Antibiotics are frequently used in veterinary medicine as growth promoters and for the treatment and prevention of diseases. Inappropriate use of antibiotics can lead to excessive accumulation of antibiotic residues in animal products (including meat and dairy products), posing a potential risk to human health.
Antibiotic residues in food can cause toxic reactions and a range of side effects, including allergic reactions, nausea, and rashes. Long-term consumption of foods containing low concentrations of antibiotics can lead to the spread of drug-resistant bacteria.
Canada and the European Union have set maximum residue levels (MRLs) for drugs in food, while the United States uses the term “tolerable limit.”
The types of antibiotic residues used in milk, including sulfonamides, fluoroquinolones, β-lactams, macrolides, and tetracyclines, have attracted the attention of regulatory agencies. Table 1 shows the residue limits for each type of antibiotic analyzed in milk in this study.
Table 1. Different classes of antibiotics and their acceptable concentration limits in milk. Source: PerkinElmer Food Safety and Quality
a. The values reported in FDA/CFSAN Office of Milk Safety Memorandum 1b, dated September 27, 2005, are “safety levels,” not acceptance levels. c. No acceptance level has been established for milk; the method target level is set at 100 ng/mL (the muscle acceptance level). d. No acceptance or safety levels have been established; the method target level is set at 5 ng/mL. Acceptable levels include total residues and individual residues; the method target level is set at 100 ng/mL. f. Overlabeling of sulfonamides (except sulfadiazine) in lactating dairy cows is prohibited.
This article presents a rapid and highly sensitive liquid chromatography-tandem mass spectrometry (LC/MS/MS) method well suited for the quantitative and confirmatory analysis of multiple antibiotics in milk. The following antibiotics were analyzed in this study:
This method uses two internal standards: ciprofloxacin-d8 and flunixin-d3. A modified version of the basic QuEChERS method was used to extract a number of antibiotics from milk.
This method exhibits good sensitivity, recovery, linearity, accuracy and selectivity, making it well suited for the analysis of antibiotic levels in milk and meeting the low tolerance levels required by numerous regulatory agencies.
Chromatographic separations were performed using a PerkinElmer Altus™ UPLC® system. This instrument includes an Altus A-30 solvent/sample module, a column thermostat, and a built-in vacuum degasser.
Detection was performed using a QSight™ 210 MS/MS triple quadrupole mass spectrometer, and instrument control, analysis, and data processing were performed using the Simplicity 3Q™ software platform.
All solvents, reagents, and diluents used in this study were LC/MS grade, and all antibiotic standards were provided by Sigma-Aldrich® (St. Louis, MO, USA). These standards were stored at 4°C to prevent degradation.
Stock solutions and mixed drug solutions used for calibration and sample spiking are prepared in methanol. All antibiotics, except β-lactam antibiotics, are prepared in methanol, while β-lactam antibiotics are prepared in water. β-lactam antibiotic standards should be stored in plastic vials, while other antibiotic standards can be stored in glass or plastic containers.
C18 columns with adsorbent caps were purchased from Supelco. Sample vials were placed in an autosampler tray and stored at 7°C to prevent analyte degradation. All stock solutions and working standards were refrigerated until use, and only brown 2 mL HPLC sample vials were used. This also prevented degradation.
Two internal standards were used for calibration and quantification: ciprofloxacin-d8 for sulfonamides, fluoroquinolones and tetracyclines, and flunixin-d3 for macrolides and β-lactams.
In this study, organic whole milk samples from a local market were used as a control matrix. Five milk samples with different fat contents were analyzed simultaneously for the presence of antibiotics. A modified QuEChERS method was used for sample pretreatment. 2,3
To calibrate for the matrix effect, after step 7, add analyte and internal standard solutions to bring the analyte concentration in the sample to a range of 0.1 ng/mL to 1000 ng/mL. Five replicates are prepared for each calibration level.
All residues were analyzed using electrospray ionization in positive ion mode. Ion source parameters (such as ion source temperature, gas flow rate, and position settings) were optimized.
The specific procedure involves adding 0.5 μg/mL sulfadiazine solution to a 90:10 water:acetonitrile solution containing 0.1% formic acid and 5 mM ammonium formate at a flow rate of 10 μL/min.
Solutions of each residue were injected at a concentration of 0.5 μg/mL to determine the optimal collision energy for each MRM transition. The width of the quadrupole peaks (Q1 and Q3) was set at 0.7 amu.
Table 3 shows the parameters for each MRM transition. For each antibiotic residue, three MRM transitions were monitored to reduce the number of false positives and false negatives in this method. 4
Table 4 shows the limit of quantification (LOQ) values and regression analysis for each antibiotic calibration curve. The analysis showed that the LOQ for each antibiotic is 5–100 times lower than the acceptable tolerance level in milk.
Table 4. Limits of quantification and linear correlation coefficients for various antibiotic residues in milk extracts. Source: PerkinElmer Food Safety and Quality
This confirms that the method is sensitive enough to analyze low concentrations of antibiotics in milk.
