Accepted Manuscript

Detection in raw cow's milk of coliform bacteria - Reservoir of antibiotic resistance

Jolanta Godziszewska, Ewelina Pogorzelska-Nowicka, Marta Brodowska, Grażyna

Jagura-Burdzy, Agnieszka Wierzbicka

PII: S0023-6438(18)30325-6
DOI: 10.1016/j.lwt.2018.04.019
Reference: YFSTL 7030

To appear in: LWT - Food Science and Technology

Received Date: 19 November 2017

Revised Date: 10 March 2018
Accepted Date: 6 April 2018

Please cite this article as: Godziszewska, J., Pogorzelska-Nowicka, E., Brodowska, M., Jagura-Burdzy,
Graż., Wierzbicka, A., Detection in raw cow's milk of coliform bacteria - Reservoir of antibiotic resistance,
LWT - Food Science and Technology (2018), doi: 10.1016/j.lwt.2018.04.019.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1

2 Detection in raw cow’s milk of coliform bacteria - reservoir of antibiotic resistance

3 Jolanta Godziszewska1, Ewelina Pogorzelska-Nowicka1*, Marta Brodowska1, Grażyna

4 Jagura-Burdzy2, Agnieszka Wierzbicka1

PT
1
5 Department of Technique and Food Development, Faculty of Human Nutrition and Consumer Sciences,

6 Warsaw University of Life Science, Nowoursynowska 159c, 02-776 Warsaw, Poland

RI
2
7 Institute of Biochemistry and Biophysics PAS, Department of Microbial Biochemistry, 02-106 Warsaw, Poland

SC
*
8 correspondence author: Ewelina Pogorzelska-Nowicka, e-mail: ewelina_pogorzelska@sggw.pl

9 Running title: Detection of bacteria in milk

U
AN
10 Keywords: antibiotic resistance, Escherichia coli, raw milk, volatile compounds

11
M
D
TE
C EP
AC

1
 ACCEPTED MANUSCRIPT
12 Abstract

13 Antibiotic resistance is one of the main threat for people, especially since food products are

14 more frequently identified as a reservoir of antibiotic resistance genes. Thus, the aim of the

15 study was to analyse the Polish raw cow’s milk in terms of the presence of coliform bacteria

PT
16 resistant to antibiotic as well as to identify volatile compounds (VC) characteristic for such

17 bacteria. Further, the identified profile of VC could be used in development of coliform

RI
18 bacteria detection method directly in the product. The obtained data showed that among

19 coliform bacteria strain isolated from Polish raw cow’s milk 88% were resistant to penicillin,

SC
20 39% to kanamycin, 43% to streptomycin, 78% to chloramphenicol and 55% to tetracycline.

U
21 Moreover, tetM gene, frequently connected with mobile genetic elements, were detected in
AN
22 42% of isolated coliform strains. Due to widespread of coliform bacteria resistant to antibiotic

23 in raw milk, simple methods of detection of coliform bacteria in such product are required.
M

24 Performed analysis suggests that lack of methional might be used as an indicator of raw cow’s

25 milk contamination with coliform bacteria.

D
TE

26
C EP
AC

2
 ACCEPTED MANUSCRIPT
27

28 1. Introduction

29 Nowadays, a significant increase in the number of strains resistant to antibiotics is being

30 widely observed, what results from the common use of antibiotics both in medicine and as

PT
31 “growth promoters”(Zycka-Krzesinska, Boguslawska, Aleksandrzak-Piekarczyk, Jopek, &

32 Bardowski, 2015). Thus, antibiotics are becoming ineffective in treatment of bacterial

RI
33 infections and antibiotic resistance is considered as one of the biggest threat to people in the

SC
34 world (Aarestrup, 2012; Godziszewska, Guzek, Gląbski, & Wierzbicka, 2016). Antibiotic

35 resistance genes can be located on mobile genetic elements and as such transferred by

U
36 horizontal gene transfer from food bacteria to bacteria living in humans body (Spanu et al.,
AN
37 2014). It is noteworthy that such transfer can occur between strains of the same or different

38 species and can relate both pathogenic and non-pathogenic bacteria (Spanu et al., 2010).
M

39 However, when donor of the antibiotic resistance module and its recipient belong to the same
D

40 family (e.g. Enterobacteriaceae) such horizontal gene transfer is extremely efficient (Kruse &
TE

41 Sørum, 1994). It was demonstrated that normal human flora can be an acceptor and donor of

42 transmissible antimicrobial resistance genes (Saenz et al., 2004). Natural commensal of

EP

43 human gastrointestinal tract is for example Escherichia coli (Eckburg et al., 2005). Tadesse et

44 al. (2012) showed that the number of multidrug resistant E. coli strains increased by 56.4% in
C

45 the last 50 years. It could be connected with the spread of antibiotic resistance among bacteria
AC

46 from different ecosystems and the localisation of some antibiotic resistance genes on mobile

47 genetic elements (Eckburg et al., 2005). Moreover, antibiotic resistant bacteria are widely

48 detected in different food products such as meat, fruit, vegetables, cheese, milk (Phillips,

49 2007; Cook et al., 2011; Kitai et al., 2005; Kwon et al., 2006; Skočková, Bogdanovičová,

50 Koláčková, & Karpíšková, 2015).

3
 ACCEPTED MANUSCRIPT
51 Nowadays, consumption of raw milk as “natural and local product” is becoming more

52 popular. Especially that lower risk for developing asthma, hay fever, allergic rhinitis, pollen

53 allergy as well as atopic sensitization might be connected with early-life exposure to

54 unprocessed cow’s milk (Claeys et al., 2013; Loss et al., 2011; Sozaska, Pearce, Dudek, &

55 Cullinan, 2013). However, raw milk can be contaminated by pathogenic bacteria and as a

PT
56 result there is a high risk of transfer of coliform pathogens from milk to consumers (Claeys et

RI
57 al. 2013). It was shown that coliform bacteria were detected in 95% of bulk tank milk samples

58 collected in 21 states of the US and that pathogenic Escherichia coli strain cause frequent

SC
59 outbreaks of human infection after raw milk consumption (Oliver, Jayarao and Almeida,

60 2005). Simultaneously, it was reported that antibiotic resistant E. coli were isolated from raw

61
U
milk (Skočková et al., 2015; Bonyadian, Moshtaghi, & Taheri, 2014; Paneto, Schocken-
AN
62 Iturrino, Macedo, Santo, & Marin, 2007). There are law regulations regarding microbial
M

63 criteria for raw cow’s milk (for example in Europe UE no. 1662/2006), however even low

64 number of pathogens, raw milk may be harmful for consumers (Claeys et al. 2013). Microbial
D

65 analyses by standard plate method (PN-EN ISO 4833-1:2013), methods based on flow
TE

66 cytometry (Gunasekera, Attifield & Veal, 2000) or enzymology are time consuming (Silva,

67 Kanugala & Weerakkody, 2015), so there is a need for new methods allowing analysis of
EP

68 microbial contamination in raw milk samples in faster, cheaper and more specific manner.

