Journal of Immunological Methods. 153 (1992) 67-71
67
© 1992 ElsevierScience Publishers B.V. All rights reserved 0022-1759/92/$05.0(I
JIM I)6387
Detection of lipopolysaccharide (LPS) and identification of its serotype by an enzyme-linked immunosorbent assay (ELISA) using poly-L-lysine K. Takahashi, M. Fukada, M. Kawai and T. Yokoehi Department of Microbiology, Aichi Medical Unirersity School of Medicine. Nagakute, Aichi 480-1I, Japan
(Received 13 January 1992, revised received31)March 1992,accepted 31)March 1992)
A new solid-phase enzyme-linked immunosorbent assay (ELISA) was developed for detection of LPS and identification of its serotype with antisera. Since LPS binds poorly to polystyrene microplates, precoating with poly-L-lysine was used before coating LPS on the surface of microplates. The small amount of LPS in complex mixtures (i.e., less than 1 t t g / m l ) could be detectable in ELISA. Use of poly-L-lysine with high molecular weight (MW) provided a higher sensitivity than poly-L-lysine with low MW. Precoating with polymyxin B, or poly-c-histidine was less effective in the sensitivity than precoating with poly-L-lysine, but it was still better than no precoating. The newly developed ELISA technique could be also applied for detection of anti-LPS antibodies in sera or for screening of monoclonal anti-LPS antibody. Key words: Lipopolysaccharide;EL1SA;Poly-L-lysine;PolymyxinB
Introduction The adequate quantification of LPS and identification of its serotype has been done by passive hemagglutination and passive hemolysis. But more recently ELISA have been described for the measurement of antibodies against different parts of LPS. However, LPS and free lipid A do not coat plastic surfaces well in this form. Galanos et al. (1972) showed that core antigens of LPS,
Correspondence to: T. Yokochi, Department of Microbiology, Aichi Medical UniversitySchool of Medicine, Nagakute, Aichi 480-11, Japan. Tel.: 0561-62-3341, ext. 2269; Fax: 056162-3348. Abbreciations: ELISA, enzyme-linked immunosorbent assay; LPS, lipopolysaccharide; MW, molecular weight; OD, optical density; T-PBS, PBS containing0.05% Tween 20.
complexed to bovine serum albumin (BSA) are biologically fully active. Appeimelk et al. (1985) applied the core region of LPS-BSA complexes for solid-phase coating in ELISA since BSA binds to plastic surfaces. There are several other reports on ELISA for detection of LPS (Scott et al., 1987; Peborde et al. 1989). This paper reports a simple and easy ELISA for detection of LPS and identification of its serotype in a single experimental set up using poly-L-lysine before coating of LPS.
Materials and methods Reagents
Poly-L-iysine with high MW ranging from 150,000 to 300,000, and with low MW ranging
from 1000 to 4000, polymyxin B, poly-L-glutamic acid (MW, 15,000-50,000), and poly-L-histidine (MW, 15,000-50,000) were purchased from Sigma Chemicals, St. Louis, MO, USA. 0.5
LPS Klebsiella 0 3 LPS was prepared from Kiebsiella pneumoniae LEN-1 ( 0 3 : K1-) by the phenol water method (Westphal et al., 1965; Yokochi et al., 1989). The LPS preparations of Escherichia coli 055, Ol11, and O128, and Sabnonella enteritidis were obtained from Difco Laboratories, Detroit, MI. The free lipid A fraction was isolated from Klebsiella 0 3 LPS by heating at 100°C for 1 h in 1% acetic acid (Kato et al., 1985).
Antibody Rabbit antiserum to O antigen of Klebsiella 0 3 LPS was prepared according to the procedure described by Edwards and Ewing (1962). Rabbit anti-lipid A serum was also prepared according to the procedure described by Galanos et ai. (1971) using the lipid A fraction extracted from Klebsiella 0 3 LPS. A panel of antisera to O antigens of Klebsiella LPS were supplied by Dr. N. Kato at Nagoya University.
