Journal of Clinical LaboratoryAnalysis 6:232-238 (1992)
A New Sensitive Microplate Assay of Plasma Endotoxin Hiroshi Tamura,’ Shigenori Tanaka,’ Taminori Obayashi,2 Minoru Y ~ s h i d a , ~ and Tadashi Kawai2
’Tokyo Research Institute, Seikagaku Corporation, Tokyo, Japan; 2Department of Clinical Pathology, Jichi Medical School, Tochigi, Japan; 3Division of Hematology, Department of Medicine, Jichi Medical School, Tochigi, Japan We have developed a microplate method for determining endotoxin in platelet-rich plasma-using Endospecy, an endotoxinspecific chromogenic Limulus test reagent. Nonspecific activators and inhibitors of the test were eliminated by exposing samples (5 FI) to the alkali reagent consisting of KOH, CaCI,, Triton X-1 00, ethyleniminepolymer and N,N-bis(2-hydroxyethyl)glycine.The recoveries of various endotoxins were almost complete and not enhanced by dilution. The doseresponse curve was linear over endotoxin concentrationsof 2-400 pg/ml with good precision (C.V. - 4 . 0 % ) . Normal human plasmas
(n =30)contained - 4 . 0 pg/ml of endotoxin in reference to that of Escherichia coliolll: 84. All plasma samples with high concentration of endotoxin by a conventional method showed high values by the microplate assay as well. Since it does not require centrifugation, the new treatment allows the whole reactions to proceed on the same microplate. This permits us to apply the Limulus test to an automated assay system, making plasma endotoxin determination simpler and more rapid than a conventional test tube method.
Limulus test, amebocyte lysate, chromogenic substrate, platelet-rich plasma (PRP),
Limulus gelation test was invented by Levin et al. (1) to detect endotoxin in various body fluids based on the finding that a trace amount of endotoxin coagulates amebocyte lysate of Limulus polyphemus. Iwanaga et al. (2) elucidated the biochemical mechanism of the limulus coagulation and paved the way for a quantitative chromogenic assay of endotoxin. The specificity of the test to endotoxin, once compromised by the discovery of the coexistence of factor G in the lysate that is highly sensitive to ( I +3)-P-D-ghcan (3), has been restored by nullifying the activity of factor G in the lysate ( 4 3 . In this study, we describe a new method for treating plateletrich plasma (PRP) to determine plasma endotoxin on a microplate with Endospecy (Seikagaku Corp., Tokyo), an endotoxin-specific chromogenic limulus test reagent prepared from amebocyte lysate of Tachypleus tridentatus. MATERIALS AND METHODS Glassware, Reagents, and Endotoxin Preparations All glassware was made endotoxin-free by heating at 250°C for 2 h. Reagents were made endotoxin-free by autoclaving at 121°C for 20-90 min. Samples and reagents were dispensed using an endotoxin- and P-glucan-free, disposable Toxipet sampler tip and Toxipet dispenser syringe (both from Seikagaku Corp.). 0 1992 Wiley-Liss, Inc.
The following 12 kinds of Westphal-type endotoxin solutions were prepared with distilled water: Escherichia coli 01 11:B4, E . coli 055:B5, E . coli 0127:B8, Shigellapexneri (Difco Laboratories, Inc., Detroit, MI); Salmonella typhimurium, S . enteritidis, Serratia marcescens (Sigma Chemical Co., St. Louis, MO); Pseudomonas aeruginosa (List Biological Laboratories, Campbell, CA); Proteus vulgaris, Enterobacter cloacae (Ribi Immunochem Research, Inc., Mont.); E. coli UKT-B (Japanese standard endotoxin, National Institute Hygienic Sciences, Osaka Branch, Osaka, Japan); and Salmonella abortus equi (Novo-Pyrexal endotoxin standard NP1, Pyroquant Diagnostik GmbH, Walldolf, FRG). The addition of endotoxin to PRP was carried out with the latter placed on melting ice throughout the experiments to prevent its inactivation by plasma factors. Endotoxin was assayed by incubating samples with Endospecy on an endotoxin- and P-glucan-free, disposable 96well microplate (Toxipet plate 96F; Seikagaku Corp.) kept warm on a dry microplate-incubator (Hotplate CT-961, Seikagaku Corp.).
