Similar Requirements of a Plant Symbiont and a Mammalian Pathogen for Prolonged Intracellular Survival

See allHide authors and affiliations

Science  31 Mar 2000:
Vol. 287, Issue 5462, pp. 2492-2493
DOI: 10.1126/science.287.5462.2492


Brucella abortus, a mammalian pathogen, andRhizobium meliloti, a phylogenetically related plant symbiont, establish chronic infections in their respective hosts. Here a highly conserved B. abortus homolog of the R. meliloti bacA gene, which encodes a putative cytoplasmic membrane transport protein required for symbiosis, was identified. An isogenicB. abortus bacA mutant exhibited decreased survival in macrophages and greatly accelerated clearance from experimentally infected mice compared to the virulent parental strain. Thus, thebacA gene product is critical for the maintenance of two very diverse host-bacterial relationships.

Rhizobia establish agriculturally important symbioses with leguminous plants (1), whereas brucellae are highly infectious pathogens of animals and cause the human disease brucellosis (2,3). Despite leading to exceedingly different outcomes in interactions with their respective eukaryotic hosts, the in-host life-styles of these closely phylogenetically related bacteria (4) show striking parallels. In the establishment of chronic infection, both rhizobia and brucellae are endocytosed by host cells, where they then undergo adaptive changes and ultimately live for prolonged periods in intracellular, acidic, host-membrane–bound compartments (1–3, 5,6).

Rhizobium meliloti bacA mutants invade alfalfa nodules like wild-type bacteria, but lyse upon release into plant cells before they can differentiate and establish a chronic host infection (7). BacA is predicted to be a cytoplasmic membrane transport protein with seven transmembrane domains (7,8). BacA is 64% identical to, and functionally interchangeable with, the Escherichia coli SbmA protein inferred to be a transporter of bleomycin and microcins B17 and J25 (9).R. meliloti bacA mutants also have increased resistance to bleomycin (9).

We identified a B. abortus DNA fragment that included a monocistronic 1248-nucleotide open reading frame (ORF) (GenBankAF244996) (10) that encodes a predicted protein of 47.3 kD with 68.2% identity to R. meliloti BacA. In both B. abortus and R. meliloti, bacA is flanked by an upstream gene for a putative transporter and a downstream gene, transcribed in the opposite direction, that has similarity to a putative bacterial secreted protein. We constructed an allele (bacA1) of the bacA gene in which 41% of thebacA ORF was replaced with a drug resistance cassette (strain KL7) (11). As with R. meliloti, disruption of bacA function resulted in increased resistance to bleomycin (Fig. 1) (12).

Figure 1

Increased resistance of the B. abortus bacA mutant to killing by bleomycin. Strains 2308 (wild-type), KL7 (ΔbacA ), and KL7R (reconstructed ΔbacA) were spread onto Schaedler agar plates and overlayed with filter paper disks containing bleomycin. The diameter of cleared zones of B. abortus growth inhibition in response to the drug was measured after 72 hours. Data are presented as the mean ± SD (n = 5).

The capacity of the brucellae to survive and replicate in host macrophages is critical to their ability to produce disease (2). The bacA1 mutant and its wild-type parent were opsonized and used to infect cultured murine macrophages (13). During the first 24 hours post infection (p.i.), both strains showed net intracellular killing by the phagocytes (Fig. 2). The virulent wild-type strain showed characteristic net replication in macrophages at 36 hours p.i. and beyond, but the bacA1 mutant did not appear to recover after the initial period of killing.

Figure 2

Decreased intracellular survival of theB. abortus bacA mutant in cultured murine macrophages. Isolated macrophages were infected with opsonized B. abortusstrains 2308 (○), KL7 (•), and KL7R (□). Percent survival of bacteria in macrophages was determined at various times after infection. Data are presented as the mean ± SD (n= 5).

To determine if this defect in intracellular replication in macrophages correlated with an inability to establish a chronic infection in the host, we experimentally infected BALB/c mice with the bacA1mutant and its wild-type parent (14). BALB/c mice represent the classic model for chronic B. abortus infection in the host (15): Substantial numbers of brucellae can be recovered from the spleens and livers of mice infected with virulent strains for beyond 20 weeks p.i. For the first 2 weeks p.i., thebacA1 mutant and its wild-type parent colonized the spleens and livers of mice at virtually equivalent levels (Fig. 3). However, at 3 weeks p.i. and beyond, the bacA1 mutant was recovered from spleens and livers of the mice at markedly lower levels than the isogenic parental strain. Thus, the onset of clearance of thebacA1 mutant seems to parallel the induction ofBrucella-specific humoral and cellular immune responses, which lead to enhanced brucellacidal activity of host macrophages (2). By 8 weeks p.i., the mutant strain was present in spleens at levels five orders of magnitude lower than in the wild-type strain. Thus, the B. abortus BacA protein is critically required for the maintenance of chronic infection in this mouse model, and the basis for the attenuation of the bacA1 mutant is likely to be its defect in intracellular survival in host macrophages. Mice infected with the bacA1 mutant and the wild-type parent exhibited equivalent delayed-type hypersensitivity and serologic responses to antigens derived from the parental strain (16). Thus, the bacA1 mutant was able to induce Brucella-specific cellular and humoral immune responses, making it an interesting vaccine candidate; currently, there is no safe and effective vaccine for human brucellosis (2,3).

Figure 3

Accelerated clearance of the B. abortus bacA mutant from experimentally infected BALB/c mice. (A) Spleen colonization. B. abortus strains 2308 (○) and KL7 (•) were injected intraperitoneally into BALB/c mice. At each time point, mice were killed, spleens dissected, and total number of brucellae per organ determined. Arrows denote level of inocula. Data are presented as the mean ± SD (n = 5). (B) Liver colonization. Counts were determined as in (A), except that left lateral lobes of the liver were analyzed. KL7R was present in tissues at levels equivalent to those in 2308 at all time points (22).

Because elements of the host innate immune response are conserved among mammals, insects, and plants (17), all of which share common antimicrobial defense strategies (18), BacA function may be required to overcome a host defense response necessary for the prevention of chronic infection, irrespective of whether that results in a pathogenic or symbiotic relationship. It has previously been shown that pathogens that can infect widely differing hosts require common virulence functions for these interactions (19).

Numerous studies with Brucella have failed to identify virulence factors typically found in pathogenic bacteria (e.g., exotoxins and adhesins). Thus, it is notable that we were able to identify a gene crucial for B. abortus pathogenesis on the basis of knowledge gained by genetic analysis of a bacterial-plant symbiosis. Recently, a B. abortus two-component regulatory system necessary for pathogenicity was found to be closely related to a symbiotically important R. meliloti two-component system (20). It will be interesting to learn whether the downstream targets of these regulatory systems are also conserved and are mechanistically required for chronic host infection, as BacA appears to be.

  • * To whom correspondence should be addressed. E-mail: gwalker{at}


View Abstract

Stay Connected to Science

Navigate This Article