Antimicrobial resistance and virulence genes of Streptococcus agalactiae isolated from mastitis milk samples in China

Streptococcus agalactiae, or group B Streptococcus (GBS), is a Gram-positive, facultatively anaerobic bacterium which was first described as the leading causative pathogenic bacteria of bovine mastitis (45). It causes a decline in milk production and quality, serious economic losses, and poses a substantial challenge to public health, making it a key challenge for the dairy industry. Antibiotic treatment is the first line of defence against mastitis (38); however, antimicrobial resistance and antimicrobial residues are areas of increasingly serious concern in both human and veterinary medicine (37). This coccus uses virulence factors encoded by its genes to enter, replicate and persist in the host (17). Antimicrobial resistance genes confer GBS with the ability to resist the effects of antibiotic medication, and these genes are found in the environment (27). Therefore, resistance and virulence are two key factors that determine the degree of infection; they may be correlated with each other (14). There are a variety of pathogenicity-related virulence factors associated with S.

agalactiae, including adhesion-, invasion-, and immune escape-related factors. The adhesion-related virulence genes include fbsA, scpB and lmb, which encode fibrinogen binding protein A, C5a peptidase, and laminin binding protein, respectively (7, 32). Invasion-related virulence genes include cylE, cfb, hylB and bca, which encode β-haemolysin, CAMP factor and α-protein, respectively, as well as hylB, which participates in the formation of hyaluronate lyase (12, 24). Some virulence genes, such as bac, facilitate immune evasion (16). Strongly drug-resistant strains were found to carry few virulence genes, confirming that resistance and virulence were negatively correlated (35). A contrasting positive correlation was posited by Arshadi et al. (2), who showed that virulence gene–positive strains became more pathogenic and more difficult to eliminate as drug resistance increased. The inexistence of any correlation is also possible, because one study reported drug resistance which was not related to virulence (36).

Conducting antimicrobial resistance studies is imperative for selecting the most appropriate antimicrobial therapies and reducing the risk of further development and spread of antimicrobial resistance through the horizontal transfer of resistance genes. Investigations of the S. agalactiae virulence and antibiotic resistance genes are important for the development of vaccines and for understanding the resistance mechanisms used by pathogens. The objective of this study was to identify the virulence and antibiotic resistance genes found in S. agalactiae isolates from mastitis milk samples in China, as well as to study phenotypic antimicrobial resistance and the correlation between the genes of one type and those of the other.

Bacterial strains. Fifteen S. agalactiae strains were isolated from 497 bovine mastitis milk samples at the Agricultural Quality and Safety Laboratory of Xinjiang according to the NY/T 2962-2016 standard (42). Standard strains of S. agalactiae (American Type Culture Collection (ATCC)12386) were purchased from BaoxinBio Inc. (Urumqi, China).

Antimicrobial susceptibility tests. The broth dilution method recommended by the Clinical and Laboratory Standards Institute (CLSI) (9) was used to evaluate the minimum inhibitory concentrations (MICs) of 16 commonly used antimicrobial drugs to suppress growth of the 15 S. agalactiae strains (Table 1). Each antibiotic tested was serially twofold diluted and wells containing different concentrations were prepared: penicillin (PEN), ampicillin (AMP), clindamycin (CLI), ciprofloxacin (CIP), and rifampicin (RIF) were prepared at concentrations of 64.0, 32.0, 16.0, 8.0, 4.0, 2.0, 1.0, 0.5, 0.25 and 0.125 mg/mL; erythromycin (ERY), cefoxitin (CET), oxacillin (OXA), tetracycline (TE), doxycycline (DOX), gentamicin (GM), florfenicol (FFC), and vancomycin (VAN) were prepared at concentrations of 128.0, 64.0, 32.0, 16.0, 8.0, 4.0, 2.0, 1.0, 0.5, 0.25 and 0.125 mg/mL; sulfisoxazole (SMZ) was prepared at concentrations of 1024, 512, 256, 128, 64, 32, 16, 8, 4 and 2 mg/mL; amoxicillin/clavulanic acid (A/C) was prepared at concentrations of 128/64, 64/32, 32/16, 16/8, 8/4, 4/2, 2/1, 1/0.5, 0.5/0.25 and 0.25/0.12 mg/mL; and sulfamethoxazole (SXT) was prepared at concentrations of 64/1208, 32/604, 16/304, 8/152, 4/76, 2/38, 1/19, 0.5/9.5, 0.25/4.8 and 0.12/2.4 mg/mL. Streptococcus agalactiae strains (ATCC12386) were used as a quality control in this study. Because no specific resistance breakpoints for Streptococcus spp. were available for some tested antimicrobials, the resistance breakpoints for an antimicrobial of the same class were referred to as recommended by the CLSI (9).

