B. abortus strain isolate A3313 was used in this study, which was isolated from the abortus of dairy cows in Hohhot District, Inner Mongolia, China. The A3313 strain was grown in Brucella broth medium (BD Biosciences) at 37°C with 5% CO₂.

All the experiments related to the cultivation of Brucella and its biofilms, as well as the operation of viable bacteria were conducted in a Biosafety Level 3 Laboratory in the College of Veterinary Medicine, Huazhong Agricultural University (Wuhan, China). For the experiments of electron microscope observation, 2-D electrophoresis, high-throughput sequencing and reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis, the Brucella and its biofilm were effectively inactivated with glutaraldehyde or bacterial lysate before being removed from the Biosafety Level 3 Laboratory.

Brucella broth was added to 6-well cell culture plates. A clean coverslip sterilized by autoclaving (121°C, 20 min) was then put in each well, and the A3313 bacterial suspension was inoculated on the coverslip at 2 ml/well. The culture plate was placed at 37°C with 5% CO₂, and the culture medium was changed every 48 h until a complete biofilm was formed. The coverslips were taken out, gently washed three times with phosphate-buffered saline (PBS; 30 mM, pH 7.4), and then fixed immediately with 2.5% glutaraldehyde for 6–8 h at 4°C. After being washed with PBS, biofilms were stained with 200 µl 1% crystal violet (Ding Bei Biological Technology Co., Ltd.) for 20 min at room temperature. These procedures were conducted to protect biofilms from falling off from the abiotic surfaces. The biofilms were observed under a phase-contrast light microscope (magnification, ×20) (Axiovert 135; Zeiss AG).

For scanning electron microscope observation, biofilms were fixed with 2% osmic acid at room temperature until black. After washing with 0.1 M PBS for three times, the samples underwent sequential dehydration with gradient ethanol solutions (30, 50, 70 and 90%) for 15 min each. Then, samples were dehydrated with 100% ethanol twice (15 min each), and dried with a critical point dryer. The dry samples were fixed on the sample stage with conducting resin, and sprayed gold with ion sputtering equipment (15 mA) for 2 min. The biofilms were observed under a scanning electron microscope.

Biofilms and planktonic bacteria were used for 2-D electrophoresis. For biofilm culture, the A3313 strain was grown in Brucella broth medium in Petri dishes at 37°C and 5% CO₂. The culture medium was changed every 48 h for 8 days. After removing the supernatant, the plates were rinsed twice with PBS. Biofilms were detached by scraping. Planktonic B. abortus was cultured in the same condition. The culture medium was collected and centrifugally washed twice with PBS. The planktonic B. abortus was resuspended with PBS.

Protein was precipitated as previously described (10,20). The biofilm and planktonic bacteria were harvested by centrifugation (6,000 × g) at 4°C for 10 min. The bacteria were washed four times with a solution containing 3 mM KCl, 68 mM NaCl, 9 mM NaH₂PO₄ and 115 mM KH₂PO₄, and then resuspended in lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 10 mM DTT and 2% Pharmalyte pH 3–10) with protease inhibitor. After ultrasonic decomposition on ice at 40% maximum power for 90 cycles (5 sec on, 10 sec off; the ultrasound conditions were the same for biofilm disruption and planktonic cells), unbroken cells and the cell debris were incubated at 25°C for 30 min, and removed by centrifugation (10,000 × g) for 30 min at 25°C. Proteins in the supernatant were precipitated in 10% trichloroacetic acid (TCA) on ice for 30 min. Precipitated proteins were washed with chilled acetone and then centrifuged at 10,000 × g for 10 min at 4°C. The air-dried proteins were dissolved in 400 µl sample preparation solution [2-D lysate: 9.5 M urea, 4% CHAPS, 2% (v/v) ampholytes, and 60 mM DTT; 10 µl 50 XProtein Inhibitor Cocktail Set I (Merck KGaA) was added into 500 µl 2-D lysate] at 25°C for 30 min and centrifuged at 10,000 × g for 20 min at 25°C. The protein concentration was determined via a Bradford protein assay.

