmtechnology. Vol 9. 1993
95
N. En&
et al.
Ismail, G., Mohamed, R., Rohana, 1991 Antibody to Pseudomonas exposed to natural infection. 277-282.
S., Sharifah, pseudomallei Veferinary
S.M. & Embi, N. exotoxin in sheep A4icrobiolom 27,
Jackson, A.E., Moore, W.L. Jr. & Sanford J.P. 1972 Recrudescent melioidosis associated with diabetic ketoacidosis. Archives of Internal Medicine 130, 268-271. Khupulsup, K. & Petchclai, B. 1986 Application of indirect haemagglutination test and indirect fluorescent antibody test for IgM antibody for diagnosis of melioidosis in Thailand. American journal of Tropical Medicine and Hygiene 35, 36&369. Mackowiak, P.A. & Smith, J.W. 1978 Septicemic melioidosis: Occurrence following acute influenza A six years after exposure in Vietnam. Journal of the American Medical Association 240,
764-766. McLean, G. & Hulbink F. 1989 in ELISA for Brucella abortw.
Use of dried blood on filter paper journal of immunological Methods
123,39-43. Mohamed, R., Nathan, S., Embi, N., Razak, N. & Ismail, G. 1989 Inhibition of macromolecular synthesis in cultured macrophages by Pseudomonas pseudomallei exotoxin. Microbiology and Immunology 33, 811-819. Mohamed, R., Periasamy, M., Ismail, G. & Embi, N. I991 The exotoxin of melioidosis, a disease of animals and man. In Toxins and Targets, eds Walters, E. et al. pp. 110~1118. Engelwood Cliffs, NJ: Hardwood Academic. Nigg, C. 1963 Serological studies on subclinical melioidosis. ]ourm?l of Immunology 91, 18-28. Nigg, C. & Johnson, M. 1961 Complement fixation tests in experimental clinical and sub-clinical melioidosis. Journal of Bacferiology 82, 159-167.
96
WorldJoud of Microbdogy
and Btofechnology. Vol 9, 1993
Punyagupta, S. 1983 Melioidosis: the great imitator. Ramathibodi Medical ]ournal 6, 147-159. Rode, J.W. & Webling, D.D.A. 1981 Melioidosis in the Northern Territory of Australia. Medical Journal of Australia 1, 181-184. Sanford, J.P. & Moore, W.L. Jr. 1971 Recrudescent melioidosis: A South-east Asian legacy. American Review of Respiratory Diseases
104,452-457. Smith, C.J., Allen, J.C., Noor Embi M., Othman, O., Razak, N. & Ismail, G. 1987 Human Melioidosis: An Emerging Medical Problem. MlRCEN]oumal Applied Microbiology and Biotechnology
3,343-366. So, S.Y. 1985 Melioidosis-An overlooked problem in Hong Kong. H.K. Practitioners 3, 1111-1113. Strauss, J.M., Alexander, A.D., Rapmund, G., Gan, E. & Dorsey, A.E. 1969 Melioidosis in Malaysia: Antibodies to Pseudomonas pseudomallei in the human population. American Journal of Tropical Medicine and Hygiene 18, 698-702. Thin, R.N.F., Brown, M., Stewart, J.B. & Garrett, CJ. 1970 Melioidosis: a report of ten cases. Quarterly Journal of Medicine 39, 115-121. Thisyakom, U. 1986 Melioidosis in children at Children’s Hospital, Bangkok. Southeast Asian ]ournal of Tropical Medicine and Public Health 17, IOI-103. Zhuang, H., Coulepsis, A.G., Locamini, S.A., & Gust, I.D. 1982 Detection of markers of hepatitis B infection in sera dried on to filter paper: an application to field studies. Bulletin of the World Health Organization 60, 783-787.
