| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Departamento de Ciencia Animal
Departamento de Estadistica Universidad Politécnica de Valencia Camino de Vera, 14. Apartado 22012, (46071) Valencia, España
Corresponding author:
M. P. Molina; e-mail:
pmolina{at}dca.upv.es.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: ewe milk • microbial inhibitor test • detection limit • animicrobial
Abbreviation key: EU = European Union, MRL = maximum residue limits
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The presence of certain antimicrobial agent residuals in milk constitutes a potential hazard for the consumer and may cause allergic reactions, interference in the intestinal flora, and resistant populations of bacteria in the general population, thereby rendering antibiotic treatment ineffective (Dewdney et al., 1991; Currie et al., 1998). Important losses are also provoked in the fermented products, by inhibiting the bacterial processes involved in the elaboration of cheese and cultured milk products (Mourot and Loussourorn, 1981; Brady and Katz, 1988).
For these reasons, several manufacturers have developed commercially available tests both for producers and the dairy industry with the aim of detecting drug residues in milk, among these the microbial inhibitor tests (IDF, 1991; Cullor, 1992).
The microbial inhibitor test procedure for detection of drug residues in milk is based on inhibition of spore outgrowth of organisms such as Bacillus stearothermophilus var. calidolactis (Carlsson and Björck, 1987), Bacillus cereus (Suhren and Heeschen, 1993), Bacillus subtilis (Aurelli et al., 1996), noted visually by interpreting the color change of a pH-indicator present in the test medium.
Among the microbial inhibitor tests widely used for detection of veterinary medicines in cow milk, Delvotest "SP" is an economic, easy-to-use screening test giving results within a relatively short period (2.30 to 3.00 h). This method was recognized by the Association of Official Analytical Chemists (Katz, 1982; Kelley, 1982).
The Delvotest "SP" method is classified visually into three categories: "negative," "doubtful," and "positive," compared with the colors of "positive" and "negative" standard samples. The visual assessment is subjective, and the different modes of action of antimicrobial agents can influence the coloring of the indicator (Suhren and Luitz, 1995).
To avoid subjective differences in the visual interpretations and take the readings in an automated and more objective manner, some authors (Schiffmann, 1992; Luitz and Suhren, 1995; Luitz et al., 1996) propose performing photometric measurements utilizing an appropriate wavelength (590 nm) and another wavelength as reference (650 nm) in an ELISA reader.
Moreover, it must be emphasized that among the properties that should be considered in the evaluation of residual detection methods, the detection limit for each antimicrobial agent deserves special attention, with the aim of preventing milk with drug residuals in excess of Maximum Residue Limits (MRL) from being marketed and reaching the consumer.
The detection limits of Delvotest "SP" have been determined in cow milk by means of visual assessment in several works (Van Os and Beukers, 1980; Senyk et al., 1990; Luitz and Suhren, 1996; Luitz et al., 1996; Zaadhof et al., 1997), although only on very few occasions in the milk of other species such as ewe milk (Althaus, 1999).
Furthermore, no studies on Delvotest detection limits utilizing photometric measurements in any type of milk are present in the bibliography. For this reason, the aim of this research was to calculate the Delvotest "SP" detection limits for 24 antimicrobial agents in ewe milk by means of the relative absorption method.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
For this study, milk samples corresponding to the morning machine milking session (0800 h) of 20 ewes were collected in the 60 to 90-d period postpartum.
Delvotest "SP" Microbial Test
Milk samples were analyzed during the 6-hr period after collection by Delvotest "SP" (DSM Food Specialties, Dairy Ingredients, Delft, The Netherlands). The method was carried out according to the instructions of the manufacturer. Thus, 100-µl milk samples were added to individual Delvotest cups ready prepared containing B. stearothermophilus var. calidolactis and indicator in solid medium. One nutrient tablet, containing carbon and nitrogen source facilitating growth/metabolism of the test organism, was added to each cup, and the plates were sealed with the tape supplied. Plates were incubated in a water bath at 64 ± 1°C for 3 h, following the manufacturer’s instructions for ewe milk.