Genomic DNA extraction. The strains were inoculated into BHI broth and cultured for 18–24 h at 36°C, after which genomic DNA was extracted according to the Bacterial DNA Extraction kit manufacturer’s instructions (Tiangen BioTech, Beijing, China). The concentration and mass of extracted DNA were determined using a Thermo NanoDrop 2000C spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). All DNA samples were stored at −20°C until use.

PCR amplification of drug resistance genes and virulence genes. The primer sequences for the drug resistance and virulence genes of S. agalactiae were synthesised at the Beijing Genome Institute (Beijing, China) (Table 1). The PCR amplification reaction was performed in a 25 μL mixture volume comprised of 12.5 μL of 2 × Taq PCR Master Mix enzyme (Takara BioTech, Kusatsu, Japan), 1 μL of upstream and downstream primers, 2 μL of DNA template and 8.5 μL of sterilised ultrapure water. The programme for amplification of resistance genes was as follows: pre-denaturation at 94°C for 4 min, denaturation at 94°C for 30 s, annealing at a temperature determined individually for each gene (Table 2), and extension at 72°C for 90 s. Thirty cycles were performed, followed by incubation at 72°C for 10 min.

The program for amplification of the virulence gene was pre-denaturation at 95°C for 5 min, followed (as in the reaction for resistance genes) by denaturation at 94°C for 30 s, annealing at a temperature determined individually for each gene (Table 3), and extension at 72°C for 1 min. As previously, thirty cycles were performed, followed by incubation at 72°C for 10 min. The amplified products were electrophoresed on a 1% agarose gel with 1× tris-acetate-ethylenediaminetetraacetic acid buffer. Electrophoresis was performed at 120 V for 30 min with gel pictures taken afterwards using a Bio-Rad Gel Doc XR molecular imaging system (Bio-Rad, Hercules, CA, USA).

Gene sequencing and analysis. The PCR products and upstream and downstream primers of each gene were sent to Sangon Biotech Co. Ltd. (Shanghai, China) for sequencing. Homology for the obtained DNA sequence was then searched for in NCBI databases (www.ncbi.nlm.nih.gov/) to determine the genotype.

Correlation between phenotype and genotype of S. agalactiae. The consistency rate between the genotype and phenotype of the drug resistance of S. agalactiae to the 16 antibiotics was determined according to the positivity rate of the resistance genotype and the resistance phenotype of S. agalactiae defined as follows:

Resistance to three or more antibiotics is the definition for a strain to be designated multi drug resistant (MDR). The result interpretation was performed according to the CLSI guidelines (9).

Statistical analysis. All data entry and analyses were performed using the Statistical Package for the Social Sciences (SPSS) software, version 24.0 (IBM, Armonk, NY, USA). Binary logistic regression was used to analyse the relationship between antibiotics and virulence genes, the antibiotic resistance was adopted as the dependent variable and the detection of virulence genes as the independent variable to be included in the binary logistic regression model. The cut-off for statistical significance was set at P <0.05.

Results of the antibiotic sensitivity test of S. agalactiae isolates. Susceptibility testing of 16 antibiotics on 15 S. agalactiae strains (Table 4) showed that collectively they were 100% (n = 15) susceptible to RIF and VA, highly susceptible to SMZ and SXT (93.33%, n = 14), and moderately susceptible to A/C, AMP, CET, DOX, and CIP (all >70.00%, n = 10 and n = 11). The strains were highly resistant to OXA, TE, and ERY (80.00%, n = 12), and moderately resistant to CLI (66.67%, n = 10). The drug classes which encountered the highest resistance were macrolides (80.00%), lincosamides (66.67%), and tetracyclines (46.67%). The lowest and highest MIC values were 0.5 and 32 μg/mL, respectively for CET, 0.5 and 32 μg/mL for OXA, 2 and 256 μg/mL for SMZ, 4 and 64 μg/mL for TE, 0.5 and 32 μg/mL for DOX, 0.5 and 64 μg/mL for ERY, 0.25 and 16 μg/mL for CLI, 1 and 16 μg/mL for GM, 0.5 and 4 μg/mL for CIP and 2 and 16 μg/mL for FFC.

