Differential Use of the Two High-Oxygen-Affinity Terminal Oxidases of Brucella suis for In Vitro and Intramacrophagic Multiplication

Expression of FnrN, a potential oxygen sensor in B. suis, is increased in bacteria grown under microaerobiosis.

Analysis of B. suis and Brucella melitensis genome sequences detected a unique copy of a gene predicted to encode a transcriptional regulator that possessed distinctive features of oxygen sensors. In a phylogenetic study (14), the B. suis Fnr-like factor was found to cluster with the members of the FnrN class, which comprises regulators of some rhizobial species. They have in common the conserved cysteine motif of Fnr, which is the general transcription factor of Escherichia coli under anaerobic conditions. These amino acids are ligands to the iron-sulfur cluster, whose oxidative state drives the transcriptional activity of Fnr (7). The 5′ region and the upstream sequence of fnrN were produced by PCR from B. suis 1330 (ATCC 23444) genomic DNA with primers FnrN5′ (TGGTACCCGGCTGATTTCGC) and FnrN3′ (GGATTTACGGGAGCTACGGC). Expression of fnrN in B. suis was monitored using a transcriptional fusion of the promoter region located on the 305-bp KpnI-HindIII fragment from the PCR product, with the promoterless gfp gene of the pBBR1-CGFP plasmid construct. This plasmid was obtained after replacement of the kanamycin resistance gene by a chloramphenicol resistance cassette (from plasmid pBlueCm-2) in plasmid pBBR1-KGFP (13). Fluorescence intensity in cultures grown in tryptic soy broth until mid-log phase (approximately 4 × 10⁹ bacteria ml⁻¹) was quantified with a FACScalibur scanner (Becton Dickinson, San Jose, CA) (10). To perform cultures under microaerobiosis, 3-ml cultures of brucellae were grown in loosely capped 10-ml tubes placed in a jar with GENbox generators (bioMérieux, Marcy l'Etoile, France) of microaerobic atmosphere (oxygen concentration ranging from 6.2 to 13.2% after 1 h). The jar was shaken at 80 rpm to keep the bacteria in suspension. The fnrN promoter was active in B. suis cultures under aerobic conditions, exhibiting an 8- ± 1-fold-higher fluorescence than the promoterless control plasmid pBBR1-CGFP, and this activity was increased to a 30- ± 5-fold-higher fluorescence under microaerobic conditions (not shown). We then analyzed the expression of fnrN in bacteria multiplying inside human THP-1 macrophage-like cells. Briefly (10), 4 × 10⁶ VD3-differentiated cells were infected with 1 ml of bacteria suspension, corresponding to a multiplicity of infection of 80. At 48 h postinfection (p.i.), the cells were scraped off, washed with phosphate-buffered saline (PBS), and lysed in 1 ml 0.1% Triton X-100 by incubation for 10 min on ice. After centrifugation (1,000 rpm, 5 min) to pellet cellular debris and nuclei, bacteria contained in the supernatant were recovered by centrifugation (13,000 rpm, 10 min) and diluted in 500 μl PBS for flow cytometry analysis. Measurement of fluorescence intensity revealed that the intracellular promoter activity was 10- ± 2-fold higher than that of the negative control, which was not statistically different (P = 0.5; Student's t test) from that obtained in vitro under aerobic conditions. This result can be interpreted according to two hypotheses: first, fnrN may not be involved in the transcriptional regulation within intracellular brucellae, and second, oxygen tension may not differ significantly inside and outside the host cell. To discriminate between these two hypotheses, we decided to study expression of the two high-oxygen-affinity terminal oxidases in the wild-type and fnrN mutant strains of B. suis. The ΔfnrN mutant was obtained by allelic exchange between chromosomal fnrN and a PCR product (see above) of this gene cloned into a suicide plasmid, with a deletion of a 480-bp HindIII fragment which was replaced by the kanamycin resistance gene from plasmid pUC4K (10). Levels of intracellular expression of high-oxygen-affinity terminal oxidases would be indicative of the intramacrophagic oxygen tension.