Immediately after the incubation period, absorbance was measured with the ELISA reader (Wallac 1420, Victor Multilavel Counter) using 590 and 650 nm as measuring and reference wavelengths, respectively. The optimum wavelength for measuring color change is 590 nm for the Delvotest bromoCresol Purple pH indicator system. The wavelength of 650 nm is taken as a reference wavelength to correct for cell-length differences (Luitz and Suhren, 1995). Dual readings (590 to 650 nm) were used in the calculations.
In each microplate, the photometric measurements were expressed in relative absorbances, according to the following transformation:
 |
where: A = relative absorbance, Ax = milk sample absorbance with "x" antibiotic concentration, A0 = absorbance of antibiotic free milk (negative control), A100 = milk sample absorbance with the antibiotic concentration that produces 100% of positive results. The positive results (purple color) were assessed visually by two trained persons.
Preparation of Antimicrobial Test Solutions
Table 1
summarizes the antimicrobial agents and the concentrations used for the preparation of their solutions. These drugs were stored and handled according to the manufacturers’ instructions before being used. All dilutions were prepared in 100-ml volumetric flasks at the day of analysis, in order to avoid possible inconvenience due to instability of the solutions.
|
Detection limits of the antimicrobial agents were established in line with the IDF indications (IDF, 1999). To this end, eight concentrations were prepared with different levels of each drug. For each concentration, 20 replicates were prepared using antibiotic-free milk samples obtained from individual animals (160 analyses for antiinfectious agents).
Statistical Analysis
The results were achieved using SAS LOGISTIC procedure (SAS, 1998). The logistic regression for analyzing the effect of concentration of antimicrobial agents upon the relative absorption of Delvotest "SP" was the following:
 |
where: Lij = logit variable (ln Aij/1-Aij); ß0, ß1 = coefficients estimated for logistic regression model; ACi = antimicrobial concentration; 
ij = residual error.
The concordance coefficient (C) was applied as rank correlation between the observed and predicted relative absorption (SAS, 1998).
The detection limit of the photometric measurements of the Delvotest "SP" was calculated as the antimicrobial concentrations that produces 45% of the maximum relative absorption (Luitz and Suhren, 1995; Luitz et al., 1996).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Also shown are the detection limits of Delvotest "SP" calculated by means of logistic model and the 45% relative absorbance approach (Luitz and Suhren, 1995; Luitz et al., 1996), along with the values of the MRL set out by the European Union (EU-MRL).
|
Given the interpretation of the "ß1" parameter of the model, the greater values of the same for penicillin "G," ampicillin and amoxycillin (Table 2
) indicate a greater increase in relative absorbance with the concentration, i.e., a higher sensitivity in its detection than for the rest of the betalactamics assayed (cephalosporin "C," cephalexin, Ceftiofur, cefadroxil, and cloxacillin; Table 2).
Figures 1 and 2 show the effect of the concentrations of penicillins and other beta-lactam antibiotics on the Delvotest method relative absorption, as well as the curves constructed by means of the logistic model ("ß0" and "ß1" coefficients; Table 2). In these figures, each point represents the mean value of 20 determinations.
|
|
), whereas the beta-lactam antibiotics shown in Figure 2 presented lower "ß1" coefficient values than the penicillin group, requiring greater increments in their concentrations in order to reach 100% absorbance (between 50 and 120 µg/kg).