Multidrug resistance pattern. Analysis of S. agalactiae showed that 100% of the strains were resistant to more than three antimicrobial agents. The major MDR profile observed in the multidrug-resistant isolates was OXA-TE-ERY-CLI-GM. Four isolates (26.67%) were resistant to three antimicrobials, one isolate (6.67%) was resistant to four of the preparations, five isolates (33.33%) were resistant to five of them, one isolate (6.67%) was resistant to six, three isolates (20.00%) were resistant to seven, and one isolate (6.67%) was resistant to nine antimicrobials. Therefore, resistance to three and five antibiotics was the most common, and resistance to up to nine antibiotics was observed (Fig. 1).

Fig. 1

Consistency between resistance genotype and phenotype of S. agalactiae. Table 5 shows the occurrence of different resistance genes among the S. agalactiae isolates. Of the eight resistance genes screened for, ermB (n = 11) was detected as the most prevalent by PCR, followed by ermA (n = 10), and lnuA (n = 9).

The strains’ positivity rates for ermB, ermA and ermC were 73.33%, 66.67%, and 33.33%, respectively. Of the 12 isolates that demonstrated resistance to macrolide antibiotics, 11 (91.66%) carried ermB, 10 (83.33%) ermA and 5 (41.66%) ermC. The prevalence rate of lnuA was 60%. Isolates demonstrating resistance to lincosamide antibiotics, of which there were 10, were in 9 instances (89.99%) positive for lnuA genes. The detection rates for tetM, tetK, tetS and tetO were 46.67%, 40.00%, 40.00% and 33.33%, respectively. Overall, we found three resistance genes encoding macrolide resistance (ermB, ermA and ermC), one gene for lincosamide resistance (lnuA), and four genes (tetM, tetK, tetS and tetO) for tetracycline resistance.

Distribution of virulence genes of S. agalactiae. Fifteen isolates were screened for 12 genes potentially involved in virulence using PCR. The lmb and bac genes were not present in S. agalactiae isolates, while the fbsA gene was harboured by 14 isolates (93.33%). The clyE, hylB, bibA and iagA genes were discovered in S. agalactiae at incidences of 53.33%, 80.00%, 73.33%, and 86.67%, respectively. In contrast, glnA, bca, cfb and scpB were identified in only six (40.00%), four (26.67%), three (20.00%), and one (6.67%) of the isolates, respectively (Fig. 2). In relation to the virulence genes screened for in this study, eleven virulence profiles were detected, which were glnA+hylB+bibA+iagA+fbsA+clyE, glnA+hylB+iagA+fbsA, bca+glnA+hylB+iagA+fbsA+clyE, bca+cfb+hylB+bibA+iagA+ fbsA+clyE, bibA+iagA+fbsA+clyE, bca+hylB+bibA+iagA+fbsA, hylB+bibA+iagA+fbsA, bca+bibA+iagA+fbsA, hylB+bibA+iagA+ fbsA+clyE, cfb+hylB+bibA+iagA+fbsA and cfb+hylB+bibA+ fbsA+clyE (Table 6). The raw mastitic milk samples contained S. agalactiae strains with glnA+hylB+bibA+iagA+fbsA+clyE and glnA+hylB+iagA+fbsA as the virulence gene combinations most commonly detected.

Fig. 2

Correlation between drug resistance and virulence genes in multidrug resistant S. agalactiae. The multiple-drug-resistant strains carried different virulence genes (Table 7). However, analysis of drug resistance and virulence genes of multidrug-resistant S. agalactiae showed that all such strains carrying the hylB, iagA and fbsA virulence genes, as well all of as those carrying the ermB and lnuA resistance genes, were resistant to both tetracycline (90%) and clindamycin (70%) (Table 6). However, there was no significant correlation between virulence genes and the MDR ofS. agalactiae strains (P >0.05) (Table 8).

Bovine mastitis is often caused by S. agalactiae, which is a highly contagious obligate bacterial pathogen of the mammary gland resulting in milk contamination and causing potential harm to human health (20, 22). Contributions to the literature on S. agalactiae as the aetiological agent of bovine mastitis originate from many parts of the world: previous studies have shown that bovine mastitis outbreaks caused by S. agalactiae occurred more frequently in Denmark and Norway than elsewhere (1, 20) and that it is also the most common pathogen causing bovine mastitis in China (47).