It was shown that some organoleptic features of milk, such as aroma and taste, can be
C

69
AC

70 affected by the coli type bacteria (Ziarno & Czapska, 2008). Godziszewska et al. (2017)

71 suggested that changes of organoleptic features of food products caused by bacteria can be

72 identified by volatile compounds (VC) method. The VC analysis is fast, cheap and does not

73 require reagents and was applied for analyses of microbial induced changes of milk during

74 storage or recognition of spoilage of milk cause by some strains of bacteria or yeast (Silcock

4
 ACCEPTED MANUSCRIPT
75 et al., 2014). However, no information are available regarding the profile of volatile

76 compounds associated with the contamination of raw cow’s milk by coliform bacteria.

77 Thus, the aim of this study was to develop fast method of the detection of coliform

78 bacteria in raw cow’s milk by analysis of volatile compounds profile as well as to determine

79 antimicrobial resistance phenotype of coliform bacteria isolated from the products with

PT
80 emphasis on antibiotics used in veterinary, i.e. penicillin, kanamycin, streptomycin,

RI
81 chloramphenicol, tetracycline.

82

SC
83 2. Materials and Methods

U
84 2.1. Characterization of milk samples
AN
85 Milk was obtained from fifteen farms in Poland. Samples were collected three-four hours
M

86 after morning milking from bulk tank milk in aseptic condition. Sterile, glass bottles were

87 used as containers. The milk samples (250 mL each) were transported refrigerated, under
D

88 aerobic conditions to the Laboratory of Food Chemistry at Warsaw University of Life Science
TE

89 and stored at 4ºC. The first analyses were performed 24 hours after milking Randomly

90 selected milk samples were stored up to 4 days at 4ºC and were examined on the 5th day after
EP

91 milking. The pH was measured by a hand-held pH meter (model 205, Teso AG, Lenzkirch,
C

92 Germany). Two buffers solution (pH=4.01 and pH=7.00) were used for calibration of the pH
AC

93 meter. Each milk sample was analysed in triplicate.

94 2.2. Volatile compounds analyses

95 2.2.1. Volatile compounds analyses of the raw milk samples

96 Each milk sample (2 mL) was transferred into glass vials (20 mL) capped with a teflon faced

97 silicon rubber cap. Volatile compounds analyses were performed by using the Heracles II

5
 ACCEPTED MANUSCRIPT
98 electronic nose (Alpha M.O.S., Toulouse, France) based on ultrafast gas chromatography

99 according the method described by Brodowska et al. (2016) and Wojtasik-Kalinowska et al.,

100 (2016). The identification of characteristic volatile compounds was done on the basis of

101 Kovats’ relative retention index (Goodner, 2008) and AroChemBase database (Alpha M.O.S.,

102 Toulouse, France). Solutions of alkanes (n-butane to n-hexadecane) were used for method

PT
103 calibration. Six randomly selected milk samples were analysed in three independent replicates

RI
104 on the first and the fourth day of the experiment.

105 2.2.2. Volatile compounds analyses of overnight cultures of Escherichia coli DH5α

SC
106 Escherichia coli DH5α strain was grown for 24 h at 37°C in LB-broth (tryptone – 10 g L-1,

U
107 yeast extract 5 g L-1, NaCl – 10 g L-1) at 37°C. Overnight culture of E.coli (NC) - it is LB-
AN
108 broth after incubation of E. coli for 24 h at 37°C. LB-broth incubated for 24 h at 37°C was

109 marked as broth. In the next step, 2 mL of prepared NC and broth were transferred into glass
M

110 vials (20 mL) and capped with teflon faced silicon rubber cap. The identification of volatile
D

111 compounds of broth and NC was conducted as for the volatile compounds analyses of milk.
TE

112 The analyses were performed in three independent experiments, each in three repetitions.

113 2.3. Microbiological analysis

EP

114 2.3.1. The total number of aerobic bacteria in raw milk

C

115 Microbiological analyses were conducted according to PN-EN ISO 4833-1:2013

AC

116 recommendations (“Norm PN-EN ISO 4833-1:2013) with some modifications. Milk sample

117 (1 mL) was diluted in 0.9% NaCl. Ten microliters of serial 10 fold dilutions of the milk were

118 spotted onto L agar plates (PCA plates – GRASO BIOTECH, Starogard Gdański, Poland).

119 After plates were incubated for 24 h at 37°C, colony forming units (CFU) of total aerobic

120 bacteria were counted and presented as CFU L-1 of the milk. Each sample was analysed in

121 four repetitions.

6
 ACCEPTED MANUSCRIPT
122 2.3.2. Enterobacteriaceae family representatives in raw milk

123 Bacteria from Enterobacteriaceae family were determined using violet red bile lactose agar

124 (VRBA) plates (GRASO BIOTECH, Starogard Gdański, Poland) according to PN-EN ISO

125 4832:2007 recommendations (Norm PN-EN ISO 4832:2007) with modifications. Analyses

126 were performed as described above. VRBA plates are recommended for detection and

PT
127 determination of the number of coliform bacteria in food and dairy products. Each sample was

RI
128 analysed in four repetitions.

129 2.4. The antibiotics susceptibility of the strains isolated from VRBA plates

SC
130 The strains isolated from VRBA plates were analysed for resistance to penicillin, kanamycin,

U
131 streptomycin, chloramphenicol, tetracycline. The antibiotic resistance of the strains were
AN
132 determined on VRBA plates with the addition of the appropriate concentration of the

133 antibiotics (Table 1). The minimum inhibitory concentration (MIC) of the antibiotics were
M

134 estimated based on analysis of reference strain – Escherichia coli DH5α. The colonies
D

135 (usually from 104 – fold dilution of the milk sample) were re-streaked onto VBRA plates with
TE

136 the selective antibiotic. Resistance of the isolated coliform bacteria was expressed as the

137 percentage of resistance colonies.

