Identification of S. agalactiae in dairy cows

Milk samples were collected from a total of 619 cows with subclinical mastitis from 33 dairy farms located in six provinces (Jiangsu, Anhui, Shandong, Zhegjiang, Jiangxi and Fujjian) and one municipality (Shanghai). S. agalactiae was detected in milk samples from 102 cows from 21 farms located at five provinces (except Fujian province) and one municipality. In total, 102 bovine isolates (one isolate per cow) of S. agalactiae were selected and analyzed (a complete list of all isolates is provided in Table S1).

Capsular genotype and alpha-like protein

All 102 bovine S. agalactiae isolates belonged to capsular genotype Ia and II (Table 1) according to the multiplex PCR assay. Overall, type Ia was the most prevalent accounting for approximately 89% (91/102) of isolates. Type II was represented by 11% (11/102) of isolates. A large proportion (93%; n = 95/102) of the bovine S. agalactiae isolates was non-typeable (NT) to any of the detected Alp genes according to the multiple PCR assay (Table 1). Only 5% (5/102) of Alp1 gene-positive strains and 2% (2/102) Alp4 gene-positive strains were characterized. The alpha-like protein genes associated with the capsular genotype were also identified (Table 1): alp1 and alp4 associated with genotype II, NT associated with genotype II and Ia (χ²=36.595, P < 0.001).

Pilus type

Each strain was detected in the three PCR assays for pilus typing. PI-1 and PI-2a genes were absent in all of the investigated bovine strains although the corresponding genes were detected in the positive controls. All 102 bovine S. agalactiae strains carried the PI-2b gene alone (Table 1).

MLST sequence types and PFGE analysis

All 102 isolates were characterized using MLST; one novel allele (sdhA allele 51, GenBank accession number is KF006268) and two new STs (ST 568 and ST570) were identified. ST568 is a single-locus variant (SLV, in which one allele differs from the ST) of ST103, ST570 is a double-locus variant (DLV, in which two alleles differ from the ST) of ST64 and ST67 according to eBURST analysis. ST568 was assigned based on the presence of a novel allele (sdhA allele 51) and ST570 was assigned based on new combinations of allele profiles. In general, all 102 bovine isolates were clustered in six STs: ST103, ST568, ST67, ST301, ST313 and ST570, and were grouped in three clonal complexes (CC): CC67, CC64 and CC103 by eBURST analysis (Figure 1). Figure 1 also shows the close correlation between CC67 and CC64, which are both subgroups of CC17. This was further supported by the relationship between ST570 and ST67 (ST570 was a DLV of ST67) as described previously. However, CC103 remained as a distinct group.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. eBURST diagram of the Streptococcus agalactiae population.

Each sequence type (ST) is represented as a dot. The dots positioned centrally in the cluster are primary founders (blue) or subgroup founders (yellow). Pink circles indicate STs detected from the strains in this study and are marked by solid line arrows. The 102 bovine isolates were clustered within sequence types ST103, ST568, ST67, ST301, ST313 and ST570 and grouped within three clonal complexes (CCs): CC103, CC67 and CC64. Both ST67 and ST64 were subgroups of CC17. ST568 was a single-locus variant (SLV) for ST103. ST301 and ST313 were SLVs and DLVs (two-locus variants), respectively of ST67. ST570 was a DLV of both ST64 and ST67, derived from ST64. The predicted founders that were not obtained in this study are marked by dashed arrows. For clarity, ST labels have been removed.

https://doi.org/10.1371/journal.pone.0067755.g001

The isolates were grouped in 27 PFGE types in this study, of which 16 common (represented by dashed rectangles) and 11 unique types were identified, with a similarity between 43.9% and 100% (Figure 2). Eight major PFGE types (7, 8, 14, 15, 16, 18, 23 and 25 shown in Figure 2) comprised close to 70% (71/102) of all the isolates. A total of eight more minor types, comprising only two or three isolates, corresponded to almost 20% (20/102) of all the isolates.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Dendrogram of PFGE profiles of 102 bovine Streptococcus agalactiae

The Dendrogram was constructed based on BioNumerics analysis of the S. agalactiae PFGE patterns and cluster analysis using the Dice coefficient and UPGMA of the digitalized PFGE patterns for the S. agalactiae strains. Clustering settings of 0.00% optimization and 1.5% band position tolerance were used. 568 * and 570* were identified as novel sequence types in this study.

https://doi.org/10.1371/journal.pone.0067755.g002

Table 1 shows the 11 capsular genotype II isolates that were distributed among four STs (ST67, ST301, ST313, ST570) and eight PFGE types. The 91 type Ia isolates belonged to three STs (ST67, ST103, ST568) and 21 different PFGE types. ST103 and novel sequence type ST568, both belonging to CC 103 (Figure 1), were the predominant sequence types. These types comprised genotype Ia isolates associated with NT Alp genes and represented 57% (58/102) and 31% (32/102) of the isolates, respectively (Table 1). These isolates were clustered into 20 types according to the PFGE analysis. ST67, ST313 (DLV of ST67) and ST301 (SLV of ST67), were grouped in CC67 (Figure 1). ST67 represented 5% (5/102) of the isolates assigned to capsular genotype II (4 isolates) and Ia (1 isolate) associated with NT Alp genes, and were further clustered in four PFGE types. ST313 and ST301 comprised a total of 5% (5/102) of the isolates, which were assigned to capsular genotype II carrying Alp1 genes and belonged to five PFGE types. ST570, belonging to CC64, represented only two isolates of serotype II carrying the Alp4 gene and clustered in PFGE type 9 (Table 1).

