Full Article - PDF

AGRICULTURE AND BIOLOGY JOURNAL OF NORTH AMERICA
ISSN Print: 2151-7517, ISSN Online: 2151-7525, doi:10.5251/abjna.2013.4.6.623.628
© 2013, ScienceHuβ, http://www.scihub.org/ABJNA
Optimization of analytical parameters and adaptation of a extraction
technique of penicillin for the detection and quantification of residues of
this antibiotic by High Performance Liquid Chromatography (HPLC) in
cow’s milk in Constantine (Algeria)
Boultif. L1, Chebira. B1, Chouder. N1, Mekroud. A1
1
Padesca Laboratory, institute of Veterinary science, University of Constantine, Algeria
ABSTRACT
In order to detect and quantify the penicillin residues in milk by High Performance Liquid
Chromatography (HPLC), the authors have, previously, proceeded to the optimization of various
analysis parameters (stationary phase, mobile phase, absorption spectrum (UV detector), flow
rate and injected volume) and the adaptation of an extraction technique of this antibiotic.
Keywords: Residues, milk, penicillin, HPLC.
INTRODUCTION
Today, the antibiotics have an important place in the
treatment of production animals. Nevertheless, their
usage raises always many interrogations. We attend
currently to the reduction of their use, particularly the
recent ban, of almost all food additives antibiotic
growth factors (AFSSA , 2006), in the European
community.
Their necessity in the therapeutic arsenal and their
economical usefulness are however undeniable. It is
necessary therefore to wonder, on one hand, the
risks that affect the consumers of animal
commodities, and, on the other hand, the implications
for food processing industries. Indeed, the presence
of antibiotics residues constitutes of multiple risks on
consumer: risk of direct toxicity (Guy et al, 2004;
Leroy and Fanir, 2005), allergic risk (FedericciMathieu, 2000), carcinogenic risk, pathology risks
related to the modification of intestinal flora (Broutin.
C et al, 2005; Châtaigner and Stevens, 2005; Fabre
and Joyes, 2000), risks of appearance of selection
and of spread of bacterial resistance to antibiotics
within the human populations and animals
(Châtaigner and Stevens, 2005; Sanders, 2005). For
food processing industry, the risks concern notably
manufacturing of fermented products (Brouillet, 2002;
Form, 2003; Maghuin-Rogister et al, 2001. In fact,
the lactic bacteria being sensitive to very weak doses
of antibiotics (Brouillet, 2002; Fabre and Joyes,
2000), the presence of antibiotics residues inhibit
partially or totally the growth of these ferments and
results in many defects (Robb, 2006; Vera, 2005),
including accidents in cheese manufacturing, yogurt
and other fermented milk products (Broutin et al,
2005; Ryckaert et al, 2003; Zinedine et al, 2007). It is
necessary to know that a single intra mammary
treatment can make it unusable over 100 000 liters of
milk, which generates significant economical losses
that impact throughout the system (Abidi, 2001;
Fabre and Joyes, 2000; Mitchell, 2005).
In this context, methods must be developed for the
monitoring of antibiotics residues notably in milk. For
that two types of tests are used: microbiological tests
(screening methods), and physicochemical tests
(confirmation
methods
(Fergusson
et
al,
2002). Among the latter High Performance Liquid
Chromatography is recognized as a preferred method
for the quantitative analysis of antibiotics residues in
milk (Oliveira et al , 2006; Reig and Toldra, 2008)
since this one detects very small amounts of these
residues, presents an extreme sensitiveness,
possesses a great strength of separation and an
excellent reproducibility. This work has for principal
objective: the detection and the quantification by High
Performance
Liquid
Chromatography
(HPLC)
residues of penicillin in the cow milk in Constantine
region. We present, in this study, the optimization of
various parameters of analysis (stationary phase,
mobile phase, absorption spectrum (UV detector),
flow rate and injected volume) and the adaptation of
a technical method for extraction of this
antibiotic. The choice of this antibiotic imposes itself,
the penicillin being one of the most used antibiotics in
veterinary medicine, notably in the treatment of the
mammary gland pathology (Guerin and GuerinFaublee, 2007). This antibiotic is also directly implied
in the appearance of the problems in dairy industry in
Agric. Biol. J. N. Am., 2013, 4(6): 623-628
a Foodstuff, that is, itself considered as the principal
source of animal protein in Algeria (Amellal, 1995).