(Received
in revised
form 6 July
1992;
accepted
17 ]uly
1992)
Journal
of Microbiology
and Biotechnology,
9, W-96
An ELBA-disc procedure for antibodies to Pseudomonas pseudoma//ei: application for a serological study of melioidosis in an endemic area N. Embi,* D. Devarajoo, R. Mohamed and G. lsmail The optimization and development of an ELISA-disc procedure for the detection of antibodies to whole cell surface antigens and purified exotoxin of pseudomonas pseudomallei is described. Comparison of the serum agglutination test (SAT), the serum based enzyme-linked immunosorbent assay (ELISA) and the ELBA-disc procedures used on goat and human sera demonstrated a high correlation in their ability to detect antibodies specific for P. pseudomallei antigens. A serological survey using the ELISA-disc method was carried out on a normal human population in Sabah, Malaysia, an area known to be endemic for melioidosis. The prevalances of antibodies towards cell surface antigens and exotoxin of P. pseudomallei were 28% and S%, respectively. As a procedure, the ELISA-disc technique reported here is technically simple and provides savings in costs and is thus deemed suitable for seroepidemiological surveillance of melioidosis in remote areas of South-East Asia. Key wards: ELISA, exotoxin,
immunodiagnosis,
melioidiosis,
Melioidosis offers a great challenge to clinicians, epidemiologists and laboratory technicians serving in endemic areas of South-East Asia (Howe et al. 1971; Punyagupta 1983; So 1985; Thisyakorn 1986; Smith ef al. 1987) and tropical Australia (Rode & Webling 1981). Information on the prevalance and incidence of melioidosis in Malaysia is much needed to know precisely the extent of morbidity and mortality due to Pseudomonas pseudomallei infections. Various assays have been developed for the serological surveillance of P. pseudomallei infections in these endemic areas, including indirect haemagglutination and complement fixation tests utilizing heat-killed or sonicated bacterial suspensions (Nigg 1963; Nigg & Johnson 1961; Strauss et al. 1969; Alexander ef al. 1970; Khupulsup & Petchclai 1986). These procedures, however, hardly provided an acceptable level of specificity and sensitivity because non-specific reactions occurred with P. aeruginosa and other diverse antibacterial sera. The ELISA-based procedure developed for the detection of antibody to P.
N,. Embi. S.D. Devarajoo, R. Mohamed of Life Science, Universiti Kebangsaan aysia. *Corresponding author. @ 1993 Rapid
Communicafions
and G. Ismail are with the Faculty Malaysia, Bangi, Selangor, Mal-
of Oxford
Pset.tdomanas psetrdomallei, serum agglutination.
pseudomallei exotoxin (Ismail et al., 1987b) is very sensitive and specific but still suffers from some drawbacks: it requires collection tubes, serum processing and serum samples which are hard to handle and transport. These factors limit the usefulness of serum-based ELISA to mainly urban hospitals, precluding its use in interior areas of South-East Asia where an epidemiological study of melioidosis is most crucial and urgently needed. The collection of blood on filter paper for combined use in the ELISA procedure has been described for several infectious diseases, including hepatitis B (Zhuang et al. 1982), brucellosis (McLean 81 Hulbink 1989), leprosy (Douglas & Worth 1984)‘ schistosomiasis (Evengard & Hagi 1988) and human irnmunodeficiency virus infection (Farzadegan et al. 1987). This ELISA-disc method provides a simple, cheap and accurate screening test that does not require elaborate confirmatory procedures and can be carried out on a mass scale in a central laboratory. Its minimal range of testing variability over time further renders this approach especially suitable for sero-epidemiological studies in remote areas where transport of blood samples is a complicating factor. This paper describes an investigation of an ELISA-disc method in comparison with a serum-based ELISA procedure
Lfd W&i
]oumal of Mxrobdogy
and Biotechnology. Vol 9, 1993
91
N. Embi et al. for
the
detection
and purified
Materials
exotoxin
of antibodies of
against
whole
P. pseudomallei in humans
cell
antigens and goats.