The isolates clustered in the same ST were clearly divided into several distinct PFGE types. For example, ST103 and 568, were divided into nine and eleven different PFGE types, respectively (Table 1 Figure 2). It was also were observed that the isolates belonging to different capsular genotypes were clustered in the same ST or PFGE type, while the same capsular genotypes were divided into different ST or PFGE types (Table 1 Figure 2).

Distribution of S. agalactiae between and within farms

The STs, molecular genotypes, Alp genes of S. agalactiae among the 21 S. agalactiae positive farms are presented in Figure 3. Seven different genotypes were found when combining these three types. The predominant type, ST568, a novel ST which is a SLV of ST103, combined with Ia and NT and was found exclusively among isolates from farms located in northern provinces (Jiangsu and Shandong) only. ST103, the predicted founder of ST568, combined with Ia and NT, and with few exceptions, was found exclusively among isolates in the farms located in the southern provinces (including Shanghai, Anhui and Zhejiang).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Typing of 102 Streptococcus agalactiae isolates present in 21 commercial dairy farms in Eastern China.

The percentage of Streptococcus agalactiae positive cases is shown above each bar. ST indicates the MLST sequence type; the corresponding capsular genotype/alpha-like protein type is shown in parentheses. NT, non-typeable. The PFGE types are represented by the Arabic numerals in each bar.

https://doi.org/10.1371/journal.pone.0067755.g003

On the basis of the combination of genotypes described, the isolates were characterized into 31 types when combined with the PFGE genotypes (Figure 3). Comparison of S. agalactiae isolates between herds revealed similar genotypes (10 types) distributed in different herds, and herd specific types (21 types) were also observed (Figure 3). Comparison of S. agalactiae isolates from the same herd showed S. agalactiae isolates displaying either different types (in 12 herds) or the same type (in 9 herds) within a same herd. Clustering of multiple isolates of the same type within an individual herd was also observed on several farms.

Ethics

Milk samples were obtained with consent from animals with subclinical mastitis under the ethical approval granted by College of Veterinary Medicine, the Nanjing Agricultural University Veterinary College. The protocol was permitted by the owners of the dairy farms under investigation. All efforts were made to minimize animal suffering.

S. agalactiae reference strains

Four reference strains of S. agalactiae: ATCC 13813 (serotype II), ATCC 12403 (Serotype III), ATCC BAA-611 (Serotype V) and A909 (Serotype Ia), used in this study were obtained from the American Type Culture Collection (ATCC) as controls.

Identification of S. agalactiae from milk samples

A total of 619 cows from 33 large-scale diary farms located in Eastern China were selected to participate in this study according to the willing of the farmers (information about the farms were provided in Table S2). Holsteins dairy cows were preferably kept on the 33 dairy farms with a mean yield at 305 d of 7,539 kg of milk. For each farm, milk somatic cell counts (SCC) of each cow were routinely carried out twice per month by electronic counting (Fossomatic 5000^TM, Foss Electric, Hillerød, Denmark). Subclinical mastitis is suspected when SCC ≥300,000 cells/ml, with decreased milk production and no inflammation of the udder. Milk samples were aseptically collected from dairy cattle with subclinical mastitis in these farms from January to December in 2011. In brief, each udder of the subclinical mastitis cow was washed and dried with a clean face towel. Each teat was disinfected with swabs soaked in 70% ethyl alcohol. The first few streams were discarded before milk samples were collected in 10-ml sterile plastic tubes and numbered. Each teat was disinfected with 70% ethyl alcohol after collection. The samples were taken to the laboratory as soon as possible, or stored at -20°C for later analysis.

For isolation of S. agalactiae, 20 µl of milk was streaked on 5% sheep blood agar plates and incubated at 37°C for 24 h. S. agalactiae were identified by conventional methods [27] and were further confirmed by PCR with 16S DNA species-specific primers as described by Martinez et al. [28]. A single isolate from each sampled cow was selected and stored at -70°C for further experimental use.