EP

138 2.5. PCR analysis

C

139 The colonies from the VRBA plates were suspended in water and subjected to heat treatment
AC

140 (100°C, 10 min). The released genomic DNA was used in the PCR reactions. PCR reactions

141 were performed according procedure described previously (Mullis et al., 1986). The primers

142 were: 5’GTGGACAAAGGTACAACGAG, 5’CGGTAAAGTTCGTCACACAC as described

143 before (Chajęcka-Wierzchowska, Zadernowska, Nalepa, Sierpińska, & Łaniewska-

144 Trokenheim, 2015). Finally, the PCR reactions were performed with an initial denaturation

145 step (95°C for 5 minutes) and 25 cycles of denaturation at 95°C for 30 seconds, annealing at

7
 ACCEPTED MANUSCRIPT
146 (52°C-62°C) for 30 seconds and elongation at 72°C for 60 seconds. Reactions ended with a

147 final elongation step (72°C for 7 minutes). PCR products were separated on 1% agarose gels,

148 stained with the DNA/RNA SimplySafeTM (EURx Molecular Biology Products, Gdańsk,

149 Poland) and visualized under UV light.

150 2.6. Statistical analyses

PT
151 All analyses were made in Statistica Software version 12.0 (StatSoft, Tulsa, Oklahoma, USA).

RI
152 The normality of data distribution was analysed using the Shapiro-Wilk tests. The statistical

153 differences were determined using one-way analysis of variance (ANOVA) or U-Mann-

SC
154 Whitneya test. The correlations were analysed using Spearmann’s rank correlation. The plot

U
155 of the Figure 2 was prepared using BoxPlotR (Spitzer, Wildenhain, Rappsilber, & Tyers,
AN
156 2014). All analyses were performed with significance level α=0.05.

157 3. Results and Discussion

M

158 3.1. Characterisation of milk samples

D

159 The samples brought to the Laboratory of Food Chemistry, WULS-SGGW were analysed. It
TE

160 was shown that pH value of the analysed samples were in range 5.98-6.61. In the next step,

161 the total number of aerobic bacteria in the milk sample was examined. In the analysed
EP

162 samples the number of CFU in mL of raw milk varied between 1x103-5x104. The most of the

163 samples met the law regulation since according committee regulation of UE no. 1662/2006
C

164 the number of bacteria in cow’s milk should not exceed 1x105 CFU mL-1 of milk (UE, 2006).
AC

165 Since coliform bacteria were reported previously to be frequently isolated in raw Polish milk

166 (Jarosinska, Barlowska, Wolanciuk, Pastuszka, & Barlowska 2014) the number of coliform

167 bacteria were analysed in the collected milk samples. It was shown that the milk samples

168 differed in quantity of the coliform bacteria, which varied between 1x103-1x104 CFU mL-1. It

169 was described previously that cooling procedure of raw milk had no effect on coliform

170 bacteria and total aerobic microbial counts (Lee, Barbano & Drake 2016) and that an

8
 ACCEPTED MANUSCRIPT
171 occurrence of coliform bacteria was connected with contamination of milk (Ziarno & Czapska

172 2008). It may be assumed that differences in the number of coliform bacteria result from

173 variable level of hygiene on the farms.

174 3.2. Susceptibility of isolated coliform bacteria to penicillin, kanamycin,

175 streptomycin, chloramphenicol and tetracycline

PT
176 Since it was suggested that Enterobacteriaceae, isolated from raw cow’s milk samples, might

RI
177 be a reservoir of antibiotics resistance genes (Skočková et al., 2015; Bonyadian et al., 2014;

178 Paneto et al., 2007) the antibiotic resistance phenotypes of coliform bacteria from milk

SC
179 samples were examined. The pink colonies isolated from VRBA plates were re-streaked onto

180 the same medium with the appropriate antibiotic (penicillin, kanamycin, streptomycin,

181
U
chloramphenicol or tetracycline). The tested antibiotics were selected on the bases of their
AN
182 usage in veterinary. E. coli DH5α was used as a control sensitive strain. The counted
M

183 proportions of coliform bacteria resistant to penicillin, kanamycin, streptomycin,

184 chloramphenicol and tetracycline are presented in Table 2. The obtained data pointed out that
D

185 a profile of antibiotic resistance of strains for each milk sample was different. However, the
TE

186 antibiotic resistant coliform strains were detected for each tested antibiotic in each analysed

187 milk sample. The results suggest that antibiotic resistant strains were spreading in raw cow’s
EP

188 milk in Poland. These data correspond with the observation made by other groups, showing

that coliform bacteria resistant to antibiotics were isolated from milk samples in different
C

189
AC

190 countries. Skočková et al. (2015) showed that E.coli strains isolated from raw cow’s milk in

191 Czech Republic were resistant to β-lactams and tetracycline. Moreover, milk samples

192 collected in Greece were contaminated with E.coli strains resistant to ampicillin,

193 streptomycin, gentamicin, cefuroxime as well as tetracycline. Different antibiotic resistance

194 profiles were described for coliform bacteria isolated from bovine milk in Iran were resistant

195 to oxytetracycline, cephalexin, nalidixic acid, nitrofurantoin, gentamicin, trimethoprim-

9
 ACCEPTED MANUSCRIPT
196 sulfamethoxazole, ciprofloxacin, ampicillin, streptomycin were identified (Bonyadian et al.,

197 2014). The data obtained during this study indicated that Polish milk might be also a reservoir

198 of coliform bacteria carrying genes determining multidrug resistance phenotypes. Moreover

199 the comparison of the results with literature data revealed that the antibiotics resistance

200 profiles of coliform bacteria were specific for each analysed country.

PT
201 Chajęcka-Wierzchowska et al. (2015) as well as Zycka-Krzesinska et al. (2015) showed that

RI
202 in Staphylococcus spp., Lactococcus lactis, which were isolated from Polish food products,

203 genes determining tetracycline resistance located on conjugative transposon Tn916 were

SC
204 identified.