Received July 26, 1991; accepted October 10, 1991 Address reprint requests to Dr. Hiroshi Tamura, Tokyo Research Institute, Seikagaku Corporation, 3-1253 Tateno, Higashiyarnatoshi, Tokyo, 207 Japan.
Microplate Assay of Plasma Endotoxin
Effects of Sample Pretreatment With Triton X-100 in KOH-CaCI, on Addition-Recovery Rate PRP was prepared by centrifuging heparinized blood (5 U/ml of blood) from healthy donors at 150g for 10 min. A 10-pl portion of E. coli 01 11:B4 (4 ng/ml), E . coli UKT-B (1 ng/ml), or S. abortus equi (2 ng/ml) endotoxin solution was added to three normal PRP samples (190 p1).A 5-pl aliquot of endotoxin-spiked PRP was dispensed in a well of a Toxipet plate 96E To each well was added 20 pl of 0.1 mol/L KOH containing 0.01 mol/L CaCI2 and various concentrations of Triton X-100 from 0.01 to 0.5% (w/v). The microplate was warmed at 37°C for 10 min on a Hotplate CT-961. Then, 100 p1 of Endospecy dissolved in a 4.4-ml aliquot of 0.1 mol/L Tris-HC1 buffer (pH 8.0) was added, and the mixture was further incubated at 37°C for 30 min. Released p-nitroaniline was diazo-coupled by adding 50 pl each of 0.04% (w/v) sodium nitrite in 1.0 mol/L HCI, 0.3% (w/v) ammonium sulfamate, and 0.07% (w/v) N - 1-naphthylethylenediamine dihydrochloride in 14% (v/v) N-methyl-2-pyrrolidone sequentially. The absorbance was measured with a microplate reader (Wellreader SK-601, Seikagaku Corp.) at 545 and 630 nm simultaneously, the latter being as a reference. The recovery rate was calculated against an aqueous endotoxin solution as a control.
Effects of Ethyleniminepolymer on Endotoxin Recovery To a 5 - p l aliquot of endotoxin-spiked PRP prepared as above, a 2O-pl portion of 0.1 mol/L KOH-0.01 rnol/L CaC12-0. 1% Triton X- 100 containing various concentrations [0.05-0.5% (w/v)] of ethyleniminepolymer (EIP, mean MW = 70,000, Nacalai Tesque Inc., Kyoto), was added and the mixture was incubated at 37°C for 10 min. Then Endospecy was added to measure endotoxin, and the recovery was calculated as described above.
portion of an endotoxin spiked PRP, TP-BEC (20 pl) was added, and the mixture was incubated at 37°C for 10 min. Then Endospecy was added to measure endotoxin, and the recovery was calculated.
Dose-Response Curve Three endotoxins ( E . coli Olll:B4, E . coli UKT-B, and S. abortus equi) were tested. A 10-pl aliquot of an endotoxin solution was added to a 190-pl portion of three normal PRP samples to various concentrations ranging from 15.6 to 400 pg/ml. To a 5-p1 portion of an endotoxin-spiked PRP, a 2 0 - 4 aliquot of TP-BEC was added, and the mixture was incubated at 37°C for 10 min. Endospecy was added to measure endotoxin. Endotoxins dissolved in distilled water in corresponding concentrations were served as a control. The dose-response curves were plotted with endotoxin concentration and absorbance expressed on a bilogarithmic scale.
Effects of Dilution of Treated Samples on Endotoxin Recovery A 10-1.1.3 portion of E . coli 01 1 1:B4 endotoxin (1,000 ng/ml) was added to 190 pI of three normal PRP samples. A 5-pl portion of the endotoxin-spiked PRP was dispensed into a test tube, to which a 2 0 - 4 aliquot of TP-BEC was added. After incubation at 37°C for 10 min, the mixture was diluted 1: lo3 and 1 :lo4 with distilled water. The effects of 1:lO and 1: lo2 dilution were examined with PRP spiked with the endotoxin in a final concentration of 1 ng/ml. Nondiluted spiked PRP (100 pg/ml) was also tested in the same manner. As a control, the same series of dilutions were made by substituting distilled water for PRP. A 25-pI aliquot of each sample was dispensed in a microplate well, to which Endospecy was added to measure endotoxin. Recoveries were obtained by dividing the absorbance of PRP by that of distilled water with a corresponding concentration of endotoxin.