Antibiotics remain the preferred treatment for bovine mastitis. However, in recent years, with the improper use of antibiotics, the problems of drug residues and resistance have become increasingly serious, not only threatening human health, but also bringing new challenges to the treatment of bovine mastitis. This study showed that S. agalactiae is highly resistant to antibiotics, particularly macrolides (80.00%), lincosamides (66.67%) and tetracycline (46.67%) and highly susceptible to ansamycins (100%), glycopeptides (100%), sulfonamides (93.33%) and quinolones (73.33%). We also showed that 100% of strains exhibited MDR, revealing the severity of this problem in China.

Previous studies showed that S. agalactiae was highly resistant to β-lactams, macrolides, and lincosamides, and was highly sensitive to quinolones and tetracyclines. In Yunnan province, China, S. agalactiae was resistant to multiple drugs, including β-lactams, macrolides, and sulfonamides (40). In Jordan and Brazil, S. agalactiae was reported to be extensively resistant to aminoglycosides and tetracyclines (11). According to the World Organisation for Animal Health (43), tetracyclines and macrolides were the two classes of antibiotics most commonly used in animals worldwide between 2010 and 2015. In our study, the results of the antimicrobial resistance tests indicated that S. agalactiae isolates from cows differed in their antimicrobial susceptibility patterns. The isolates were resistant to tetracycline and erythromycin, which is consistent with previous reports on bovine S. agalactiae antibiotic resistance and provides further evidence that antibiotic-resistant S. agalactiae are a global problem. We strongly recommend that tetracycline and erythromycin should not be used in the treatment of dairy cow mastitis in China, owing to their ineffectiveness against the 80% of S. agalactiae strains which can resist them. This conclusion will be shared with local dairy farms as a guide for the accurate prevention and treatment of bovine mastitis caused by this pathogen.

In Streptococcus, resistance to tetracyclines is encoded by ribosome protection genes including tetM and tetO or by the tetK and tetL efflux pump genes (34). Resistance to macrolides is due to two common mechanisms: a ribosome methylase, encoded by the erm genes, and an active efflux pump by a membrane-bound protein encoded by the mef gene. Resistance to clindamycin is encoded by a lincosamide that inactivates the lnuA nucleotidyl transferase gene. The results of this study showed that the dominant drug resistance gene of S. agalactiae was ermB and that between the resistance phenotype and the carrier rate for macrolides, lincosamides, and tetracyclines there was substantial consistency with an average rate of >70%. The ermB and ermA genes were detected in 73.33% and 66.68% of the 15 S. agalactiae isolates, respectively, suggesting that erythromycin-resistant methylase may be the major mechanism of resistance in the present study. Similar findings were reported by Hernandez et al. (18) for S. agalactiae strains isolated from Argentinean cattle with mastitis and by da Silva et al. (10). The prevalence of the ermB determinant shows that S. agalactiae often uses a target methylation mechanism for macrolide resistance.

S. agalactiae exhibits resistance to antibiotics through various metabolic mechanisms mediated by the corresponding resistance genes. The mefA gene– mediated efflux pump mechanism is the main pathway of the bacterium’s resistance to macrolides (5), erm leads to resistance to erythromycin by changing the drug target and methylation of the ribosomal 23S rRNA (41), and the tet-mediated ribosome protective protein mechanism confers resistance to tetracycline (4). This study suggests that S. agalactiae in China may have gained resistance to macrolides and tetracycline via ribosome binding protein sites and efflux pump mechanisms.

The PCR assay for the detection of virulence genes revealed that a high percentage of the S. agalactiae isolates were positive for fbsA (93.33%, 14/15), iagA (86.67%, 13/15) and hylB (80.00%, 13/15) (Fig. 2). The high prevalence of these genes in S. agalactiae has been reported previously (6, 30). However, higher frequencies were not universally detected for all virulence genes. We observed a low frequency of scpB (6.67%) and bca (26.67%) in S. agalactiae isolates, and obtained results inconsistent with the 90.1% for scpB and 86.0% for bca reported in Zimbabwe (18). Furthermore, none of the isolates in the present study possessed the bac or lmb genes. The bac gene was found with a similar frequency to that resulting from previous studies in the USA and Sweden with 20% and 12% of group B Streptococcus possessing bac, respectively (26, 31). A study conducted in Argentina in 2021 also observed high frequencies for hylB (100%) and cpsA (96%) and a lower frequency for bca (36%) in S. agalactiae; bac, lmb, and scpB genes could not be detected in any of the isolates (28). Our comparison with studies from Zimbabwe, the USA, Sweden, and Argentina suggests that discrepancies may be due to geographical location among other factors.

