The establishment and application of a one‐step multiplex real‐time polymerase chain reaction assay for the detection of Streptococcus suis, Streptococcus suis serotype 2, and Glaesserella parasuis

Abstract

This article aims to establish a multiplex real‐time polymerase chain reaction (PCR) assay for the simultaneous detection of Streptococcus suis (SS), Streptococcus suis serotype 2 (SS2), and Glaesserella parasuis (GPS). In this study, three pairs of primers and three probes were designed based on the specific sequences of SS (gdh), SS2 (cps2j), and GPS (infB). The results showed that the assay was not cross‐reacted with other swine pathogens (Escherichia coli, Pasteurella multocida, Staphylococcus aureus, Streptococcus agalactiae, Streptococcus pneumoniae, Actinobacillus pleuropneumoniae, Mycoplasma hyopneumoniae, and Enterococcus faecalis; Streptococcus pyogenes). 10⁸ to 10² copies/μL showed the R² values for SS, SS2, and GPS were 0.999, 0.992, and 0.990, respectively. The multiplex real‐time PCR efficiency was 93.816% for gdh, 105.260% for cps2j, and 93.175% for infB. The sensitivity result showed that SS, SS2, and GPS could be detected at 10 copies/μL. The repeatability result showed that intra‐assay and inter‐assay coefficients of variation of SS, SS2, and GPS were <2%. The best cutoff values for SS, SS2, and GPS were determined from ROC curves to be 35.085, 35.620, and 34.940, respectively. Areas under the curve were 0.943, 0.968, and 0.958. In total, 88 clinical samples were analyzed. The results indicated positive rates of 11.364% (10/88) for SS, 20.455% (18/88) for SS2, and 18.182% (16/88) for GPS. In conclusion, the developed one‐step multiplex real‐time PCR assay may be a valuable tool for the early detection of the SS, SS2 and, GPS with high specificity and sensitivity.

INTRODUCTION

Streptococcus suis (S. suis, SS) and Glaesserella parasuis (G. parasuis, GPS) are most common pathogens in the world of infected pigs widely; both could cause inflammatory infections such as septicemia, meningitis, polyserositis, arthritis, and pneumonia []. The number of incidences of porcine streptococcal disease and GPS disease ranked top 10 swine diseases consistently in China from 2018 to 2021 []. SS and GPS, two vital porcine respiratory resident and pathogenic bacteria, cause clinical symptoms and pathological changes by mixed infection, increasing the complexity of the disease and the difficulty of diagnosis []. There are 35 serotypes of SS depending on different capsular antigens [], of which Streptococcus suis serotype 2 (SS2) infections are classified as zoonotic infections and are the most predominant causative organisms in both swine and human populations []. It has the highest pathogenicity and prevalence and are primary targets for outbreak surveillance and pathogen identification []. The lethality of SS2 infection in humans is about 12% worldwide while as high as 18% in China [].

At present, bacterial isolation and culture with serological typing is still the gold standard for diagnosis [], but the method is time-consuming, with low sensitivity and high operational requirements, which is not conducive to promotion and outbreak monitoring, and the serological method can not distinguish between podocarpellate-free strains and self-coagulating strains []. Single or multiple polymerase chain reaction (PCR) [], multilocus sequence typing [], single real-time quantitative PCR detecting system [], and other methods have been established according to common genes of pathogenic bacteria or different serotype-specific podoconiosis antigen genes, which have realized the identification and typing of SS and GPS at the gene level, compensated for deficiencies of serological typing, and improved the accuracy and sensitivity of detection []. Jiang et al. [] established a real-time recombinant enzyme amplification assay for the detection of SS2 against the cps2j gene. In addition, GPS was detected using a Loop-Mediated Isothermal Amplification (LAMP) Assay [] and recombinase polymerase amplification and lateral flow strip []. However, these methods can not detect more than one test at a time, which increases the workload and cost of epidemic surveillance, whereas the multiplex fluorescence quantitative PCR assay not only can detect more than one test at a time but also has advantages of high sensitivity and high specificity [].

