Subscribe Now Subscribe Today
Research Article

The Detection Limits of Antimicrobial Agents in Cow`s Milk by a Simple Yoghurt Culture Test

M. Mohsenzadeh and A. Bahrainipour
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail

The aim of this study was to study performance of Yoghurt Culture Test (YCT) in the detection of antimicrobial residues in milk. For this purpose, the sensitivity of YCT for 15 antibiotics were determined. For each drug, 8 concentrations were tested. The detection limits of YCT at 2.5 h and 4 h incubation were determined (μg kg-1): 15 and 37.5, penicillin G; 4 and 5, ampicillin; 5 and 7.5, amoxycillin; 100 and 200, cephalexin; 80 and 100, cefazoline; 100 and 200, oxytetracycline; 500 and 100, chlortetracycline; 100 and 200, tetracycline; 150 and 200, doxycycline; 200 and 400, sulphadimidine; 500 and 1000, gentamycin; 1000 and 1500, spectinomycin; 400 and 500, erythromycin; 50 and 100, tylosin; 5000 and 10000, chloramphenicol. The YCT detection limits at 2.5 h incubation for ampicillin, cephalexin, tetracycline, oxytetracycline and tylosin are similar to those obtained as Maximum Residue Limit (MRL) according to Regulation 2377/90 EEC as set out by the European Union. In addition the detection limits of YCT for some antibiotics were lower than some of microbial inhibitor test.

Related Articles in ASCI
Similar Articles in this Journal
Search in Google Scholar
View Citation
Report Citation

  How to cite this article:

M. Mohsenzadeh and A. Bahrainipour, 2008. The Detection Limits of Antimicrobial Agents in Cow`s Milk by a Simple Yoghurt Culture Test. Pakistan Journal of Biological Sciences, 11: 2282-2285.

DOI: 10.3923/pjbs.2008.2282.2285



Antimicrobial agents are administered in therapeutic treatment of cattle and constitute a common cause of the presence of chemotherapeutic drug residues in milk. Mastitis is the most prevalent disease of milk-producing cattle which requires antimicrobial treatment (Suhren, 2002). 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).

From a technological point of view, the residues of antimicrobial agents in milk can produce important losses in fermented products; for example, they inhibit the bacterial fermentation processes involved in the production of some dairy products, such as cheese or yoghurt (Nouws et al., 1999; Suhren, 2002).

To ensure human food safety, Maximum Residue Limits (MRLs) have been set out for many antimicrobial agents and different methods of analysis developed for the swift detection of residuals of inhibitors present in milk. 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 (Reichmuth et al., 1997; Suhren, 2002; Suhren and Walte, 2003).

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 (Suhren, 2002), 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. In general, microbiological inhibition tests are used for the screening stage, many of them using Bacillus stearothermophilus var. calidolactis, such as BRT-AiM®, Delvotest®, CH® -ATK microplate. These screening methods were mainly developed and used with cow milk (Scannella et al., 1997). In the case of penicillin-G, most of these procedures can detect between 0.004 and 0.006 IU mL-1 of milk, but the responses of the microbial inhibitor tests to other antibiotics or inhibitory residues varies with the compound in question.

The aim of the present research was to study performance of Yoghurt Culture Test (YCT) in the detection of different antimicrobial agents belonging to the most representative groups utilized in veterinary medicine in milk.


