- Email: [email protected]
Comparison of the solid phase enzyme receptor assay (SPERA) and the microbiological assay for teicoplanin
Journal
of Hospital
Infection
(1986) 7 (Supplement
A), 85-89
Comparison of the solid phase enzyme assay (SPERA) and the microbiological teicoplanin Luigi
Cavenaghi,
Angelo
Corti and Giovanni
receptor assay for
Cassani
Merrell Dow Research Institute, Lepetit Research Centre, via Durando 38, 20158 Milano, Italy Summary: A solid phase enzyme receptor assay (SPERA) for teicoplanin has been developed and recently presented in a kit form. The assay relies on the competition between antibiotic present in the biological fluid and peroxidase labelled teicoplanin for albumin-E-amino-caproyl-D-alanyl-Dalanine (BSA-s-ACA-D-Ala-D-Ala), a synthetic analog of the biological receptor. The kit contains BSA-e-ACA-D-Ala-D-Ala coated microplates together with all the necessary reagents. The limit of detection of the SPERA is O.Smgl-‘. The microbiological assay for teicoplanin uses Bacillus subtilis ATCC 6633 as test organism and a high salt medium to ensure a low limit of detection of O.OSmgl-‘. The previously established correlation between the two methods has been confirmed using the kits on 280 serum and 216 urine samples with concentrations up to SOOmgl-‘. Introduction
Teicoplanin, a novel antibiotic (Parenti et al., 1978) currently undergoing phase 3 clinical investigation, has usually been assayed by a conventional microbiological assay with Bacillus subtilis as test organism. Recently we have developed a new method: the solid phase enzyme receptor assay (SPERA) which is based on the competition between the antibiotic present in the biological fluid and the peroxidase labelled teicoplanin I for albumin-s-aminocaproyl-D-alanyl-D-alanine (BSA-&-ACA-D-Ala-D-Ala), a synthetic analogue of the biological receptor. The correlation between the two methods had been established on 60 serum samples with concentrations up to 12mgl-’ (Corti et al., 1985). SPERA has the great advantage over the microbiological assay of giving results the same day. This is particularly important in monitoring the levels achieved in clinical use. We have prepared a kit containing the components of the assay for routine use. Since it was imperative to check the results obtained with both methods on a large number of samples of controlled origin, we analysed 280 serum samples of concentrations up to lOOmgl-’ and 216 urine samples of concentrations up to SOOmgl-’ collected from eight healthy volunteers in a study organized and carried out by Dr G. Buniva of the Clinical Research Department of the Lepetit Research Centre. 01954701/86/07A085+05
$03,00/O
0
85
1986 The Hospital
Infection
Society
86
L. Cavenaghi
Materials
et al.
and methods
Administration
of teicoplanin The subjects received 400mg teicoplanin in the first day of the study, six 400mg doses at 24h intervals from 168 to 288 h; blood sampling urine collection were carried out up to day 20.
Microbiological
then and
assay
The microbiological assay was carried out with the 3 + 3 design described in Appendix XIVA of the British Pharmacopoeia and the results calculated accordingly. We used Petri dishes of 100mm internal diameter filled with 12ml of antibiotic medium No. 2 with 6% NaCl, supplemented with 5% 1 M phosphate buffer pH4.2 and inoculated with 0.2% of a B. subtilis ATCC6633 spore suspension (2 x 109cfuml-‘). Three standard solutions were prepared in the appropriate diluent at concentrations of 0.1, 0.2 and 0*4mgl-‘. The samples were preliminarily tested to assess the range of concentrations and then diluted to obtain concentrations close to the standards. Nine 6 mm diameter wells were dug in each of the six dishes used for every assay. Two samples were run on each set of dishes against the standard by filling the wells with 0.02ml of the pertinent solution through an automatic pipette. After overnight incubation at 33°C the diameter of the inhibition zones obtained were measured to the nearest 0.1 mm.