FnrN specifically activates in vitro expression of the cytochrome cbb3-type terminal oxidase.

Expression of the cytochrome cbb3-type and cytochrome bd oxidases was analyzed by measurement of green fluorescent protein (GFP)-mediated fluorescence under control of the ccoNOQP and cydDCAB promoters, respectively. The transcriptional gfp fusions to the cco and cyd promoter regions comprised 290 bp and 1 kb of upstream sequences of ccoN and cydD, obtained by PCR with primers 5PRCcoN (GACCGCTCGAGACGGCTACAGGATCAGCAAG) and 3PRCcoN (ACGCACGGTACCGTTGCGATGACGCCATAAC) and with primers 5PRCyd (CCGCTCGAGACAGCAAGGAGTTGCCTTC) and 3PRCyd (GGTACCGCTGCATAAGCCAGAAGGGC), respectively. XhoI and KpnI restriction sites in the 5′ and 3′ primers allowed direct cloning into the pBBR1-CGFP plasmid.

Transcription of the ccoNOQP operon was strongly increased under microaerobic conditions (10-fold; P = 0.037). As induction by microaerobiosis was significantly reduced in the ΔfnrN mutant (Fig. 1A) and was not different from results obtained under aerobic conditions (P = 0.075), microaerobic activation depended strictly on FnrN. The ccoNOQP operon may therefore represent one of the potential targets of B. suis FnrN, as underlined by the finding of the perfect “FNR box” TTGAT N₄ ATCAA located at a proper distance from ATG of ccoN (3), the transcription initiation site being unknown.

• Open in new tab
• Download powerpoint

FIG. 1.

Expression of the cbb3-type cytochrome oxidase and the cytochrome bd ubiquinol oxidase in wild-type (WT) and ΔfnrN mutant strains of B. suis. In vitro expression levels of the ccoNOQP (A) and cydDCAB (B) promoters were determined by flow cytometry analysis of the gfp construct under aerobic (black bars) and microaerobic (gray bars) conditions. The control is wild-type B. suis transformed with native plasmid pBBR1-CGFP. Standard errors are reported for the means from three independent experiments.

In the wild-type strain, cytochrome bd oxidase expression was not detected under aerobic conditions, as the fluorescence intensity measured was the same as that obtained with the control (Fig. 1B). Slight induction was observed under microaerobiosis (twofold; P = 0.01). With both oxygen concentrations tested, the disruption of fnrN increased (P < 0.01) cydDCAB transcription levels by a rather similar factor compared to those observed in the wild type (1.8-fold under aerobic conditions and 1.6-fold under microaerobic conditions) (Fig. 1B). These results indicated that in contrast to the cytochrome cbb3-type oxidase, the cytochrome bd oxidase was poorly expressed in vitro under microaerobic conditions and that the fnrN regulator repressed its expression.

The cytochrome bd oxidase is the sole high-oxygen-affinity terminal oxidase expressed inside macrophages.

Activity of the promoters in bacteria multiplying within human THP-1 macrophage-like cells was analyzed as described above for the fnrN promoter. In the macrophage, the ccoNOQP promoter was inactive (Fig. 2), as no fluorescence peak could be observed, and the fnrN deletion had no effect (not shown). In contrast, the cydDCAB promoter was found to be expressed similarly in the presence (Fig. 2) or absence (not shown) of the native fnrN gene. These results suggested that inside the cells, B. suis specifically utilized the cytochrome bd oxidase, via another, unknown mechanism of regulation.

• Open in new tab
• Download powerpoint

FIG. 2.

Expression of intramacrophagic terminal cytochrome oxidases. Flow cytometry analysis of expression of the cyd and cco promoter-gfp fusions in wild-type B. suis obtained from infected macrophages is shown. The control curve corresponding to intracellular bacteria containing native plasmid pBBR1-CGFP was identical to that of pcco-CGFP and is omitted for clarity.

The cytochrome bd oxidase but not the cytochrome cbb3-type oxidase participates in intracellular growth of B. suis.