That is to regulate the different units of the HPLC
instrumentation (software, pump, and detector).
Experimental:
Optimization of analysis parameters: In order to
detect and to quantify the penicillin residues in the
milk
by
High
Performance
Liquid
Chromatography (HPLC), it was necessary to start
beforehand with the step of the optimization of the
analysis parameters namely: stationary phase,
mobile phase, solvent of dilution of the standard
penicillin, absorption spectrum (UV detector), flow
rate, injection volume. This step necessitated more
than seven hundred essays.
Instrumentation: Device HPLC Shimadzu corporate
Body. 2002 consists of: a column (type vp-ods 250
L*40, filled with grafted silica gel: column C18 in
reversed phase), two pumps (models LC-10 ATvp,
pumps 1 reference S/N: C20974009859J2, pump 2
references S/N:
C21014009440CD), two UV
detectors (models SPD-10Avp/10AVvp equipped of
a Deuterium lamp, absorption spectrum (UV detector)
190 to 350 nm, reference S/N: C21004001496LP), a
gas cleaner:
(Models DGU-20A5 reference
S/N: L20244301983CR), a control system or
integrator
(models
SCL-10
AVP,
reference
S/N: C21014009440CD), an injection loop, two
reservoirs of mobile phase and a plastic
piping allowing the progression of the liquid phase in
the different compartments of the device. Other
equipments: Centrifuge, a filtrating device and a cold
chamber.
Examination of the method’s reliability
Repetitiveness: consists in the realization of the
technique by the same team and the same
laboratory. The essay results are obtained by the
same method, on individuals of identical essays, by
the same operator, using the same equipment and
during a current interval of time (ISO 3534-1, ISO
5725-1). To test the Repetitiveness of the obtained
results, we carried out three series of 5 essays, on
the same day, with the same solution of work, the
same operator, the same place, the same equipment
and with the same conditions of analysis.
In addition to the current laboratory equipment and
accessories, we used the following specific
equipment: analytical scale, vortex, HPLC special
injection syringes (capacity 20 and 50 µl, reference
C 670-12554-03), filter ( Whatman filter and micro
porous filters; nylon porous membrane 0,45 µm X 47
mm , Supelco, reference 58067), automatic pipettes,
glass tubes and plastic bottles.
Reproducibility: in this case, the essays are realized
in different conditions. At least one of the following
parameters voluntarily is modified: operator, device,
solution setting, and differed realizations in
time... (ISO 3534-1, ISO 5725-1). To achieve this,
two different operators carried out three series of
twenty essays each, with three different work
solutions (one for every series) and in a few days
apart.
Chemicals and reagents: Standard penicillin G was
obtained from riedel-de Haën, Laborchemikalien
GMBH, sigma-aldrich, reference 46609, the methanol
from sigma-aldrich, (reference 5363C), acetonitrile
from Biochem. Chemopharma, (reference 6235A),
and water HPLC rank from sigma-aldrich, (reference
34877).
Linearity of the calibration curve: the linearity of
the variation of the heights of the peaks area
(surfaces) according to the variation of the
concentrations of the injected product must be
verified for every composite. The correlation
coefficients between heights of the peaks and
concentrations, as well as the parameters of the
corresponding lines of linear regression must be
calculated (ISO 3534-1, ISO 5725-1). To establish
the curve of linearity and to determine the regression
line for the penicillin molecule, we prepared four
dilutions ranging from 0.10 mg/ml to 0,0375 mg/ml of
the standard solution (concentration: 0,15 mg/ml).
Sampling: The milk sample was collected at the
farm situated in the town of Constantine. According
to the person in charge of the farm, the cows did not
receive any antibiotics treatment for more than two
months. The milking was done manually in a cleaned
and disinfected container, the 200 ml collected
quantity was transported in an isothermal tray
towards our laboratory where it was preserved at 4°
c.
METHODS
Determination of the limit detection: it corresponds
to the minimum concentration of an analyte that, in a
given matrix and with a specific method, has a
probability of 99% to be identified qualitatively or
Instrumentation calibration: A mixture of four
molecules was used (Anthracite, Tetracene,
acenaphthene and acenaphtylene) prepared by a
specialized firm in well-predetermined conditions.