205 3.3. Identification of the tetM genes in the isolated coliform bacteria

206
U
The tetM gene (connected with resistance to tetracycline) is frequently located on Tn916
AN
207 transposon, which can be transferred naturally between a variety of gram-positive and gram-
M

208 negative bacteria (Bertram, Strätz, & Dürre, 1991). Such localisation is associated with higher

209 risk of natural transfer of tetracycline resistance to other bacteria. Therefore, to verify the
D

210 presence of tetM gene in the isolated coliform bacteria PCR reactions were performed. In the
TE

211 TetR strains, tetM genes were probed using specific primers described by Chajęcka-

212 Wierzchowska et al. (2015). The photograph of exemplary gel visualising PCR product
EP

213 analysis is shown in Figure 1. The presence of tetM gene was detected in 42% of the analysed

TetR coliform bacteria. The obtained results suggest that raw milk contaminated by coliform
C

214
AC

215 bacteria might be of public health significance due to the possible spread of tetracycline

216 resistance. Thus, quick detection of milk contamination with coliform is important in to

217 prevention of spread of antibiotic resistance. To make it the most effective, a presence of

218 antibiotics resistant of coliform bacteria in raw milk should be analysed as good

219 manufacturing practise (GMP) as well as hazard analysis and critical point (HACCP) in dairy

220 factories. Although HACCP implementation in small daily factory needs larger financial

10
 ACCEPTED MANUSCRIPT
221 expenses, such control could contribute to improve the quality of product and consumer

222 satisfaction with special attention on consumers health (Cusato et al., 2013; Cusato et al.,

223 2014).

224 3.4. Analysis of the volatile compound profile of E. coli DH5α

225 Growth of microorganisms in raw milk leads to induction bacterial enzymes activity like

PT
226 lipases, proteases or esterases that interfere with fermentation pathways. Therefore,

RI
227 determination of presence of specific microorganisms with undesirable phenotypes in milk,

228 has become increasingly important (Murphy, Martin, Barbano & Wiedmann, 2016).

SC
229 Moreover, coliform bacteria might be dangerous for consumers of raw milk, what creates a

230 need of faster and cheaper methods of detection of coliform, which should be applied as

231
U
element of monitoring of quality of the product (Oliver, Jayarao and Almeida, 2005). The
AN
232 electronic nose may be useful equipment for detection of bacterial contamination
M

233 (McEntegart, Penrose, Strathmann, & Stetter 2000). Simultaneously, it was verified that

234 analysis of a volatile compound profile of the food products allowed to detect a spoilage
D

235 caused by bacteria (Godziszewska et al., 2017). The usage of electronic nose in factory may
TE

236 have some disadvantages, because requires specific and time-consuming training for staff

237 (Górska-Horczyczak et al., 2016). Moreover, control samples should be frequently analysed
EP

238 and monitored, because its volatile compound profile may be affected by many factors like

animal diet or storage near aromatic products (Wojtasik-Kalinowska et al. 2016). It was
C

239
AC

240 shown that fatty acids composition and its metabolize influence on a volatile compound

241 profile of analysed product (Brodowska et. al. 2016, Wojtasik-Kalinowska et al., 2016).

242 However, the methodology of usage of electronic nose is simple, no-expensive as well as fast

243 and thank to that features very useful (Górska-Horczyczak et al., 2016).

244 In order to determine changes of volatile compounds in raw cow’s milk caused by coliform

245 bacteria, the volatile compound profile of one of the coliform bacteria – E. coli was

11
 ACCEPTED MANUSCRIPT
246 determined. The number of E. coli cells in examined overnight culture was near 1x109 CFU

247 mL-1, the sterile broth was used as a control. Previous analysis showed that characteristic

248 volatile organic compounds for E. coli are indoles (Yu, Hamilton-Kemp, Archbold, Collins, &

249 Newman 2000, Zhu et al. 2010). Since, indoles were also identified in raw milk (Moid,

250 Etievant, Langlois, Dekimpe, & Addeo, 1994) analysis of volatile compounds was done with

PT
251 use of appropriate columns.

RI
252 The analysis of volatile compounds of broth and overnight culture of E. coli was performed

253 by electronic nose Heracles II based on ultra-fast gas chromatography with two columns

SC
254 MXT-5 and MXT-17. Indole were identified near the end of the chromatographic separation

255 of the aroma (1289, 1564 - Kovats index for columns MXT-5 and MXT-17 respectively) and

256
U
were not analysed during the study. On the basis of chromatographic separation on MXT-5
AN
257 and MXT-17 columns and analysis of the data in AroChemBase database volatile compounds
M

258 profiles of overnight culture of E.coli in LB-broth (NC) and LB-broth itself were compared in

259 order to identify characteristic substances for the growth of E.coli.. It was shown that the
D

260 volatile compounds identified in broth and overnight culture of E. coli partially overlapped
TE

261 (Table 3). Statistical analysis of the obtained data showed that relative concentration of

262 trimethylamine, propanal, 1-propanol; 2-methyl, methional, (E)-2-penten-1-ol, butanal was

EP

263 lower in the profile of NC than in broth. Such lower values might suggest that E.coli

metabolises the compounds during growth. Simultaneously, it should be noted that an amount
C

264
AC

265 of methanol and 2-methylpropanal were higher in the NC profile than in the broth profile. It

266 might imply that these volatile compounds are produced during growth of E. coli. The

267 comparison of volatile compounds identified in the broth and overnight culture makes it

268 possible to propose the volatile compounds characteristic for E.coli (without indole) by

269 eliminating substances found only in broth. The 24 identified volatile substances are listed in

270 Figure 2 accompanied with graph indicating their amounts. The graph was created in

12
 ACCEPTED MANUSCRIPT
271 BoxPlotR (n.d.). Such type of plot includes statistical analysis of obtained data. It was shown

272 that in the volatile compound profile of E. coli dominated 1-propanol, 2-methyl, 2-

273 methylpropanol, methanol, propanol and butanal. These data correspond with the observations

274 that E. coli cells produce methanol (Toews & Adler, 1979) but not with data of other authors

275 (Patrignani et al., 2008; Yu et al., 2000). However, the differences may be the result of the

PT
276 use of columns with different specificity.

