JAHP_14_3_996-1006

INTRODUCTION

Mastitis is a dominant pathology in dairy cattle. Besides the potential risk for food safety (Artursson et al., 2018), mastitis induces severe economic losses in all dairy systems. These losses are due to the decrease in milk production (Heikkilä et al., 2018; Aparicio-Roque et al., 2025), reduction in milk quality (Narváez-Semanate et al., 2022) and higher costs of treatment products (Nicholas et al., 2009). It is the primary cause of the use of antimicrobial in dairy cattle (Deng et al., 2020). Mastitis can be caused by a variety of bacteria, which differ depending on the pathogenicity and prevalence frequency. These bacteria can be broadly classified into two groups; contagious (such as Staphylococcus aureus and Streptococcus agalactiae) or environmental (Streptococcus uberis, Streptococcus dysgalactiae and coliforms), depending on their primary reservoir and mode of transmission (Ruegg, 2012). The contagious bacteria are transmitted among cows by contact with infected milk during the milking process. Environmental bacteria arise from the environment of cows (manure, mud and dirty bedding materials) entering into the udder between milking or during the milking process, they can multiply and cause mastitis. Several studies in different countries showed that the most frequently isolated bacteria in cases of clinical and subclinical mastitis are staphylococci (Staphylococcus aureus and Coagulase negative Staphylococci), Streptococcus uberis and Escherichia coli (Fadlelmula et al., 2009; Mekonnen et al., 2017; Vakkamaki et al., 2017; Heikkilä et al., 2018; Maalaoui et al., 2021). The frequency and origin of udder infection vary from country to country and even between herds in the same country (Olde Riekerink et al., 2008), this is due to the difference in the preventive measures and milking procedures on each farm (Bradley, 2002).

Mastitis diagnostic methods such as clinical inspection, somatic cell counting, the California mastitis test (CMT), electrical conductivity or the detection of body enzymes released due to tissue damage, are important for determining mastitis cases in a dairy farm (Stanek et al., 2024) but are unable to provide information on the identity of the causative agent and severity of the infection. Therefore, the identification and characterization of the various microorganisms responsible for this disease are important steps for mastitis control, better herd management and improving the treatment strategy, which decreases the use of inappropriate antimicrobials (Taponen et al., 2009) and the infection incidence. The widely used identification methods are based on bacterial culture and biochemical tests. Bacterial culture is the gold standard for routine analysis of milk samples (Koskinen et al., 2010). It allows the identification of a large number of bacteria and costs are low compared with more advanced molecular methods. However, this standard method is heavy, long and ambiguous (Hegde, 2011; Prabhu et al., 2013). In addition, it may not show any bacterial growth in milk samples from truly infected glands due to low concentration of pathogens (Phuektes et al., 2001). Furthermore, the lack of unique biochemical markers for identification of species makes identification based on phenotypic differences unreliable (Capurro et al., 2009). The DNA-based molecular method, such as multiplex PCR, has been developed for rapid detection of major pathogens causing mastitis (Hegde, 2011; Shome et al., 2011; Kajdanek et al., 2024). Indeed, in multiplex PCR two or more loci are simultaneously amplified in the same reaction tube that helps to save time and minimize reagents used. In addition, PCR may improve the level of detection due to its high sensitivity (Phuektes et al., 2003).

In this context, the aim of the current study was to evaluate the multiplex PCR method for the simultaneous detection of mastitis pathogens collected from cow’s milk from smallholder farms and to compare it with the conventional culture method. Finding a less expensive, specific and fast method to facilitate and improve the detection of bacteria causing bovine mastitis in smallholder farms in Tunisia is an important issue.

MATERIALS AND METHODS

Sampling

Two hundred milk samples were collected aseptically from cows suffering from clinical and subclinical mastitis belonging to smallholder farmers, according to the guidelines of the National Mastitis Council (NMC, 1999) for the bacteriology of milk. The farmers were informed about the purpose and the methods of the study. Oral consent was obtained from the owner of the animals before commencement of the study. All the procedures used were non invasive and in addition all the results were communicated to animal owners. There is no specific law for milk sample collection and hence no approval was mandatory.

In subclinical mastitis cases, milk samples were taken from cows with a score of California Mastitis Test (CMT) greater than or equal to two. In clinical mastitis cases, samples were collected from cows with clinical signs such as altered milk secretion (clots, flakes and bloody or purulent appearance), swollen, painful udder and/or with visible injury. These milk samples were originated from one central-eastern region (Mahdia, 77 samples) and three northern regions of Tunisia (Bizerte, Beja and Jendouba, 123 samples). Milk samples were subjected to isolation and identification of mastitis pathogens using bacterial culture in the diagnostic bacteriology laboratory at Veterinary Research Institute of Tunis (IRVT). After that, an aliquot of each sample was deep-frozen (-20°C) immediately after culturing for one week. Then, it was sent to the research laboratory Toxicology, Environmental Microbiology and Health at the Faculty of Sciences of Sfax to identify mastitis bacteria by multiplex PCR.

Bacterial culture

To identify bacteria present in mastitis milk samples, 10µL of milk were plated onto blood agar (tryptone soy agar with 5% sheep blood) and BromoCresol Pourpre agar (BCP). The inoculated plates were incubated at 37°C for 18-24 hours. After incubation, the number and the morphology (form and hemolytic activity) of the colonies were evaluated. From each agar, one colony per morphology was Gram stained to define the gram (+) and gram (-) bacteria and to choose the type of rapid identification gallery (API20E, API20NE, API20Staph, API20Strep, bioMérieux, France) to be used next. In addition, other tests have been carried out such as catalase and oxidase tests, and also the mobility test to confirm the mobility of some gram (-) bacteria. Other characteristics have been sought such as free coagulase to confirm the presence of S. aureus. Finally, bacterial isolates were stored at -80°C in brain/heart infusion broth (BHI) supplemented with 15% glycerol.