Solid phase enzyme receptor assay Each SPERA kit contained five BSA-a-ACA-D-Ala-D-Ala coated 96 well microplates, one freeze-dried vial of buffer (when reconstituted: pH7.3, 0.05 M Na phosphate, 0.15 M NaCl, 2mgml-’ of 8-anilino-1 -naphtalenesulphonic acid, 0.5% of polyoxyethylene sorbitan monolaurate and 2% of bovine serum albumin), five freeze-dried vials of teicoplanin standard (each containing 8pg), five freeze-dried vials of horseradish peroxidase labelled teicoplanin (HRP-teico) and five freeze-dried vials of substrate (when reconstituted: 0.1 M Na citrate buffer with 1 mg ml-’ of PHI, o-phenylenediamine). Each kit is sufficient for up to 55 assays. The design adopted was based on a seven-point standard curve prepared with teicoplanin dissolved in the appropriate biological fluid (serum or urine) diluted tenfold with the buffer. From the high concentration of 4mgl-’ obtained from one of the standard vials, six twofold dilutions were prepared. The samples were routinely diluted tenfold in buffer. Where necessary, the samples were further diluted with the appropriate biological fluid diluted tenfold in buffer. A portion of the appropriate biological fluid diluted tenfold in buffer is saved for use as reference (maximum colour: BO). 0.2ml aliquots of each standard and sample solutions were diluted with 0*2ml of the solution of a freeze-dried vial of HRP-teico in buffer. Three wells of the microplate were filled with 0.1 ml of the standards and of the samples. The remaining wells were filled with the reference solution. After
Assay
of teicoplanin
by SPERA
87
2h incubation at room temperature, the microplates were washed with washing buffer (pH 7.3,0.05 M Na phosphate buffer, 0.15 M NaCl and 0.05% polyoxyethylene sorbitan monolaurate) and 0.2 ml of the reconstituted substrate added to all the wells; the colour development was stopped with 4.5 M sulphuric acid after 30min and the absorbance in each well measured, then expressed as percentage of the ‘zero standard’ (BO). The standard points were plotted on logit-log paper and the amount of teicoplanin in the samples calculated by interpolation and corrected for the dilution factor. Results
Table I shows the results of the determination of teicoplanin in sera and urines of eight subjects at different times of sampling using microbiological assay and SPERA. The distribution of the results obtained by SPERA expressed as per cent of the value obtained by microbiological assay were plotted as an histogram and closely approximate a Gaussian distribution (Figure 1). The regression 100
r
80
0
83
87
I SPERA /MICRO
(%)
Figure 1. Histogram distribution of 496 determinations of teicoplanin in biological fluids by SPERA, as a function of the ratio between the value obtained by SPERA and that obtained by microbiological assay expressed as percentage.
88
L. Cavenaghi
et al.
Table I. Mean of teicoplanin concentration in sera and urines of eight subjects as measured by microbiological assay and SPERA Urine Time 0/4h 4/8h 8/12h 12/24h 24148 h 48172 h 72196 h 96/120h 120/144h 144/168h 168/192h 192/216h 216/240h 2401264 h 264/288 h 2881292 h 2921296 h 2961300 h 300/312h 312/336h 336/360h 360/384h 384/408 h 4081432 h 4321456 h 456/480h