Proteins (200 µg) were separated by 2-D electrophoresis. Isoelectric focusing for the first dimension was performed in precast Immobiline DryStrips (GE Healthcare Life Sciences) with a nonlinear gradient of pH 3 to 10 in an Ethan IPGphor Isoelectric Focusing System (GE Healthcare Life Sciences) according to the manufacturer's instructions. The electrophoresis conditions were 30 V for 12 h, 500 V for 1 h, 1,000 V for 1 h, 8,000 V for 8 h and 500 V for 4 h. The second dimension (SDS-PAGE) was conducted vertically in a Hofer SE 600 (GE Healthcare Life Sciences) using 12.5% polyacrylamide gels. The resolved proteins were then stained with silver for 30 min at room temperature and scanned with UMax Powerlook 2110Xl (GE Healthcare Life Sciences). All experiments were performed in triplicate. The gels were analyzed with the Image Master Platinum version 5.0 software (GE Healthcare Life Sciences). The normalized protein amount for each protein spot was calculated as the ratio of that spot volume to the total spot volume on the gel. Significant differences between two groups were determined using the Student's t-test, and a fold change ≥1.5 was considered the threshold value.

The differentially expressed protein spots were excised from the 2-D gels and then subjected to MALDI-TOF/TOF-MS analysis (Shanghai Applied Protein Technology Co. Ltd.). Before MS analysis, the protein spots were digested in-gel by 0.1 mg/ml trypsin for 2 h at 37°C and desalinated by Ziptip (EMD Millipore). The digested samples were then freeze-dried. After being re-dissolved, 1 µl samples were spotted on the target plate (Immobiline DryStrips; 13 cm, pH 3–10 NL) with air drying, and then 0.5 µl supersaturated α-cyano-4-hydroxycinnamic acid matrixes (Sigma-Aldrich; Merck KGaA) prepared in 50% acetonitrile with 0.1% trifluoroacetic acid were spotted and naturally dried. All MALDI-MS and MS/MS data were acquired in the positive reflectron ion mode on a 4800 Plus MALDI TOF/TOF Analyzer (AB Sciex LLC). Samples were irradiated by a 355-nm Nd:YAG laser (355 nm), and the acceleration voltage was 2 kV. The scanned area for MS was 800–4,000 Da, and the parent ion with signal:noise ratio >50 was selected for MS/MS analysis. Data from MALDI-MS and MS/MS were subjected to a Database Search from NCBI or Uniprot (NCBI preferentially) using Mascot version 2.2 software (Matrix Science), and a Mascot score was calculated. The MS/MS spectra were subjected to similarity searches using the BLASTX algorithm. The parameter settings were trypsin digestion, fixed modification of carbamidomethyl, dynamical modification of oxidation (M), unrestricted protein mass, peptide mass tolerance for monoisotopic data of ±100 ppm, fragment mass tolerance of ±0.4 Da, peptide charge state of 1+, and one maximum missed cleavage.

RNA sequencing of the biofilms and planktonic bacteria were performed at Genergy Biotechnology Co., Ltd. Total RNA was isolated and examined by NanoDrop spectrophotometer (NanoDrop Technologies; Thermo Fisher Scientific, Inc.) and 1% agarose gel electrophoresis (Tanon Science and Technology Co., Ltd.). An RNA library was then constructed and sequenced on the Illumina HiSeq 2500 platform (Illumina, Inc.). The data are available at National Center for Biotechnology Information under the accession numbers SRX1604658 (biofilm conditions; http://www.ncbi.nlm.nih.gov/sra/?term=SRX1604658) and SRX1604659 (planktonic conditions; http://www.ncbi.nlm.nih.gov/sra/SRX1604659/).

The raw reads were evaluated using RSeQC 2.3.2 (http://rseqc.sourceforge.net/) (21), and sequence alignment was conducted with TopHat 2.0.10 (http://ccb.jhu.edu/software/tophat/index.shtml) (22). The remaining reads were used for the following analyses.