Bacterial strains

Bacterial strains derived from milk samples from clinical and subclinical mastitis cases were used in this study to standardize multiplex PCR and assess its accuracy in identifying species. These strains were previously well identified by phenotypic and molecular methods. These strains are S. aureus, S. chromogenes, K. pneumoniae, E. coli, S. epidermidis, S. haemolyticus, Str. agalactiae and Str. uberis. All bacterial strains were cultured on sheep blood agar and nutrient agar at 37°C for 24 hours to be used for further DNA extraction.

Primers

The oligonucleotide primer sequences used to amplify and identify the main bacteria causing mastitis were derived from Shome et al. (2011). Detailed information on targeted organisms, primer sequences and size of PCR amplicons is presented in Table 1.

Extraction of bacterial genomic DNA

Bacterial genomic DNA was extracted from bacterial cultures using Zymo «Quick-gDNA™ MiniPrep» Kit (Zymo Research, D 3024, USA) according to manufacturer’s instructions. Afterwards, the quantity and quality of genomic DNA was determined using Nanodrop 2000 spectrophotometer (IMPLEN) and then DNA was stored at -20°C until further use.

Multiplex PCR assay

A multiplex PCR assay was used for simultaneous detection of eight bacterial species causing bovine mastitis. A two- tube multiplex PCR assay was developed. Each tube includes four sets of primers targeting different bacteria. The final result of sample was a combination of the results found in the two tubes.

Table 1: Oligonucleotide primers sequences and predicted sizes of PCR products (Shome et al., 2011).

Organisms	Gene	Oligonucleotide primer 5´– 3´	Location within gene	Amplicon size (bp)
Staphylococcus aureus	23S rRNA	AGCGAGTCTGAATAGGGCGTTT CCCATCACAGCTCAGCCTTAAC	678-699 1571-1550	894
Staphylococcus chromogenes	sodA	GCGTACCAGAAGATAAACAAACTC CATTATTTACAACGAGCCATGC	134-157 355-334	222
Staphylococcus epidermidis	rdr	AAGAGCGTGGAGAAAAGTATCAAG TCGATACCATCAAAAAGTTGG	400 016-400 039 400 145-400 125	130
Staphylococcus haemolyticus	sodA	CAAATTAAATTCTGCAGTTGAGG AGAGCCCCATTGTTCTTTGA	63-85 276-257	214
Streptococcus uberis	cpn60	TCGCGGTATTGAAAAAGCAACAT TGCAATAATGAGAAGGGGACGAC	57-79 456-434	400
Streptococcus agalactiae	16S rRNA	GCTAATACCGCATAAGAGTAATTAAC GGTAGATTTTCCACTCCTACCAA	132-158 448-425	317
Klebsiella pneumoniae	Khe	GGAAGTGTGGATAAACGGCT CTCCTGCTCGGTGTTATTGA	4800987-4801006 4801076-4801094	108*
Escherichia coli	phoA	GGTAACGTTTCTACCGCAGAGTTG CAGGGTTGGTACACTGTCATTACG	433-456 900-877	468

*(Cressier and Bissonnette, 2011).

Optimization of multiplex PCR assay

Individual PCR assays for S. aureus, S. chromogenes, S. epidermidis, S. haemolyticus, Str. agalactiae, Str. uberis, E. coli and K. pneumoniae were initially optimized separately. Different concentrations of Taq polymerase, MgCl2, dNTP and primers were used in 25 µL reaction volume, to define the optimal PCR conditions for each individual PCR assay. To optimize the multiplex PCR assay, different combinations of individual PCR with varying concentrations of primers and template DNA were used (Figure 1). The final protocol included two separate tubes. Each tube was comprised of four sets of primers targeting different bacteria. The first tube reaction consisted of S. aureus, E. coli, S. chromogenes and K. pneumoniae and the second tube reaction included Str. uberis, S. haemolyticus, Str. agalactiae and S. epidermidis. Several assays were made until a final result was obtained which was a combination of the results found. In this study, we relied on the protocol of Shome et al. (2011) with slight modifications.

Each optimized reaction mixture for PCR had a total volume of 25 µL composed of 1.25U of Taq polymerase (Promega, France), 1.5 mmol/l MgCl2, 0.2 mmol/l dNTPs,1× reaction buffer (Promega, France), combination of optimized primer concentrations (0.5µmol/l of S. chromogenes, S. aureus and K. pneumoniae and 0.2µmol/l of E. coli for tube 1 reaction and 0.5µmol/l of S. epidermidis, S. haemolyticus, Str. agalactiae and 0.2µmol/l of Str. uberis for tube 2 reaction) and 100ng of DNA extract. The amplification was performed in an automated thermocycler (UNO96,VWR) with initial denaturation at 94°C for 5 minutes, followed by 40 cycles at 94°C for 30 seconds; 60°C for 30 seconds, 72°C for 45 seconds and finally once at 72°C for 10 minutes. The amplified products were analysed by electrophoresis on 1.5% agarose gel containing 0.5% ethidium bromide. After an electrophoresis run-time of 40 minutes, the gels were photographed under UV light.