ELISA The technique described by Engvall and Perlmann (1972) was used as the basis for the assay. Below are the steps used in our modified ELISA. Precoating of poly-1.-lysine. A solution (100 p.1) of poly-L-lysine in 0.01 M phosphate-buffered saline at pH 7.2 (PBS) was placed in disposable polystyrene microplates (Nunc lmmunoplate, InterMed, Denmark). The solution was incubated overnight at room temperature. A solution of polymyxin B, poly-L-histidine, or poly-L-glutamic acid (10 p.g/ml) was also used for precoating. The negative control well was treated with PBS alone. Binding of LPS. Aliquots (100 p.l) of various concentrations of LPS suspended in PBS were placed in poly-L-lysine-precoated plates, and then incubated for 1 h at 37°C. The plates were washed three times with PBS containing 0.05% Tween 20 (T-PBS). Antiserum. An optimum dilution (1/250) of rabbit immune serum (100/.tl) was added to each
0
..........................
10~2
10 -1
100
101
102
Concentration (/Jg/ml) Fig. 1. Optimizationof precoatingwith poly-L-lysinein ELISA. Plates were precoated with varyingconcentrations of poly-klysine with high MW, coated with Klebsiella 03 LPS or irrelevant LPS at 10/.tg/ml, and then treated with antisera to Klebsiella 03 LPS. The OD unit was read relative to that of irrelevant LPS. The OD values of the wells coated directly with LPS and the LPS blank wells were less than 0.01 OD unit.
well, incubated for 1 h at 37°C, and washed three times with T-PBS. Conjugate. Aiiquots (100/~1) of a 1/500 dilution of peroxidase-conjugated protein A (EY Labs, San Mateo, USA) were added to each well, incubated for 1 h at 37°C, and washed three times with T-PBS. Substrate. The peroxidase substrate solution containing 0.01% o-phenylene diamine and 0.03% H 2 0 2 (100/~1) was added to each well. Reactions were stopped about 30 min later by 1 N H2SO4. The optical density (OD) unit was read relative to a LPS blank or irrelevant LPS at 492 nm on an automatic microplate reader (MTP-32, Corona Electric Co., Japan). Less than 0.02 OD unit was judged as the non-specific background.
Results
Optimization of precoating with poly-L-lysine The optimization of precoating with poly-Llysine was examined (Fig. 1). Microplates were precoated with varying concentrations of poly-Llysine with MW ranging from 150,000 to 300,000, and then coated with Klebsiella 0 3 LPS at 10 p.g/ml. Definite reactions were detectable at the poly-L-lysine concentration more than 1 p.g/ml.
The maximum response was obtained with a polyL-lysine concentration of more than 10 /~g/ml. Therefore, precoating of microplates with poly-Liysine at 10 p,g/ml was used for further experiments. No significant reactions were detectable on the wells that LPS was directly coated up to 100 p,g/mi and on the LPS blank well in precoated microplates. Irrelevant negative control antisera did not show significant non-specific backgrounds, either. Their OD values were less than 0.01, Next, precoating with poly-L-lysine with high MW (150,000-300,000) or low MW (10004000) were compared for binding of LPS. Poly-Llysine with high MW gave higher OD values than that with lower MW. The differences were statistically significant (data not shown).
Detection of LPS and identification of the serotype in ELISA Quantitative detection of LPS was examined when poly-L-lysine was precoated at the given concentration (10 p,g/ml)(Fig. 2) Precoating with poly-L-lysine at 10 p,g/ml could detect less than 1 p,g/ml of LPS. Subsequently, we tried to identify the serotypes of various kinds of LPS using the ELISA technique. Antisera to the O antigen of Klebsiella 0 3 LPS specifically reacted with Klebsiella 0 3 LPS, whereas the antisera did not detect the other LPS, such as E. coil 055, Ol11, O128 and S. enteritidis. Antisera to various O antigens
o ~
1f 0
0.5
Oo-__, ..... 10 : -~
100
101
102
Concentration (pglml) Fig. 3. Comparison in detection of LPS on microplatesprecoated with poly-L-lysinewith the high MW (e), polymyxinB (ll), poly-L-hystidine(-). and poly-L-glutamicacid. Precoated microplateswere coated with Klebsiella03 LPS, and treated with antisera. The OD unit was read relative to irrelevant LPS on poly-L-lysine-precoatedmicroplates.No significant reactions were detected on microplates precoated with poly-L-glutamicacid.