In order to evaluate a possible involvement of the cytochrome cbb3-type and cytochrome bd oxidases in bacterial replication within the host cell, we examined the effect of ccoN and cydB deletions on the intracellular survival of B. suis. These genes, produced by PCR with primers CcoN5′ (GACGGCTACAGGATCAGCAAG) and CcoN3′ (CGGCGAATTCTTATTCGGCAGGCTGCATGG) and primers CydB5′ (CGGCAAGCTTGCAATGCGACGGACGAACAG) and CydB3′ (CGCTCTAGACCAAGCGCAGGCGCGATCAG) were cloned, and the 870-bp NaeI and 124-bp StyI fragments within the ccoN and cydB open reading frames, respectively, were deleted and replaced by the kanamycin or chloramphenicol resistance cassette. B. suis mutants were obtained as described for ΔfrnN.

The multiplication rates of wild-type and mutant strains of B. suis 1330 were determined after infection of differentiated human macrophage-like THP-1 cells. Infection of 5 × 10⁵ cells was performed in 24-well plates as indicated above, at a multiplicity of infection of 20. At 1.5, 7, 24, and 48 h p.i., cells were washed with PBS and lysed in 0.2% Triton X-100. CFU were determined by plating serial dilutions on TS agar.

Deletion of cydB resulted in a reduced multiplication rate of 0.27 ± 0.05 compared to that of the wild-type strain (P = 0.01; Student's t test) at 48 h p.i. (Fig. 3). Similar results were obtained with the murine macrophagic cell line J774.A1 (not shown), in agreement with results previously published for B. abortus (6). This rather small effect could be considered incompatible with our previous work showing that the cydD mutant was more severely affected than the cydB mutant (11). Deletion of cydB, the last gene of the cydDCAB operon, however, does not prevent expression of cydD. cydDC mutants of E. coli are pleiotropic and show highly reduced levels of all periplasmic cytochromes (19). A cydD deletion in B. suis may have a similar effect, since there is good sequence conservation of cydDC genes and their products (4).

• Open in new tab
• Download powerpoint

FIG. 3.

Intracellular growth of B. suis mutants in human THP-1 cells. Cells were infected with the wild-type strain (•) or the ccoN (⋄) or cydB (▪) mutant. Experiments were performed in triplicate, and error bars represent the standard errors of the means. Asterisks indicate statistical significance (P = 0.03 and P = 0.01 at 24 h and 48 h p.i., respectively, as determined by Student's t test).

The ccoN mutation had no effect on the replication of the bacteria inside the macrophages (Fig. 3). This is consistent with the absence of expression of the ccoNOQP promoter at the intracellular state. No significant reduction in intracellular survival was observed for the strain with fnrN deleted (not shown).

Conclusion.

Brucella possesses a set of genes which allows the bacteria to adapt to low oxygen tension (16), among which are two operons encoding high-oxygen-affinity terminal oxidases. The cytochrome cbb3-type and cytochrome bd oxidases were found to have differential expression patterns in B. suis. The first one was specifically expressed in vitro, with maximal activation under microaerobiosis and dependent on FnrN. The second one was preferentially expressed during intracellular multiplication and was involved in adaptation to the replicative niche. Since the cytochrome bd oxidase was found to be repressed by FnrN, its constant expression rate within cells infected by the ΔfnrN strain indicated that FnrN was not involved in the regulation at the intracellular state. This was in accordance with the fact that this regulator was not necessary for multiplication of B. suis within cells. Intracellular expression of the cytochrome bd oxidase may indicate low oxygen tension inside the macrophage. The affinities of B. suis cytochrome oxidases for oxygen have not yet been established, but in Rhodobacter capsulatus, which possesses both cytochrome cbb3 and cytochrome bd oxidases, the latter was proposed to have the higher affinity (22). The important attenuation of the cydB mutant of B. abortus in mice (6) suggests that a very low oxygen tension is encountered by the pathogen in vivo. The use of the cytochrome bd oxidase most likely facilitates Brucella survival in the host, where the bacteria reside in various tissues previously described as containing low but variable oxygen concentrations (1, 15).