Detection of pathogens in milk samples by multiplex PCR assay

The aliquot of each 200 milk samples frozen at -20°C was thawed out and processed for DNA extraction directly from milk using the Zymo “Quick-gDNA ™ MiniPrep” extraction Kit (Zymo Research, D 3024, USA) with a few modifications to eliminate fats. Before starting the DNA extraction protocol, two essential steps were carried out due to the milk fat content:

Of each milk sample, centrifuge 1 mL of milk for 20 minutes at 13000 rpm. After centrifugation, discard the surface lipid layer and the supernatant and conserve the pellet.
Add 200 µL of sterile water to the pellet then shake with an intense vortex to homogenize the mixture.
After DNA extraction from the milk samples, a multiplex PCR assay was applied to all samples to identify the mastitis causative agents. The detection of PCR products was carried out by electrophoresis on agarose gel (1.5%). The obtained results were compared with those of culture method.

Statistical analysis

Data were transferred to an Excel spreadsheet (Microsoft Corp., Redmond, WA) for analysis. The diagnostic sensitivities of culture and PCR on milk samples were compared using an adjusted McNemar’s X2 test. The test result agreement between the two methods was evaluated using Kappa statistics (Kappa value below 0.40 “weak” agreement, 0.4-0.6 “moderate”, 0.6-0.8 “substantial”, and greater than 0.8 “almost perfect”; McGINN et al., 2004). Analysis was performed using SPSS version 20.0. A P-value < 0.05 was considered statistically significant.

RESULTS

Bacterial culture enabled the isolation of several bacteria (Gram positive and Gram negative) directly from the milk, including the eight target organisms detected by multiplex PCR. These organisms are Staphylococcus aureus, Staphylococcus chromogenes, Staphylococcus epidermidis, Staphylococcus haemolyticus, Streptococcus uberis, Streptococcus agalactiae, Escherichia coli, and Klebsiella pneumoniae.

Bacteriological culture of milk samples

Out of 200 milk samples detected positives for mastitis by CMT and clinical examinations of the udder and milk, 37 were from clinical cases of bovine mastitis and 163 were from subclinical cases. Conventional bacterial culture identified one or more udder pathogens in 84% (168/200) of total milk samples.

Out of the 37 clinical samples, 89.2% (n = 33) were found positive for the presence of bacteria. Hence, 10.8% (n = 4) provided culture-negative results (Table 2). Forty bacterial isolates could be cultured and identified by biochemical tests that consist of 48.6% (n= 18) isolates of S. aureus, 27% (n= 10) isolates of CNS and 5.4% (n= 2) isolates for each of Str. uberis, Aerococcus viridians, Enterococcus spp., Micrococcus spp., Pseudomonas luteola and Pasteurella spp. CNS species identified from all samples (Table 2) were as follows: S. chromogenes (10.8%), S. epidermidis (2.7%), S. haemolyticus (2.7%), S. saprophyticus (2.7%), S. lentus (2.7%), S. warneri (2.7%) and S. spp. (2.7%).

From the 163 subclinical samples, 82.8% (n = 135) had a detectable bacterial infection. However, no growth was observed in the remaining samples. One hundred eighty six strains were isolated. S. aureus was isolated from 20.9% ( n = 34) of samples and coagulase negative staphylococci from 40.5% (n = 66) of samples including S. spp (10.4%), S. xylosus (8.6%), S. chromogenes (11.7%), S. lentus (2.5%), S. epidermidis (1.8%), S. haemolyticus (1.2%), S. cohnii ssp (1.2%), S. capitis (1.2%), S. sciuri (<1%), S. saprophyticus (<1%) and S. auricularis (<1%). Str. uberis, E. coli and

Table 2: Bacterial culture results of bacteria isolated from subclinical and clinical mastitis samples.

Organism	Mastitis				Total
	Clinical mastitis (N=37)		Subclinical mastitis (N=163)		N= 200
	N	%	N	%	N	%
Positive samples	33	89.2	135	82.8	168	84
S. aureus	18	48.6	34	20.9	52	26
Coagulase-negative staphylococci	10	27	66	40.5	76	38
S. chromogenes	4	10.8	19	11.7	23	11.5
S. epidermidis	1	2.7	3	1.8	4	2
S. haemolyticus	1	2.7	2	1.2	3	1.5
S. sciuri	0	0	1	0.6	1	0.5
S. xylosus	0	0	14	8.6	14	7
S. saprophyticus	1	2.7	1	0.6	2	1
S. auricularis	0	0	1	0.6	1	0.5
S. cohnii ssp	0	0	2	1.2	2	1
S. lentus	1	2.7	4	2.5	5	2.5
S. capitis	0	0	2	1.2	2	1
S. warneri	1	2.7	0	0	1	0.5
S. spp	1	2.7	17	10.4	18	9
Str. uberis	2	5.4	3	1.8	5	2.5
E. coli	0	0	3	1.8	3	1.5
K. pneumoniae	0	0	3	1.8	3	1.5
Aerococcus viridians	2	5.4	24	14.7	26	13
Enterococcus spp	2	5.4	4	2.5	6	3
Pseudomonas luteola	2	5.4	9	5.5	11	5.5
Aeromonas hydrophila	0	0	4	2.5	4	2
Micrococcus spp	2	5.4	4	2.5	6	3
Pasteurella spp	2	5.4	8	4.9	10	5
Kocuria varians/rosea	0	0	2	1.2	2	1
Burkholderia cepacia	0	0	4	2.5	4	2
No growth	4	10.8	28	17.2	32	16

Table 3: Multiplex PCR results of target bacteria isolated from subclinical and clinical mastitis samples.

