Introduction

Bacterial contamination in fresh boar semen plays an important role in semen quality. Negative impacts affect fertility rate, embryonic or foetal death, and endometritis in sows/gilts after insemination1,2. The clinical appearance of endometritis is commonly observed with vaginal discharge3,4 which may occur due to different causes, e.g. hormonal imbalance5 or post-ovulatory insemination6,7. Although the severity of acute endometritis can be alleviated with antibiotics, acute endometritis can be progressive and turn into chronic endometritis, resulting in significant impacts on reproductive performance3.

Preservation of boar semen is a routine process for artificial insemination (AI) in the swine industry8. There are numerous advantages of AI, such as transmitted disease prevention, genetic improvement, piglet production, and quality enhancement1,8. Although AI can potentially reduce the rate of disease transmission from boar semen, microbial contamination in boar semen is an issue of concern since it also plays a major role in reproductive performance1.

Microbial contamination occurs during the process of semen collection and is derived from either animal or non-animal origin9,10. In addition, Gram-negative bacteria are the most predominant (more than 80%) from fresh boar semen isolation11. The significant impacts of bacterial contamination on semen quality include: (i) sperm motility reduction; (ii) increased sperm agglutination; (iii) acrosomal damage sperm; and (iv) plasma membrane disruption9,10. The relationship between boar semen quality and farm production reveal that sperm agglutination due to Escherichia coli (E. coli) contamination can markedly reduce litter size12. In practice, numerous antibiotics are mixed into the semen extender with the aims of inhibiting bacterial growth and limiting the negative impacts from the contamination13,4,5,6,7,8,9,10,11,12,13,14,15. For instance, gentamicin, neomycin, streptomycin, and other antibiotics are commonly used in boar semen extender16,17,18. In addition, more than one antibiotic can be combined with the boar semen extender, for example gentamicin and florfenicol or gentamicin and polymyxin B combinations have been used19. Consequently, the popularity of antibiotics to maintain farm production raises questions about the reasonableness for antibiotics use as well as the concern about antibiotic resistant bacteria20.

Colistin (polymyxin E) was discovered in 1947 as a secondary metabolite of Paenibacillus polymyxa subsp. Colistinus21,22. The use of colistin is conserved as a last-line antibiotic for humans in the treatment of serious infection caused by multidrug resistant Gram-negative bacteria23. Due to excessive use of colistin, there is an incidence of mobile colistin resistance (mcr) genes which is developed with chromosomal mutations and is plasmid-mediated in numerous bacterial species. Recently, the identification of mobilized colistin resistance genes has been reported as mcr-1 to mcr-10 and the mcr-1 gene is the most predominant23,24. By considering the mechanisms, the mcr-1 gene encodes phosphoethanolamine transferase which plays an important role in modification of lipopolysaccharides (LPS) in the outer surface of Gram-negative bacteria by adding phosphoethanolamine (PEA) to lipid A moieties. It then causes a lower affinity of colistin to its primary target25,26,27. The integron-integrase gene is an essential source of gene cassettes with horizontal gene transfer of antibiotic resistance. Consequently, this gene plays a crucial role in the spread and transmission of antibiotic-resistant determinants in resistant bacteria28. There are three classes of integron-integrase genes, with the class 1 integron-integrase (int1) gene is the most predominant29,30.

Some compounds (i.e. antimicrobial peptides and bacteriocins) in probiotics cell free supernatants (CFS) can inhibit the growth of other bacteria31. CFS with its antimicrobial peptides and/or bacteriocin constituent derived from probiotics, especially lactic acid bacteria (LAB), exhibit the inhibitory activities against various pathogenic bacteria. Previous studies of the antimicrobial activities of CFS have investigated various pathogens, including E. coli32, Salmonella Typhi and Salmonella Typhimurium33,34, Listeria monocytogenes35, and Staphylococcus aureus34. Although many authors have carried out the antimicrobial effects of CFS, there are vast gaps in CFS antimicrobial properties in bacterial isolation on animals. In particular, the pathogenic bacteria carrying antimicrobial resistance genes isolated from fresh boar semen are scarce and not comprehensive. The effect of CSF on inhibition the growth of other bacteria may guide the direction of finding a promising alternative method of using CFS-origin antimicrobial peptides and/or bacteriocin in the semen extender preservation of the pig industry.