Experimental design and milk samples: The present study was conducted during June to September 2006 in Mashhad, Northeast of Iran. Fresh and antibiotic-free milk sample was drawn from cow known to be free from any form of medication and sample was transported to the laboratory at 4°C. Appropriate volumes of the milk were then dispensed into two clean and sterile jars and, while one jar was held as a control, the pH of another sample was adjusted to 6.0 using 1 N HCl. The milk in each jar was then warmed in an oven for a period of time known to give a temperature of 45°C and was inoculated with 4% (w/w) of yoghurt culture containing equal mixtures of Streptococcus thermophilus and Lactobacillus delbrueckii ssp. bulgaricus that are in regular use in Mashhad dairy plants. The yoghurt culture was prepared by mixing 1 g of well-mixed, fresh yoghurt culture with 99 mL of skim-milk (10% dry milk solids, w/v) that had been heat treated at 95°C for 5 min. After thorough mixing of milk samples with culture, each portion was placed in a water bath at 42°C. Measurements of pH were made immediately using a pH meter and after 1.5, 2.0, 2.5 and 4 h incubation. As an alternative to measurements of acidity, the use of pH indicators was examined in a further trial by adding 0.1 mL of Chlorophenol Red (0.2 in 50% ethanol) to the milk either before incubation.

Antimicrobial solutions and test samples: Antibiotics for preparation of the antimicrobial solutions were stored and handled according to the manufacturer`s instructions before being used. Each antimicrobial agent was tested at seven different concentrations and in every case a negative control was included.

Table 1 shows the 15 antimicrobial agents and the concentrations used for the preparations of their solutions. For each concentration, 10 replicates were prepared using antibiotic-free milk samples obtained from individual animals. A total of 80 test samples were analysed for each antimicrobial agent. The pH value of each sample of milk was then adjusted to 6.0 with 1 N HCl and the sample warmed to 45°C. A fresh starter culture was then used to inoculate the contaminated milks, along with a negative control sample (zero antibiotic), at rate of 4% and Chlorphenol Red was added as an indicator. After mixing well, each test sample was placed in a water bath at 42°C. Measurements of pH and consideration of coagulum formation were made at 2.5 and 4.0 h and the colour changes were recorded as well: samples without any change in colour/crud formation were suspected being contaminated with antibiotics and were considered positive.

Table 1: Antibiotic concentrations employed for Yoghurt Culture Test detection limits in cow`s milk
Image for - The Detection Limits of Antimicrobial Agents in Cow`s Milk by a Simple Yoghurt Culture Test

Statistical analysis: For each concentration, 10 replicates were prepared using antibiotic-free milk samples obtained from individual animals. Data were averaged and analysed statistically using SPSS software (Version 10.0.5). The detection limit of the visual interpretation of the YCT method was estimated as concentrations in which 95% of the results were positive (Molina et al., 2003).


The minimum concentrations of the different antibiotics giving positive results in the YCT, i.e., no curd formation or colour shift in the presence of Chlorophenol Red, are shown in Table 2, along with MRL values in accordance with EU regulations.

The detection limit of amoxycillin at 2.5 h incubation was lower than the 7 μg kg-1 determined in ewe milk samples by Eclipse 100® (Montero et al., 2005) and 6 μg kg-1 determined by Suhren and Knappstein (1998). In the case of ampicillin, the level detected in this study was lower than 6 μg kg-1 determined by Molina et al. (2003). For cephalexin residues, the detection limit was lower than 270 μg kg-1 detected by Molina et al. (2003) and at 2.5 h incubation was lower than 115 μg kg-1 detected by Montero et al. (2005). The sensitivity of the test to penicillin was rather disappointing because, at 2.5 h, the YCT appeared less sensitive than any other microbial inhibitor tests, it may be that the Lac. delbrueckii ssp. bulgaricus component of the culture was not affected immediately by the inhibitor. It has also been reported that mixed cultures of Str. thermophilus and Lac. delbrueckii ssp. bulgaricus are less sensitive than the individual species growing alone and this effect might have altered the results as well (Robinson and Tamime, 2002).

The sensitivity of the YCT to oxytetracycline and tetracycline at 2.5 and 4 h incubation was better than BRT- AiM® (Molina et al., 2003), delvotest photometric measurement (Althaus et al., 2003) and Eclipse 100® (Montero et al., 2005 ). In addition the detection limit of YCT at 2.5 h incubation for these antibiotics can be at similar levels to EU-MRLs.