Organism	Mastitis				Total
	Clinical mastitis (N=37)		Subclinical mastitis (N=163)		N= 200
	N	%	N	%	N	%
Positive samples	26	70.3	74	45.4	100	50
S. aureus	9	24.3	31	19	40	20
Coagulase-negative staphylococci	11	29.7	38	23.3	49	24.5
S. chromogenes	10	27	31	19	41	20.5
S. epidermidis	0	0	2	1.2	2	1
S. haemolyticus	1	2.7	5	3.1	6	3
Str. uberis	10	27	8	4.9	18	9
Str. agalactiae	0	0	2	1.2	2	1
E. coli	0	0	7	4.3	7	3.5
K. pneumoniae	3	8.1	12	7.4	15	7.5
No growth	11	29.7	89	54.6	100	50

K. pneumoniae were found in 1.8% (n= 3) of samples. Aerococcus viridians, Pseudomonas luteola and Pasteurella spp. were isolated from 14.7%, 5.5% and 4.9% of samples, respectively. Other samples contained Micrococcus spp. (2.5%), Enterococcus spp. (2.5%) and Aeromonas hydrophila (2.5%), Kocuria varians/ rosea (1.2%) and Burkholderia cepacia (2.5%).

Optimization of multiplex PCR assay

Individual PCR assays specifically amplified the eight bacterial strains tested in this study in the same conditions. Species of each reaction mixture were tested individually and then in pairs until the detection of four target pathogens in a single reaction. The evaluation of eight bacterial strains showed that all the pairs of primers designed were specific to the species. PCR product sizes obtained for S. aureus, S. chromogenes, S. epidermidis, S. haemolyticus, Str. uberis, Str. agalactiae, K. pneumoniae and E. coli were 894, 222, 130, 214, 400, 317, 108, 468 bp, respectively, as shown in Figure 1. Finally, all the eight target pathogens were detected simultaneously in two reactions (Figure 2). In each reaction mixture, when the primers were added at equal concentrations, some target species were not detected. So, the concentrations of the primers were modified to detect all target strains in each reaction.

Pathogen detection in milk samples by multiplex PCR assay

The multiplex PCR assay was tested to detect target pathogens using DNA directly extracted from milk. The test effectively identified target bacteria. In clinical mastitis cases, S. aureus, Str. uberis, S. haemolyticus, S. chromogenes and K. pneumoniae were detected in 24.3%, 27%, 2.7%, 27% and 8.1% of samples, respectively. None of these milk samples was positive for S. epidermidis, E. coli, Str. agalactiae. From 163 subclinical samples, all target bacteria are isolated such as 19% S. aureus, 19% S. chromogenes, 3.1% S. haemolyticus, 1.2% S. epidermidis, 4.9% Str.uberis, 1.2% Str. agalactiae, 4.3% E. coli and 7.4% K. pneumoniae (Table 3).

According to the multiplex PCR assay, 35% of the total milk samples were positive for a single target species, 15% were positive for more than one target species, while 50% were negative for all target bacterial species.

Comparison of bacteriological culture and PCR results

Our results showed that of 200 milk samples cultured, 89 samples (44.5%) yielded one or two of the eight target pathogens, 79 samples presented microorganisms that were not targeted by the PCR test and 32 samples with negative culture. Comparing PCR with bacterial culture, 100 samples (50%) were positive and 131 strains of target bacteria were detected by PCR. Furthermore, comparing

Table 4: Comparative detection of target pathogens in 200 milk samples by bacterial culture and multiplex PCR.

Bacterial culture

Multiplex PCR

McNemar

(P)

Kappa value

(95 % CI)

Organism

Results

Positive

Negative

Staphylococcus aureus

Positive

Negative

Total

116

168

115

128

0.002

0.803

[0.705-0.897]

Staphylococcus chromogenes

Positive

Negative

Total

145

168

129

132

0.004

0.625

[0.447-0.769]

Staphylococcus epidermidis

Positive

Negative

Total

164

168

163

166

0.625

0.324

[-0.017-0.798]

Staphylococcus haemolyticus

Positive

Negative

Total

165

168

162

0.250

0.660

[0-1]

Streptococcus uberis

Positive

Negative

Total

163

168

153

154

0.012

0.399

[-0.009-0.673]

Streptococcus agalactiae

Positive

Negative

Total

168

166

Escherichia coli

Positive

Negative

Total

165

168

160

162

0.453

0.206

[-0.027-0.660]

Klebsiella pneumoniae

Positive

Negative

Total

165

168

156

157

0.021

0.268

[ -0.015-0.558]

No growth

* 4 samples: Str. uberis, 1 sample: S. chromogenes, 3 samples: S. chromogenes + K. pneumoniae and 1 sample: K. pneumoniae + S. chromogenes + E. coli

the results obtained by the two methods for the detection of the eight target bacteria using the McNemar test (Table 4), we found no significant difference between conventional methods and multiplex PCR for the detection of S. epidermidis (P= 0.625), S. haemolyticus (P=0.250), and E. coli (P=0.453). However, a significant difference was reported between the two methods for the detection of S. aureus (P=0.002), S. chromogenes (P= 0.004), Str. uberis (P=0.012) and K. pneumoniae (P=0.021). Also, the kappa value between the two methods varies from weak to almost perfect agreement depending on the species (Table 4). Despite the relatively good concordance observed between culture and PCR results, PCR identified species that were not detected in culture in many samples. The PCR test detected S. chromogenes in 16 samples, Str. uberis in 10 samples, K. pneumoniae in nine samples, E. coli in five samples, S. haemolyticus in three samples and S. aureus and S. epidermidis in one sample (Table 4). In addition, amongst the target species, Str. agalactiae was not isolated in any sample by bacterial culture but it was detected by PCR in two samples. In total, PCR detected 14 microorganisms in nine samples showing no growth in culture. These included Str. uberis (n=4), S. chromogenes (n=5), K. pneumoniae (n=4) and E. coli (n=1). Twenty-three samples were negative for both cultural and PCR detection. Despite the higher detection of target bacteria observed in PCR when compared to culture, some positive-culture samples showed negative PCR results for the corresponding pathogens. For most species, except S. aureus, these samples ranged from one to three. For S. aureus, the PCR assay did not detect this pathogen in 13 culture-positive samples, eight of which showed no species and five presented species other than S. aureus including S. chromogenes (n=1), S. haemolyticus (n=1), Str. uberis (n=1), association between S. chromogenes and Str. uberis (n=1) and association between Str. uberis and S. epidermidis (n=1, Figure 3).