Bacterial culture and DNA extraction

The isolates, including the reference strains, were cultured in 3 ml of Tryptic soya broth (TSB, MO BIO, Laboratories. Inc.) overnight at 37°C. The bacterial culture was centrifuged (14,000 ×g for 5 min at room temperature) and the pellets were harvested and resuspended in 200 µl TE buffer (10 mM Tris-HCl, pH 8.0; 1 mM EDTA, pH 8.0) supplemented with 20 mg/ml (final concentration) of lysozyme and incubated at 37°C for 30 min. Genomic DNA was extracted using the Bacterial DNA extraction kit following manufacturer’s specifications for Gram-positive bacteria (Omega Bio-Tek, USA). The extracted DNA was used as the template for PCR.

Capsular genotyping

The capsular genotype Ia, Ib, II–IX of S. agalactiae was identified by a multiplex PCR assay as previously described [29]. The PCR system (total volume, 25 µl) contained 50 ng DNA template, 1× PCR buffer; 2 mmol/1 MgCl₂, 200 µmol/l dNTPs (dATP, dTTP dCTP, and dGTP), 400 nmol/l primers cpsI-Ia-6-7-F and cpsI-7-9-F, 250 nmol/l of each other primer and 0.3 U of HotMaster Taq DNA Polymerase (Tiangen, China). The PCR amplification conditions were as follows: preheating at 95°C for 5 min, followed by 15 cycles of 95°C for 1 min, 54°C for 1 min, and 72°C for 2 min and then by an additional 25 cycles of 95°C for 1 min, 56°C for 1 min, and 72°C for 2 min with a final cycle of 72°C for 10 min. Serotypes of strains were identified by analyzing the unique banding pattern following agarose gel (1.5% wt./vol.) electrophoresis.

Alpha-like protein (Alp) family

The alpha-like protein genes bca, alp1 (Epsilon), alp2/3, Rib, and alp4 in the strains were detected by a simple multiplex PCR assay described by Creti, et al. [17]. In brief, the PCR system (total volume, 25 µl) contained the following: 50 ng DNA template, 1× PCR buffer; 2 mmol/l MgCl₂, 200 µmol/l dNTPs, 400 nmol/l of each of the five pairs of primers; 0.3 U of HotMaster Taq DNA Polymerase (Tiangen, China). The amplification conditions were as follows: preheating at 96°C for 3 min, followed by 30 cycles of 95°C for 1 min, 58°C for 45 s, and 72°C for 45 s, with a final cycle of 72°C for 10 min. Amplification of the alpha-like protein genes was evaluated by agarose gel (2% wt./vol.) electrophoresis of the PCR products. Strains that tested negative for any of the Alp genes under investigation were considered non-typeable (NT).

Pilus typing

To identify the presence of pilus islands in the strains, three PCR assays were used to detect the PI-1, PI-2a or PI-2b genes as described previously [20]. In brief, the PCR system (total volume, 25 µl) contained 10 ng DNA template, 1× PCR buffer; 2 mmol/l MgCl₂, 200 µmol/l dNTPs, 400 nmol/l of each of the six primers; and 0.25 U of TaqDNA polymerase (Tiangen, China). The amplification conditions were as follows: preheating at 94°C for 5 min, followed by 35 cycles of 94°C for 45 s, 54°C for 45 s, and 72°C for 1 min and 30 s to 2 min (according to the lengths of the amplicons) and concluding with a cycle of 72°C for 10 min. The identity of the PCR products was analyzed by agarose gel (1% wt./vol.) electrophoresis and deemed to be positive based on the expected size of the PCR fragment.

MLST

All the isolates were typed using multilocus sequence typing (MLST) in this study. The seven housekeeping genes (adh, pheS, atr, glnA, sdhA, glcK and tkt) were amplified by PCR and internal fragments sequences were obtained as described previously [7]. For each isolate, the allele number and sequence types (STs) were defined by analysis of the alleles sequence in the MLST database (http://pubmlst.org/sagalactiae/). The allele sequences or previously undescribed ST were assigned new numbers and the data were deposited in the MLST database. CC analysis was performed using the entire S. agalactiae MLST database and eBURST program (http://eburst.mlst.net) [30].

PFGE

DNA was extracted and digested with the SmaI restriction enzyme (Takara, China.) as previously described [31,32]. The PFGE program was performed according to Chen et al. [32]. A Salmonella serotype Braenderup H9812 DNA digested with XbaI was used as the molecular size standard as recommended in PulseNet [33]. The S. agalactiae PFGE patterns were analyzed with BioNumerics (Applied Maths BVBA, Belgium) using an optimization setting of 0.00% and band position tolerance of 1.5%. Cluster analysis was performed using the Dice coefficient and UPGMA of the digitalized PFGE patterns for the 102 S. agalactiae strains. Genotypic related groups, characterized at 80% similarity or above are represented by dashed rectangles in the dendrogram.

Statistical analysis

Categorical data were analyzed by using the Pearson chi-squared test. When data were insufficient for the test demands, Fisher’s exact test and Likelihood-ratio tests were used. P values of <0.05 were considered to indicate statistical significance.