Here, our work aims to identify the bacterial species, conduct antibiotic susceptibility testing (AST), and detect antimicrobial resistance genes (mcr-1 to mcr-10 and int1) isolated from fresh boar semen in Thailand. Furthermore, the study of inhibitory activities of CFS derived from Bacillus spp., Enterococcus spp., Weissella spp., Lactobacillus spp., and Pediococcus spp. against pathogens isolated from fresh boar semen are also investigated.

Results

Bacterial identification

Bacterial from fresh boar semen (n = 10) were successfully identified and confirmed using 16S rRNA gene sequencing. The bacterial identification and percentage of identity were presented in Table 1. The semen sample were identified as 10 bacterial species and classified into three species for Gram-positive bacteria and seven species of Gram-negative bacteria (Table 1). The three major pathogens identified of this study were P. aeruginosa (5/10; 50%), E. coli (4/10; 40%), and P. mirabilis (3/10; 30%). The other Gram-negative bacteria including Citrobacter koseri, Enterobacter hormaechei, Providencia stuartii, and P. alcaligenes showed one isolate (1/10; 10%). In addition, the Gram-positive bacteria were identified as Staphylococcus spp. (S. chromogenes, S. sciuri, and S. warneri).

Table 1 Bacterial identification, antimicrobial susceptibility against 10 antimicrobial agents, and antimicrobial resistant genes detection of pathogens isolated from fresh boar semen
Full size table

Antibiotic susceptibility test (AST)

Bacterial identification from boar semen were tested for antimicrobial susceptibility with 10 selected antibiotics. Most Gram-negative bacterial isolates were resistant to amoxicillin and ceftriaxone except for P. alcaligenes (Table 1). All P. aeruginosa isolates were resistant to ceftiofur (MIC > 8 μg/mL), amoxicillin (MIC > 32 μg/mL), ceftriaxone (MIC > 4 μg/mL), oxytetracycline (MIC > 6 μg/mL), and trimethoprim:sulfamethoxazole (1:19) (MIC > 8 μg/mL) (Table 1). Meanwhile P. aeruginosa was susceptible to colistin (MIC < 4 μg/mL), enrofloxacin (MIC < 2 μg/mL), and amoxicillin trihydrate:potassium clavulanate (4:1) (MIC < 32 μg/mL) (Table 1).

Similarly, all E. coli strains were resistant to amoxicillin (MIC > 32 μg/mL), ceftriaxone (MIC > 4 μg/mL), and trimethoprim:sulfamethoxazole (1:19) (MIC > 8 μg/mL), but susceptible to ceftazidime and colistin (MIC < 4 μg/mL) (Table 1). Furthermore, the 50% of E. coli isolates was susceptible to gentamicin (MIC < 16 μg/mL), ceftiofur (MIC < 8 μg/mL) as well as enrofloxacin (MIC < 2 μg/mL) (Table 1). P. mirabilis was resistant to colistin (MIC > 4 μg/mL), amoxicillin (MIC > 32 μg/mL), and ceftriaxone (100 %) (MIC > 4 μg/mL). All Gram-positive bacteria isolated from boar semen were susceptible to gentamicin (MIC < 16 μg/mL), enrofloxacin (MIC < 2 μg/mL), amoxicillin (MIC < 32 μg/mL), and trimethoprim:sulfamethoxazole (1:19) (MIC < 4 μg/mL) (100 %) (Table 1).

Detection of mcr and int1 genes

The detection of mcr-1 to mcr-10 and int1 in all pathogens isolated from fresh boar semen were performed using multiplex PCR and showed in Table 1. The int 1 gene positive was detected in 3 strains of P. mirabilis, 1 isolate of each P. aeruginosa, E. coli, C. koseri and E. hormaechei from both Farm A and B. Most pathogens in this study were not contained mcr-1 to mcr-10 genes except only one E. coli isolate from Farm B showed positive to mcr-3. This E. coli strain also exhibited the int 1 positive and was classified as multidrug resistant according to MIC result. All pathogens which are carried either int 1 or mcr-3 were classified as multidrug resistant strains based on the MIC result as shown in Table 1.

Cell free supernatants (CFS) against pathogens from boar semen

The three selected pathogens from boar semen (P. aeruginosa, E. coli, and P. mirabilis) were conducted to find out the inhibitory effect of CFS from probiotics. The CFS was collected from eight interesting probiotics and their characteristics were declared in Table 2. The antimicrobial activities of CFS were determined as inhibition zone by performing agar well diffusion assay (Fig. 1).