DISCUSSION

Since several bacterial species can be involved in mastitis, inadequate techniques for detecting or identifying pathogens often delay timely interventions in disease treatment and control. The current study evaluated a multiplex PCR assay, which uses many primers in the same reaction, for rapid and accurate detection of eight bacterial pathogens frequently associated with bovine mastitis. The multiplex PCR method was compared with bacterial culture and the difference between these tests was noted. Results reported a corresponding level of agreement between Bacteriological Culture and PCR diagnosis for some samples. These results are in agreement with those reported by Shome et al. (2011). However, PCR detected species in samples found culture-negative for the corresponding pathogens and others showing no growth. Str. agalactiae was diagnosed only by multiplex PCR in two milk samples but not detected by bacterial culture. In the same way, S. chromogenes, Str. uberis, K. pneumoniae detection presented higher sensitivity by multiplex-PCR (41, 18 and 15 positive milk samples detected, respectively), when compared to the detection by microbiological culture (23, 5 and 3 positive samples, respectively). Furthermore, the PCR assay improved the detection of mastitis bacteria in nine milk samples that may not show any bacterial growth under bacterial culture conditions. These results are in agreement with those found by Kahya Demirbilek et al. (2024), who showed that a bacterial and/or yeast gene was found by rPCR in 187 of 246 (76.01%) samples with no bacterial growth. In addition, several studies have reported the presence of a large proportion of culture-negative milk samples from cows with clinical or subclinical mastitis ranging from 27 % to 50% (Makovec and Ruegg, 2003; Barrett et al., 2005; Bradley et al., 2007; Taponen et al., 2009). These negative results might be explained by the use of non-specific culture media for these bacteria, the low bacterial concentration, and the unknown use of antibiotics prior to collection or to the presence of several substances that could have inhibited the growth of bacteria on agar plates (Phuektes et al., 2001; Taponen et al., 2009). In addition, the host immune system may have effectively affected the viability of a microorganism, but this microorganism could maintain the status of inflammation by releasing toxins (Taponen et al., 2009; Krômker et al., 2010). The case of two species samples detected only by PCR might be due to a competition between germs for nutrients and space and a low concentration of some pathogens compared to others or as in culture. These results might be also explained by bacterial colonies that are selected based on morphological characteristics, species with similar phenotype can be found and, in this case, the risk of missing a species is probable (Shome et al., 2011). Similarly, possible reasons for no growth in milk samples may be due to the absence of viable bacterial cells. PCR can detect viable or non-viable bacteria present in milk. In this case, although the non-viable bacteria have no role in the initiation of infection, they are identified as positive by PCR. Thus, these samples may be considered as false positives. Therefore, multiplex PCR can be used as an alternative method to bacteriological culture in the routine diagnosis of all mastitis pathogens regardless of fastidious, slow growing and non-culturable microorganisms or easy-to-culture pathogens.

Despite the higher detection of target bacteria observed in PCR when compared to culture, 23 positive-culture samples showed negative PCR results. Discordance was observed mainly for S. aureus which was not detected in 13 samples. These results agree with studies of Troncarelli et al. (2015) and Singh et al. (2019) where S. aureus was isolated by culture but not by multiplex PCR. These culture-positive/PCR-negative cases could be explained by the presence of PCR inhibitors that have been found in milk samples such us antibiotics, detergents, enzymes, polysaccharides, fats, proteins, plasmin and bacterial debris (Wilson, 1997; Murphy et al., 2002). Several protocols of DNA-extraction in milk were developed to eliminate these PCR inhibitors and to improve its sensitivity (Kim et al., 2001; Cremonesi et al., 2006; Kubota et al., 2007). Phuektes et al. (2001) reported that applying different DNA extraction methods showed different PCR sensitivities. In addition, Cressier and Bissonnette (2011) showed a lower sensitivity of PCR due to the use of spin-columns method for DNA-extraction from milk. In our study, before extraction protocol we tried to remove fats from milk that make it difficult to extract DNA in high quantity and quality (Amills et al., 1997; Cremonesi et al., 2006) and therefore, the inhibition of PCR.

Besides, incorrectly identified species by bacterial culture can be a cause of PCR-negative results (Pitkälä et al., 2005; Koskinen et al., 2010) where in several negative cases for the corresponding pathogen, we find the identification of other bacteria than those found by culture. In case of S. aureus, five samples positive for this pathogen in culture showed other pathogens by PCR. Indeed, we found that these five strains identified as S. aureus by conventional culture, using the API STAPH identification gallery, were identified as S. chromogenes in the first case, S. haemolyticus in the second case, Str. uberis in the third case, an association between S. chromogenes and Str. uberis in the fourth case and an association between S. epidermidis and Str. uberis in the fifth case. In addition, the PCR-negative results for some gram-positive bacteria can be related to the complexity of the cell wall, which may hamperlysis during the extraction of DNA (Cremonesi et al., 2006). The cell wall of gram-positive bacteria contain several molecules which provide a rigid exoskeleton for protection against both mechanical and osmotic lysis (Singh et al., 2019).