of Klebsiella LPS could not crossreact with Klebsiella 0 3 LPS (data not shown).
Comparison of carious reagents for precoating in ELISA Precoating with polymyxin B, poly-L-histidine, poly-L-glatamic acid, and bovine serum albumin were tested for binding of LPS in ELISA. The results are shown in Fig. 3. Poly-L-lysine gave the highest OD values in ELISA. Precoating with
1f
0.5 os
00---2" "'1~'-1 10 2
10 1
100
101
100
101
102
102
Concentration ( p g/ml) Fig. 2. Quantitative detection of LPS on microplates precoated with poly-L-lysineat l0 /zg/ml in ELISA. Precoated microplates were coated with various concentrationsof Klebdella 03 LPS, and treated with antisera. The OD unit was read relative to the LPS blank and irrelevant LPS.
Concentration ( p g/ml) Fig. 4. Detection of the free lipid A in ELISA. Microplates were precoated with poly-L-lysine(e) or polymyxin B (ll), cc~atedwith the free lipid A, and then treated with anti-lipid A antibody. The OD unit was read relative to the lipid A blank.
poly-t.-histidine, or polymyxin B was less efficient in detection of LPS than precoating with poly-Llysine with low MW. Reactions were not detectable in precoating at any concentrations of poly-L-glutamic acid and bovine serum albumin (data not shown). Detection o f the free lipid A in ELISA it was tested whether or not our ELISA method could detect the free lipid A fraction prepared by acid treatment (Fig. 4) The free lipid A obtained from Klebsiella 0 3 could be detected by anti-lipid A antibody in ELISA, although precoating with polymyxin B gave higher O D units than that with poly-L-lysine.
follow them. As compared with the complexity of previously reported ELISA methods, this new technique is a very simple and easy method with good sensitivity for detection of LPS and anti-LPS antibody, Furthermore, this technique would be able to detect rough LPS because our ELISA could detect free lipid A as well as smooth LPS. However, precoating with polymyxin B seemed to be preferable in the case of the free lipid A.
Acknowledgements This work was supported in part by a grant from the Ministry of Education, Science and Culture.
Discussion References The present study demonstrates that ELISA using poly-L-lysine as a precoating reagent before coating of LPS was useful for detection of LPS and anti-LPS antibody. This technique provided much higher sensitivity on detection of LPS because ELISA with no precoating could not detect LPS. There are several reports on ELISA improved for detection of LPS or an,i-LPS antibody, since LPS does not coat the surfaces of plastic microplate wells. Appelmeik et al. (1985) reported that LPS-BSA complexes were used for the coating of ELISA plates. Further, Pebode et al. (1989) developed an inhibition ELISA technique using LPS-BSA complexes. Our ELISA technique does not require glutaraidehyde activation for preparing LPS-BSA complexes. The use of LPS-polymyxin B complexes was also reported by Scott and Barclay (1987). They prepared LPSpolymyxin B complexes, dialyzed, and then coat the complexes on microplates. But we showed that it was unnecessary to make LPS-polymyxin B complexes before coating. Moreover, our ELISA technique using poly-L-lysine was more sensitive than that using polymyxin B. The ELISA using Mg 2+ (Ito et ai., 1980) or whole bacteria (Luk et al., 1990; Nelson et al, 1990) for detection of LPS were also reported. Although there were several reports on conventional ELISA technique for detection of LPS (Bruins e t a ! . , 1978; Eskenazy et al., 1982; Gagliardi et al. 1986), we could not