Micro SD 223.1 f 87.4f 50.4f 59.8f 48.1 f 38.4f 19.1 f 12.31t 9.8f 5.2f 105.3 f 153.3 f 158.4& 160.1 f 198.2 f 366.1 f 162.6 f 156.4f 122.1 f 85.8f 48.3+ 41.0f 33.4f 24.6f 27.2f 13.9f
78.7 47.4 22.2 18.8 10.5 14.2 6.9 4.4 3.1 1.9 35.7 48.1 43.4 59.3 59.2 113.2 49.7 67.6 76.7 11.6 14.9 12.3 11.6 10,s 14.0 4.9
Serum
SPERA SD
Time
Micro SD
SPERA SD
221.8 f 74.6 87.6zk 50.1 49.1 f 22.3 58.7f 19.9 49.4f 13.2 36.9f 13.1 2O.Ozt 6.9 12.3& 4.6 9.1 f 3.1 5.3f 2.3 103.6f 37.4 148.4f 43.9 156.4f 42.2 161.9f 62.6 192.9 f 64.6 355.6f 103.3 158.0 f 50.4 153.6f 68.6 119.9f 74.3 86.9f 42.9 4Clf 16.9 41.8f 12.6 34.5f 11% 24.8f 11.3 27.6f 14.2 14.5f 4.6
5 min lh 2h 4h 6h 8h 10h 12h 24h 48h 72h 96h 192h 216h 240h +5min 264h +Smin 288h +.5min 289h 290h 292h 294h 296h 298h 300h 312h 336h 360h 384h 408h 432h 456h
71.0f 4-9 27.1 f 3.4 19.6f 2.2 13.3 f 2.2 9.7f 1.4 8.4f 0.9 7.3f 0.7 6.7* 0.7 3.6f 0.4 2.5f 0.2 1.5f 0.2 1.3f 0.2 4.6f 0.7 7.2f 1.1 9.4f 1.2 68+$f 6.3 ll.Of 2.1 63.7f 6.4 11.2f 1.0 74.1f11.9 34.8zk 8.2 28*8* 2.9 17.3f 1.2 13+31k 1.6 13.0f 1.3 12.5f 2.0 12.4f 1.4 8.3f 0.9 7.2f 0.9 4.8f 0.5 4.0f 0.9 3.1 f 0.4 2.5f 0.6 2.1 f 0.4
72.4f11.3 23.7f 3.5 l&4* 3.6 12.1 f 2.3 9.0f 1.0 8.1 f 1.1 6.7f 0.8 6.6f 1.7 4.2f 1.3 2.2f 0.5 lG3f 0.6 l*Sf 0.4 4-s* 1.4 7*0f 1.9 8.5f 1.7 70.1 f 7.1 lO.Of 2.1 6571t 6.5 10.5f 2.0 71.9* 17.3 30.5 f 7.6 27.2f 5.8 17.Sf 1.1 152f 3.9 13.1 f 2.3 12.3f 1.7 11.9f 1.8 8.7f 1.9 6.2f 1.8 4.6f 1.1 4.5f 14 3.0f 1.0 2.2f 0.7 2.1 f 0.7
line determined by the method of least squares has a slope of 0.97946 and an intercept of 0.28105 while the correlation coefficient is 0.99768 indicating that the teicoplanin concentrations obtained by SPERA correlates well with those obtained by microbiological assay (Figure 2). Conclusions
In conclusion, the good correlation shown on 296 samples between the results obtained with microbiological assay and with SPERA shows that either method can be used when assessing the levels of teicoplanin in serum or urine.
Assay
of teicoplanin
by SPERA
400
T-
300 r
2 if v) 200
I00
MICRO (mg I-‘)
Figure 2. Correlation curve between teicoplanin concentrations in human sera and urines as determined by SPERA and by microbiological assay. The regression line has been calculated with the method of least squares. T = 0.997675; slope = 0.97946; intercept = 0.28105 1;
N=496.
However, in conditions that may interfere with a straightforward microbiological assay, such as the presence of other antibiotics or lysed red blood cells (which inhibit the growth of the B. subtilis, e.g., human blood freeze-thawed four times gives inhibition zones on the dishes used for the assay even when diluted 512 times with bovine serum: Cavenaghi L., unpublished observation), SPERA is preferable. Moreover, SPERA is superior to microbiological assay when surfactants have to be used, as in the determination of the teicoplanin content of bones where the antibiotic is strongly bound to bone matrix. References Corti,
A., Rurali, C., Borghi, A. & Cassani, G. (1985). Solid phase enzyme receptor assay (SPERA) for glycopeptide antibiotics of the vancomycin class. Clinical Chemistry (in press). Parenti, F., Beretta, G., Berti, M. & Arioli, V. (1978). Teichomycins, new antibiotics from Actinoplanes teichomyceticus nov. sp. I. Description of the producer strain, fermentation studies and biological properties. Journal of Antibiotics 31, 276-281.