Our study showed that twenty-three mastitis-samples were negative for both cultural and PCR detection. These results can be mainly explained by the fact that mastitis in these cases could be of traumatic origin due to the milking machine, or less probably caused by yeasts, algae or fungi (Krukowski et al., 2006; Fadlelmula et al., 2009; Lassa and Smulski, 2013).

Apart from the target species, we found several cases where bacterial culture method identified species not targeted by the PCR assay as for example Aerococcus viridians, Enterococcus spp., Pseudomonas luteola, Aeromonas hydrophila, Micrococcus spp. and Pasteurella spp. For this reason, it would be important to increase the number of target species to be detected by the multiplex PCR test. In this case many researchers developed multiplex PCR assay for simultaneous detection of several pathogens causing bovine mastitis in milk such us ten bacterial species in a study of Shome et al. (2011) and nine in a study of Ashraf et al. (2017).

The present study is in agreement with several researches, which confirmed that the diagnosis by PCR could be carried out in a few hours (Phuektes et al., 2001; Ashraf et al., 2017; Kajdanek et al., 2024). It takes less than 24 hours (while conventional culture requires more than 72 hours) thus it enables us to do the test monthly or even weekly, which allows the detection of the first cases of mastitis in herds. Rapid results can improve mastitis control and reduce costs through improved treatment strategy and oportum-time intervention (Pyörälä, 2002; Barkema et al., 2006) which improves the cure rates and decreases the use of inappropriate antimicrobials (Taponen et al., 2009) and discarded milk. However, despite these advantages and the concordance between multiplex PCR and bacterial culture shown in our study, the discordance found for the identification of S. aureus by PCR multiplex, should not be overlooked. In this case, it is important to revise this multiplex PCR protocol and attempt to improve its efficiency, by finding a solution to the problems of PCR inhibition and by developing efficient DNA extraction protocols so that it can be used for routine diagnosis of bovine mastitis and in herd decision pathways for smallholders

CONCLUSIONS

The current study has shown that, as compared to the conventional culture method, the multiplex PCR assay can be applied for rapid and sensitive detection of S. aureus, S. chromogenes, S. haemolyticus, S. epidermidis, Str. uberis, Str. agalactiae, E. coli and K. pneumoniae simultaneously in milk samples collected from mastitis cows from smallholder farms. The widespread and systematic use of PCR tests to help in the management and control of mastitis on farm and particularly on small dairy farms, would allow rapid diagnosis of the mastitis origins, making treatment earlier and more effective and with lower losses. However, despite the benefits, this method has shown unacceptably low sensitivity for the identification of S. aureus. Furthermore, PCR methods are limited to detecting only the targeted bacterial species. In this case, it would be advantageous to improve the multiplex PCR protocol by finding a solution to the problems of PCR inhibition and by increasing the number of targeted species for detection of mastitis by this method.

ACKNOWLEDGEMENTS

Authors would like to thank all those who helped us to carry out this work (ODESYPANO, Vitalait, Randa Akacha). They also thank the Tunisian smallholder farms who agreed to participate in this study.

NOVELTY STATEMENT

The development and evaluation of a multiplex PCR assay for simultaneous detection of eight mastitis-causing bacteria directly in milk. The assay will be useful for the detection of mastitis, testing bacteriological safety of milk and for species level differentiation. Finding a less expensive, specific and fast method to facilitate and improve the detection of bacteria causing bovine mastitis in smallholder farms is an important issue.

AUTHOR’S CONTRIBUTION

All authors contributed to this study. AM designed the study, conducted all experiments, analysed data and wrote the manuscript. HS designed, conducted the experiments and revised the manuscript. AT supervised and coordinated the study and revised the manuscript. RG examined and evaluated the results and revised the manuscript. NS contributed to the experiments. PGM and AM revised the manuscript. All authors read and approved the final manuscript.

Generative AI and AI assisted technology statement

The authors declare that no generative AI and AI assisted technology was used in the creation of this manuscript.

Conflict of interest

The authors have declared no conflict of interest.

REFERENCES

Amills M, Francino O, Jansa M, Sanchez A (1997). Isolation of genomic DNA from milk samples by using Chelex resin. J. Dairy Res., 64(2): 231-238. https://doi.org/10.1017/S0022029997002161

Aparicio-Roque C, Alaniz-Gutiérrez L, Jiménez-Jiménez RA, Rendón-Rendón MC, Mendoza-Núñez MA, González-Álvarez VH (2025). Bovine mastitis: Prevalence and economic losses. Agro Productividad. https://doi.org/10.32854/605sx166

Artursson K, Schelin J, Thisted Lambertz S, Hansson I, Olsson EE (2018). Foodborne pathogens in unpasteurized milk in Sweden. Int. J. Food. Microbiol., 284: 120–127. https://doi.org/10.1016/j.ijfoodmicro.2018.05.015

Ashraf A, Imran M, Yaqub T, Tayyab M, Shehzad W, Thomson PC (2017). A novel multiplex PCR assay for simultaneous detection of nine clinically significant bacterial pathogens associated with bovine mastitis. Mol. Cell. Probes., 33: 57-64. https://doi.org/10.1016/j.mcp.2017.03.004

Barkema HW, Schukken YH, Zadoks RN (2006). The role of cow, pathogen, and treatment regimen in the therapeutic success of bovine Staphylococcus aureus mastitis. J. Dairy Sci., 89(6): 1877-1895. https://doi.org/10.3168/jds.S0022-0302(06)72256-1