Three genes were used to design the primers and probes as specific targets to establish a multiplex real-time PCR assay to detect SS, SS2, and GPS simultaneously. The first one is the gdh gene of SS; it is highly conserved and can be used as target genes for genus identification []. The second one is cps2j, which only exists in serotype 2 and 1/2 strains [] among the genes in the capsular polysaccharide biosynthesis locus. The third one is the infB gene, which can serve as a target gene for the identification of GPS []. In this study, a multiplex real-time PCR assay for the simultaneous detection of SS, SS2, and GPS has been established. The sensitivity, specificity, and reproducibility of the method were evaluated, and the method was carried out for clinical application.

MATERIALS AND METHODS

Bacterial strains and DNA extraction

SS was grown at 37°C in brain–heart infusion broth supplemented with 0.5% yeast extract (Difco Laboratories). GPS was grown at 37°C in tryptic soy broth supplemented with 5% newborn calf serum and 0.01% nicotinamide adenine dinucleotide (Difco Laboratories). Other bacterial strains were grown in appropriate media under recommended conditions. The used bacterial strains are displayed in Table . The total DNA of bacterial strains was extracted using Genomic DNA TIANamp Bacteria DNA Kit (TIANGEN). Finally, the extracted DNA was stored at −80°C for further use.

TABLE 1 Bacteria strains used in this study and determination of results.

Bacterial strains	Source	FAM	VIC	TAMRA
Streptococcus suis serotype 1	CVCC608	+	-	-
S. suis serotype 2	CVCC3308	+	+	-
S. suis serotype 2	CVCC3312	+	+	-
S. suis serotype 2	CVCC3307	+	+	-
S. suis serotype 2	CVCC3311	+	+	-
S. suis serotype 2	CVCC3313	+	+	-
S. suis serotype 2	CVCC3309	+	+	-
S. suis serotype 2	CVCC3930	+	+	-
S. suis serotype 2	CVCC3928	+	+	-
S. suis serotype 2	CVCC3929	+	+	-
S. suis serotype 3	CVCC1933	+	-	-
S. suis serotype 4	CVCC1937	+	-	-
S. suis serotype 7	CVCC603	+	-	-
S. suis serotype 9	CVCC1939	+	-	-
S. suis serotype 14	CVCC595	+	-	-
S. suis serotype 16	CVCC1934	+	-	-
Glaesserella parasuis serotype 4	CVCC3962	-	-	+
G. parasuis serotype 5	CVCC4121	-	-	+
Escherichia coli	CVCC192	-	-	-
Pasteurella multocida	CVCC398	-	-	-
Staphylococcus aureus	CVCC546	-	-	-
Streptococcus agalactiae	CVCC586	-	-	-
Streptococcus pneumoniae	CVCC1929	-	-	-
Actinobacillus pleuropneumoniae	CVCC259	-	-	-
Enterococcus faecalis	CVCC1927	-	-	-
Streptococcus pyogenes	CVCC593	-	-	-
Mycoplasma hyopneumoniae	CVCC354	-	-	-

Primer and probe design

According to the specific regions of SS (gdh, GeneBank: AY853916.1), SS2 (cps2j, GeneBank: AM946016.1), and GPS (infB, GeneBank: DQ410886.1), the multiple pairs of primers and probes were prepared using the Beacon Designer 8 and verified for specificity by Primer-BLAST (Table ). Primers and probes were synthesized by Beijing Liuhe Huada Gene Technology Co.

TABLE 2 Primers and probes of SS, SS2, and GPS.

Pathogens	Gene	Primers and probes	Sequence (5′ end to 3′ end)	Length (bp)	Accession no.
SS	gdh	Gdh-F	GATGGATTGGTTTACAGTTG	117	AY853916.1
Gdh-R	TGGACAGATAAAGATGGAAA
Gdh-probe	FAM-AAGTCAACCGTGGCTACCGT-BHQ1
SS2	cps2j	cps2j-F	CGGAAAAGTTTTATTCTACC	113	AM946016.1
cps2j-R	TCCCAATCGTATACTAATCTA
cps2j-probe	VIC-ACGGTTCTTCAGATTCATCAACGGATA-BHQ2
GPS	infB	GPS-F	CTGCTTTGATTTCGTTTAATAG	115	DQ410886.1
GPS-R	GTAGGTGGTATTACGGAAA
GPS-probe	TAMRA- AACCTTAGCGGCAGCGTCTA-BHQ2