and Methods
Bacteria Pseudomonas pseudomallei was obtained from the Microbiology Department, Faculty of Medicine, National University of Malaysia, Kuala Lumpur. The strain was originally isolated from a male patient who died of melioidosis. The organisms were propagated on Brain/Heart Infusion Agar (Sigma Chemical Co., St. Louis, MO, USA) slants and stored at 4’C. Antiserum Production The P. pseudomallei whole-cell antigen was prepared as previously described (Ismail et al. 1987~) and the exotoxin was purified by the method employed by Mohamed et al. (1989). Rabbit sera specific for whole cell antigens and exotoxin were used as controls in the SAT and ELISA. Rabbits were intravenously inoculated with 1.0 ml of heat-killed or sonicated whole-cell antigens of P. pseudomallei, (lo6 cells/ml) in an equal volume of Freund’s incomplete adjuvant and sampled at weekly intervals. For antitoxin sera, rabbits were immunized with IOO pgg/ml of purified exotoxin. The antigen for the SAT was prepared from a 7-day broth culture of P. pseudomallei that has been autoclaved and centrifuged. The supematant fluid was removed, preserved with phenol (0.5% by vol) and stored at 4’C before use. Blood Samples Human blood samples from normal individuals were collected directly onto filter paper discs by the finger-prick method. A total of 418 samples was collected from the Kota Kinabalu, Tuaran, Tambunan id Penampang Districts of Sabah, Malaysia. Human cord blood samples of newborns were obtained from the maternity ward of Queen Elizabeth Hospital, Kota Kinabalu, Sabah, Malaysia. Goat sera that were positive for P. psedomallei infection utilizing the SAT were obtained from the Veterinary Department, Sabah, Malaysia. ELlSA Procedures The detection of specific antibody to P. pseudodlei exotoxin was performed according to method previously described (Ismail et al. 1987b). Polystyrene, flat-bottomed, 96-well microtitration plates were used for the assay (NUNC Immunoplate II, Gibco Laboratories, Grand Island, NY, USA). Wells were coated with 100 ~1 of purified exotoxin (0.5 $/ml) in carbonate-bicarbonate binding buffer, 0.1 M, pH 9.6. After overnight incubation at 4’C, the plates were washed five times with phosphate-buffered saline containing 0.05% (w/v) Tween 80 (PBST) and stored at 4’C. A 100-p sample of antiserum diluted in phosphate-buffered saline (PBS) was added to each appropriate well and the plates were incubated, with shaking, at room temperature for 30 min. After incubation, the plates were washed five times with PBST, and 100 ~1 of peroxidase-labelled rabbit anti-mouse immunoglobulin G (Sigma Chemical Co.) diluted I in 200 in PBST was added. The plates were incubated for 20 min with shaking at room temperature, washed five times with PBST and then 100 ~1 of substrate solution was added. The substrate solution was 0.012% H,O1 in citrate buffer (pH 5.0) containing 6 mg 2-2’-azinobis (3-ethylbenz-thiazoline) sulphonic acid (ABTS)/ml. Reaction mixtures were incubated with shaking at room temperature for 10 min. Control wells were treated identically, except that PBS
92
World ]ouml
of Microbiology and Biotechnology. Vol 9, 1993
replaced the serum. The turbidity was read at 405 nm in a MicroELISA Reader (Dynatech). The ELISA-disc method was a modification of the method of McLean & Hulbink (1989). Discs were made out of filter paper for all blood collections in the study. Wells of flat-bottomed microtitre plates were coated with 100 ~1 of purified exotoxin (5 hg/ml) or sonicated heat-killed cells (5 pg/ml) in carbonate/bicarbonate binding buffer, 0.1 M, pH 9.6. After an overnight incubation at 4’C, plates were washed three times with PBST. To minimize non-specific adsorption, 100 ~1 of 2% (v/v) skimmed milk was used to coat the wells and incubated at 37’C for 1 h. Plates were stored at 4°C until used. Filter paper discs (diameter 5 mm) saturated with the blood samples obtained by the finger-prick method were put into each well. A volume of 100 ~1 PBST was added to each well and incubated on an orbital shaker apparatus at room temperature for 1 h. After incubation, discs were removed and plates were washed three times with PBST before incubation with peroxidase-labelled anti-IgG antibody (Sigma). The assay conditions for the ELISA-disc method and several physical parameters were compared. Discs made out of Whatman No. 4 and Whatman No. 5 filter paper were compared with the discs originally used in the study, made out of Schleicher & Schull filter paper. The effect of detergent (Tween 20) in the elution step of blood-soaked discs was also investigated. Other physical parameters studied included: (i) sample volume effect: two volumes of samples (5 ~1 and 10 ~1) were compared; (ii) sample storage temperature: samples were stored at room temperature, 4’C and - 20°C prior to testing and (iii) sample storage-duration: samples were kept for 1, 2, 5, 10 and 15 days before use in the assay.