Table 3 summarizes the values of the statistical parameters, detection limits, and EU-MRL for sulphonamides. The effect of sulphonamide concentrations upon the relative absorbance of Delvotest "SP" is shown in Figure 3, where the logistic curves constructed on the basis of the "ß0" and "ß1" coefficient in Table 3 are also represented. It may be observed that sulfametoxypyridazine requires greater increments in concentration than the other sulphonamides assayed in order to achieve 100% absorbance.
|
|
It may be seen that the "ß1" coefficient values are lower than those obtained for beta-lactam antibiotics (Table 2). This demonstrates the low sensitivity of Bacillus stearothermophilus in detecting residues of these antimicrobial agents compared with the beta-lactam antibiotics.
|
Beta-Lactam Antibiotics
The detection limit of amoxycillin (3 µg/kg, Table 2) was lower than the 6 µg/kg determined by Honkanen-Buzalski and Reybroeck (1995) and Suhren and Reichmuth (1998). In the case of ampicillin, the level detected in this work (2 µg/kg, Table 2) was similar to the 3 µg/kg calculated by Luitz et al. (1996), while the 18 µg/kg of cloxacillin (Table 2) was lower than the 30 µg/kg detected by Van Os and Beukers (1980) and Gardner et al. (1996).
In the present study, penicillin "G" presented a detection limit of 1 µg/kg (Table 2), lower than the values reported by Charm and Ruth (1993); Honkanen-Buzalski and Reybroeck (1995); Gardner et al. (1996) and Suhren and Reichmuth (1998), which ranged from 2.5 to 4 µg/kg of penicillin.
The fact that the detection limit of penicillin "G" calculated by means of photometric measurements in ewe milk were lower than those obtained by means of visual interpretation in cow milk by other authors could be attributed to the different approaches used in calculations.
Indeed, Althaus et al. (2001), utilizing the Brilliant Black Reduction Test (BRT), obtained lower values in the detection limits when calculated as the concentration producing 45% of the maximum relative absorption by photometric measurement than when calculating the detection limit as the concentration producing 95% of "positive results" by visual evaluation.
This makes it necessary to reconsider the calculation criteria for detection limits depending on whether photometric readings or visual evaluation are employed, in order to obtain similar values in both cases.
On the other hand, Ceftiofur also presented a detection limit (33 µg/kg, Table 2) in ewe milk lower than the 50 µg/kg obtained by other authors (Charm and Ruth, 1993; Honkanen-Buzalski and Reybroeck, 1995; and Gardner et al., 1996).
It must be emphasized that the detection limits calculated by photometric measurements for ampicillin, amoxycillin, penicillin "G," cephalexin, cefoperazone, and Ceftiofur were lower than the EU-MRL (Table 2). No detection limit values for cefadroxil, cephalosporin "C", cephalexin, cefoperazone and cefuroxime in milk of any other species were found in the consulted literature.
Sulphonamides
The detection limits for sulphonamides calculated by photometric measurements (Figure 3) were similar to EU-MRL (100 µg/kg, Table 3). It should be noted that the sulfadiazine detection limit (88 µg/kg) in ewe milk was lower than the 1000 µg/kg reported by Charm and Ruth (1993) with cow milk samples. In the literature consulted, no detection limits were found for sulfamethoxazole, sulfametoxypyridazine, and sulfaquinoxaline in milk of any other species.
Aminoglycosides
Table 4 shows the detection limits for aminoglycosides. Detection limits calculated by photometric measurements in ewe milk were 6100 (streptomycin) and 1200 µg/kg (gentamycin). Said values were similar to the 6000 µg/kg (streptomycin) and 1200 µg/kg (gentamycin) observed by Senyk et al. (1990). For neomycin residues in ewe milk, the detection limit was 2600 µg/kg, greater than the 1000 to 2000 µg/kg range obtained by Van Os and Beukers (1980) in cow milk.
When comparing the detection limits calculated for the three aminoglycosides in ewe milk with the EU-MRL (Table 3), it was proved that the Delvotest "SP" was unable to detect these levels due to the sensitivity drop of Bacillus stearothermophilus var calidolactis for aminoglycosides. Aurelli et al. (1996) obtained good sensitivity for streptomycin (125 µg/kg), lower than the 200 µg/kg established as EU-MRL, when using Bacillus subtilis ATCC 6633 microorganism.