of Hospital
Infection
(1986) 7 (Supplement
A), 85-89
Comparison of the solid phase enzyme assay (SPERA) and the microbiological teicoplanin Luigi
Cavenaghi,
Angelo
Corti and Giovanni
receptor assay for
Cassani
Merrell Dow Research Institute, Lepetit Research Centre, via Durando 38, 20158 Milano, Italy Summary: A solid phase enzyme receptor assay (SPERA) for teicoplanin has been developed and recently presented in a kit form. The assay relies on the competition between antibiotic present in the biological fluid and peroxidase labelled teicoplanin for albumin-E-amino-caproyl-D-alanyl-Dalanine (BSA-s-ACA-D-Ala-D-Ala), a synthetic analog of the biological receptor. The kit contains BSA-e-ACA-D-Ala-D-Ala coated microplates together with all the necessary reagents. The limit of detection of the SPERA is O.Smgl-‘. The microbiological assay for teicoplanin uses Bacillus subtilis ATCC 6633 as test organism and a high salt medium to ensure a low limit of detection of O.OSmgl-‘. The previously established correlation between the two methods has been confirmed using the kits on 280 serum and 216 urine samples with concentrations up to SOOmgl-‘. Introduction
Teicoplanin, a novel antibiotic (Parenti et al., 1978) currently undergoing phase 3 clinical investigation, has usually been assayed by a conventional microbiological assay with Bacillus subtilis as test organism. Recently we have developed a new method: the solid phase enzyme receptor assay (SPERA) which is based on the competition between the antibiotic present in the biological fluid and the peroxidase labelled teicoplanin I for albumin-s-aminocaproyl-D-alanyl-D-alanine (BSA-&-ACA-D-Ala-D-Ala), a synthetic analogue of the biological receptor. The correlation between the two methods had been established on 60 serum samples with concentrations up to 12mgl-’ (Corti et al., 1985). SPERA has the great advantage over the microbiological assay of giving results the same day. This is particularly important in monitoring the levels achieved in clinical use. We have prepared a kit containing the components of the assay for routine use. Since it was imperative to check the results obtained with both methods on a large number of samples of controlled origin, we analysed 280 serum samples of concentrations up to lOOmgl-’ and 216 urine samples of concentrations up to SOOmgl-’ collected from eight healthy volunteers in a study organized and carried out by Dr G. Buniva of the Clinical Research Department of the Lepetit Research Centre. 01954701/86/07A085+05
$03,00/O
0
85
1986 The Hospital
Infection
Society
86
L. Cavenaghi
Materials
et al.
and methods
Administration
of teicoplanin The subjects received 400mg teicoplanin in the first day of the study, six 400mg doses at 24h intervals from 168 to 288 h; blood sampling urine collection were carried out up to day 20.
Microbiological
then and
assay
The microbiological assay was carried out with the 3 + 3 design described in Appendix XIVA of the British Pharmacopoeia and the results calculated accordingly. We used Petri dishes of 100mm internal diameter filled with 12ml of antibiotic medium No. 2 with 6% NaCl, supplemented with 5% 1 M phosphate buffer pH4.2 and inoculated with 0.2% of a B. subtilis ATCC6633 spore suspension (2 x 109cfuml-‘). Three standard solutions were prepared in the appropriate diluent at concentrations of 0.1, 0.2 and 0*4mgl-‘. The samples were preliminarily tested to assess the range of concentrations and then diluted to obtain concentrations close to the standards. Nine 6 mm diameter wells were dug in each of the six dishes used for every assay. Two samples were run on each set of dishes against the standard by filling the wells with 0.02ml of the pertinent solution through an automatic pipette. After overnight incubation at 33°C the diameter of the inhibition zones obtained were measured to the nearest 0.1 mm.