Results Correlation with Serum-based ELISA and SAT The conventional serum-based ELISA, ELISA-disc and SAT procedures were compared with respect to the sensitivity of each assay in detecting antibodies specific to P. psecdomallei antigens (Table I). Comparisons were made on eight serum samples obtained from goats, four of them being positive for P. pseudomallei infection as indicated by SAT. Generally, the absorbance values obtained from the ELISA-disc procedure were consistently higher than the values from the conventional ELISA. All SAT-positive sera gave high absorbance readings in both the ELISA and ELISA-disc methods. For SAT-negative serum samples, the absorbance values were lower but well above the cut-off level value of 0.002 for negative controls. The relative sensitivity of the ELISA-disc compared with the conventional ELISA was studied using 71 goat sera that were positive for P. pseudomallei infection as indicated by SAT (Table 2). Both ELISA and ELISA-disc procedures employed in the study showed similar sensitivity in detecting the presence of antibodies against P. psetrdomallei antigens. Data indicate that 54% of the serum samples were reactive towards cell surface antigens whereas only 26% were positive for exotoxin antibody. Further, not all the serum samples that were positive for exotoxin antibody showed reactivity towards cell surface antigens.
EllSA
Table 1. Comparlson of serum-based ELISA, ELISAdisc procedures in the detection of specific antlbodies pseudomallel whole cell antigens in goat sera. Sample
Absorbance
at 405 nm
ELlSA
ELISA-disc 0.101 0.102
+ +
0.060 0.081 0.058 0.044
0.084 0.129 0.065 0.036
+ + -
0.085 0.063
0.091 0.064
-
of serum-based ELISA and detection of antibodies against of P. pseudomalleiin goat sera.
No. of sera tested
Sera Whole
ELISA ELISA-disc
Table 3. ELISA-disc
71 71
Comparison procedure
16 (25.8%) 16 (25.8%)
Mean
absorbance at 405 mn*
0.012 0.009
+ 0.004 * 0.005
Whatman
No.
0.006
+ 0.004
NS-not
analysis significantly
values obtalned types of filter paper.
in
SignificanceT
N.S
represents
means
was
using
done
Exotoxln
38 (53.5%)
& Schull No. 5
values
to:
38 (53.5%)
Schleicher Whatman
4
positive
cell antigens
of absorbance using three different
Type of filter paper
l Absorbance samples. TStatistical
SAT
0.099 0.094
Table 2. Relative sensltivities ELISAdisc procedures in the whole cell antigens and exotoxin Method
and SAT to P.
of the
21
goat
serum
Student’s
t-test.
different.
Table 4. The effect of the presence of detergent Tween-20 the buffer solution used in the ELISA-disc procedure. Buffer
Mean
absorbance at 405 nm’
Phosphate-buffered saline (PBS)
0.003
_+ 0.001
Phosphate-buffered saline with
0.005
f
Tween-20 ‘Absorbance samples. T Statistical
Significancet
P = 0.01 0.001
(PBST) values analysis
represent was
done
means using
the
of
21
Student’s
goat t-test.
serum
in
procedures for melioidosis
Blood samples from umbilical cords of newborn babies gave negative results when tested for the presence of antibodies against P. pseudomallei whole cell antigens or purified exotoxin. Similar results were obtained when processed sera from cord blood instead of whole blood samples were used in both assay procedures (data not shown). Comparison of Physical Parameters Results from the comparison of absorbance values obtained in the ELISA-disc method using three different types of filter paper are given in Table 3. Macroscopical examination was done to assess the soaking capacity of sampling discs and all the three types of filter paper exhibited even blood spreading and absorbance. Filter paper discs from Schleicher & Schull possessed the optimum soaking capacity followed by the Whatman No. 5 and Whatman No. 4 filter paper. The former were therefore used routinely. The effect of Tween-20 in the buffer solution used to elute the samples from the discs was investigated. Samples eluted with buffer containing detergent gave higher absorbance values than those washed with buffer alone (Table 4). indicating the consistently-improved yield of sampled antibody in the presence of detergent. Two volumes of samples (5 and 10 ~1) were tested on the filter discs to determine whether different sample volumes adsorbed to the discs can affect the absorbance values of the assay. For the two different volumes tested, there was no observable difference in the absorbance readings, indicating that the total amount of antibodies extractable from the discs, even at the lower 5 ~1 volume, had reached the optimal practical level detectable by the ELISA-disc procedure (data not shown). Figure I shows that blood samples adsorbed to filter paper discs could be kept up to a maximum of 10 days at a suitable temperature before use in the ELISA-disc assay. The absorbance values however tended to decrease after storage for more than 10 days. Further, samples on discs could be stored at room temperature and 4°C for up to 10 days without showing a decline in the absorbance readings. The absorbance values however showed a downward trend when samples were kept at - 20°C before use in the assay. Application for Sero-epidemiological Study Figure 2 gives the results of serological surveys carried out on human population from four districts in Sabah, Malaysia in relation to their age groups. In general, the data show that 28% (117/418) of the normal population tested were positive for the presence of antibodies towards the cell surface antigens but only 8% (341418) were positive for anti-exotoxin antibodies. The incidence of antibody to P. pseudomallei antigens was highest in the age groups ranging from 21-35 years.