624
Agric. Biol. J. N. Am., 2013, 4(6): 623-628
quantitatively, as being greater than zero (ISO 35341, ISO 5725-1). We proceeded to the determination
of the limit detection of our analysis method HPLC by
the preparation of penicillin dilutions. From a
standard solution of 0, 15 mg/ml, we proceeded to
-3
-10
dilutions ranging from 15.10 mg/ml to 15.10
mg/ml.
then centrifuged at 6000-rotation / min for 10
min. Next, the supernatant (aqueous portions) are
collected, filtered with the watman filters. We added
20 ml of acetonitrile to the filtrate, and then the
mixture is divided up into 4 tubes, which are
subjected to the vortex during 1 Min and centrifuged
for the second time at 3000 rotations/min for 5
min. At last the aqueous portions are harvested and
filtered two times: the first time with a Watman filter,
and the second time with a nylon membrane filter of
45 µm/ 47 mm. This method was applied on a milk
sample presumed exempt of antibiotics. A half of the
quantity of this sample was doped with the penicillin
standard; the other half was used as a control.
Statistics study
Repeatability calculation: we realized an average
comparison according to Student Test [16]. We
calculated the average, the standard deviation and
the variance from the obtained surfaces to calculate
the next “t” (calculated) and to compare it with the “t”
table of Student to the respective risks of 1 and
5%. The “t” (calculated) is measured between every
two series (a and b, a and c, b and c) according to
the following formula:
RESULTS AND DISCUSSION
Withheld
Parameters:
Our
investigations allowed keeping
parameters of HPLC as follows:
t = |Xa - Xb|/√ σ²/Na + σ²/Nb
experimental
the analysis
Sample concentration: 0, 15 mg/ml (diluted in
acetonitrile), stationary phase: column C18 reversed
phase (Grafted silica gel), mobile phase: methanol,
injected volume: 20 µl, absorption spectrum (UV
detector): 250 nm, flow rate: 1ml/min and analysis
time: 8 min.
Xa being the average of the first series (series a)
Xb being the average of the second series (series b)
σ² being the variance between the two series a and
b
Na being the number of the effective of the series a
The isocratic mode (single mobile phase) was
applied during all the experimentation.
Nb being the number of the effective of the series b
The figure below shows the obtained chromatogram
while using the withheld parameters, the retention
time of the penicillin being of 3.90 min.
The obtained t calculated is compared next with the t
table:
- If t calculated is inferior to t table; the difference
between the two series is not significant that means
that our method is repeatable (or reproducible)
- If t calculated is superior to t table; the difference
between the two series is significant that means that
our method is not reliable
Reproducibility calculation: we applied the same
preceding formula (test of Student) on the obtained
surfaces in every series.
Extraction method of penicillin in cow milk: the
second section of our study consists in the adaptation
of an extraction method of penicillin in milk. It is a
key-step that will allow us afterward to work on a
wider sampling. In order to be able to do that, we
adopted a technique from different recommended
methods (Diaz et al, 2007; Helio et al, 2007). This
technique rests on the following principle: mix 10 ml
of milk with 20 ml of acétonitrile and 6 ml of HPLC
water. The mixture is divided up into four tubes,
which are homogenized in the vortex for 1 minute
625
Agric. Biol. J. N. Am., 2013, 4(6): 623-628
Fig. 1: Chromatogram representative of the used
parameters
demonstrative. The correlation coefficient is equal to
0.9982.
Reliability
examination
of
the
withheld
parameters: The repeatability: the obtained results
(picture 1), expressed in peak area, are subjected to the
tests statistics of the Student for the verification of the
Repetitiveness and while comparing the series two by
two. After statistic analysis, it is proven that there is no
significant difference between the three series (to the risks
of 1% "t" (table) = 4,6 and 5% "t" (table) = 2,78); the t
calculated in the three cases (table I) is inferior to the “t”
table. That means that the analysis method is repeatable
in the conditions of required analysis.