Construction of standard plasmids

The genome DNA of SS, SS2, and GPS were used as the template and primers in Table  were used to amplify the target fragment, and then, the amplicons were purified and cloned into the pET-32a vector (Sangon Biotech) with ClonExpress Ultra One Step Cloning Kit (Vazyme Biotech). The constructed plasmids were subjected to positive transformation into a Fast-T1 competent cell (Vazyme Biotech). After being cultured for 16–20 h at 37°C, the plasmids were extracted utilizing the EndoFree Mini Plasmid Kit II (TIANGEN) and quantified using a NanoPhotometer® (Thermo Fisher). The plasmids were named pET-SS, pET-SS2, and pET-GPS and then stored at −80°C until their use as standard plasmids. Copy numbers of plasmids were calculated with the following formula: $$\text{Pl}\text{asmid}\,\text{copies}/\mu \mathrm{L}=\left(6.02\times {10}^{23}\right)\times \left({\mathrm{X}}^{\ast }\text{ng}/\mu \mathrm{L}\times {10}^{-9}\right)\,/\text{constructed}\,\text{plasmids}\,\text{length}(\text{bp})\times 660$$ where *X: recombinant plasmid concentration [].

TABLE 3 Primers used for standard plasmid construction.

Pathogens	Gene	Primers	Sequence (5′ end to 3′ end)	Length (bp)	Accession no.
SS	gdh	F	acggagctcgaattcggatccATGTCAAATGCCAAAGCTTACATC	1347	AY853916.1
R	gacaaggccatggctgatatcTTATACCAAACCTTGGGCAATCA
SS2	cps2j	F	acggagctcgaattcggatccATGGAAAAAGTCAGCATTATTGTACC	999	AM946016.1
R	gacaaggccatggctgatatcTTAATCATTATTTTTTTCTTCCCTAATTAA
GPS	infB	F	gccatggctgatatcggatccCTCATTATTAGACTATATCCGTAAAGCGA	1360	DQ410886.1
R	ctcgagtgcggccgcaagcttGTTACGGACTTCTGAAACGTCGT

Optimization of the multiplex real-time PCR assay

The single control variable method (SCVM) was adopted to optimize multiplex real-time PCR assay parameters, such as the annealing temperature, the primer concentration, and the probe. The total volume of the multiplex real-time PCR reaction is 20 μL, including 2× Animal Detection U⁺ Probe Qpcr Super Premix (Vazyme Biotech); final concentrations of primers and probes were 0.1–0.5 μmol/μL, 2 μL of positive plasmid template, 0.4 μL 50 × ROX Reference Dye 2, and distilled water to a total volume of 20 μL. All reactions were amplified by an ABI 7500 fast Real-Time PCR System (Applied Biosystems), and the amplification parameters were 37°C for 2 min, 95°C for 30 s; and then 40 cycles of 95°C for 15 s and annealing temperature and extension temperature (56°C, 57°C, 58°C, 59°C, 60°C, 61°C, and 62°C) for 30 s.

Standard curve generation

The three standard plasmids were mixed in equal volumes; 10-fold-diluted standard plasmids from 1 × 10⁸ to 1 × 10² copies/μL were selected for standard curve analysis. At the end of the reaction, the amplification curve and the standard curve were generated directly from quantitative fluorescence software. The correlation coefficient (R²), amplification efficiency (E), and standard equation were calculated from the amplification curve.

Specificity of the multiplex real-time PCR assay

The DNA of all bacteria in Table  were deemed as templates for the established multiplex real-time PCR to verify the specificity of the assay. In addition, the three standard plasmids (pET-SS, pET-SS2, and pET-GPS) and distilled water were considered as the positive and negative controls, respectively.

Sensitivity of the multiplex real-time PCR assay

To evaluate the sensitivity of the method, the standard plasmids of pET-SS, pET-SS2, and pET-GPS were mixed, and then, the 10-time serial diluent from 1 × 10⁸ to 1 × 10⁰ copies/μL was used as a template under optimized conditions.

Repeatability of the multiplex real-time PCR assay

Three positive plasmids with different gradients (1 × 10⁷, 1 × 10⁵, and 1 × 10³ copies/μL) were equivalently mixed as templates and then detected by the established multiplex PCR. All intra- and inter-assay runs were performed in triplicate with a 2-week interval. The intra- and inter-assay coefficient of variation (CVs) were used as an index to evaluate the repeatability of the assay.