Table 2 Characteristics of CFS from selected probiotics.
Full size table
Figure 1
figure 1

The figure shows the diameter of the inhibition zone of CFS from probiotic bacteria against Escherichia coli at 8 h (a), 10 h (b), 12 h (c), 14 h (d) and 16 h (e) incubation time. The diameter of inhibition zone decreased with increasing incubation time. Note Ba Bacillus amyloliquefaciens, Bs Bacillus subtilis, Bx Bacillus axarquiensis, Nc Negative control, Wc Weissella confusa.

Full size image

The inhibitory activity of CFS against P. aeruginosa could be initially noticed at 8 h incubation except for the CFS from B. subtilis (Fig. 2a). At 10 h of incubation, there were no longer inhibitory effects of CFS from B. axarquiensis and B. amyloliquefaciens. On the contrary, the CFS produced from E. faecium, L. plantarum, P. acidilactici, P. pentosaceus, and W. confusa could remain the inhibitory effect with the presence of similar diameter of inhibition zone at 8, 10, and 12 h incubation. In addition, the inhibitory effect since 12 h incubation was declined in tendency. Interestingly, after 8 hours after incubation, comparing the inhibition zone of CFS from P. acidilactici and L. plantarum against P. aeruginosa, it was found that CSF from P. acidilactici showed a larger inhibition zone (26.50 to 30.50 mm) than CFS from L. plantarum (25.50 to 27.50 mm) (p value < 0.05).

Figure  2
figure 2

The antimicrobial activity at different time of CFS of probiotic against Pseudomonas aeruginosa (a), Escherichia coli (b) and Proteus mirabilis (c) isolated from boar semen.Note a,b,c,dSignificant difference between the CFS from probiotics at the same of incubation time (p value < 0.05).

Full size image

The interesting probiotics could produce CFS with inhibitory effects against E. coli except for CFS derived from B. axarquiensis (Fig. 2b). In details, the CFS produced from B. amyloliquefaciens and B. subtilis could inhibit E. coli in short period of time, 8-10 h and 8 h respectively. On the contrary, the CFS produced from E. faecium, L. plantarum, P. pentosaceus, and W. confusa thoroughly expressed the inhibitory effects from 8-16 h incubation time. Interestingly, it was apparent that CFS from P. acidilactici exhibited the largest inhibition zone (27.00 to 32.00 mm) against E. coli. Furthermore, the inhibitory effects of CFS produced from probiotics against E. coli shared the same tendency to P. aeruginosa.

Finally, the CFS produced by B. axarquiensis and B. subtilis could not show the inhibitory activities against P. mirabilis at 8-16 h incubation time (Fig. 2c). At 8-10 h incubation time, the diameter of inhibition zone expressed from L. plantarum, P. acidilactici, P. pentosaceus, and W. confusa were statistic equivalently. Interestingly, the CFS produced by P. acidilactici still exhibited the largest diameter of inhibition zone (27.00 to 28.00 mm) thorough the incubation period. In contrary, the CFS from B. amyloliquefaciens were presented inhibition activity with a smallest inhibition zone (13.00 to 17.00 mm).

Ultimately, it was apparent that the inhibitory activities of probiotics derived CFS shared the same tendency against P. aeruginosa, E. coli, and P. mirabilis. In other words, the inhibitory activities became less effective from the decline of inhibition zone when it had continued for a long period of incubation time.

Discussion

It was apparent that the bacteria in fresh boar semen could be identified as both Gram-negative and Gram-positive bacteria. P. aeruginosa, E. coli, and P. mirabilis are the predominant bacteria in this study. The results of this survey are found to be similar to a previous survey undertaken in Brazil38.In addition, P. aeruginosa and E. coli caused negative effects on boar spermatozoa, whether to induce sperm agglutination or decrease sperm motility10,38. That effect could be limited by using gentamicin antibiotics in semen extender10. The antimicrobial susceptibility test revealed that P. aeruginosa and E. coli show susceptibility rates to gentamicin 25% and 50%, respectively, while P. mirabilis was susceptible to gentamicin for 100%. Additionally, it was observed that sows with endometritis had higher rates of the antimicrobial resistant bacteria. These findings should be of concern since endometritis can be transmitted by bacterial contamination in boar semen1,37. In their study, Burch and Sperling40 found that 41% of endometritis sows were caused by a single bacterial infection, 72.3% of which were identified as E. coli. From the result, E. coli was the most resistant to common antibiotics including amoxicillin and tetracycline, which are used in pig farms and boar semen extender.