N. Em&i et al.
O.OOi
t
2 5 10 15
12
51015
Incubation m
Room Temperature
1 2 5 1015
Time 1Days) qoc
-2yJc
m
Figure 1. The effect of sample storage at different temperatures for a maximum period of 15 days before use in the ELBA-disc procedure.
Discussion Infection by P. psetrdomallei may not cause any form of clinical manifestations or may result in different symptomatologies affecting various organs (Smith et al. 1987; Brown et al. 1990). Moreover, melioidosis can be further complicated by the fact that these clinical manifestations or recrudescences of chronic infections have been reported in
40
L
(6) n - o-1
1-5
6-10
m
11-15
Whole
16-20
21-25
association with a variety of stressful events, e.g. thermal injury, surgery, pneumococcal pneumonia, diabetic ketoacidosis (Jackson ef a/. 1972) and influenza A infection (Mackowiak & Smith 1978). Therefore, a sero-epidemiological investigation that can facilitate measuring the intensity and geographical distribution of l? psetrdomullei infections is required to permit optimal use of the available resources in public health programmes. The ELISA-disc method of detecting antibodies to P. pseztdomullei antigens reported here provides a useful technique for such investigation. Comparison of SAT, serum-based ELISA and ELISA-disc procedures gave a high correlation in the ability of these three serological tests to detect antibodies specific for P. pseudomullei antigens. However, not all SAT-positive sera were reactive for antibodies to whole cell antigens and exotoxin of P. pseudomullei when assayed using either conventional ELISA or ELISA-disc procedures. This may indicate that SAT is less specific for P. psetldomullei antigens and that the false serum-positive reactions observed may be attributed to immunological cross-reactions with other bacterial antigens, as previously reported (Nigg & Johnson 1961). The detection of antibodies against exotoxin is of particular relevance in serological surveillance of melioidosis because several studies have shown that the exotoxin possesses a variety of biological activities in vitro (Ismail ef al. 1987a; Mohamed ef al. 1989, 1991) and is produced in viva in naturally-infected animals (Ismail et ul. 1991), suggesting its possible role as a virulence factor in the pathogenicity of P. pseudomullei. Small discs, with a diameter of 5 mm, fit into the wells of microtitre plates, resulting in an even extraction of blood
26-30
Age group cell
antigen
31-3536-40
41-4546-56
(years) m
Exotoxin
Figure 2. Prevalence of antibodies to P. pseudomallei whole cell antigens and exotoxin in normal healthy individuals from four districts of Sabah, Malaysia, according to age group. Numbers in parentheses indicate the numbers of individuals tested in each age group.