RI
277 3.5. Analysis of volatile compounds of raw cow’s milk samples during storage

278 To determine typical volatile compounds of milk, various samples of raw milk from

SC
279 different farms were analysed using the electronic nose also with columns MXT-5 and MXT-

280 17. To follow the changes of volatile compounds associated with growth of bacteria the

281
U
samples were examined 24 hours after milking and 5 days after milking (samples were stored
AN
282 four days). Pasteurised or UHT milk could not be use as a sterile control for this experiment,
M

283 because performed analysis indicated that the volatile compound profile of pasteurised milk

284 or UHT milk was completely different from the volatile compound profile of raw milk (data
D

285 not shown). Moreover, literature data suggests that based on routinely performed
TE

286 microbiological and sensory tests of raw milk, shelf-life of pasteurized milk obtained from

287 analysed raw milk could not be predicated (Martin et al., 2011). In milk analysed 24 hours
EP

288 after milking 10 typical volatile substances were identified (data not shown). In the milk

stored four days presence of 13 typical volatile compounds were observed (data not shown).
C

289
AC

290 The volatile compounds, which changed concentration during the storage of raw milk, are

291 presented in Figure 3. Statistically significant increase in methanol was observed during the

292 storage. On the other hand, on the fourth day of the storage lower concentration of propanal,

293 phenol, 2-methyl, 2-methylpropanal, dimethyl disulphide and methional were noted. It was

294 shown before that E. coli may be an efficient fermenter changing the volatile profile of UHT

295 milk, what was visualised by PCA analysis (Ali, O’Hare, & Theaker, 2003). Such

13
 ACCEPTED MANUSCRIPT
296 characteristic might explain the changes of concentration of volatile compounds as a result of

297 coliform growth in the milk samples.

298 To select a volatile compound, which could be indicator of milk contaminated by

299 coliform bacteria, the comparison of volatile compounds of E. coli with the typical volatile

300 compounds of raw cow’s milk during the storage was made. The volatile compounds present

PT
301 in overnight culture of E.coli and in milk stored four days and absent in broth and in milk

RI
302 analysed 24 hours after milking were not identified. However, volatile compounds, which

303 concentration changed in the same direction (increased or decreased) during the storage of

SC
304 broth and of milk were noticed. Correlations between those compounds in pairs: broth to

305 milk, (day 0) as well as NC : milk, (day 4) were performed (Table 4). Statistically significant

306
U
correlations were observed for methanol, and moderate but, close to significance negative
AN
307 correlations were noticed for propanal. It is noteworthy that content of methional decrease
M

308 during growth of E.coli in broth (Table 3). After the storage of the milk methional was hardly

309 detected. It might suggest that high content of methanol, low content of propanal and lack of
D

310 methional result from coliform growth in raw milk. Especially as it was reported that
TE

311 methional is a known compound produced by Lactococcus lactis from methionine (Amrita,

312 Ferndez-Espl, Requena, & Pelaez, 2001). Based on performed analyses it was concluded that
EP

313 the lack of methional in the volatile compound profile of samples may be used as the first

indicator of coliform contamination of raw milk, since absence of substance is better indicator
C

314
AC

315 than change in the concentration of the compound.

316 4. Conclusions

317 The presented study showed that Polish raw cow’s milk is contaminated with coliform

318 bacteria, which are resistant to antibiotics used in veterinary. Moreover, it was noted that

319 tetracycline resistance were widely spread among coliform bacteria in analysed milk samples

320 and that this phenotype frequently was connected with tetM gene characteristic for mobile

14
 ACCEPTED MANUSCRIPT
321 genetic element - conjugative transposon Tn916. Transfer of antibiotic resistance from milk

322 coliform bacteria to coliform bacteria of human gastrointestinal tract may occur with high

323 risk. Moreover, the performed volatile compound analyses allow to propose lack of methional

324 as the first indicator of raw cow’s milk contamination with coliform bacteria.

325

PT
326 5. Acknowledgements

RI
327 Research financed by Polish Ministry of Science and Higher Education within funds of

328 Faculty of Human Nutrition and Consumer Sciences, Warsaw University of Life Sciences

SC
329 (WULS-SGGW), for scientific research. We thank Krzysztof Glabski PhD, Department of

330 Microbial Biochemistry, PAS for valuable suggestions during writing of the manuscript.

331
U
AN
332 6. References
M

333 *Aarestrup, F. (2012). Sustainable farming: Get pigs off antibiotics. Nature, 486(7404), 465–

334 466.
D

335 Ali, Z., O’Hare, W., & Theaker, B. (2003). Detection of bacterial contaminated milk by
TE

336 means of a quartz crystal microbalance based electronic nose. Journal of Thermal

337 Analysis and Calorimetry, 71(1), 155–161.

EP

338 Amrita, F., Fernndez-Espl, D., Requena, T., & Pelaez, C. (2001). Conversion of methionine to

methional by Lactococcus lactis. FEMS Microbiology Letters, 204(1), 189–195.

C

339
AC

340 https://doi.org/10.1111/j.1574-6968.2001.tb10884.x

341 *Bertram, J., Strätz, M., & Dürre, P. (1991). Natural transfer of conjugative transposon Tn916

342 between gram-positive and gram-negative bacteria. Journal of Bacteriology, 173(2),

343 443–448.

344 Bonyadian, M., Moshtaghi, H., & Taheri, M. A. (2014). Molecular characterization and

345 antibiotic resistance of enterotoxigenic and entero-aggregative Escherichia coli

15
 ACCEPTED MANUSCRIPT
346 isolated from raw milk and unpasteurized cheeses. Veterinary Research Forum, 5(1),

347 29–34.

348 Brodowska, M., Guzek, D., Ko\lota, A., G\ląbska, D., Górska-Horczyczak, E., Wojtasik-

349 Kalinowska, I., & Wierzbicka, A. (2016). Effect of diet on oxidation and profile of

350 volatile compounds of pork after freezing storage. Journal of Food & Nutrition

PT
351 Research, 55(1). Retrieved from

RI
352 http://search.ebscohost.com/login.aspx?direct=true&profile=ehost&scope=site&authty

353 pe=crawler&jrnl=13368672&AN=113242504&h=Q4o6Xveg72oJ6cIWxB%2F3rGm

SC
354 UKb9F09qkptSHam5dXFx8zK9mqL%2FWYmfVzijv0QRB4u8EelKZb57YKQo%2F

355 4t9NTA%3D%3D&crl=c

356
U
*Chajęcka-Wierzchowska, W., Zadernowska, A., Nalepa, B., Sierpińska, M., & Łaniewska-
AN
357 Trokenheim, Ł. (2015). Coagulase-negative staphylococci (CoNS) isolated from
M

358 ready-to-eat food of animal origin – Phenotypic and genotypic antibiotic resistance.