Determination of detection thresholds

The standard plasmids of pET-SS, pET-SS2, and pET-GPS were mixed and diluted to 10¹, 10², and 10³ copies/μL and then mixed with the total DNA of true-negative samples in equal volumes of 1:1 (V/V) to serve as true-positive samples. Thirty true-negative samples and thirty true-positive samples were tested by the multiplex real-time PCR assay. The diagnostic sensitivity (Se) and specificity (Sp) of the developed the multiplex real-time PCR assay were measured by receiver operating characteristic (ROC) analysis using SPSS statistical software. The area under the curve (AUC) was obtained, and the Youden index (Youden index = Se + Sp − 1) was calculated. The value corresponding to the larger value of the Youden index was used as the detection threshold.

Detection of clinical samples by the multiplex real-time PCR assay

Approximately 88 (nasal swabs, lungs, arthritis, tonsils, etc.) samples were collected from the pig with sign of respiratory diseases and arthritis in Henan, China, from July 2022 to December 2022. Nasal swabs collected were preserved using Sample Preservation Buffer (TIANGEN), and then, DNA was extracted according to the Hi-Swab DNA Kit (TIANGEN); Tissue samples were weighed 0.5 g; 500 μL PBS was added, ground thoroughly, and DNA was extracted using the TIANamp Genomic DNA Kit (TIANGEN) and subsequently tested for the presence of SS, SS2, and GPS using the established multiplex real-time PCR assay. In addition, using a singlex real-time PCR method, which has been reported to detect the above DNA (the singlex real-time PCR method for the detection of SS []; Method of the real-time PCR for the detection of SS type 2 (GB/T 19915.7–2005) []; Detection Methods for Haemophilus parasuis (GB/T 34750-2017) []).

RESULTS

Optimal reaction conditions for the multiplex real-time PCR assay

The optimal primer concentration, probe concentration, and annealing temperature were obtained by the SCVM, and the multiplex real-time PCR assay was established (Figure ). The developed assay utilized a 20 μL reaction mixture, consisting of 10 μL of 2× Animal Detection U + Probe qPCR Super Premix (Vazyme Biotech) and 2.0 μL of total DNA (DNA of three pathogens), together with the optimum primer and probe concentration for three pathogens shown in Table , 0.4 μL 50× ROX Reference Dye 2, and distilled water to a total volume of 20 μL. The amplification parameters were 37°C for 2 min, 95°C for 30 s, and then 40 cycles of 95°C for 10 s and 60°C (annealing temperature and extension temperature) for 30 s. In addition, if only the FAM channel has a signal, the test result is SS; if the FAM and VIC channels have only one signal and the TAMRA channel has no signal, the test result is SS2; if only the TAMRA channel has a signal, the test result is GPS.

[IMAGE OMITTED. SEE PDF]

TABLE 4 Optimum primer and probe concentrations.

Pathogens	Final concentration (μmol/μL)
Forward and reverse primer	Probe
SS	0.15	0.20
SS2	0.20	0.25
GPS	0.30	0.30

Construction of the standard curve

Three standard plasmids with dilution from 1 × 10⁸ to 1 × 10² copies/μL were mixed in equal volume and then amplified under the optimized multiplex real-time PCR condition. The corresponding slope of the equation and correlation coefficient (R²) amplification efficiency (E) were −3.480%, 0.999%, and 93.816% for SS, respectively, −3.200%, 0.992%, and 105.260% for SS2, respectively, and −3.497%, 0.990%, and 93.175% for GPS, respectively (Figure ). Overall, an excellent linear relationship (R² ≥ 0.990) was confirmed between initial template concentrations and the corresponding threshold cycle (Ct) values, and amplification efficiencies were all in the range of 90%–110%.

[IMAGE OMITTED. SEE PDF]

Specificity of the multiplex real-time PCR assay

As shown in Figure , SS, SS2, and GPS had specific amplification curves, while other pathogens and the negative control had no amplification or fluorescent signal. The multiplex real-time PCR assay was shown to be specific for SS, SS2, and GPS. As shown in Table , if only the FAM channel has a signal, the test result is SS; if FAM and VIC channels have signals and the TAMRA channel has no signal, the test result is SS2; if only the TAMRA channel has a signal, the test result is GPS.