94
World Journal of Microbiology and Biotechnology. Vol 9, I%X
ELLSA procedures for melioidosis samples. The Schleicher & Schull filter paper made the best sampling discs, probably due to its pore size, trapping the antibody molecules more efficiently. The presence of detergent helped in enhancing the absorbance values for the ELISA-disc tests by lessening the non-specific adsorption of serum components to the wells and causing lysis of interfering red blood cells during the elution process. Each finger-prick droplet of blood is approximately 25 to 50 ,ul and this appears to be far in excess of that required for optimal absorbance reading in our ELISA-disc test. Blood samples on discs can be kept at either room temperature or 4’C for up to 10 days without showing any decline in the absorbance values obtained. Thus, as a procedure, ELISA-disc possesses substantial advantages over the serum-based ELISA method. It provides savings in costs because transport of samples to the testing laboratory is greatly facilitated by simply mailing the filter paper samples without refrigeration. For sera obtained from normal healthy individuals in four different localities in Sabah, Malaysia, the prevalence rate of antibodies towards cell surface antigens of P. pseudomallei was 28% and for exotoxins, 8%. The reason(s) for this difference is not clear. The positive reaction to anti-exotoxin may indicate more recent infection or exposure to the organisms whereas the presence of anti-whole cell antibody may represent the occurrence of natural agglutinins to P. pseudomnllei in individuals living in endemic areas. The high positive rate for antibodies to either exotoxin or whole cell antigens is indicative of the high incidence of asymptomatic infection in the human population surveyed. Previous studies by Nigg (1963) and Strauss ef al. (1969) gave values of 15 to 30% for apparently healthy populations in Malaysia and Thailand. Another study, recently carried out in Malaysia and utilizing the serum-based ELISA, reported higher prevalences of antibodies in military personnel; 54% were positive for antibodies to exotoxin and 66% for whole cell antigens (Embi et al. 1992). These high rates were, however, expected since melioidosis is renowned for its high incidence in military personnel serving in endemic areas (Thin et al. 1970; Sanford & Moore 1971; Everette & Nelson, 1975). It is also noteworthy that these individuals showing the highest positive rates for antibodies against P. pseudomallei antigens in the present study were 21 to 35 years old and also formed the major work force, serving as road construction labourers, rice and vegetable farmers etc, in the four districts surveyed. These jobs expose the workers to P. pseudornallei found in the natural habitats (e.g. in soil and surface water) and place them at high risk to melioidosis infection. Newborns surveyed in our study were negative for P. pseudomallei antigens, suggesting that prior exposure to the causative organisms found in the natural environment is required before any detectable level of specific antibody to the bacteria appears in the host. In conclusion, the ELISA-disc method described here can
be used as an alternative tool for serological surveillance of P. pseudomullei infection in endemic areas. The good correlation of results from the ELISA-disc method with those from the serum-based ELISA and the technical simplicity of the ELISA-disc method make the latter particularly suitable for use in sero-epidemiological studies of melioidosis in remote endemic areas of South-East Asia.
Acknowledgements This study was supported by IRPA Grant 07-03-012 and EEC Grant TS2-0230-UK. The authors are thankful to the Superintendent of the Queen Elizabeth Hospital; the Director of Animal Disease Center, Sabah and the Flying Doctors of the Sabah Foundation for their assistance throughout this study. We also thank MS F. Sanusi for typing and the secretarial assistance rendered during preparation of the manuscript.
References Alexander, A.D., Huxsoll, D.L., Warner, A.R. Jr., Shepler, V. & Dorsey, A. 1970 Serological diagnosis of human melioidosis with indirect hemagglutination and complement fixation tests. Applied Microbiology 20, 82.5-833. Brown, A.E., Dance, D.A.B., Chaowagul, W., Webster, H.K. & White, N.J. 1990 Activation of cellular immune responses in melioidosis patients as assessed by urinary neopterin. Trunsactions of the Royal Sociefy for Tropicul Medicine and Hygiene 17, 183-191. Douglas, J.T. & Worth, R.M. 1984 Field evaluation of an ELISA to detect antibodies in leprosy patients and their contacts. International ]ournal of Leprosy 52, 2&32. Embi, N., Suhaimi, A., Mohamed, R. & Ismail, G. 1992 Prevalence of antibodies to Pseudomonas pseudomaliei exotoxin and whole cell antigens in military personnel in Sabah and Sarawak, Malaysia. n/iicrobio/ogy and Immunology (in press). Evengard, B. & Hagi, H. 1988 A filter paper technique for the detection of IgG and IgM class schistosome specific antibodies in an endemic area. Annals of Tropical Medicine and Parasitology 82, 307-309. Everette, E.D. & Nelson, R.A. 1975 Pulmonary melioidosis: Observations in 39 cases. American Review of Respiratory Diseases 112, 331-340. Farzadegan, H., Quinn, T. & Polk, B.F. 1987 Detecting antibodies to human immunodeficiency virus in dried blood on filter papers. Journal of Internal Diseases 155, 1073-1074. Howe, C., Sampath, A. & Spotnitz, M. 1971 The Pseudomallei Group: A Review. Journal of Infectious Diseases 124, 598-607. Ismail, G., Embi, M.N., Omar, O., Mohamed, R. & Razak, N. 1987a Toxigenic properties of Pseudomonas pseudomallei extracellular products. Tropical Biomedicine 4, 101-109. Ismail, G., Embi, M.N., Omar, O., Razak, N., Allen, J.C. & Smith, C.J. 1987b Enzyme immunoassay for the detection of antibody to Pseudomonas pseudomallei exotoxin in mice. FEMS Microbiology Letters 40, 27-31. Ismail, G., Embi, M.N., Omar, O., Razak, N., Allen, J.C. & Smith, C.J. 1987~ A competitive immunosorbent assay for the detection of Pseudomonas pseudomallei exotoxin. Journal of Medical Microbiology 23, 253-257.