359 Food Microbiology, 46, 222–226. https://doi.org/10.1016/j.fm.2014.08.001

D

360 Claeys, W. L., Cardoen, S., Daube, G., De Block, J., Dewettinck, K., Dierick, K., … Herman,
TE

361 L. (2013). Raw or heated cow milk consumption: Review of risks and benefits. Food

362 Control, 31(1), 251–262. https://doi.org/10.1016/j.foodcont.2012.09.035

EP

363 Cook A, Reid-Smith RJ, Irwin RJ, McEwen SA, Young V, Butt K, & Ribble C. (2011).

Antimicrobial resistance in Escherichia coli isolated from retail milk-fed veal meat
C

364
AC

365 from Southern Ontario, Canada. J Food Prot, 74(8), 1328–1333.

366 Cusato, S., Gameiro, A. H., Corassin, C. H., Sant’Ana, A. S., Cruz, A. G., Faria, J. de A. F.,

367 & de Oliveira, C. A. F. (2013). Food Safety Systems in a Small Dairy Factory:

368 Implementation, Major Challenges, and Assessment of Systems’ Performances.

369 Foodborne Pathogens and Disease, 10(1), 6–12.

370 https://doi.org/10.1089/fpd.2012.1286

16
 ACCEPTED MANUSCRIPT
371 Cusato, S., Gameiro, A. H., Sant’Ana, A. S., Corassin, C. H., Cruz, A. G., & de Oliveira, C.

372 A. F. (2014). Assessing the costs involved in the implementation of GMP and HACCP

373 in a small dairy factory. Quality Assurance and Safety of Crops & Foods, 6(2), 135–

374 139. https://doi.org/10.3920/QAS2012.0195

375 De Silva, S. A. S. D., Kanugala, K. A. N. P., & Weerakkody, N. S. (2016). Microbiological

PT
376 Quality of Raw Milk and Effect on Quality by Implementing Good Management

RI
377 Practices. Procedia Food Science, 6, 92–96.

378 https://doi.org/10.1016/j.profoo.2016.02.019

SC
379 Eckburg, P. B., Bik, E. M., Bernstein, C. N., Purdom, E., Dethlefsen, L., Sargent, M., …

380 Relman, D. A. (2005). Diversity of the human intestinal microbial flora. Science,

381 308(5728), 1635–1638.

U
AN
382 *Godziszewska, J., Guzek, D., G\ląbski, K., & Wierzbicka, A. (2016). Mobilna
M

383 antybiotykooporność–o rozprzestrzenianiu się genów determinujących oporność

384 bakterii poprzez produkty spożywcze. Postepy Hig Med Dosw (Online), 70, 803–810.
D

385 *Godziszewska, J., Guzek, D., Pogorzelska, E., Brodowska, M., Górska-Horczyczak, E.,
TE

386 Sakowska, A., … Wierzbicka, A. (2017). A simple method of the detection of pork

387 spoilage caused by Rahnella aquatilis. LWT - Food Science and Technology, (in
EP

388 press). https://doi.org/doi.org/10.1016/j.lwt.2017.05.049

Goodner, K. L. (2008). Practical retention index models of OV-101, DB-1, DB-5, and DB-
C

389
AC

390 Wax for flavor and fragrance compounds. LWT - Food Science and Technology, 41(6),

391 951–958. https://doi.org/10.1016/j.lwt.2007.07.007

392 Górska-Horczyczak, E., Guzek, D., Molęda, Z., Wojtasik-Kalinowska, I., Brodowska, M., &

393 Wierzbicka, A. (2016). Applications of electronic noses in meat analysis. Food

394 Science and Technology, 36(3), 389–395. https://doi.org/10.1590/1678-457X.03615

395 Gunasekera, T. S., Attfield, P. V., & Veal, D. A. (2000). A flow cytometry method for rapid

17
 ACCEPTED MANUSCRIPT
396 detection and enumeration of total bacteria in milk. Applied and Environmental

397 Microbiology, 66(3), 1228–1232.

398 Jarosinska, A., Barlowska, J., Wolanciuk, A., Pastuszka, R., & Barlowska, K. (2014). Sk\lad

399 chemiczny i jakość mikrobiologiczna mleka towarowego dostarczanego do 5

400 mleczarni z regionu lubelskiego, z uwzględnieniem sezonu skupu. Roczniki Naukowe

PT
401 Polskiego Towarzystwa Zootechnicznego, 10(2). Retrieved from

RI
402 http://agro.icm.edu.pl/agro/element/bwmeta1.element.agro-f3b90bd1-0c32-46c6-90f3-

403 33a4fbce6f0c

SC
404 Kitai S, A Shimizu, J Kawano, E Sato, C Nakano, T Uji, & H Kitagawa. (2005).

405 Characterization of methicillin-resistant Staphylococcus aureus isolated from retail

406
U
raw chicken meat in Japan. J Vet Med Sci, 67(1), 107–110.
AN
407 Kruse, H., & Sørum, H. (1994). Transfer of multiple drug resistance plasmids between
M

408 bacteria of diverse origins in natural microenvironments. Applied and Environmental

409 Microbiology, 60(11), 4015–4021.

D

410 Kwon, N. H., Park, K. T., Jung, W. K., Youn, H. Y., Lee, Y., Kim, S. H., … Park, Y. H.
TE

411 (2006). Characteristics of methicillin resistant Staphylococcus aureus isolated from

412 chicken meat and hospitalized dogs in Korea and their epidemiological relatedness.
EP

413 Veterinary Microbiology, 117(2–4), 304–312.

https://doi.org/10.1016/j.vetmic.2006.05.006
C

414
AC

415 Lee, A. P., Barbano, D. M., & Drake, M. A. (2016). Short communication: The effect of raw

416 milk cooling on sensory perception and shelf life of high-temperature, short-time

417 (HTST)–pasteurized skim milk. Journal of Dairy Science, 99(12), 9659–9667.

418 https://doi.org/10.3168/jds.2016-11771

419 Loss, G., Apprich, S., Waser, M., Kneifel, W., Genuneit, J., Bchele, G., … Braun-Fahrlnder,

420 C. (2011). The protective effect of farm milk consumption on childhood asthma and

18
 ACCEPTED MANUSCRIPT
421 atopy: The GABRIELA study. Journal of Allergy and Clinical Immunology, 128(4),

422 766–773.e4. https://doi.org/10.1016/j.jaci.2011.07.048

423 Martin, N. H., Ranieri, M. L., Murphy, S. C., Ralyea, R. D., Wiedmann, M., & Boor, K. J.

424 (2011). Results from raw milk microbiological tests do not predict the shelf-life

425 performance of commercially pasteurized fluid milk. Journal of Dairy Science, 94(3),

PT
426 1211–1222. https://doi.org/10.3168/jds.2010-3915

RI
427 McEntegart, C. M., Penrose, W. R., Strathmann, S., & Stetter, J. R. (2000). Detection and

428 discrimination of coliform bacteria with gas sensor arrays. Sensors and Actuators B:

SC
429 Chemical, 70(1), 170–176.