Appelmelk, B.J., Verweij-Van Vught, A.MJ.J., MacLaren, D.M. and Thijs. L.G. (1985) An enzyme-linked immunosorbent assay (EL1SA) for the measurement of antibodies to different parts of the Gram-negativelipopolysaccharide core region. J. Immunol. Methods 82, 199. Bruins, S.C., lngwer, !., Zeckel, M.L. and White, A.C. (1978) Parameters affecting the enzyme-linked immunosorbent assay of immunoglobulinG antibody to a rough mutant of Salmonella minnesota. Infect. Immun. 21,721. Edwards, P.R. and Ewing, W.H. (1962) Identification of Enterobacteriaceae, 2rid edn., Burgess Publishing Co. Minneapolis, MN, pp, 214-215. Engvall, E. and Perlmann, P. (1972) Enzyme-linked immunosorbent assay, Elisa. 3. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes. J. lmmunol. 109, 129. Eskenazy, M., Naumova, F., Tekelieva, R. and Konstantinov, G. (1982) Quantitation of rabbit immunoglobufinG antibodies to Salmonella minnesota Re by enzyme-linked immunosorbent assay. J. Clin. Microbiol. 16, 276. Gagliardi, N.C., Nolan, J.P., Feind, D.M. and DeLissio, M. (1986) A rapid sensitive monoclonal assay for lipid A in solution. J. lmmunol. Methods 91,243. Galanos, C., Luderitz, O. and Westphal, O. (1971) Preparation and properties of antisera against the lipid A component of bacterial lipopolysaccharide.Eur. J. Biochem. 24, 116. Galanos, C., Rietschel, O. Luderitz, O. and Westphal, O. (1972) Biological activities of lipid A complexed with bovine-serum albumin. Eur. J. Biochem.31,230. Ito, J.l., Wunderlich, A.C., Lyons, J., Davis, C.E., Guiney, D.G. and Braude, A.I. (1980) Role of magnesium in the enzyme-linked immunosorbent assay for lipopolysaccharides of rough Escherichia coli strain J5 and Neisseria gonorrhea. J. Infect. Dis. 142, 532.
Kate, N., Kido, N. Ohta, M. and Naito, S. (1985) Comparative studies on adjuvanticity of Klebsiella 0 3 lipopolysaccharide and its |ipld A and polysaccharide fractions. Immunology 54, 317-324. Luk, J.M.C., Nnalue, N.A. and Lindberg, A.A. (1990) Efficient production of mouse and rat monoclonal antibodies against the O antigens of Salmonella serogroup C I, using LPS-coated bacteria as immunogen. J. Immunol. Methods 129, 243. Nelson, D., Neill, W. and Poxton, l.R, (1990) A comparison of immunoblotting, flow cytometry and ELISA to monitor the binding of anti-lipopolysaccharide monoclonal antibodies. J. lmmunol. Methods 133, 227. Peborde, J.P., Laporte, M. and Samain, D. (1989) An inhibition enzyme-linked immunosorbent assay technique for
the detection of endotoxins in proteins extracted from Escherichia colt KI2 recombinant DNA. J. Immunol. Methods 120, 259. Scott. B.B. and Barclay, G.R. (1987) Endotoxin.polymyxin complexes in an improved enzyme-linked immunosorbent assay for IgG antibodies in blood donor ~ r a to Gramnegative endotoxin core glycolipids. Vex Sang. 52, 272. Westphal, O. and .Iann, K. (1%5) Baclcr;al lipop~lysaccharides - Extraction with phenol water and further applicalion of the procedure. In: R.L. Wisller (Ed.), Methods in Carbohydrate Chemistry, Vol. 5. Academic Press, New York, pp. 83-91. Yokochi, T.. Inoue, Y., Yokoo, J., Kimura, Y. and Kate, N. (1989) Retention of bacterial lipopolysaccharide at lb.: :;ire of subcutaneous injection. Infect. lmmun, 57, 1786.