[IMAGE OMITTED. SEE PDF]

Sensitivity and repeatability of the multiplex real-time PCR assay

The sensitivity studies showed that the detectability of the multiplex real-time PCR assay for SS, SS2, and GPS was 10 copies/μL (Figure ). As shown in Table , the CV was <2%, indicating that the assay had good repeatability.

[IMAGE OMITTED. SEE PDF]

TABLE 5 Repeatability of the multiplex real-time PCR assay.

Standard plasmid	Concentration of template (copies/μL)	Intra-coefficient of variation	Inter-coefficient of variation
X ± SD	CV (%)	X ± SD	CV (%)
pET-SS	107	18.964 ± 0.038	0.200	18.951 ± 0.061	0.322
105	26.141 ± 0.194	0.742	26.241 ± 0.090	0.343
103	32.747 ± 0.134	0.409	33.081 ± 0.514	1.554
pET-SS2	107	20.320 ± 0.170	0.837	20.420 ± 0.089	0. 436
105	27.301 ± 0.365	1.337	27.201 ± 0.260	0.956
103	33.047 ± 0.422	1.277	32.713 ± 0.498	1.548
pET-GPS	107	17.356 ± 0.233	1.342	17.523 ± 0.066	0.377
105	24.431 ± 0.036	0.147	24.308 ± 0.249	1.024
103	31.382 ± 0.170	0.542	31.516 ± 0.072	0.228

Determination of critical values

Based on the results of the plotted ROC curves (Figure ), the CT critical values of this method for SS, SS2, and GPS were 35.085 (Sp: 0.800, Se: 0.920, Youden index: 0.720, and AUC = 0.943), 35.620 (Sp: 0.880, Se: 0.960, Youden index: 0.840, and AUC = 0.968), and 34.940 (Sp: 0.880, Se: 0.920, Youden index: 0.800, and AUC = 0.958), respectively. AUC values were all greater than 0.800, indicating good accuracy of the assay.

[IMAGE OMITTED. SEE PDF]

Clinical sample testing

Among the total 88 clinical samples (Table ), positivity rates for SS, SS2, and GPS were 11.364%, 20.455%, and 18.182%, respectively. In addition, 4.545% of the samples showcased a combination of both SS and GPS and 7.955% of the samples, a combination of SS2 and GPS. A high rate of compliance was observed with the singlex real-time PCR assay that was established in this study. Thus, the establishment of the multiplex real-time PCR assay with higher accuracy is essential for the timely prevention and control of the spread of the disease.

TABLE 6 Test results of clinical samples.

Pathogens	The multiplex real-time PCR	Reference methods	Agreement (%)
Positive	Positive
SS	11.364% (10/88)	11.364% (10/88)	100
SS2	20.455% (18/88)	19.318% (17/88)	94.444
GPS	18.182% (16/88)	18.182% (16/88)	100
SS + GPS	4.545% (4/88)	4.545% (4/88)	100
SS2 + GPS	7.955% (7/88)	7.955% (7/88)	100

DISCUSSION

In this study, a multiplex real-time PCR assay was established based on SS gdh gene, SS2 cps2j gene, and GPS infB gene. The method showed a good linear relationship between standard curves (R² > 0.99) and a high amplification efficiency (90%–110%); This method showed high specificity, sensitivity, and reliability, with no cross-reactivity with other pathogens infecting pigs; the lowest lower limit of detection was 10 copies/μL; coefficients of variation for both inter- and intra-batch testing were less than 2%. Moreover, infections with SS, SS2, and GPS were detected efficiently and with high concordance with reported singlex real-time PCR assays, facilitating early clinical diagnosis and rapid molecular epidemiological investigations.

SS is divided into 35 serotypes based on capsular polysaccharides []. It is a conditionally pathogenic organism []. SS2 infections are considered zoonotic infections []. According to reports, between 2005 and 2021, 75.76% of SS isolates in humans from Shenzhen, China, were identified as SS2, while 24.24% comprised of SS14 []. Currently, due to the continuous development of pig farm intensification, there is a prevalence of GPS-caused diseases in many regions of China, posing a serious threat to the pig farming industry. In recent years, there has been an increase in morbidity and mortality caused by mixed or secondary infections of SS and GPS []. In a study conducted by Rao et al. in Guangxi Province, China, it was reported that SS accounted for 65.21% of pig respiratory pathogens, GPS accounted for 48.19% of pig respiratory pathogens, whereas mixed infections of both were found to account for 13.10% of all cases []. Infections of SS and GPS pose a significant threat to the safety of food and livestock production, as well as the related workers in China [].