Table I: Results of the statistical calculations of the
repeatability
Series a
Series b
Series c
(n=5)
(n=5)
(n=5)
Average
9536934,2
9518480,6
9544828,6
Standard
deviation
132591,55
171943,39
93079,48
Variance
17580520256
29564530647
8663791447
t (a & b)= 0.1
Fig. 2: penicillin calibration curve
Detection limit: To determine the corresponding
concentration of the limit detection of penicillin,
several dilutions at decreasing concentrations were
-10
done. A concentration of 15.10 mg/ml corresponds
to the limit of detection of this molecule. In fact, at
this concentration, the area of the peak tends
towards zero as the figure 3 confirms.
t (b & c)= 0.11
Calculated t
t (a&c)= 0.02
The reproducibility: The areas of the peaks (table II) are
subjected to the statistics tests of Student. As for the
repeatability, it was necessary to compare the series two
by two. We also noticed that there is no significant
difference between the three series (t calculated inferior to t
table). That confirms that the method is reproducible in our
analysis conditions.
Table II: Statistical results of the reproducibility
Série A
Série B
Série C
(n=20)
(n=20)
(n=20)
Average
1025413,9
1028475,35
1025029,2
Standard
deviation
57089,36
52778,67
21708,88
Variance
325919597
6
2785588238
471275526
Confidence
Interval
998040,753
<m<105278
7,05
1003169,09<
m<1053781,6
1
1014620,25<m
<1035438,15
t (A&B) = 0.17
t (B&C)= 0.27
Fig. 3: detection limit of penicillin (concentration 15.10
10
mg/ml).
Calculated t
t (A&C)= 0.04
-
Extractions method of the penicillin in milk: The
preparation of the sample is usually necessary in
order to eliminate the unwelcome components, and
to extract the antibiotics in milk. The different
The linearity of the calibration curve: The
calibration curve represented in figure 2 is
626
Agric. Biol. J. N. Am., 2013, 4(6): 623-628
methods of extraction can be applied such as: the
liquid-liquid extraction, the ultra-filtration, the
precipitation of the proteins and the solid phase
extraction (Oliveira et al, 2006). After the
implementation of the adopted extraction method in
the milk sample, we obtained a transparent and
homogenous liquid; the latter is injected directly in the
device HPLC. After the doping of the milk sample by
the penicillin standard, we obtained a corresponding
peak after 4.2 min (figure 4). As for the control
sample, no peak could be identified (figure 5).
CONCLUSION: This study involved the optimization
of various analysis parameters by high performance
liquid chromatography (stationary phase, mobile
phase, absorption spectrum (UV detector), flow rate
and injection volume) of penicillin in cow's milk from
the region of Constantine and adapting a technique
for extracting this antibiotic.
The achieved results in these preliminary steps have
already allowed the detection and quantification of
penicillin residues on a reduced sampling of milk.
This allows us to initiate a thorough study on the
presence and quantity of the antibiotic residues in
milk produced in the region of Constantine.
REFERENCES
1.
Abidi. K, Résidus d’antibiotiques dans le lait de
boisson. Thesis: veterinary medicine. Sidi Thabet,
Tunisia. (2004), p. 6-23.
2.
Amellal. R. La filière lait en Algérie : entre objectif
de la sécurité alimentaire et la réalité de la
dépendance. Review : Options Méditerranéennes.
Series B. N°14. (1995). p. 229-230.
3.
AFSSA (Agence Française de la Sécurité
Sanitaire des Aliments). Usages vétérinaires des
antibiotiques,
résistance
bactérienne
et
conséquences pour la santé humaine. (2006). p.
49-55.
4.
Brouillet. P. Résidus de médicaments dans le lait
et tests de détection. Revue : Bulletin des GVT.
N°15. Mai-Juin 2002, (2002). p. 25-41.
5.
Broutin. C et al. Maîtrise de la qualité dans la
transformation laitière. Guide de bonnes pratiques
d’hygiène, (2005). p. 29-31.
6.
Châtaigner. B, Stevens. A. Investigation sur la
présence de résidus d’antibiotiques dans les
viandes commercialisées a Dakar. Institut
Pasteur de Dakar. (2005). p. 6-9.
7.
Diaz. T. G, Cabanillas. A. G. Rapid
Determination of Sulfathiazole, Oxytetracycline
and Tetracycline in Honey by High-Performance
Liquid Chromatography. Revue : Analytical
Letters. N°23 (4). (2007). p. 607-616
8.