Table 2: The detection limits of antibiotics in cow`s milk by the YCT (μg kg-1)
Image for - The Detection Limits of Antimicrobial Agents in Cow`s Milk by a Simple Yoghurt Culture Test
1: Council Regulation 2377/90 EEC, 2: Sum of all substances of this group, 3: Not allowed

Other microorganisms could also be assayed in other to be able to detect tetracyclines at levels close to EU-MRLs (100 μg kg-1). Suhren and Heeschen (1993) pointed out that the Bacillus cereus var. mycoides ATCC 9634 is sensitive to concentrations of less than 100 μg kg-1 of different tetracyclines, while Nouws et al. (1998) detected between 10 and 30 μg kg-1 of tetracyclines when using B. cereus ATCC 11778.

The sensitivity of the YCT to gentamycin (500 μg kg-1 at 2.5 and 1000 μg kg-1 at 4 h incubation) was better than that reported by Molina et al. (2003), Althaus et al. (2003) and Montero et al. (2005).

In this study the erythromycin detection limit was lower than BRT-AiM® (Molina et al., 2003), delvotest phothometric measurement (Althaus et al., 2003) and Eclipse 100® (Montero et al., 2005 ). The detection limit for erythromycin is very high compared with the EU-MRLs (40 μg kg-1). The detection limit of tylosin at 2.5 h incubation YCT at similar levels to EU-MRLs. For chloramphenicol residues, the sensitivity of YCT was higher than that reported by Molina et al. (2003), Althaus et al. (2003) and Montero et al. (2005). The EU regulation allow zero tolerance for this antimicrobial agent. For this reason, the use of other methods will be assessed. Kolosova et al. (2000) can detect 0.08 μg kg-1 of chloramphenicol when utilizing an indirect competitive ELISA method. Whereas Gaudin and Maris (2001) achieved a detection limit of 0.1 μg L-1 by means of a biosensor immunoassay based on polyclonal antibodies.

The reduced sensitivities of the YCT at 4 h is a reflection of the fact that the concentrations that cause a failure at 2.5 h leave a percentage of cells of one or both organisms unaffected. Consequently, sufficient acidity has been generated at the end of 4 h to form a coagulum and a higher concentration is needed to ensure that too few cells survive to lower the pH to 4.8 or below (Yamani et al., 1999).

Nevertheless, the results at 4 h were useful for developing the following protocol:

Failure to change indicator in 2.5 and 4 h-unacceptable level of inhibitory substances in the milk
Failure to change indicator in 2.5 h, but change after 4 h-marginal level of inhibitory substances in the milk
Change of indicator in 2.5 h-inhibitory substances below level of detection

Clearly, the disadvantage of the YCT is that it is not so sensitive to β-lactam antibiotics as some of the commercial kits, but this criticism does not alter the value of the YCT as a practical method of assessment for a dairy.

Overall, it would appear that the YCT employing sensitive strains of Str. thermophilus and Lac. delbrueckii ssp. bulgaricus provides a test for inhibitory substances in milk that is broadly comparable in response to other commercial kits. Obviously, the YCT would not be suitable for use in the laboratory of a Regulatory Authority where the priority is to protect consumers from extremely low levels of β-lactam residues, but use of the YCT could be encouraged in countries where the testing of milk supplies for antibiotics is not mandatory.


This study was supported by a grant (FUM, 1-684) from the Research Council of Ferdowsi University of Mashhad.

1:  Althaus, R.L., A. Torrer, A. Montero and M.P. Molina, 2003. Detection limits of antimicrobials in ewe milk by delvotest photometric measurements. J. Dairy Sci., 86: 457-463.
Direct Link  |  

2:  Aurelli, P., A. Ferrini and V. Mannoni, 1996. Presumptive identification of sulphonamide and antibiotic residue in milk by microbial inhibitor test. J. Food Control, 7: 165-168.
CrossRef  |  Direct Link  |  

3:  Currie, D., L. Lynas, G. Kennedy and J. McCaughey, 1998. Evaluation of modified EC four-plate method to detect antimicrobial drugs. Food Addit. Contam., 15: 651-660.
Direct Link  |  