67
© 1992 ElsevierScience Publishers B.V. All rights reserved 0022-1759/92/$05.0(I
JIM I)6387
Detection of lipopolysaccharide (LPS) and identification of its serotype by an enzyme-linked immunosorbent assay (ELISA) using poly-L-lysine K. Takahashi, M. Fukada, M. Kawai and T. Yokoehi Department of Microbiology, Aichi Medical Unirersity School of Medicine. Nagakute, Aichi 480-1I, Japan
(Received 13 January 1992, revised received31)March 1992,accepted 31)March 1992)
A new solid-phase enzyme-linked immunosorbent assay (ELISA) was developed for detection of LPS and identification of its serotype with antisera. Since LPS binds poorly to polystyrene microplates, precoating with poly-L-lysine was used before coating LPS on the surface of microplates. The small amount of LPS in complex mixtures (i.e., less than 1 t t g / m l ) could be detectable in ELISA. Use of poly-L-lysine with high molecular weight (MW) provided a higher sensitivity than poly-L-lysine with low MW. Precoating with polymyxin B, or poly-c-histidine was less effective in the sensitivity than precoating with poly-L-lysine, but it was still better than no precoating. The newly developed ELISA technique could be also applied for detection of anti-LPS antibodies in sera or for screening of monoclonal anti-LPS antibody. Key words: Lipopolysaccharide;EL1SA;Poly-L-lysine;PolymyxinB
Introduction The adequate quantification of LPS and identification of its serotype has been done by passive hemagglutination and passive hemolysis. But more recently ELISA have been described for the measurement of antibodies against different parts of LPS. However, LPS and free lipid A do not coat plastic surfaces well in this form. Galanos et al. (1972) showed that core antigens of LPS,
Correspondence to: T. Yokochi, Department of Microbiology, Aichi Medical UniversitySchool of Medicine, Nagakute, Aichi 480-11, Japan. Tel.: 0561-62-3341, ext. 2269; Fax: 056162-3348. Abbreciations: ELISA, enzyme-linked immunosorbent assay; LPS, lipopolysaccharide; MW, molecular weight; OD, optical density; T-PBS, PBS containing0.05% Tween 20.
complexed to bovine serum albumin (BSA) are biologically fully active. Appeimelk et al. (1985) applied the core region of LPS-BSA complexes for solid-phase coating in ELISA since BSA binds to plastic surfaces. There are several other reports on ELISA for detection of LPS (Scott et al., 1987; Peborde et al. 1989). This paper reports a simple and easy ELISA for detection of LPS and identification of its serotype in a single experimental set up using poly-L-lysine before coating of LPS.
Materials and methods Reagents
Poly-L-iysine with high MW ranging from 150,000 to 300,000, and with low MW ranging
from 1000 to 4000, polymyxin B, poly-L-glutamic acid (MW, 15,000-50,000), and poly-L-histidine (MW, 15,000-50,000) were purchased from Sigma Chemicals, St. Louis, MO, USA. 0.5
LPS Klebsiella 0 3 LPS was prepared from Kiebsiella pneumoniae LEN-1 ( 0 3 : K1-) by the phenol water method (Westphal et al., 1965; Yokochi et al., 1989). The LPS preparations of Escherichia coli 055, Ol11, and O128, and Sabnonella enteritidis were obtained from Difco Laboratories, Detroit, MI. The free lipid A fraction was isolated from Klebsiella 0 3 LPS by heating at 100°C for 1 h in 1% acetic acid (Kato et al., 1985).
Antibody Rabbit antiserum to O antigen of Klebsiella 0 3 LPS was prepared according to the procedure described by Edwards and Ewing (1962). Rabbit anti-lipid A serum was also prepared according to the procedure described by Galanos et ai. (1971) using the lipid A fraction extracted from Klebsiella 0 3 LPS. A panel of antisera to O antigens of Klebsiella LPS were supplied by Dr. N. Kato at Nagoya University.