Barrett DJ, Healy AM, Leonard FC, Doherty ML (2005). Prevalence of pathogens causing subclinical mastitis in 15 dairy herds in the Republic of Ireland. Ir. Vet. J., 58(6): 333–337. https://doi.org/10.1186/2046-0481-58-6-333

Bradley A (2002). Bovine mastitis: An evolving disease. Vet. J., 164(2): 116-128. https://doi.org/10.1053/tvjl.2002.0724

Bradley AJ, Leach KA, Breen JE, Green LE, Green MJ (2007). Survey of the incidence and aetiology of mastitis on dairy farms in England and Wales. Vet. Rec., 160(8): 253-257. https://doi.org/10.1136/vr.160.8.253

Capurro A, Artursson K, Waller KP, Bengtsson B, Unnerstad EH, Aspan A (2009). Comparison of a commercialized phenotyping system, antimicrobial susceptibility testing, and tuf gene sequence-based genotyping for species-level identification of coagulase-negative staphylococci isolated from cases of bovine mastitis. Vet. Microbiol., 134(3-4): 327–333. https://doi.org/10.1016/j.vetmic.2008.08.028

Cremonesi P, Castiglioni B, Malferrari G, Biunno I, Vimercati C, Moroni P, Morandi S, Luzzana M (2006). Technical note: Improved method for rapid DNA extraction of mastitis pathogens directly from milk. J. Dairy Sci., 89(1): 163–169. https://doi.org/10.3168/jds.S0022-0302(06)72080-X

Cressier B, Bissonnette N (2011). Assessment of an extraction protocol to detect the major mastitis-causing pathogens in bovine milk. J. Dairy Sci., 94(5): 2171–2184. https://doi.org/10.3168/jds.2010-3669

Deng Z, Lam TJGM, Hogeveen H,Spaninks M, Heij N, Postema M, van Werven T, Koop G (2020). Antimicrobial use and farmers’ attitude toward mastitis treatment on dairy farms with automatic or conventional milking systems. J. Dairy Sci., 103(8): 7302-7314. https://doi.org/10.3168/jds.2019-17960

Fadlelmula A, Al-Dughaym AM, Mohamed GE, Al-Deib MK, Al-Zubaidy AJ (2009). Bovine mastitis: epidemiological, clinical and etiological study in a Saudi Arabian large dairy farm. Bulg. J. Vet. Med., 12(3).

Hegde R (2011). Rapid identification of bacterial pathogens causing subclinical bovine mastitis with special reference to Staphylococcus aureus, Escherichia coli and predominant streptococcal species by molecular methods. PhD Thesis, Karnataka Veterinary, Animal and Fisheries Sciences University, Bidar, India.

Heikkilä AM, Liski E, Pyörälä S, Taponen S (2018). Pathogen-specific production losses in bovine mastitis. J. Dairy Sci., 101(10): 9493-9504. https://doi.org/10.3168/jds.2018-14824

Kahya Demirbilek S, Yıldız M, Akkoç A, Mutlu AM, Ardıçlı Ö, Aner H (2024). Comparison of bacteriological culture method and multiplex real-time PCR for detection of mastitis. Res. Vet. Sci., 172: 105237. https://doi.org/10.1016/j.rvsc.2024.105237

Kajdanek A, Kluska M, Matusiak R, Kazimierczak J, Dastych JA (2024). Rapid and inexpensive PCR test for mastitis diagnosis based on NGS data. Pathogens., 13(5): 423. https://doi.org/10.3390/pathogens13050423

Kim CH, Khan M, Morin DE, Hurley WL, Tripathy DN, Kehrli M, OluochAO, Kakoma I (2001). Optimization of the PCR for detection of Staphylococcus aureus nuc gene in bovine milk. J. Dairy Sci., 84(1): 74–83. https://doi.org/10.3168/jds.S0022-0302(01)74454-2

Koskinen MT, Wellenberg GJ, Sampimon OC, Holopainen J, Rothkamp A, Salmikivi L, van Haeringen WA, Lam TJ, Pyorala S (2010). Field comparison of real-time polymerase chain reaction and bacterial culture for identification of bovine mastitis bacteria. J. Dairy Sci., 93(12): 5707–5715. https://doi.org/10.3168/jds.2010-3167

Krömker V, Zinke C, Paduch JH, Klocke D, Reimann A, Eller G (2010). Evaluation of increased milking frequency as an additional treatment for cows with clinical mastitis. J. Dairy Res., 77(1): 90-94. https://doi.org/10.1017/S0022029909990422

Krukowski H, Lisowski A,Rózański P, Skórka A (2006). Yeasts and algae isolated from cows with mastitis in the south-eastern part of Poland. Pol. J. Vet. Sci., 9(3): 181-184.

Kubota M, Hayashi T, Iwasaki K, Ohtsuka H, Kohiruimaki M, Kawamura S, Sakaguchi K, Abe R (2007). Rapid and effective method for separation of Staphylococcus aureus from somatic cells in mastitis milk. J. Dairy Sci., 90(9): 4100–4107. https://doi.org/10.3168/jds.2006-671

Lassa H and Smulski S (2013). Yeast like fungi as etiological agents of mastitis in cows. Życie Weterynaryjne., 88(11): 954-956

Maalaoui A, Majdoub H, Trimeche A, Souissi N, Saidani F, Marnet PG (2021). Prevalence of bovine mastitis and main risk factors in Tunisia. Trop. Anim. Health Prod., 53(5): 469. https://doi.org/10.1007/s11250-021-02925-7.