The antimicrobial susceptibility test from fresh boar semen revealed that the majority of Gram-negative bacteria were resistant to antibiotic, whereas the Gram-positive bacteria were less antimicrobial resistance. The antibiotic resistance ratio from fresh boar semen was similar to a study undertaken in Italy41, while a study in Romania found 56.52% of Gram-negative boar semen bacterial isolation were resistant to gentamicin42. This study discovered a high rate of antibiotic drug resistance which increases concerns about the problems caused by the use of antibiotics in agriculture, while the use of antibiotics with semen extender is recommended to protect the spermatozoa14. The discovery of novel antimicrobial compounds such as antimicrobial peptides and/or bacteriocin derived from CSF to replace conventional antibiotics is an interesting issue to conduct in the future. It has been shown that the alternative methods were studied to find a feasible way to reduce the use of antibiotics in boar semen extender, including: (i) bacterial removal by a physical method using single-layer centrifugation43; (ii) antimicrobial peptides (AMPs) or short antimicrobial lipopeptides44; (iii) other substances, such as lysozyme and kojic acid45,46; and (iv) semen storage at low temperature conditions (5 °C) without antibiotic supplement47. These methods have advantages and disadvantages, such as loss of spermatozoa from the physical method43 or no broad-spectrum activity with kojic acid46. Moreover, each method has strengths and weaknesses in terms of antimicrobial activity and the effect on semen quality.

Although studies on bacterial contamination and antibiotic resistant bacteria from boar semen are available, no report of antimicrobial resistance genes, particular mcr-3 gene was found. One boar semen sample was detected for mcr-3. By considering the relationship between mcr-3 detection and colistin resistance, it was found that the sample was susceptible to colistin. The emergence of this incident resembles the study of mcr-1, mcr-4, and mcr-5 presented by García et al.48 as well as mcr-1 to mcr-10 by Nguyet et al.49. These studies utilised five samples of mcr genes positive (one sample from mcr-1 and four samples from mcr-4) which were found to be susceptible to colistin48. The positive mcr gene without colistin resistant spectacle was possible because bacteria were a carrier of an inactive form of the mcr gene48. Meanwhile the result of four negative mcr genes were found to be colistin resistant by the MIC test (three samples from P. mirabilis and one sample from P. stuartii). The colistin resistant with the negative mcr gene corresponded to a previous study of E. coli49. The colistin resistance was supposed to have another mechanism that does not depend on mcr genes. Consequently, the higher incidence of mcr genes in livestock animals significantly increased the risk of mcr genes being transmitted to humans. In detail, the mcr genes could be transmitted to humans via foodborne, zoonotic, and vector-borne routes. Nevertheless, the incidence of the mcr gene in humans is higher than the incidence in animals50.

The prevalence of the int1 gene has been studied at different stages of the pig production system. The results indicate that there is a high-rate detection of the int1 gene in sows and piglets by conducting a rectal swab. Nevertheless, detection of the gene in boars has not been studied51. The present study successfully detected the int1 gene from boars. In addition, the int1 gene was detected from P. mirabilis, P. aeruginosa, E. coli, C. koseri, and E. hormaechei. The int1 gene has been detected in Gram-negative bacteria and resulted in various antimicrobial resistances, such as β-lactam, sulfonamide, and aminoglycoside28,52. According to the related literature, the int1 gene was not only detected from Gram-negative bacteria but also from Gram-positive bacteria including Staphylococcus spp.53,54. Furthermore, Stalder et al.55 found that the presence of the int1 gene increased the risk of the spread and transmission of resistance genes to environment and other bacteria.

Our findings, one E. coli isolate was detected both of the mcr-3 and int1 genes. This result corresponded with a previous study49 which had 24 out of 37 samples positive for both int1 and mcr genes, and four samples were only positive for the int1 gene.