Effects of Incubation Time on Endotoxin Recovery To a 5-p1 aliquot of endotoxin-spiked PRP prepared as above, a 2 0 - 4 portion of an alkali reagent [0.1% Triton X- 100-0.1 mol/L KOH-0.03 mol/L N,N-bis(2-hydroxyethyl)glycine (Bicine, Dojindo Laboratories, Kumamoto, Japan)0.07% EIP-0.01 moVL CaCI2] (TP-BEC) was added. Bicine, a dotite reagent, was added to the pretreatment mixture to prevent the formation of insoluble substance on standing. The mixture was incubated at 37°C for 0 , 5 , 10,20, and 30 min. Then Endospecy was added to measure endotoxin, and the recovery was calculated.
Addition-Recovery Test of Various Endotoxins Twelve different endotoxins described above were tested. Endotoxin was added to six normal PRP samples placed on melting ice in a final concentration of 100 pg/ml. To a 5-p1
Measurements on Clinical Samples PRPs from 30 healthy volunteers and 40 plasma samples that contained endotoxin by a conventional method (5) were treated with TP-BEC. Endotoxin concentrations were measured with Endospecy using E . coli 01 l 1:B4 endotoxin as a standard.
RESULTS Effects of Sample Pretreatment With Triton X-100 in KOH-CaCI, on Addition-Recovery Rate With all three endotoxins tested, the recovery was best when samples were treated with 0.1% Triton X-100. The recoveries of E . coli 01 11:B4 and UKT-B endotoxins were about loo%, while that of S. abortus equi was 60% (Fig. 1).
Tamura et al.
6 . c o l i UKT-El
H S.abortus equi
L u 0 >
Triton X-100 conc. I%(w/v)]
Fig. 1. Effect of Triton X-100 on endotoxin recoveries from human PRP.
L c o l i 0111:El4
t k S. abortus equi
M f SD, n=3 Y
EIP conc. [%(U/V)l
Fig. 2. Effect of the addition of ethyleniminepolymer to Triton X-100 on endotoxin recoveries from human PRP.
Microplate Assay of Plasma Endotoxin
1 ?! M ? SD, n=3
S.abortus equj equi
Incubation time ( m i n )
Fig. 3. Effect of incubation time on endotoxin recoveries.
Effects of EIP on Endotoxin Recovery
EIP enhanced the recovery of S. abortus equi up to 90% without affecting recoveries of the other two endotoxins (Fig. 2).
A nice linear relationship was noted between absorbance and endotoxin concentrations of the three endotoxins tested on a bilogarithmic scale. The regression lines of endotoxin in distilled water coincided neatly with those of corresponding endotoxin in PRP.The regression lines of different endotoxins were parallel one another (Fig. 4).
Effects of Incubation Time on Endotoxin Recovery The recoveries of all the three endotoxins were highest when samples were treated for 10 min (Fig. 3).
Addition-Recovery Test of Various Endotoxins The recoveries of all 12 endotoxins spiked to normal PRP were 90.2-1 15.9% (Table 1). TABLE 1. Recoveries of Endotoxins From Normal PRP After Alkali Reagent (TP-BEC) Pretreatment Endotoxha
Escherichia coli 01 1 I :B4 E. coli 055:B5 E. coli 0127:B8 E. coli UKT-B Salmonella typhimurium S. enteritidis S. abortus equi Shigella flexineri Serratia marcescens Pseudomonas aeruginosa Proteus vulgaris Enterobacter cloacae
105.0 f 5.1 107.7 2 5.8 90.7 f 7.4 96.5 f 4.7 111.2 2 7.6 90.2 f 6.7 92.6 f 5.1 98.8 f 4.5 114.1 f 6.8 105.9 ? 6.5 115.9 ? 9.8 100.2 -t 9.2
"Westphal-typeendotoxin solution were prepared with distilled water. bMean 5 SD, n = 6.
Effects of Dilution on Endotoxin Recovery Dilution up to 1:104 did not affect the recovery of E . coli O1 1:B4 added to PRP 2).
Measurements on Clinical Samples According to the present method (Fig. 5 ) , normal endotoxin values in PRP were 2.4 ? 1.O pg/ml (mean 2 SD, maximum 4.2 pg/ml, n = 30) in reference to E . coli 01 11:B4 endotoxin. The standard curve using E . coli 01 11:B4 endotoxin dissolved in distilled water was linear over endotoxin concentration from 2 to 400 pg/ml with good precision (C.V.