Makovec JA, Ruegg PL (2003). Results of milk samples submitted for microbiological examination in Wisconsin from 1994 to 2001. J. Dairy Sci., 86(11): 3466–3472. https://doi.org/10.3168/jds.S0022-0302(03)73951-4

McGINN T, Wyer PC, Newman TB, Keitz S, Leipzig R (2004). Tips for learners of evidence-based medicine: 3. Measures of observer variability (kappa statistic). Cmaj, 171(11): 1369-1373. https://doi.org/10.1503/cmaj.1031981

Mekonnen SA, Koop G, Melkie ST, Getahun CD, Hogeveen H, Lam TJGM (2017). Prevalence of subclinical mastitis and associated risk factors at cow and herd level in dairy farms in North-West Ethiopia. Prev. Vet. Med., 145: 23-31. https://doi.org/10.1016/j.prevetmed.2017.06.009

Murphy MA, Shariflou MR, Moran C (2002). High quality genomic DNA extraction from large milk samples. J. Dairy Res., 69(4): 645-649. https://doi.org/10.1017/S0022029902005848

Narváez-Semanate JL, Daza-Bolaños CA, Valencia-Hoyos CE, Hurtado-Garzón DT, Acosta-Jurado DC (2022). Diagnostic methods of subclinical mastitis in bovine milk: An overview. Rev. Fac. Nacional Agron. Medellín., 75(3): 10077-10088. https://doi.org/10.15446/rfnam.v75n3.100520

National Mastitis Council (NMC) (1999). Laboratory Handbook on Bovine Mastitis National Mastitis Council Inc, Madison, WI

Nicholas RA, Ayling RD, McAuliffe L (2009). Vaccines for Mycoplasma diseases in animals and man. J. Comp. Pathol., 140(2-3): 85-96. https://doi.org/10.1016/j.jcpa.2008.08.004

Olde Riekerink RG, Barkema HW, Kelton DF, Scholl DT (2008). Incidence rate of clinical mastitis on Canadian dairy farms. J. Dairy Sci., 91(4): 1366-1377. https://doi.org/10.3168/jds.2007-0757

Phuektes P, Browning GF, Anderson G, Mansell PD (2003). Multiplex polymerase chain reaction as a mastitis screening test for Staphylococcus aureus, Streptococcus agalactiae, Streptococcus dysgalactiae and Streptococcus uberis in bulk milk samples. J. Dairy Res., 70(2): 149-155. https://doi.org/10.1017/S0022029903006010

Phuektes P, Mansell PD, Browning GF (2001). Multiplex polymerase chain reaction assay for simultaneous detection of Staphylococcus aureus and Streptococcal causes of bovine mastitis. J. Dairy Sci., 84(5): 1140–1148. https://doi.org/10.3168/jds.S0022-0302(01)74574-2

Pitkälä A, Gindonis V, Wallin H, Honkanen-Buzalski T (2005). Interlaboratory proficiency testing as a tool for improving performance in laboratories diagnosing bovine mastitis. J. Dairy Sci., 88(2): 553-559. https://doi.org/10.3168/jds.S0022-0302(05)72717-X

Prabhu KN, Isloor S, Hegde R, Rathnamma D,Veeregowda BM, Narasimha Murthy HN, Shome R, Suryanarayana VVS (2013). Development of polymerase chain reaction for detection of predominant streptococcal isolates causing subclinical bovine mastitis. Ind. J. Biotechnol., 12: 208-212.

Pyörälä S (2002). New strategies to prevent mastitis. Reprod. Domest. Anim., 37, 211-216. https://doi.org/10.1046/j.1439-0531.2002.00378.x

Ruegg PL (2012). New perspectives in udder health management. Vet. Clin. North Am. Food Anim. Pract., 28: 149-163. https://doi.org/10.1016/j.cvfa.2012.03.001

Shome BR, Das Mitra S, Bhuvana M, Krithiga N, Velu D, Shome R, Isloor S, Barbuddhe SB, Rahman H (2011). Multiplex PCR assay for species identification of bovine mastitis pathogens. J. Appl. Microbiol., 111(6): 1349-1356. https://doi.org/10.1111/j.1365-2672.2011.05169.x

Singh K, Chandra M, Kaur G, Narang D, Gupta DK, Arora AK, Sharma NS (2019). Development of a multiplex PCR for identification of mastitis causing organisms. Ind. J. Dairy Sci., 72(2): 176-181. https://doi.org/10.33785/IJDS.2019.v72i02.008

Stanek P, Żółkiewski P, Januś E (2024). A review on mastitis in dairy cows research: Current status and future perspectives. Agriculture, 14(8): 1292. https://doi.org/10.3390/agriculture14081292

Taponen S, Salmikivi L, Simojoki H, Koskinen MT, Pyörälä S (2009). Real-time polymerase chain reaction-based identification of bacteria in milk samples from bovine clinical mastitis with no growth in conventional culturing. J. Dairy Sci., 92(6): 2610-2617. https://doi.org/10.3168/jds.2008-1729

Troncarelli MZ, Langoni H, Richini-Pereira VB, Marson PM, Da Silva RC (2015). Accuracy of a multiplex PCR protocol for Staphylococcus aureus, Streptococcus agalactiae and Escherichia coli detection in bulk tanks. Vet. Zoot., 22(4): 625-633.

Vakkamäki J, Taponen S, Heikkilä AM, Pyörälä S (2017). Bacteriological etiology and treatment of mastitis in Finnish dairy herds. Acta. Vet. Scand., 59(1): 33. https://doi.org/10.1186/s13028-017-0301-4

Wilson IG (1997). Inhibition and facilitation of nucleic acid amplification. Appl. Environ. Microbiol., 63(10): 3741–3751. https://doi.org/10.1128/aem.63.10.3741-3751.1997