4:  Dewdney, J.M., L. Maes, J. P. Raynaud, F. Blanc and J.P. Scheid et al., 1991. Risk assessment of antibiotic residues of beta-lactams and macrolides in food products with regard to their immunoallergic potential. Food Chem. Toxicol., 29: 477-483.
CrossRef  |  Direct Link  |  

5:  Gaudin, V. and P. Maris, 2001. Development of a biosensor based immunoassay for screening of chloramphenicol residues in milk. Food Agric. Immunol., 13: 77-86.
Direct Link  |  

6:  Kolosova, A.Y., J.V. Samsonova and A.M. Egorov, 2000. Competitive ELISA of chloramphenicol: Influence of immunoreagent structure and application of the method for the inspection of food of animal origin. Food Agric. Immunol., 12: 115-125.
Direct Link  |  

7:  Molina, M.P., R.L. Althaus, A. Molina and N. Fernandez, 2003. Antimicrobial agent detection in ewe’s milk by the microbial inhibitor test brilliant black reduction test-BRT AiM®. Int. Dairy J., 13: 821-826.
Direct Link  |  

8:  Montero, A., R.L. Althaus, A. Molina, I. Berruga and M.P. Molina, 2005. Detection of antimicrobial agents by a specific microbiological method (Eclipse 100®) for ewe milk. Small Ruminant Res., 57: 229-237.
CrossRef  |  Direct Link  |  

9:  Nouws, J., H. Van Egmond, I. Smulders, G. Loeffen and J. Schouten et al., 1999. A microbiological assay systemfor assessment of raw milk exceeding EU maximum residue levels. Int. Dairy J., 9: 85-90.
Direct Link  |  

10:  Nouws, J., G. Loeffen, J. Schouten, H. van Egmond, H. Keukens and H. Stegeman, 1998. Testing of raw milk for tetracycline residues. J. Dairy Sci., 81: 2341-2345.
Direct Link  |  

11:  Reichmuth, J., G. Suhren and R. Beukers, 1997. Evaluation of microbial inhibitor tests- The IDF approach. Milchwissenschaft, 52: 691-694.
Direct Link  |  

12:  Robinson, R.K. and A.Y. Tamime, 2002. Microbiology of Fermented Milks. In: Dairy Microbiology, Robinson, R.K. (Ed.). Chapman and Hall, London, ISBN: 9780471385967.

13:  Scannella, D., P. Neaves, K. Keedy and C. Bell, 1997. An evaluation of the Delvo X-Press test for detecting β-lactams in ex-farm raw milks. Int. Dairy J., 7: 93-96.
CrossRef  |  Direct Link  |  

14:  Suhren, V.G., 2002. Inhibitors and residues of veterinary drugs in milk`legal basis, detection methods and detection systems. Kieler Milchwirtschaftliche Forschungsberichte, 54: 35-71.
Direct Link  |  

15:  Suhren, G. and W. Heeschen, 1993. Detection of tetracyclines in milk by Bacillus cereus microlitre test with indicator. Milchwissenschaft, 48: 259-263.
Direct Link  |  

16:  Suhren, G. and H.G. Walte, 2003. Experiences with the application of method combinations for the detection of residues of antimicrobial drugs in milk from collecting tankers. Milchwissenschaft, 58: 536-540.
Direct Link  |  

17:  Suhren, G. and K. Knappstein, 1998. Detection of incurred dihydrostreptomycin residues in milk by liquid chromatography and preliminary confirmation methods. Analyst, 123: 2797-2801.
Direct Link  |  

18:  Yamani, M.I., L.M.A. Al-Kurdi, M.S.Y. Haddadin and R.K. Robinson, 1999. A simple test for the detection of antibiotics and other chemical residues in ex-farm milk. Food Control, 10: 35-39.
CrossRef  |  Direct Link  |  

©  2021 Science Alert. All Rights Reserved