430 Moid, L., Etievant, P., Langlois, D., Dekimpe, J., & Addeo, F. (1994). Detection of powerful

431
U
odorants in heated milk by use of extract dilution sniffing analysis. Journal of Dairy
AN
432 Research, 61(3), 385–394.
M

433 Mullis K, Faloona F, Scharf S, Saiki R, Horn G, & Erlich H. (1986). Specific enzymatic

434 amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harb Symp
D

435 Quant Biol, 1, 263–273.

TE

436 Murphy, S. C., Martin, N. H., Barbano, D. M., & Wiedmann, M. (2016). Influence of raw

437 milk quality on processed dairy products: How do raw milk quality test results relate
EP

438 to product quality and yield? Journal of Dairy Science, 99(12), 10128–10149.

https://doi.org/10.3168/jds.2016-11172
C

439
AC

440 Norm PN-EN ISO 4832:2007 Mikrobiologia żywności i pasz. Horyzontalna metoda

441 oznaczania bakterii z grupy coli. Metoda płytkowa (in Polish). (2007).

442 Norm PN-EN ISO 4833-1:2013 Mikrobiologia żywności i pasz. Horyzontalna metoda

443 oznaczania liczby drobnoustrojów. Metoda płytkowa w temperaturze 30 stopni C (in

444 Polish). (2013).

445 Oliver, S. P., Jayarao, B. M., & Almeida, R. A. (2005). Foodborne pathogens in milk and the

19
 ACCEPTED MANUSCRIPT
446 dairy farm environment: food safety and public health implications. Foodbourne

447 Pathogens & Disease, 2(2), 115–129.

448 Paneto, B. R., Schocken-Iturrino, R. P., Macedo, C., Santo, E., & Marin, J. M. (2007).

449 Occurrence of toxigenic Escherichia coli in raw milk cheese in Brazil. Arquivo

450 Brasileiro de Medicina Veterinária E Zootecnia, 59(2), 508–512.

PT
451 Patrignani, F., Iucci, L., Belletti, N., Gardini, F., Guerzoni, M. E., & Lanciotti, R. (2008).

RI
452 Effects of sub-lethal concentrations of hexanal and 2-(E)-hexenal on membrane fatty

453 acid composition and volatile compounds of Listeria monocytogenes, Staphylococcus

SC
454 aureus, Salmonella enteritidis and Escherichia coli. International Journal of Food

455 Microbiology, 123(1–2), 1–8. https://doi.org/10.1016/j.ijfoodmicro.2007.09.009

456
U
Phillips, I. (2007). Withdrawal of growth-promoting antibiotics in Europe and its effects in
AN
457 relation to human health. International Journal of Antimicrobial Agents, 30(2), 101–
M

458 107. https://doi.org/10.1016/j.ijantimicag.2007.02.018

459 Saenz, Y., Brinas, L., Dominguez, E., Ruiz, J., Zarazaga, M., Vila, J., & Torres, C. (2004).
D

460 Mechanisms of Resistance in Multiple-Antibiotic-Resistant Escherichia coli Strains of

TE

461 Human, Animal, and Food Origins. Antimicrobial Agents and Chemotherapy, 48(10),

462 3996–4001. https://doi.org/10.1128/AAC.48.10.3996-4001.2004

EP

463 Silcock, P., Alothman, M., Zardin, E., Heenan, S., Siefarth, C., Bremer, P. J., & Beauchamp,

J. (2014). Microbially induced changes in the volatile constituents of fresh chilled

C

464
AC

465 pasteurised milk during storage. Food Packing and Shelf Life, 2(2), 81–91.

466 https://doi.org/10.1016/j.fpsl.2014.08.002

467 Skočková, A., Bogdanovičová, K., Koláčková, I., & Karpíšková, R. (2015). Antimicrobial-

468 Resistant and Extended-Spectrum β-Lactamase–Producing Escherichia coli in Raw

469 Cow’s Milk. Journal of Food Protection, 78(1), 72–77.

470 Sozańska, B., Pearce, N., Dudek, K., & Cullinan, P. (2013). Consumption of unpasteurized

20
 ACCEPTED MANUSCRIPT
471 milk and its effects on atopy and asthma in children and adult inhabitants in rural

472 Poland. Allergy, 68(5), 644–650. https://doi.org/10.1111/all.12147

473 Spanu, V., Scarano, C., Cossu, F., Pala, C., Spanu, C., & De Santis, E. P. L. (2014).

474 Antibiotic Resistance Traits and Molecular Subtyping of Staphylococcus aureus

475 Isolated from Raw Sheep Milk Cheese: Staphylococcus aureus drug resistance….

PT
476 Journal of Food Science, 79(10), M2066–M2071. https://doi.org/10.1111/1750-

RI
477 3841.12590

478 Spanu, V., Virdis, S., Scarano, C., Cossu, F., De Santis, E. P. L., & Cosseddu, A. M. (2010).

SC
479 Antibiotic resistance assessment in S. aureus strains isolated from raw sheep’s milk

480 cheese. Veterinary Research Communications, 34(S1), 87–90.

481
U
https://doi.org/10.1007/s11259-010-9404-3
AN
482 Spitzer, M., Wildenhain, J., Rappsilber, J., & Tyers, M. (2014). BoxPlotR: a web tool for
M

483 generation of box plots. Nat Meth, 11(2), 121–122.

484 Tadesse, D. A., Zhao, S., Tong, E., Ayers, S., Singh, A., Bartholomew, M. J., & McDermott,
D

485 P. F. (2012). Antimicrobial Drug Resistance in Escherichia coli from Humans and
TE

486 Food Animals, United States, 1950–2002. Emerging Infectious Diseases, 18(5), 741–

487 749. https://doi.org/10.3201/eid1805.111153

EP

488 Toews, M. L., & Adler, J. (1979). Methanol formation in vivo from methylated chemotaxis

proteins in Escherichia coli. Journal of Biological Chemistry, 254(6), 1761–1764.