Gene amplification is considered to be one of the most dependable methods for identifying bacteria with advancements in molecular biology techniques []. Studies have demonstrated that the molecular biology method detects pathogenic bacteria at a higher rate than microscopy []. A LAMP method was developed by Meng et al. [] using the cpsK genes of SS2 and SS14 as target genes. The assay could detect sample concentrations as low as 18.4 CFU []. LAMP methods based on the infB gene of GPS was developed by Pilchová et al. []. This method had a minimum detection limit of 36 fg/uL and was specific []. However, these methods were unable to simultaneous differentiate between SS, SS2, and GPS.

The selection of the target gene is a crucial factor that affects the sensitivity and specificity of fluorescence quantitative PCR. The gdh gene of SS encodes an essential virulence factor, glutamate dehydrogenase []. Besides, the nucleotide sequence of this gene is highly conserved between serotypes, demonstrating similarity ranging from 96% to 100% []. The cps2j gene encodes the capsular polysaccharide of SS2 and is highly specific, as it is unique to this particular strain of bacteria []. Although detection methods for GPS often rely on the 16S rRNA gene as the target [], numerous studies are shown that target the 16S rRNA gene and lacks specificity when differentiating GPS from closely related species, for instance, Actinobacillus indolicus []. In this situation, infB genes could be a better choice rather than 16S rRNA. Hedegaard et al. founded the usefulness of infB genes as genetic markers for phylogenetic studies and distinguishing species from GPS []. Turni et al. also confirmed that the infB gene as a target for the real-time PCR assay in distinguishing GPS from closely related species [].

Determining the detection threshold is one of the factors that affect test accuracy. In most studies, the default threshold CT value is 40, or the CT value corresponding to the lower limit of detection is chosen. However, this approach is statistically insufficient and susceptible to multiple experimental factors, which can result in high or low thresholds. The ROC curve, also known as the subject operating characteristic curve, is a combined measure of continuous variables that represents the sensitivity and specificity of an assay []. The ROC curve is presently utilized in clinical practice to determine the ideal threshold for various diagnostic techniques []. The critical CT value of a reverse transcription real-time quantitative PCR method for GETV was determined by Cao et al. based on ROC curve analysis []. This finding serves as a reference for using the ROC curve to determine the critical value of the real-time PCR assay. This study analyzed the results of a multiplexed real-time PCR assay on true positive and true negative samples using ROC curves. The assay's optimal critical CT values were determined to be 35.085, 35.620, and 34.940, respectively, with AUCs greater than 0.9. This suggests that the assay has high accuracy.

CONCLUSION

The study established a one-step multiplex real-time PCR assay for simultaneous identification of SS, SS2, and GPS successfully. The assay has strong specificity, high sensitivity, and good reproducibility, which can support clinical detection, epidemic prevention and monitoring, epidemiological investigation, and vaccine evaluation.

AUTHOR CONTRIBUTIONS

Liangquan Zhu and Youlong Lu conceived and designed the experiments. Lingxiang Xin, Haojie Wang, and Yunhao Hu performed the experiments, analyzed the data, and prepared figures and tables. Yan Liu and Wensheng Yao were responsible for project management. Xiuli Wang, Jian Li, and Yuanjie Liu provided clinical samples. Rendong Tong and Qi Wang were responsible for editing manuscripts. All authors reviewed drafts of the paper and approved the final draft.

ACKNOWLEDGEMENTS

This work was supported by the National Key Research and Development Program of China (2022YFD1800903) and the Public Welfare Special Project of the China Institute of Veterinary Drug Control (GY202103).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT

The animal study was reviewed and approved by Animal Ethics Committee of the China Institute of Veterinary Drug Control.

References

Gao, M., Zuo, J., Shen, Y., Yuan, S., Gao, S., Wang, Y., & Yi, L. (2023). Modeling co‐infection by Streptococcus suis and Haemophilus parasuis reveals influences on biofilm formation and host response. Animals: An Open Access Journal from MDPI, 13(9), [eLocator: 1511]. [DOI: https://dx.doi.org/10.3390/ani13091511]