The results of the antimicrobial effect of CFS from selected probiotics exhibited a decreased diameter of the inhibition zone as the incubation time continued. To support this finding, our results are consistent with previous studies32,34,56. In addition, Kaewchomphunuch et al.32 reported the CFS from L. acidophilus, L. plantarum, and P. pentosaceus which expressed inhibitory activity only to pathogenic E. coli isolated from pigs. Nevertheless, the previous study did not undertake a comprehensive study of the inhibitory activity against other pathogens or specific E. coli strain isolated from semen origin.

The CFS produced from L. acidophilus could inhibit the growth of P. aeruginosa while another activity was able to disrupt biofilms from P. aeruginosa56. The difference between El-Mokhtar’s study and the present study are the probiotics for CFS collection. Our study collected from other probiotics in lactic acid bacteria (LAB) including P. acidilactici, P. pentosaceus, and L. plantarum which showed similar results.

From the results of the present study, P. mirabilis could be inhibited by CFS from selected probiotics except for B. axarquiensis and B. subtilis. The inhibitory results are similar to Shaaban et al.57 study which mentioned the action from L. casei and L. reuteri. Besides, the inhibitory activities of pathogen growth inhibition derived from the CFS of L. casei and L. reuteri were able to inhibit P. mirabilis biofilm formation as well.

Factors that affect the ability to inhibit pathogens that depend on pH or concentration of antimicrobial compound including lactic and acetic acid or antimicrobial peptides (AMPs)57,58. The comparable findings of CFS from L. johnsonii was between pH 3.5 and pH 6.0. In addition, it has been reported that lower pH (acidic condition) could inhibit growth of B. cereus while higher pH causes the activity to disappear58. The reduced antimicrobial activity of CFS from Lactobacillus spp. and Enterococcus spp. cultures were found when their pH value was greater than 4.559. However, a study of CFS from L. plantarum adjusted the pH up to 6.5 and the antimicrobial activity was still maintained34. In the same direction, Soria and Audisio58 revealed that the different compound in CFS also influenced the inhibitory activity of bacterial growth. From the present results, the pH value was observed as a main feature in the antimicrobial activity of CFS, the key compound in CFS for inhibiting pathogen growth might be that of AMPs and/or bacteriocins. With regard to the pH influence on inhibitory activity, our 3 CFSs produced from Lactobacillus plantarum NN31-5B, Pediococcus acidilactici NN82-7M and Pediococcus pentosaceus NN115-6M were neutralized to pH 6.0 and performed the antimicrobial activities with boar semen pathogens by agar well diffusion assay (data not shown). The inhibition zone was decreased but still inhibit the pathogens which showed similar results to other studies34,56,57. This can be suggested that our CFS might be contained antimicrobial peptides and/or bacteriocin with their antimicrobial properties. The AMPs were isolated from CFS (subtilosin) derived from B. amyloliquefaciens and could inhibit the bacterial vaginosis associated bacteria similar with the subtilosin from B. subtilis60. In addition, organic acid was found to be an essential compound of CFS derived from LAB. Nevertheless, Arrioja-Bretón et al.34 mentioned that the acid was not the only factor inhibiting bacterial growth. To support this hypothesis, Tenea61 demonstrated that the AMPs extracted from LAB-producing CFS (L. plantarum and Lactococcus lactis) could also inhibit against Salmonella enterica.

Although the previous studies were conducted to investigate the inhibitory effect of CFS derived from probiotics with some food-poisoning bacteria, our recent study successfully declares the inhibitory effect against pathogenic bacteria carrying antimicrobial resistance genes isolated from boar semen. However, further studies are needed to identify the constituent antimicrobial compound in this CFS, construct and synthetic these AMPs. Furthermore, the interaction between synthetic antimicrobial peptides and spermatozoa and the synthetic antimicrobial peptides-based semen extender without antibiotics will be included in a further study to determine its effect on the qualities of spermatozoa and field fertility.

Conclusions

In conclusion, using bacterial contaminated fresh boar semen for artificial insemination is a possible cause of sow endometritis. Consequently, antibiotics are added to the boar semen extender for artificial insemination. Fortunately, CFS derived from probiotics can effectively inhibit bacteria carrying antimicrobial resistance genes isolated from fresh boar semen, in particular the CFS produced from the LAB. However, the ability against bacteria is observed and decreased with more extended incubation periods. In the present results, CFS with its antimicrobial peptides and/or bacteriocin constituent inhibits semen-origin pathogenic bacteria provide the direction as a promising alternative antibiotics method used in the semen extender preservation of the pig industry.