ELISA The technique described by Engvall and Perlmann (1972) was used as the basis for the assay. Below are the steps used in our modified ELISA. Precoating of poly-1.-lysine. A solution (100 p.1) of poly-L-lysine in 0.01 M phosphate-buffered saline at pH 7.2 (PBS) was placed in disposable polystyrene microplates (Nunc lmmunoplate, InterMed, Denmark). The solution was incubated overnight at room temperature. A solution of polymyxin B, poly-L-histidine, or poly-L-glutamic acid (10 p.g/ml) was also used for precoating. The negative control well was treated with PBS alone. Binding of LPS. Aliquots (100 p.l) of various concentrations of LPS suspended in PBS were placed in poly-L-lysine-precoated plates, and then incubated for 1 h at 37°C. The plates were washed three times with PBS containing 0.05% Tween 20 (T-PBS). Antiserum. An optimum dilution (1/250) of rabbit immune serum (100/.tl) was added to each
0
..........................
10~2
10 -1
100
101
102
Concentration (/Jg/ml) Fig. 1. Optimizationof precoatingwith poly-L-lysinein ELISA. Plates were precoated with varyingconcentrations of poly-klysine with high MW, coated with Klebsiella 03 LPS or irrelevant LPS at 10/.tg/ml, and then treated with antisera to Klebsiella 03 LPS. The OD unit was read relative to that of irrelevant LPS. The OD values of the wells coated directly with LPS and the LPS blank wells were less than 0.01 OD unit.
well, incubated for 1 h at 37°C, and washed three times with T-PBS. Conjugate. Aiiquots (100/~1) of a 1/500 dilution of peroxidase-conjugated protein A (EY Labs, San Mateo, USA) were added to each well, incubated for 1 h at 37°C, and washed three times with T-PBS. Substrate. The peroxidase substrate solution containing 0.01% o-phenylene diamine and 0.03% H 2 0 2 (100/~1) was added to each well. Reactions were stopped about 30 min later by 1 N H2SO4. The optical density (OD) unit was read relative to a LPS blank or irrelevant LPS at 492 nm on an automatic microplate reader (MTP-32, Corona Electric Co., Japan). Less than 0.02 OD unit was judged as the non-specific background.
Results
Optimization of precoating with poly-L-lysine The optimization of precoating with poly-Llysine was examined (Fig. 1). Microplates were precoated with varying concentrations of poly-Llysine with MW ranging from 150,000 to 300,000, and then coated with Klebsiella 0 3 LPS at 10 p.g/ml. Definite reactions were detectable at the poly-L-lysine concentration more than 1 p.g/ml.
The maximum response was obtained with a polyL-lysine concentration of more than 10 /~g/ml. Therefore, precoating of microplates with poly-Liysine at 10 p,g/ml was used for further experiments. No significant reactions were detectable on the wells that LPS was directly coated up to 100 p,g/mi and on the LPS blank well in precoated microplates. Irrelevant negative control antisera did not show significant non-specific backgrounds, either. Their OD values were less than 0.01, Next, precoating with poly-L-lysine with high MW (150,000-300,000) or low MW (10004000) were compared for binding of LPS. Poly-Llysine with high MW gave higher OD values than that with lower MW. The differences were statistically significant (data not shown).
Detection of LPS and identification of the serotype in ELISA Quantitative detection of LPS was examined when poly-L-lysine was precoated at the given concentration (10 p,g/ml)(Fig. 2) Precoating with poly-L-lysine at 10 p,g/ml could detect less than 1 p,g/ml of LPS. Subsequently, we tried to identify the serotypes of various kinds of LPS using the ELISA technique. Antisera to the O antigen of Klebsiella 0 3 LPS specifically reacted with Klebsiella 0 3 LPS, whereas the antisera did not detect the other LPS, such as E. coil 055, Ol11, O128 and S. enteritidis. Antisera to various O antigens
o ~
1f 0
0.5
Oo-__, ..... 10 : -~
100
101
102
Concentration (pglml) Fig. 3. Comparison in detection of LPS on microplatesprecoated with poly-L-lysinewith the high MW (e), polymyxinB (ll), poly-L-hystidine(-). and poly-L-glutamicacid. Precoated microplateswere coated with Klebsiella03 LPS, and treated with antisera. The OD unit was read relative to irrelevant LPS on poly-L-lysine-precoatedmicroplates.No significant reactions were detected on microplates precoated with poly-L-glutamicacid.
of Klebsiella LPS could not crossreact with Klebsiella 0 3 LPS (data not shown).
