Table I lists the strains and constructed plasmids used in this study. Brucella abortus 2308 was obtained from the Chinese Center of Disease Prevention and Control (CDC; Beijing, China). B. melitensis 027 strain was isolated from Xinjiang, China, and was identified by the CDC. Brucella strains were cultured in tryptic soy agar (TSA) or tryptic soy broth (TSB; Sigma-Aldrich, St. Louis, MO, USA). Plates were incubated at a temperature of 37°C in an atmosphere enriched with 5% CO₂. Escherichia coli strain JM109 (Promega Corporation, Madison, WI, USA) was grown in Luria-Bertani (LB) media. The culture media were supplemented with 50 µg/ml ampicillin (Invitrogen; Thermo Fisher Scientific, Inc., Carlsbad, CA, USA). The plasmid pMD18-T Simple Vector was purchased from Takara Bio, Inc. (Otsu, Japan). The standard curves were constructed using pMD18-T Simple Vector.

Murine macrophages (RAW 264.7) and human trophoblasts (HPT-8) were used in this study. HPT-8 cells and RAW 264.7 murine macrophage were purchased from the Cell Resource Center, IBMS, CAMS/PUMC (Beijing, China).

Deletion of the ery operon in B. abortus 2308 and B. melitensis 027 (2308Δery and 027Δery) was performed as previously described (13).

To monitor the growth of B. abortus 2308, B. melitensis 027, 2308Δery and 027Δery, cells were cultured in TSB to an optical density at 600 nm (OD600) of 0.6, then diluted with TSB to an OD600 of 0.05 and cultured in rotary shaker (100.62 × g) at 37°C for 48 h. Aliquots of cultures were collected at 4-h intervals, and bacterial growth was measured at OD600.

To detect erythritol sensitivity in 2308Δery and 027Δery, stationary phase pre-cultures of B. abortus 2308, B. melitensis 027, 2308Δery and 027Δery were diluted in TSB containing 20 mM erythritol (Sigma-Aldrich), and grown for 48 h. The bacterial growth was measured at an OD600.

RAW 264.7 murine macrophages were used to assess the intracellular survival of B. abortus 2308, B. melitensis 027, 2308Δery and 027Δery. RAW 264.7 cells were infected as previously described (14). Briefly, 5×10⁵ cells/well were cultured in 24-well plates for 16 h at 37°C and infected with Brucella at a multiplicity of infection (MOI) of 100. Culture plates were centrifuged for 5 min at 350 × g at room temperature and placed in an incubator at 37°C with 5% CO₂ atmosphere. At 45 min post-infection, the cells were washed twice with media and then incubated with Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc., Rockville, MD, USA) containing 50 µg/ml gentamicin (Invitrogen; Thermo Fisher Scientific, Inc.) for 1 h to kill extracellular bacteria. The media was then replaced with DMEM containing 25 µg/ml gentamicin (incubation point 0 min). At 4, 12, 24 and 48 h post-infection, the number of colony-forming units (CFU) was obtained by plating serial dilutions of the lysates on TSA plates. All assays were performed in triplicate and repeated at least three times.

HPT-8 cells were infected by B. abortus 2308 and B. melitensis 027 in media with or without erythritol (20 mM). HPT-8 cells were grown at 37°C in a 5% CO₂ atmosphere in DMEM containing 20% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.). Cells were seeded (1×10⁶) in 12-well culture dishes 24 h prior to each infection assay. HPT-8 cells were infected at a MOI of 100 bacteria per cell as previously described (14,15). Culture plates were centrifuged for 5 min at 350 × g at room temperature. Post-infection, cells were grown in the presence of erythritol at a concentration of 1% as a nutritional supplement or at 20 mM for induction of the ery operon and placed in an incubator at 37°C with 5% CO₂ atmosphere. Cells were washed three times with phosphate-buffered saline (PBS) and monolayers of cells were further incubated with culture media supplemented with 50 µg/ml gentamicin for 1 h to kill extracellular bacteria. The cells were washed with DMEM containing 10% fetal bovine serum to remove gentamicin, then the cells were lysed with TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.).

Total RNA (1 µg) from HPT-8 cells at 0 min (bacterial culture), 20 min, 1 h, 2 h, 3 h, 4 h and 12 h post-infection was isolated (Qiagen RNeasy Mini-kits; Qiagen, Hilden, Germany) and cDNA was generated using random hexamer primers and MMLV-RT according to the manufacturer's recommendations (Gibco; Thermo Fisher Scientific, Inc.). Genomic DNA was removed using a DNase RT kit (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. The 10 µl reaction mixture system containing 1 µg RNA, 1U DNase, 1 µl DNase buffer and ddH2O was added to 10 µl. The mixture system was mixed and allowed to rest for 10 min at room temperature. A total of 1 µl 25 mM EDTA was added and incubated for 10 min at 65°C. The DNA polymerase was obtained from Invitrogen (Thermo Fisher Scientific, Inc.). cDNA was stored at −80°C and used as a template for reverse transcription-quantitative polymerase chain reaction (RT-qPCR).

TaqMan primers for 16S rRNA (housekeeping gene), eryA, eryB, eryC and eryD genes were designed using Primer Express 5.0 software (Applied Biosystems, Palo Alto, CA, USA) according to sequences in GenBank (http://www.ncbi.nlm.nih.gov/genbank/) (Table II).