Materials and methods

Sample collection

Total semen samples (n = 10) were collected from 10 individual boars in three distinct AI centers in commercial pig farms in Thailand. Boars were housed in an individual pen in an evaporative cooling house system. They were fed with a commercial feed 3 kg once a day and the water were ad libitum. The sampling locations were conducted in Chai Nat province (farm A; n = 5), Chon Buri province (farm B; n = 3), and Chachoengsao province (farm C; n = 2). The boar semen samples were collected using the gloved hand technique and the whole ejaculate was filtrated with sterile gauze to eliminate the gel-rich fraction7. Then, only a fresh sperm-rich fraction was stored in sterile container. All specimens were preserved under the sterile repository at 4 °C and immediately shipped to the Laboratory of Bacteria, Veterinary Diagnostic Center, Faculty of Veterinary Science, Mahidol University. The research ethics was approved by the Faculty of Veterinary Science, Mahidol University-Institute Animal Care and Use Committee (FVS-MU-IACUC-Protocol No. MUVS-2021-10-41), Animal use license No. U1-01281-2558. All methods were performed in accordance with the relevant guidelines and regulations.

Bacterial isolation and species identification

All semen samples were cultured on tryptone soy agar (Oxoid, UK) with 5% sheep blood and MacConkey agar (Oxoid, UK) incubated at 37 °C for 18-24 h. All different colonies were identified using standard biochemical tests followed by 16S rRNA sequencing and stored in Brain Heart Infusion (BHI) (Oxoid, UK) with 20% glycerol at -80 °C. Genomic DNA of all isolates was performed using G-spinTM genomic DNA extraction kit (iNtRON, Republic of Korea) and amplified 16S rRNA by PCR with a BiometraTOne96G thermal cycler (AnalytikJena, Germany) using UFUL (5’- GCCTAACACATGCAAGTCGA-3’) and 800R (5’-TACCAGGGTATCTAATCC-3’) primers. The PCR was performed with the following protocol: initial denaturation at 94 °C for 3 min followed by 30 cycles of denaturation at 94 °C for 30 sec, annealing at 55 °C for 30 sec, and extension at 72 °C for 45 sec, with a final extension step at 72 °C for 5 min. The PCR products were purified by MEGAquick-spinTM Plus Total Fragment DNA purification kit (iNtRON, Republic of Korea) and sequenced with an Applied Biosystems 3730XL DNA Analyzer (Bionics, Republic of Korea). Each 16S rRNA sequences was blasted against the NCBI nucleotide database (https://blast.ncbi.nlm.nih.gov) to identify all isolates.

Antimicrobial susceptibility testing (AST)

All isolates were streaked onto blood agar to obtain single colony. Following incubation at 37 °C for 18-24 h, one to three colonies with similar morphological appearance were transferred into normal saline solution (0.85% NaCl) and thoroughly mixed. The turbidity of bacterial suspension was measured using 0.5 McFarland standard (approximately 108 CFU/mL). The minimum inhibitory concentrations (MIC) were conducted by the broth microdilution with following a guideline from the Clinical and Laboratory Standards Institute (CLSI). The assays were performed in triplicate with 96 well plates; in each well, 100 μL of bacterial suspension previously diluted in Mueller Hinton broth (Difco, USA) to 106 CFU/mL were added to 100 μL of appropriate dilutions of antimicrobials. A total of 10 antimicrobials were tested in the following concentrations by means of two-fold dilution: amoxicillin (TCI, Japan) 1-128 μg/mL, amoxicillin trihydrate:potassium clavulanate (4:1; Sigma, Germany) 1-128 μg/mL, ceftazidime (Sigma, Germany) 1-128 μg/mL, ceftriaxone (TCI, Japan) 0.25-32 μg/mL, ceftiofur (TCI,Japan) 0.25-32 μg/mL, colistin (Sigma, Germany) 0.25-32 μg/mL, enrofloxacin (Fluka Biochemika, Japan) 0.06-8 μg/mL, gentamicin (TCI, Japan) 0.5-64 μg/mL, oxytetracycline (AppliChem, USA) 0.25-32 μg/mL, and trimethoprim:sulfamethoxazole (1:19; TCI, Japan) 0.25-32 μg/mL. The 96 well plates were incubated at 37 °C for 16-20 h. Medium without antimicrobials was conducted as control and inoculated prior to and following each antimicrobial-containing series of plates. MIC values were recorded after incubation and defined as the lowest concentration of each antibiotic without a visible growth of bacteria. The reference strains Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were conducted in each experiment to assess the reliability of methodology.