Comparison of carious reagents for precoating in ELISA Precoating with polymyxin B, poly-L-histidine, poly-L-glatamic acid, and bovine serum albumin were tested for binding of LPS in ELISA. The results are shown in Fig. 3. Poly-L-lysine gave the highest OD values in ELISA. Precoating with
1f
0.5 os
00---2" "'1~'-1 10 2
10 1
100
101
100
101
102
102
Concentration ( p g/ml) Fig. 2. Quantitative detection of LPS on microplates precoated with poly-L-lysineat l0 /zg/ml in ELISA. Precoated microplates were coated with various concentrationsof Klebdella 03 LPS, and treated with antisera. The OD unit was read relative to the LPS blank and irrelevant LPS.
Concentration ( p g/ml) Fig. 4. Detection of the free lipid A in ELISA. Microplates were precoated with poly-L-lysine(e) or polymyxin B (ll), cc~atedwith the free lipid A, and then treated with anti-lipid A antibody. The OD unit was read relative to the lipid A blank.
poly-t.-histidine, or polymyxin B was less efficient in detection of LPS than precoating with poly-Llysine with low MW. Reactions were not detectable in precoating at any concentrations of poly-L-glutamic acid and bovine serum albumin (data not shown). Detection o f the free lipid A in ELISA it was tested whether or not our ELISA method could detect the free lipid A fraction prepared by acid treatment (Fig. 4) The free lipid A obtained from Klebsiella 0 3 could be detected by anti-lipid A antibody in ELISA, although precoating with polymyxin B gave higher O D units than that with poly-L-lysine.
follow them. As compared with the complexity of previously reported ELISA methods, this new technique is a very simple and easy method with good sensitivity for detection of LPS and anti-LPS antibody, Furthermore, this technique would be able to detect rough LPS because our ELISA could detect free lipid A as well as smooth LPS. However, precoating with polymyxin B seemed to be preferable in the case of the free lipid A.
Acknowledgements This work was supported in part by a grant from the Ministry of Education, Science and Culture.
Discussion References The present study demonstrates that ELISA using poly-L-lysine as a precoating reagent before coating of LPS was useful for detection of LPS and anti-LPS antibody. This technique provided much higher sensitivity on detection of LPS because ELISA with no precoating could not detect LPS. There are several reports on ELISA improved for detection of LPS or an,i-LPS antibody, since LPS does not coat the surfaces of plastic microplate wells. Appelmeik et al. (1985) reported that LPS-BSA complexes were used for the coating of ELISA plates. Further, Pebode et al. (1989) developed an inhibition ELISA technique using LPS-BSA complexes. Our ELISA technique does not require glutaraidehyde activation for preparing LPS-BSA complexes. The use of LPS-polymyxin B complexes was also reported by Scott and Barclay (1987). They prepared LPSpolymyxin B complexes, dialyzed, and then coat the complexes on microplates. But we showed that it was unnecessary to make LPS-polymyxin B complexes before coating. Moreover, our ELISA technique using poly-L-lysine was more sensitive than that using polymyxin B. The ELISA using Mg 2+ (Ito et ai., 1980) or whole bacteria (Luk et al., 1990; Nelson et al, 1990) for detection of LPS were also reported. Although there were several reports on conventional ELISA technique for detection of LPS (Bruins e t a ! . , 1978; Eskenazy et al., 1982; Gagliardi et al. 1986), we could not