Solid phase enzyme receptor assay Each SPERA kit contained five BSA-a-ACA-D-Ala-D-Ala coated 96 well microplates, one freeze-dried vial of buffer (when reconstituted: pH7.3, 0.05 M Na phosphate, 0.15 M NaCl, 2mgml-’ of 8-anilino-1 -naphtalenesulphonic acid, 0.5% of polyoxyethylene sorbitan monolaurate and 2% of bovine serum albumin), five freeze-dried vials of teicoplanin standard (each containing 8pg), five freeze-dried vials of horseradish peroxidase labelled teicoplanin (HRP-teico) and five freeze-dried vials of substrate (when reconstituted: 0.1 M Na citrate buffer with 1 mg ml-’ of PHI, o-phenylenediamine). Each kit is sufficient for up to 55 assays. The design adopted was based on a seven-point standard curve prepared with teicoplanin dissolved in the appropriate biological fluid (serum or urine) diluted tenfold with the buffer. From the high concentration of 4mgl-’ obtained from one of the standard vials, six twofold dilutions were prepared. The samples were routinely diluted tenfold in buffer. Where necessary, the samples were further diluted with the appropriate biological fluid diluted tenfold in buffer. A portion of the appropriate biological fluid diluted tenfold in buffer is saved for use as reference (maximum colour: BO). 0.2ml aliquots of each standard and sample solutions were diluted with 0*2ml of the solution of a freeze-dried vial of HRP-teico in buffer. Three wells of the microplate were filled with 0.1 ml of the standards and of the samples. The remaining wells were filled with the reference solution. After
Assay
of teicoplanin
by SPERA
87
2h incubation at room temperature, the microplates were washed with washing buffer (pH 7.3,0.05 M Na phosphate buffer, 0.15 M NaCl and 0.05% polyoxyethylene sorbitan monolaurate) and 0.2 ml of the reconstituted substrate added to all the wells; the colour development was stopped with 4.5 M sulphuric acid after 30min and the absorbance in each well measured, then expressed as percentage of the ‘zero standard’ (BO). The standard points were plotted on logit-log paper and the amount of teicoplanin in the samples calculated by interpolation and corrected for the dilution factor. Results
Table I shows the results of the determination of teicoplanin in sera and urines of eight subjects at different times of sampling using microbiological assay and SPERA. The distribution of the results obtained by SPERA expressed as per cent of the value obtained by microbiological assay were plotted as an histogram and closely approximate a Gaussian distribution (Figure 1). The regression 100
r
80
0
83
87
I SPERA /MICRO
(%)
Figure 1. Histogram distribution of 496 determinations of teicoplanin in biological fluids by SPERA, as a function of the ratio between the value obtained by SPERA and that obtained by microbiological assay expressed as percentage.
88
L. Cavenaghi
et al.
Table I. Mean of teicoplanin concentration in sera and urines of eight subjects as measured by microbiological assay and SPERA Urine Time 0/4h 4/8h 8/12h 12/24h 24148 h 48172 h 72196 h 96/120h 120/144h 144/168h 168/192h 192/216h 216/240h 2401264 h 264/288 h 2881292 h 2921296 h 2961300 h 300/312h 312/336h 336/360h 360/384h 384/408 h 4081432 h 4321456 h 456/480h