The mRNA expression levels were detected based on Cufflinks 2.2.1 (http://cole-trapnell-lab.github.io/cufflinks/) software (23). Differentially expressed mRNAs were defined based on strict criteria [q value ≤0.05, and log2(fold change) ≥1] using the Cuffdiff program (23). Functional annotation of differentially expressed genes was carried out using various bioinformatics procedures, including Gene Ontology (GO) (24) and Kyoto Encyclopedia of Genes and Genomes (KEGG, Kolmogorov-Smirnov value <0.05) (25).

The mRNA levels of differential proteins identified through 2-D electrophoresis and function-associated genes identified by high-throughput sequencing and bioinformatics analyses were detected via RT-qPCR according a previously described method (26). In brief, total RNA was isolated from biofilms and planktonic cells (1×10⁶ cells) of the A3313 strain using a Takara MiniBEST Universal RNA Extraction kit (Takara Biotechnology Co., Ltd.). The RNA quality and concentration were determined by a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific, Inc.). The RNA was reverse-transcribed into cDNA using a PrimeScript RT reagent kit (Takara Biotechnology Co., Ltd.; 42°C for 15 min). The 16S rRNA housekeeping gene was amplified as the internal control. The specific primers are listed in Table SI. The SYBR Green PCR method was performed using an SYBR Premix Ex Taq kit (Takara Biotechnology Co., Ltd.). The qPCR reaction was carried out under the following conditions: 95°C for 30 sec, 40 cycles of 95°C for 3 sec and 60°C for 30 sec. Relative mRNA expression ratios of selected genes were calculated with the 2^−ΔΔCq method (27). The experiment was performed with three replications.

The levels of differentially expressed proteins identified through 2-D electrophoresis, described above, were measured by western blotting. Briefly, cells were lysed in RIPA buffer (Sigma-Aldrich; brand of Merck KGaA), and the protein concentration was measured using BCA Protein Assay Kit (Thermo Fisher Scientific, Inc.). Protein samples (20 µg) were separated using 12% SDS-PAGE and transferred onto a polyvinylidene fluoride membrane (GE Healthcare), and the membrane was blocked with 100 mM Tris, 150 mM NaCl, 0.05% Tween-20 (TBST), containing 5% dry milk powder for 2 h at room temperature. Then, the blocked membrane was incubated with sera from primary antibodies [hypothetical protein BAbS19_I16470 (1:1,000; cat. no. orb309412; Biorbyt Ltd.); chaperone protein DnaJ (1:1,000; cat. no. PA3-018; Invitrogen; Thermo Fisher Scientific, Inc.); elongation factor Tu (1:1,000; cat. no. ab210089; Abcam); Chaperonin Cpn60/TCP-1 (1:1,000; cat. no. 3094R-100; BioVision, Inc.); polyprenyl synthetase (1:1,000; cat. no. ab80647; Abcam); periplasmic binding protein (1:1,000; cat. no. M30934-1; Wuhan Boster Biological Technology, Ltd.); enolase (1:1,000; cat. no. sc-271384; Santa Cruz Biotechnology, Inc.); acetyl-CoA carboxylase, a subunit (1:1,000; cat. no. MAB6898; R&D Systems, Inc.); tryptophanyl-tRNA synthetase (1:1,000; cat. no. ab31536; Abcam); aspartate-semialdehyde dehydrogenase (1:1,000; cat. no. EM1708-10a; Jingke Huaxue; exosporium protein B (1:1,000; cat. no. ab92932; Abcam); enoyl-(acyl carrier protein) reductase (1:1,000; cat. no. abx109426; Abbexa Ltd.); Omp16 (1:1,000; cat. no. ab93127; Abcam)] for 2 h at room temperature and then incubated with horseradish peroxidase-labeled secondary antibodies (1:5,000; cat. no. 29139; Invitrogen; Thermo Fisher Scientific, Inc.) in blocking buffer for 1 h at room temperature. After washing with 0.05% Tween-20 (TBST), the membranes were incubated with DAB substrate (Tiangen Biotech Co., Ltd.) for 10 min at room temperature. Outer membrane protein 16 was used as a loading control. The western blot bands were visualized using the Millipore ECL Western Blotting Detection System (EMD Millipore).