Detection of mcr and int1 genes

Bacterial plasmid DNA was extracted from all pathogens by using QIAprep Spin Miniprep Kit (Qiagen, Germany) with following the manufacturer’s instruction. DNA concentrations were measured using a BioDrop DUO (DKSH, UK). The plasmid-mediated colistin resistance genes (mcr-1 to mcr-10) and the class 1 integron-integrase gene (int1) were detected by multiplex PCR using Green PCR master mix kit (Biotechrabbit, Germany) with following the Nguyet et al. protocol49 of the primers and PCR conditions. Briefly, the amplification steps were performed using BiometraTOne96G (AnalytikJena, Germany) with the following thermal cycles: the initial denaturation at 94 °C for 3 min, followed by 25 cycles denaturation at 94 °C for 30 sec, annealing at 58 °C for 90 sec, and extension at 72 °C for 60 sec, a final extension step at 72 °C for 5 min. The PCR products were separated using 1.5% agarose gel electrophoresis, stained with 1X GelRed (Sigma Aldrich, USA), and visualized under an UV transilluminator UVP GelStudio (AnalytikJena, USA). The reference strain E. coli ATCC 25922 was conducted as a negative control strain, while E. coli harboring mcr genes were conducted as a positive control strain.

Probiotic strains used

The probiotic strains used in our study were thoroughly considered and then selected the probiotics which were supported and based on our preliminary and previous studies. To exemplify, Kaewchomphunuch et al.32 reported on the inhibitory activities of Lactobacillus spp., Pediococcus spp., and Enterococcus spp. against pathogenic E. coli isolated from pigs in Thailand32. Aupad et al.36 also carried out the antibacterial activities of isolated Bacillus spp. against bacteria found in food. The multidrug-resistant E. coli subjected to the Weissella confusa activity were investigated by Dey et al.37. Thus, the probiotic candidates were listed in Table 2.

Cell free supernatants (CFS) preparation from selected probiotics

The probiotic strains used in this study were obtained from a collection of bacterial stock cultures which stored in glycerol at -80 °C from the Laboratory of Bacteria, Veterinary Diagnostic Center, Faculty of Veterinary Science, Mahidol University. The CFS was prepared according to the Kaewchomphunuch et al. protocol32. Briefly, overnight MRS culture broth of eight selected probiotics (Table 2) was transferred into 1.5 mL microcentrifuge tube and centrifuged for 2 min at 5,000 rpm (Denville Micro 260D Microcentrifuge, Denville Scientific, Inc., Metuchen, USA). Supernatants were collected by pass through 0.22 μm sterile syringe filter (Guangzhou Jet Bio-Filtration Co., Ltd., Guangzhou, China). The filtrated CFS was either used freshly in agar well diffusion assay or stored at -20 °C for further analysis.

Agar well diffusion assay

Pseudomonas aeruginosa, E. coli, and Proteus mirabilis isolated from boar semen were subjected for testing the inhibitory effect of CFS from probiotics. All bacteria were cultured in BHI broth at 37 °C for 20-24 h. Bacterial suspension was initially diluted into to 0.5 McFarland standard and performed spread plate method onto nutrient agar. Then, the inoculated nutrient agar was pierced with the sterile 8 mm diameter cork borer to create wells. The volume of 100 μL of CFS was loaded into wells and incubated at 37 °C for 8, 10, 12, 14, and 16 h. After incubation, the inhibition zone will be measured in each well. To validate the result, MRS broth (pH 6.0) was conducted as a negative control.

Statistical analysis

The descriptive statistic was used in this study. In addition, the data analysis was performed by using one-way analysis of variance (ANOVA) and compared means by using Duncan’s test by The PASW Statistics for Windows, version 18.0 (SPSS Inc., Chicago, IL, USA). A statistical significance is determined as p value < 0.05.

Ethics declarations

The study was conducted in compliance with the ARRIVE guidelines. The research ethics was approved by the Faculty of Veterinary Science, Mahidol University-Institute Animal Care and Use Committee (FVS-MU-IACUC-Protocol No. MUVS-2021-10-41), Animal use license No. U1-01281-2558.