Micro SD 223.1 f 87.4f 50.4f 59.8f 48.1 f 38.4f 19.1 f 12.31t 9.8f 5.2f 105.3 f 153.3 f 158.4& 160.1 f 198.2 f 366.1 f 162.6 f 156.4f 122.1 f 85.8f 48.3+ 41.0f 33.4f 24.6f 27.2f 13.9f
78.7 47.4 22.2 18.8 10.5 14.2 6.9 4.4 3.1 1.9 35.7 48.1 43.4 59.3 59.2 113.2 49.7 67.6 76.7 11.6 14.9 12.3 11.6 10,s 14.0 4.9
Serum
SPERA SD
Time
Micro SD
SPERA SD
221.8 f 74.6 87.6zk 50.1 49.1 f 22.3 58.7f 19.9 49.4f 13.2 36.9f 13.1 2O.Ozt 6.9 12.3& 4.6 9.1 f 3.1 5.3f 2.3 103.6f 37.4 148.4f 43.9 156.4f 42.2 161.9f 62.6 192.9 f 64.6 355.6f 103.3 158.0 f 50.4 153.6f 68.6 119.9f 74.3 86.9f 42.9 4Clf 16.9 41.8f 12.6 34.5f 11% 24.8f 11.3 27.6f 14.2 14.5f 4.6
5 min lh 2h 4h 6h 8h 10h 12h 24h 48h 72h 96h 192h 216h 240h +5min 264h +Smin 288h +.5min 289h 290h 292h 294h 296h 298h 300h 312h 336h 360h 384h 408h 432h 456h
71.0f 4-9 27.1 f 3.4 19.6f 2.2 13.3 f 2.2 9.7f 1.4 8.4f 0.9 7.3f 0.7 6.7* 0.7 3.6f 0.4 2.5f 0.2 1.5f 0.2 1.3f 0.2 4.6f 0.7 7.2f 1.1 9.4f 1.2 68+$f 6.3 ll.Of 2.1 63.7f 6.4 11.2f 1.0 74.1f11.9 34.8zk 8.2 28*8* 2.9 17.3f 1.2 13+31k 1.6 13.0f 1.3 12.5f 2.0 12.4f 1.4 8.3f 0.9 7.2f 0.9 4.8f 0.5 4.0f 0.9 3.1 f 0.4 2.5f 0.6 2.1 f 0.4
72.4f11.3 23.7f 3.5 l&4* 3.6 12.1 f 2.3 9.0f 1.0 8.1 f 1.1 6.7f 0.8 6.6f 1.7 4.2f 1.3 2.2f 0.5 lG3f 0.6 l*Sf 0.4 4-s* 1.4 7*0f 1.9 8.5f 1.7 70.1 f 7.1 lO.Of 2.1 6571t 6.5 10.5f 2.0 71.9* 17.3 30.5 f 7.6 27.2f 5.8 17.Sf 1.1 152f 3.9 13.1 f 2.3 12.3f 1.7 11.9f 1.8 8.7f 1.9 6.2f 1.8 4.6f 1.1 4.5f 14 3.0f 1.0 2.2f 0.7 2.1 f 0.7
line determined by the method of least squares has a slope of 0.97946 and an intercept of 0.28105 while the correlation coefficient is 0.99768 indicating that the teicoplanin concentrations obtained by SPERA correlates well with those obtained by microbiological assay (Figure 2). Conclusions
In conclusion, the good correlation shown on 296 samples between the results obtained with microbiological assay and with SPERA shows that either method can be used when assessing the levels of teicoplanin in serum or urine.
Assay
of teicoplanin
by SPERA
400
T-
300 r
2 if v) 200
I00
MICRO (mg I-‘)
Figure 2. Correlation curve between teicoplanin concentrations in human sera and urines as determined by SPERA and by microbiological assay. The regression line has been calculated with the method of least squares. T = 0.997675; slope = 0.97946; intercept = 0.28105 1;
N=496.
However, in conditions that may interfere with a straightforward microbiological assay, such as the presence of other antibiotics or lysed red blood cells (which inhibit the growth of the B. subtilis, e.g., human blood freeze-thawed four times gives inhibition zones on the dishes used for the assay even when diluted 512 times with bovine serum: Cavenaghi L., unpublished observation), SPERA is preferable. Moreover, SPERA is superior to microbiological assay when surfactants have to be used, as in the determination of the teicoplanin content of bones where the antibiotic is strongly bound to bone matrix. References Corti,
A., Rurali, C., Borghi, A. & Cassani, G. (1985). Solid phase enzyme receptor assay (SPERA) for glycopeptide antibiotics of the vancomycin class. Clinical Chemistry (in press). Parenti, F., Beretta, G., Berti, M. & Arioli, V. (1978). Teichomycins, new antibiotics from Actinoplanes teichomyceticus nov. sp. I. Description of the producer strain, fermentation studies and biological properties. Journal of Antibiotics 31, 276-281.