All experiments were repeated three times except for high-throughput sequencing, and the results were presented as the mean ± standard deviation. Statistical analyses were performed using GraphPad 6.0 (GraphPad Software, Inc.). P-values were calculated using Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Biofilm formation was observed under a phase-contrast light microscope. A large number of bacteria adhered to the coverslips and formed a community (Fig. 1). The structure of biofilms on coverslips was observed under a scanning electron microscope. The biofilms formed flake or microcolony clusters on the coverslips. Colonies were wrapped together by the mucus they secreted, forming an uneven, dense membrane structure (Fig. 2).

A total of 1,930 protein spots were detected in all gels. The representative 2-D electrophoresis images of 20 differentially expressed protein spots (7 upregulated and 13 downregulated protein spots) between biofilms and planktonic cells are presented in Fig. 3. Most proteins were distributed in the range of isoelectric point 4–7. MALDI MS and MS/MS analysis identified 20 protein spots corresponding to 18 individual proteins, including 6 upregulated (including catalase, extracellular solute-binding protein and ubiquinol-cytochrome C reductase iron-sulfur subunit) and 12 downregulated ones (including elongation factor Tu, enolase, isocitrate dehydrogenase and Chaperonin Cpn60/TCP-1; Table I).

The mRNA and protein expression levels of the 18 identified proteins in 2-D electrophoresis were detected by RT-qPCR and western blot analyses, respectively. RT-qPCR analysis revealed that all of the 18 genes were downregulated in biofilms compared with planktonic cells, including elongation factor Tu (fold change = −100.0), enolase (fold change = −8.6), isocitrate dehydrogenase (fold change = −9.9), and Chaperonin Cpn60/TCP-1 (fold change = −12.5; Fig. 4A). The consistency rate with the 2-D electrophoresis results at the transcriptional level was 66.67% (12/18). Thus, the proteins of the 12 genes that had consistent expression trends in 2-D electrophoresis and RT-qPCR were further detected by western blot analysis. The results showed that 9 proteins were significantly downregulated and 3 were upregulated in biofilms (Fig. 5). Therefore, the 9 downregulated proteins were considered differentially expressed proteins between biofilms and planktonic cells of the A3313 strain. The fold changes of the 18 mRNAs are shown in Table II.

A total of 22,039,653 and 37,506,048 reads were generated from biofilms and planktonic cells, respectively. After pre-processing, 16,624,091 and 34,222,222 aligned reads were obtained from biofilms and planktonic cells respectively.

Based on the thresholds of q value ≤0.05 and log2(fold change) ≥1, 157 differentially expressed mRNAs were identified. These mRNA species were grouped in three categories defined by GO, including biological processes (including protein glycosylation, nitrogen compound metabolic process, and cell wall organization), cellular compartment (such as integral component of membrane, plasma membrane, and cytoplasm), and molecular function (including DNA binding, ATP binding, and oxidoreductase activity; Fig. 6A). In addition, these differentially expressed mRNAs were significantly involved in six pathways (annotated by KEGG), including RNA degradation, sulfur metabolism, butanoate metabolism, aminoacyl-tRNA biosynthesis, aminobenzoate degradation, and selenocompound metabolism (Fig. 6B).

RT-qPCR was performed to confirm 14 function and pathway-associated genes identified by bioinformatics analyses, including BAbS19_I10210 [log2(fold change) = 2.67], BAbS19_I13070 [log2(fold change) = 2.51], BAbS19_I02060 [log2(fold change) = 1.92], BAbS19_I03220 [log2(fold change) = 2.79], and the results demonstrated that all of the 14 genes were upregulated in biofilms, in accordance with the sequencing results (Fig. 4B). For instance, the fold changes for the genes above (BAbS19_I10210, BAbS19_I13070, BAbS19_I02060, and BAbS19_I03220) were 13.9, 1.4, 2.3 and 9.9, respectively (Table III). These genes were considered differentially expressed genes between biofilms and planktonic cells of A3313 strain.