Fabre. J.M et Joyes. D. Résidus dans le lait :
observation des inhibiteurs bien utiliser les
médicaments.
Proceedings : lait, qualité et
santé. (2000). p. 10-12.
9.
Federicci-Mathieu. C. Résidus dans le lait et
sécurité alimentaire : quels risques ? quels
moyens de maîtrise ? Revue : Bulletin des GVT.
Avril-Mai 2000. N°7. (2000). p. 21-22.
Fig. 4: Doping milk sample
Fig. 5: Control sample
627
Agric. Biol. J. N. Am., 2013, 4(6): 623-628
denrées alimentaires. Services scientifiques du
premier
Ministre
Affaires
scientifiques,
techniques et culturelles (SSTC) France.
Rapport Final SSTC, (2001). p. 13-58.
10. Fergusson J.P, Baxter G.A, Mac Evoy J.D.
Detection
of
streptomycin
and
dihydrostreptomycin residues in milk, honey and
meat. Revue: The Analyst. N°127, (2002). p 951956
17. Mitchell. M. Détection des résidus d'antibiotiques
dans le lait de chèvre. Revue : Artificial
Intelligence. N° 170(18). (2005). p. 1194-1212.
11. Form. G. Les résidus inhibiteurs dans le lait.
Evolution des méthodes de détection Facteurs de
risques en région Rhône-Alpes. Thèse de docteur
vétérinaire. Université Claude-Bernard, Lyon.
(2003) p 26-30.
18. Oliveira. R, De Pietro. A, Cass. Q Quantification of
cephalexin as residue levels in bovine milk by
high-performance liquid chromatography with online sample cleanup. Revue : Talanta. N° 71.
(2006). p. 1233–1238.
12. Guerin. P, Guerin-Faublee. V. Mammites de la
vache laitière. Thèse docteur vétérinaire. Ecole
nationale vétérinaire de Lion. (2007). p. 77-117.
19. Reig. M, Toldra. F. Veterinary drug resides in
meat: concerns and rapid methods for detection.
Meat Science. N° 78. (2008). p 60-67.
13. Guy. P, Royer. D, Mottier. P, Gremaud. E,
Perisset. A, Stadler. R. Quantitative determination
of chloramphenicol in milk powders by isotope
dilution liquid chromatography coupled to tandem
mass
spectrometry.
Revue :
Journal
of
Chromatography A. N°1054. (2004). p. 365–371.
20. Robb. Ed. Pour un fromage de meilleure qualité.
Revue : pour l’amour des vaches. Vol 5. N°1.
(2006). p. 6.
21. Ryckaert. I et al. 42 questions sur le lait. Édition
IMP Bruxelles. (2003). p. 13-56.
14. Helio. A, Martins. J, Tereza. A, Kussumi. A, Wang.
D, Lebre. T. A rapid method to determine
antibiotic residues in milk using liquid
chromatography coupled to electrospray tandem
mass spectrometry. Revue: Journal of the
Brazilian Chemical Society, volume 18, N°2 São
Paulo. Mars-Avril 2007. (2007). p. 52-71.
22. Sanders. P. L’antibiorésistance en médecine
vétérinaire : enjeux de santé publique et de santé
animale. Revue : Bulletin de l’académie
vétérinaire de France. tome 158. N°2. (2005). p.
139-140.
23. Vera. G. La rationalisation des traitements des
mammites en exploitation laitière. Thèse de
docteur vétérinaire. Ecole Nationale d’Alfort.
(2005). p. 26-40.
15. Leroy. P et Fanir. F. Méthode statistique en
médecine vétérinaire. Faculté de médecine
vétérinaire de liège. p. 49-133.
24. Zinedine. A, Faid. M, Benlemlih. M. Détection des
résidus d’antibiotiques dans le lait et les produits
laitiers par méthode microbiologique. Revue :
REMISE. Vol 1. N°1. (2007). p. 1-9.
16. Maghuin-Rogister. G, Janosi. A, Helbo. V, Van
Peteghem. C, Sanders. E, Van Eeckhout. E,
Cornelis. M et Jouret. M. Stratégie intégrée
d’analyse qualitative et quantitative des résidus
de substances antimicrobiennes dans les
628