World ~oournaiof Microbiology and Bmtechnology. Vol 9. 1993
95
N. En&
et al.
Ismail, G., Mohamed, R., Rohana, 1991 Antibody to Pseudomonas exposed to natural infection. 277-282.
S., Sharifah, pseudomallei Veferinary
S.M. & Embi, N. exotoxin in sheep A4icrobiolom 27,
Jackson, A.E., Moore, W.L. Jr. & Sanford J.P. 1972 Recrudescent melioidosis associated with diabetic ketoacidosis. Archives of Internal Medicine 130, 268-271. Khupulsup, K. & Petchclai, B. 1986 Application of indirect haemagglutination test and indirect fluorescent antibody test for IgM antibody for diagnosis of melioidosis in Thailand. American journal of Tropical Medicine and Hygiene 35, 36&369. Mackowiak, P.A. & Smith, J.W. 1978 Septicemic melioidosis: Occurrence following acute influenza A six years after exposure in Vietnam. Journal of the American Medical Association 240,
764-766. McLean, G. & Hulbink F. 1989 in ELISA for Brucella abortw.
Use of dried blood on filter paper journal of immunological Methods
123,39-43. Mohamed, R., Nathan, S., Embi, N., Razak, N. & Ismail, G. 1989 Inhibition of macromolecular synthesis in cultured macrophages by Pseudomonas pseudomallei exotoxin. Microbiology and Immunology 33, 811-819. Mohamed, R., Periasamy, M., Ismail, G. & Embi, N. I991 The exotoxin of melioidosis, a disease of animals and man. In Toxins and Targets, eds Walters, E. et al. pp. 110~1118. Engelwood Cliffs, NJ: Hardwood Academic. Nigg, C. 1963 Serological studies on subclinical melioidosis. ]ourm?l of Immunology 91, 18-28. Nigg, C. & Johnson, M. 1961 Complement fixation tests in experimental clinical and sub-clinical melioidosis. Journal of Bacferiology 82, 159-167.
96
WorldJoud of Microbdogy
and Btofechnology. Vol 9, 1993
Punyagupta, S. 1983 Melioidosis: the great imitator. Ramathibodi Medical ]ournal 6, 147-159. Rode, J.W. & Webling, D.D.A. 1981 Melioidosis in the Northern Territory of Australia. Medical Journal of Australia 1, 181-184. Sanford, J.P. & Moore, W.L. Jr. 1971 Recrudescent melioidosis: A South-east Asian legacy. American Review of Respiratory Diseases
104,452-457. Smith, C.J., Allen, J.C., Noor Embi M., Othman, O., Razak, N. & Ismail, G. 1987 Human Melioidosis: An Emerging Medical Problem. MlRCEN]oumal Applied Microbiology and Biotechnology
3,343-366. So, S.Y. 1985 Melioidosis-An overlooked problem in Hong Kong. H.K. Practitioners 3, 1111-1113. Strauss, J.M., Alexander, A.D., Rapmund, G., Gan, E. & Dorsey, A.E. 1969 Melioidosis in Malaysia: Antibodies to Pseudomonas pseudomallei in the human population. American Journal of Tropical Medicine and Hygiene 18, 698-702. Thin, R.N.F., Brown, M., Stewart, J.B. & Garrett, CJ. 1970 Melioidosis: a report of ten cases. Quarterly Journal of Medicine 39, 115-121. Thisyakom, U. 1986 Melioidosis in children at Children’s Hospital, Bangkok. Southeast Asian ]ournal of Tropical Medicine and Public Health 17, IOI-103. Zhuang, H., Coulepsis, A.G., Locamini, S.A., & Gust, I.D. 1982 Detection of markers of hepatitis B infection in sera dried on to filter paper: an application to field studies. Bulletin of the World Health Organization 60, 783-787.
(Received
in revised
form 6 July
1992;
accepted
17 ]uly
1992)