Appelmelk, B.J., Verweij-Van Vught, A.MJ.J., MacLaren, D.M. and Thijs. L.G. (1985) An enzyme-linked immunosorbent assay (EL1SA) for the measurement of antibodies to different parts of the Gram-negativelipopolysaccharide core region. J. Immunol. Methods 82, 199. Bruins, S.C., lngwer, !., Zeckel, M.L. and White, A.C. (1978) Parameters affecting the enzyme-linked immunosorbent assay of immunoglobulinG antibody to a rough mutant of Salmonella minnesota. Infect. Immun. 21,721. Edwards, P.R. and Ewing, W.H. (1962) Identification of Enterobacteriaceae, 2rid edn., Burgess Publishing Co. Minneapolis, MN, pp, 214-215. Engvall, E. and Perlmann, P. (1972) Enzyme-linked immunosorbent assay, Elisa. 3. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes. J. lmmunol. 109, 129. Eskenazy, M., Naumova, F., Tekelieva, R. and Konstantinov, G. (1982) Quantitation of rabbit immunoglobufinG antibodies to Salmonella minnesota Re by enzyme-linked immunosorbent assay. J. Clin. Microbiol. 16, 276. Gagliardi, N.C., Nolan, J.P., Feind, D.M. and DeLissio, M. (1986) A rapid sensitive monoclonal assay for lipid A in solution. J. lmmunol. Methods 91,243. Galanos, C., Luderitz, O. and Westphal, O. (1971) Preparation and properties of antisera against the lipid A component of bacterial lipopolysaccharide.Eur. J. Biochem. 24, 116. Galanos, C., Rietschel, O. Luderitz, O. and Westphal, O. (1972) Biological activities of lipid A complexed with bovine-serum albumin. Eur. J. Biochem.31,230. Ito, J.l., Wunderlich, A.C., Lyons, J., Davis, C.E., Guiney, D.G. and Braude, A.I. (1980) Role of magnesium in the enzyme-linked immunosorbent assay for lipopolysaccharides of rough Escherichia coli strain J5 and Neisseria gonorrhea. J. Infect. Dis. 142, 532.
Kate, N., Kido, N. Ohta, M. and Naito, S. (1985) Comparative studies on adjuvanticity of Klebsiella 0 3 lipopolysaccharide and its |ipld A and polysaccharide fractions. Immunology 54, 317-324. Luk, J.M.C., Nnalue, N.A. and Lindberg, A.A. (1990) Efficient production of mouse and rat monoclonal antibodies against the O antigens of Salmonella serogroup C I, using LPS-coated bacteria as immunogen. J. Immunol. Methods 129, 243. Nelson, D., Neill, W. and Poxton, l.R, (1990) A comparison of immunoblotting, flow cytometry and ELISA to monitor the binding of anti-lipopolysaccharide monoclonal antibodies. J. lmmunol. Methods 133, 227. Peborde, J.P., Laporte, M. and Samain, D. (1989) An inhibition enzyme-linked immunosorbent assay technique for
the detection of endotoxins in proteins extracted from Escherichia colt KI2 recombinant DNA. J. Immunol. Methods 120, 259. Scott. B.B. and Barclay, G.R. (1987) Endotoxin.polymyxin complexes in an improved enzyme-linked immunosorbent assay for IgG antibodies in blood donor ~ r a to Gramnegative endotoxin core glycolipids. Vex Sang. 52, 272. Westphal, O. and .Iann, K. (1%5) Baclcr;al lipop~lysaccharides - Extraction with phenol water and further applicalion of the procedure. In: R.L. Wisller (Ed.), Methods in Carbohydrate Chemistry, Vol. 5. Academic Press, New York, pp. 83-91. Yokochi, T.. Inoue, Y., Yokoo, J., Kimura, Y. and Kate, N. (1989) Retention of bacterial lipopolysaccharide at lb.: :;ire of subcutaneous injection. Infect. lmmun, 57, 1786.