Research has found certain regional differences in the prevalence of S. agalactiae drug resistance phenotypes and virulence genes. Besides the relevance of this in bovine mastitis treatment, it is also an aspect to take into account in healthcare provision in humans. The results for the drug resistance and virulence of bovine S. agalactiae in our study were compared with those from analysis of strains isolated from human samples. Associated research in humans has often focused on S. agalactiae infection in pregnancy, because it can cause abortion and premature rupture of membranes and because when postpartum intrauterine infection with these bacteria becomes severe, it can lead to neonatal infection and other risks (8). Penicillin is often used when S. agalactiae causes adverse reactions during pregnancy, but erythromycin and clindamycin are often used for treatment in patients with allergic reactions to penicillin, resulting in a gradual increase in the drug-resistance rate of human S. agalactiae, which is consistent with our study. According to research results on human S. agalactiae from Guangzhou (25), Urumqi (19), Shenzhen (48), Beijing (41), Taiwan and Zhejiang (46) in China, the ermB gene is the main mediating gene of S. agalactiae resistance to erythromycin from both human and bovine sources. However, there are some differences in the genes related to drug resistance in S. agalactiae isolated from different regions. In addition to ermB, ermA, ermC, and ermTR, there are other drug resistance genes. When we compared the virulence genes of S. agalactiae with those identified in studies on S. agalactiae isolated from people in Guangzhou (48) and Hainan (40), we found that they carried cylE, hylB, bibA, iagA, bca, cfb and scpB, but not fbsA, cyl or glnA. A possible reason for this may be that different virulence genes are carried in different regions and sources.

The genetic mechanisms of virulence and drug resistance are achieved through the transfer of genes and movable elements between species and genera, and virulence and drug resistance are also interrelated (3). Several studies have confirmed the relationship between drug resistance and bacteria virulence. Streptococcus agalactiae isolated from hospitals carried fbsB, cfb, hylB, lmb, cylE, cpsA, bca, scpB, and fbsA virulence genes aggregated in antibiotic sensitivity-related gene–rich fragments (48). This study showed that MDR strains of S. agalactiae ubitquitously had hylB, iagA, and fbsA virulence genes and had hylB, iagA, and fbsA in 80% of cases. It also noted that 60% of glnA-positive strains were resistant to TE+ERY, and 60% of the resistance genes were ermA+ermB+lnuA+tetK. The TE and ERY resistance may be due to a virulence gene. This association was supported by our analysis of antibiotic-resistance and virulence genes, but this was not statistically significant according to the binary logistic-regression model. In Shenzhen, China, 89.5–100% of S. agalactiae isolates from humans were resistant to tetracycline, and these isolates revealed their drug resistance gene spectrum to be tetO+tetM and their main virulence gene spectrum to be hylB+lmb+scpB (44). The high resistance to macrolides, lincosamides and quinolones of GBS isolates was revealed in pregnant women in southern China, and all GBS isolates harboured the hylB and cylE genes as a common virulence gene profile (15). In China, the antimicrobial-resistance and virulence gene spectra of human S. agalactiae evidenced by previous research are inconsistent with our results, and the relationship between the antimicrobial resistance and virulence of S. agalactiae is not clear and requires further exploration. However, there are serious problems with antimicrobial resistance in China, which strongly calls for the country to strengthen the monitoring of drug resistance.

In conclusion, our study showed a high prevalence of multidrug-resistant S. agalactiae isolated from cattle with mastitis in China. These streptococci exhibited high resistance to oxacillin, tetracycline, and erythromycin. The dominant resistance and virulence genes were ermB and fbsA, respectively. Multiple drug resistance was frequently observed in S. agalactiae strains expressing iagA and fbsA. However, no statistically significant correlation was observed between drug-resistance and virulence gene spectra. This study highlights the need for antimicrobial management in clinical veterinary medicine to avoid the increase and dissemination of antimicrobial resistance arising from the use of antimicrobial drugs in animals.