Figure 1 shows three MRM ion pairs for a single antibiotic, sulfadiazine. These antibiotics, added to a milk matrix at a concentration of 5 ng/mL, demonstrated a good signal-to-noise ratio. The retention times of each analyte demonstrated good reproducibility within ±0.1 min over 24 hours.
Figure 1. Chromatograms of three MRM transformations (one quantitative ion A and two qualitative ions B and C) of sulfadiazine (5 ng/mL) in milk extract. Image source: PerkinElmer Food Safety and Quality
To construct calibration curves corresponding to the sample matrix, antibiotics were added to antibiotic-free milk samples at various concentrations (0.1–1000 ng/ml). Nine different concentration levels were established during this process, and an internal standard was used simultaneously.
Figure 2 shows an approximate calibration curve for sulfadiazine, displaying four orders of magnitude.
Figure 2. Calibration curves for sulfadiazine in milk extracts at concentrations from 0.1 to 1000 ng/mL (n=5 for each concentration level). Image source: PerkinElmer Food Safety and Quality
The calibration curves were linear, and the R² coefficient of determination for all analytes exceeded 0.9958. Table 5 shows that each analyte exhibited extremely high precision of response across five different concentration levels.
Table 5. Reproducibility of results at different concentrations of antibiotic residues in milk extracts. Source: PerkinElmer Food Safety and Quality
The data show that for each analyte present in the milk extract, the relative standard deviation of the response is less than 5% when the concentration reaches half or more of the acceptable concentration.
Table 6 presents data on the recovery of 13 antibiotics from milk at concentrations close to their acceptable limits. Absolute recovery of macrolides, sulfonamides, and β-lactams ranged from 70% to 120%, with relative standard deviations (RSD%) of less than 20% for replicate measurements (n=5).
The absolute recovery of fluoroquinolones ranged from 60% to 70%, while the recovery of tetracyclines was less than 30%. The recovery of the internal standards, ciprofloxacin-d8 and flunixin-d3, was approximately 70% and 90%, respectively.
Since fluoroquinolones and deuterated fluoroquinolones may undergo similar losses during the extraction process, internal standards can be used to compensate for the lower absolute yield of fluoroquinolones.
The recovery of fluoroquinolones and tetracyclines (Table 1) was adjusted for losses during the extraction process using the internal standard ciprofloxacin-d8.
After correction using deuterated fluoroquinolones as internal standards, the recovery of fluoroquinolones ranged from 80% to 110%, while the recovery of tetracyclines was approximately 30%. In future experiments, deuterated tetracyclines could be used to compensate for the loss of tetracyclines during the extraction process.
Table 6 lists the effects of using various organic solvents in the protein precipitation process, indicating how these solvents affect analyte recovery and matrix ion suppression.
Table 6. Mean values and relative standard deviations of the recovery and matrix effects of various antibiotics in milk extracts during protein extraction by precipitation using different extraction solvents (acetonitrile and methanol). Source: PerkinElmer Food Safety and Quality
It should be noted that, with the exception of erythromycin, the recovery of all analytes was similar in both solvents: erythromycin was recovered approximately 27% in acetonitrile and 98% in methanol.
Matrix inhibition and enhancement effects were calculated by comparing the analyte response in a milk matrix with that of a standard solvent. When using methanol or acetonitrile for sample preparation, matrix inhibition and enhancement effects caused by sulfonamides, macrolides, and β-lactam antibiotics were negligible.
Both fluoroquinolones and tetracyclines enhance ionic interactions with the drug matrix to some extent. On average, using methanol instead of acetonitrile results in a more pronounced enhancement of ionic interactions. This effect is associated with the matrix of both fluoroquinolones and tetracyclines.
Internal standards were added to five milk samples with varying fat contents (from 1% to 5%). These samples were then tested for antibiotics using the method described here.
Retention time and quantitative ion response confirmed that the target antibiotic analyte was not detected even at concentrations three times lower than the limit of quantification of the method (0.1–1 ng/mL).
The relative standard deviation (RSD) of the recoveries of the two internal standards in milk samples with different fat contents was less than 15%, indicating that milk fat content does not affect the recovery of these analytes using the sample preparation method described here.
To ensure accurate identification and confirmation, we monitored three ion transitions instead of two. This method ensures a ratio of two product ions instead of one, guaranteeing higher selectivity and better identification specificity.
The ion ratio is determined by calculating the ratio of the peak area of the ion with the lower intensity to the peak area of the ion with the higher intensity, resulting in an ion ratio of less than 100%.
Table 7 shows the average ion ratios of the target analytes. These ratios correspond to concentrations ranging from half to ten times higher than the permissible concentration in milk extracts.
Table 7. Mean values and ranges of two-ion ratios of antibiotic residues in milk extracts at various concentration levels. Source: PerkinElmer Food Safety and Quality
At various concentration levels, the relative ion ratios were within ±11% of the mean ion ratio, which is below the ±30% relative tolerance limit specified in the SANCO guidelines for the analysis of chemical residues in food .
Post time: Oct-16-2026