Ery operon (eryA, eryB, eryC and eryD) and 16S rRNA open reading frames were amplified by PCR with specific primers (see Table II; Premier Biosoft, Palo Alto, CA, USA) from the B. abortus 2308 genome. The amplified DNA fragments and pMD18-T simple vectors were ligated overnight at 16°C using T4 DNA ligase (Takara Bio, Inc.). The ligation reaction was transformed into E. coli JM109, and insert-containing plasmids were identified by restriction analysis or PCR. Positive recombinant plasmids were sequenced to confirm the correct construction.

The concentration and purity of recombinant plasmids (dilution, ×100) were measured using a Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific, Inc.) and used to compute target gene copy numbers. A three-step, 45-cycle RT-qPCR method was conducted using a LightCycler^® 480 System (Roche Diagnostics, Basel, Switzerland). A linear standard curve was created with the LightCycler^® 480 System Software and serial dilutions of recombinant plasmids containing the genes encoding 16S rRNA, eryA, eryB, eryC and eryD. RT-qPCR was conducted with the following reaction conditions: 5 min at 95°C, followed by 45 cycles at 58°C for 30 sec and 72°C for 30 sec. A negative control (no cDNA) and RT control (no reverse transcription) were used in the experiments. All assays were performed in triplicate and repeated at least three times. The expression levels of the target genes were calculated by comparison with cycle threshold (Ct) and determined at 0 min, 20 min, 1 h, 2 h, 3 h, 4 h and 12 h post-infection. 16S rRNA expression was used as a reference value to compare the relative expression levels at the various time points. Target genes were amplified in triplicate using the LightCycler^® 480 System, and the data were presented as the mean of each triplicate with standard deviations (SDs). All assays were repeated a minimum of three times.

The data were analyzed using Student's t-test and expressed as the mean value ± SD. The differences between groups were analyzed by analysis of variance using SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

The ery operon deletion was confirmed by PCR in the 2308Δery and 027Δery clones (Fig. 1A). Bacteriological analysis and typing of the mutant showed that deletion of the ery gene was stable after passage in culture media (Fig. 1B).

To test whether deletion of the ery operon affected the growth of 2308 or 027, we measured the bacterial growth in nutrient-replete (TSB 7.0) media. When cultured in normal TSB media, 2308Δery and 027Δery displayed a similar lag phase and reached the stationary phase at a similar optical density compared with 2308 and 027 (Fig. 2A).

To test whether the 2308Δery and 027Δery strains were sensitive to erythritol when grown in broth containing erythritol (20 mM), we measured the bacterial growth. The results showed that the virulent strains 2308 and 027 grew well in broth containing erythritol; however, the 2308Δery and 027Δery mutants grew at a slower rate (Fig. 2B). These results indicated that 2308Δery and 027Δery do not respond to erythritol, and that the virulent strains 2308 and 027 may utilize erythritol for growth.

To assess whether the ery operon influences virulence, RAW 264.7 macrophages were infected with 2308Δery, 027Δery, 2308 and 027. The surviving bacteria were enumerated, and there was no difference in the number of surviving bacteria 4 h post-infection (Fig. 3; P>0.05). This indicated that deletion of the ery operon does not affect macrophage invasion. However, at 12 h post-infection, there was a 1.50-log decrease (P<0.05) in the number of 2308Δery compared to 2308, and there was a 1.55-log decrease (P<0.05) in the number of 027Δery compared to 027 (Fig. 3). At 24 h post-infection there was a 2.70-log decrease in the number of 2308Δery compared to 2308 (Fig. 3; P<0.01); and there was a 2.70-log decrease in the number of 027Δery compared to 027 (Fig. 3; P<0.01). At 48 h post-infection, there was a 4.10-log decrease in the number of 2308Δery compared to 2308 (Fig. 3; P<0.01); and there was a 4.00-log decrease in the number of 027Δery compared to 027 (Fig. 3; P<0.01). Therefore, 2308Δery and 027Δery mutants had a replication defect in RAW264.7 macrophages.

The equations for the linear regression line for the standard curves of different genes generated by RT-qPCR assay and the corresponding R² value are as follows: 16S rRNA, y=0.238x+11.041, R²=0.986; eryA, y=−0.267x+12.217, R²=0.986; eryB, y=−0.266x+11.081, R²=0.983; eryC, y=−0.216x+10.165, R²=0.977; and eryD, y=−0.297x+11.414, R²=0.995. Based on the slope rates of these regression lines, the levels of amplification efficiency (E=10^−a-1) were 72.98% for 16S rRNA, 84.93% for eryA, 84.50% for eryB, 64.44% for eryC, 98.15% for eryD. Melt curve (unpublished) analysis of amplification products indicated that there was a single peak with a Tm of 86.8°C for 16S rRNA, 86.2°C for eryA, 84.6°C for eryB, 90.06°C for eryC and 91.75°C for eryD.

We analyzed gene expression at 0 min, 20 min, 1 h, 2 h, 3 h, 4 h and 12 h post-infection. The relative expression levels of eryA, eryB, eryC and eryD were significantly higher in Brucella grown with medium containing erythritol. Furthermore, the relative expression levels of eryA, eryB and eryC were highest (P<0.01) at 2 h post-infection (Fig. 4A-C), but were highest (P<0.01) at 3 h post-infection for eryD (P<0.05) (Fig. 4D). Gene expression levels were calculated based on the comparison to the expression of 16S rRNA. Based on these calculations, eryB was the most highly expressed gene followed by eryA, eryC and eryD (Fig. 5).