C

489
AC

490 UE. (2006). Rozporządzenie Komisji (we) nr 1662/2006.

491 Wojtasik-Kalinowska, I., Guzek, D., Górska-Horczyczak, E., Głąbska, D., Brodowska, M.,

492 Sun, D.-W., & Wierzbicka, A. (2016). Volatile compounds and fatty acids profile in

493 Longissimus dorsi muscle from pigs fed with feed containing bioactive components.

494 LWT - Food Science and Technology, 67, 112–117.

495 https://doi.org/10.1016/j.lwt.2015.11.023

21
 ACCEPTED MANUSCRIPT
496 Yu, K., Hamilton-Kemp, T. R., Archbold, D. D., Collins, R. W., & Newman, M. C. (2000).

497 Volatile Compounds from Escherichia coli O157:H7 and Their Absorption by

498 Strawberry Fruit. Journal of Agricultural and Food Chemistry, 48(2), 413–417.

499 https://doi.org/10.1021/jf990576b

500 Zhu, J., Bean, H. D., Kuo, Y.-M., & Hill, J. E. (2010). Fast Detection of Volatile Organic

PT
501 Compounds from Bacterial Cultures by Secondary Electrospray Ionization-Mass

RI
502 Spectrometry. Journal of Clinical Microbiology, 48(12), 4426–4431.

503 https://doi.org/10.1128/JCM.00392-10

SC
504 Ziarno, M., & Czapska, M. (2008). Skład jakościowy mikroflory mleka krowiego surowego i

505 pasteryzowanego. Przegląd Mleczarski, (5), 4–8.

506
U
Zycka-Krzesinska, J., Boguslawska, J., Aleksandrzak-Piekarczyk, T., Jopek, J., & Bardowski,
AN
507 J. K. (2015). Identification and characterization of tetracycline resistance in
M

508 Lactococcus lactis isolated from Polish raw milk and fermented artisanal products.

509 International Journal of Food Microbiology, 211, 134–141.

D

510 https://doi.org/10.1016/j.ijfoodmicro.2015.07.009
TE

511

512 The key references were chosen based on topic described in the publications.
EP

513 Moreover, the methods described in the publications with some modifications were

used to obtained described in the manuscript data.

C

514
AC

515

22
 ACCEPTED MANUSCRIPT
516 Figure 1. The identification of tetM gene in the coliform bacteria isolated from the milk

517 samples - agarose electrophoresis of PCR amplification products

518 M-GeneRuler 1 kb Plus DNA (Thermo Scientific, Waltham, MA USA), line 1 – negative

519 control, PCR without DNA temple, line 2 – positive control, PCR product 406bp, lines 3-19,

520 PCR reactions performed on random chosen dark pink colonies from VRBA plates isolated

PT
521 from the milk samples. PCR products near 406 bp (characteristic for tetM genes ) were

RI
522 amplified on DNA isolated from the colonies numbered 13-18.

523

SC
524 Figure 2. The volatile compound profile of Escherichia coli DH5α in LB-broth

525 * volatile compounds were identified based on Kovats indices for column MXT-5 and MXT-

526
U
17 and information from AroChemBase database; for each compound Kovats indices for
AN
527 MXT-5 and MXT-17 column are indicated in the bracket, respectively for MXT-5 and MXT-
M

528 17,** relative concentration was calculated based on relative peak area of peaks for MXT-5

529 column; BoxPlotR (boxplot.tyerslab.com) were used for preparation of the plot. In the graph
D

530 box limits indicate the 25th and 75th percentiles, center line in box show the median, whiskers
TE

531 extend 1.5 times the interquartile range from the 25th and 75th percentiles.

532
EP

533 Figure 3. Typical volatile compounds identified in raw milk during the storage
a,b
– significantly different concentration are presented with different letter for each volatile
C

534
AC

535 compound,

536 * Kovats’ indices for column MXT-5 and MXT-17, respectively ** volatile compounds were

537 identified based on Kovats indices for column MXT-5 and MXT-17 and information from

538 AroChemBase database; *** relative concentration was calculated based on relative peak area

539 of peaks for MXT-5 column

540

23
 ACCEPTED MANUSCRIPT
Table 1. Concentration of the antibiotics used during analyses

concentration
[µg mL-1]

penicillin 300
kanamycin 50
streptomycin 30

PT
chloramphenicol 10
tetracycline 10

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 2. Antibiotic resistance of the isolated coliform bacteria

percentage of antibiotic
resistance strains

Mean Min Max

[%] [%] [%]
penicillin 88.3 36 100
kanamycin 38.6 8.6 95

PT
streptomycin 42.6 13 86.6
chloramphenicol 77.8 11.1 100
tetracycline 55.1 15.7 93.3

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 3. The relative content of volatile compounds, which were identified both in overnight

culture of E. coli and broth

PT
identified volatile broth overnight
compound mean ±SD culture of
[%] E. coli
mean ±SD

RI
[%]
trimethylamine 1.207±0.068a 0.309±0.034b
(389,457)*
methanol 0.083±0.006a 17.791±0.456b

SC
(421,534)
propanal 11.454±0.193a 8.688±0.163b
(450,567)
69.403±0.620a 33.919±0.394b

U
1-propanol, 2-methyl-
(634,734)
6.199±0.071a 0.344±0.022b
AN
methional
(911,1033)
2-methylpropanal 0.695±0.197a 19.594±0.272b
(515,642)
M

(E)-2-penten-1-ol 0.155±0.011a 0.135±0.006b

(764,817)
butanal 5.780±0.174a 4.710±0.166b
(565,663)
D

a,b
– significantly different contents are presented with different letter for each volatile compound
* - for each compound Kovats indices for MXT-5 and MXT-17 column are indicated in the bracket, respectively
TE

for MXT-5 and MXT-17

Relative concentration was calculated based on relative peak area of peaks for MXT-5 column.
C EP
AC
 ACCEPTED MANUSCRIPT

Table 4. Correlation of the selected compounds

milk, day 0 milk, day 4

culture of
overnight
broth

E. coli
methanol propanal methanol propanal

PT
methanol

methanol
R (-0.71) R (-0.76)

RI
propanal

propanal
R (0.51) R (0.57)

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT
Highlights

• Raw milk is an reservoir of antibiotic resistance gene

• tetM gene identified in coliform bacteria isolated from raw milk
• Indicator of contamination of raw milk by coliform bacteria was identified

PT
RI
U SC
AN
M
D
TE
C EP
AC