Macrolides
The erythromycin detection limit (830 µg/kg, Table 4) in ewe milk is within the range of 400 to 900 µg/kg reported by Van Os and Beukers (1980) in cow milk. The detection limit for erythromycin is very high compared with the EU-MRL (40 µg/kg); it is, therefore, necessary to study other microorganisms, such as Streptococcus salivarius ssp. thermophilus, which is sensitive to 75 µg/kg of erythromycin (Honkanen-Buzalski and Reybroeck, 1995).
The tylosin residues in ewe milk must be present at a concentration of 100 µg/kg (Table 4) in order to be detected photometrically by Delvotest "SP". This level coincides with the detection limit reported by Charm and Ruth (1993) in cow milk and is not far from the 50 µg/kg set out for EU-MRL.
Tetracyclines
The detection limit of oxytetracycline (320 µg/kg, Table 4) in ewe milk using ELISA plate reader was lower than 400 µg/kg (Zaadhof et al., 1997) and 500 µg/kg (Luitz and Suhren, 1996), whereas the detection limit of tetracycline (590 µg/kg, Table 4) was similar to the 600 µg/kg reported by Senyk et al. (1990).
As seen in Table 4, the detection limits of tetracyclines were above the EU-MRL (100 µg/kg). Thus, it would be convenient to enhance the conditions of the method in order to detect values approaching the MRL, or assess the utilization of other microorganisms with greater sensitivity for tetracyclines, such as Bacillus cereus var. mycoides (Suhren and Heeschen, 1993; Nouws et al., 1998).
Other Chemotherapeutics
The Delvotest "SP" method presents a low sensitivity for detection of chloramphenicol (12,000 µg/kg) and trimethoprim (290 µg/kg) residues in ewe milk (Table 4). However, other authors also reported high chloramphenicol detection limits; Senyk et al. (1990) and Van Os and Beukers (1980) observed ranges in cow milk of 9000 to 21,000 and 8000 to 10,000 µg/kg, respectively.
Considering the "zero tolerance" stipulated by the EU for chloramphenicol and the high detection limit calculated by means of the microbial inhibitor tests, Kolosova et al. (2000) assayed the indirect competitive ELISA method which allows detection of 0.08 µg/kg of chloramphenicol in cow milk.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
For those antimicrobial drugs whose detection limits were similar to those set out as EU-MRL, the following values were obtained (µg/kg): 3, amoxycillin; 2, ampicillin; 18, cloxacillin; 1, penicillin "G;" 40, cephalexin; 20, cefoperazone; 33, Ceftiofur; 100, tylosin; 88, sulfadiazine; 44, sulfamethoxazole; 140, sulfametoxypyridazine, and 48, sulfaquinoxaline.
The Delvotest method is an economic test, easy to use, and with good detection limits for a wide range of beta-lactam antibiotic, sulphonamide, and tylosin residuals in ewe milk.
In contrast, the Delvotest "SP" method did not detect streptomycin, gentamycin, neomycin, erythromycin, oxytetracycline, tetracycline, chloramphenicol, and trimethoprim at EU-MRL. For this reason, we would recommend improvement in the sensitivity of Delvotest "SP" in order to detect a greater number of residues of veterinary medicines that may be present in ewe milk.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Received for publication April 26, 2002. Accepted for publication September 8, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Linage, C. Gonzalo, J. A. Carriedo, J. A. Asensio, M. A. Blanco, L. F. De La Fuente, and F. San Primitivo Performance of Blue-Yellow Screening Test for Antimicrobial Detection in Ovine Milk J Dairy Sci, December 1, 2007; 90(12): 5374 - 5379. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yamaki, M. I. Berruga, R. L. Althaus, M. P. Molina, and A. Molina Occurrence of Antibiotic Residues in Milk from Manchega Ewe Dairy Farms J Dairy Sci, October 1, 2004; 87(10): 3132 - 3137. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
