Keywords

5.1 Nickel Import by NikABCDE and NikKMLQO

With regards to nickel homeostasis in Brucella, nickel import has been the most studied system to date. Brucella spp. contain two loci encoding nickel import systems. NikABCDE is an ABC-type transporter that was first described in 2001 by Jubier-Maurin et al., and NikKLMQO is an energy coupling factor (ECF)-type transporter transcribed with the ure2 operon (Sangari et al. 2010; Jubier-Maurin et al. 2001). nikABCDE encodes an archetypal ABC-type transporter with nikD and nikE encoding ATPases, nikB and nikC inner membrane proteins, and nikA as a periplasmic nickel-binding protein (Fig. 5.1). The functions of these proteins have not been experimentally characterized in Brucella but can be inferred based on homology to other characterized nickel transporters (Rodionov et al. 2006). However, B. suis NikA has been crystallized and specific nickel binding sites were determined, adding to the evidence that NikA is the periplasmic binding protein of the nikABCDE system (Lebrette et al. 2014).

Fig. 5.1
figure 1

Nickel uptake and export system

Full size image

Utilizing a GFP-transcriptional fusion, it was shown that the nikA promoter in B. suis 1330 was induced during infection of J774.A1 macrophage-like cells, and similarly, the nikA promoter was activated when bacteria were grown in culture medium supplemented with EDTA or under microaerobic conditions (Jubier-Maurin et al. 2001). Despite being induced upon infection of J774.A1 cells, strains containing a partial deletion of nikA showed similar replication rates in both human monocytes and THP-1 cells, indicating that this system is not necessary for the survival and replication of B. suis during macrophage infection (Table 5.1).

Table 5.1 Nickel-related genes and connections to virulence in Brucella
Full size table

Interestingly, the nikABCDE gene locus is dissimilar amongst Brucella spp. nikD is a pseudogene in B. ovis ATCC 25840 (Tsolis et al. 2009); and in B. abortus strains 2308, 9-941, and S19, there is a nonsense mutation about halfway through nikA, rendering it a pseudogene (Sangari et al. 2010). However, the presence of a second nickel import system likely compensates for these mutations in B. ovis and B. abortus.

While studying the ure2 gene locus, Sangari et al. described the presence of an ECF-type nickel transporter gene cluster downstream of the ure2 operon that encodes the putative nickel transporter nikKMLQO (Sangari et al. 2010). ECF transporters lack a extracytoplasmic periplasmic binding protein and instead contain a membrane-embedded substrate binding protein (Wang et al. 2013; Erkens et al. 2011). Bioinformatic homology analyses predict nikM and nikQ to encode inner membrane proteins, nikO to encode an ATPase, and nikK and nikL to encode substrate-binding proteins (Fig. 5.1) (Sangari et al. 2010). Again, none of these functions have been experimentally demonstrated. Sangari et al. showed that the ure2 operon encodes 3 separate systems, ureABCEFGD2, ureT, and nikKMLQO. In their study, they constructed a nikO mutant to determine the role of this transporter on urease activity. It was shown that the nikO mutant had lower urease activity and was more sensitive to acidic pH in culture (Table 5.1). The nikO mutant was not used to infect cells in vitro.

Altogether, these data show that Brucella spp. contain two separately encoded nickel import systems, which likely compensate for one another. None of the above mutants were used to define the role of the nik genes in vivo. Further studies are necessary to characterize both import systems and to understand the potential for them to serve redundant functions in vivo. One method to do this would be to construct mutants containing mutations in each system individually and then construct a mutant containing mutations in both systems. These mutants could be used to elucidate the proposed redundancy of the systems and understand the role of nickel in Brucella virulence. As discussed below, the studies provide evidence that they are necessary for the import of nickel and for the efficient activity of urease during infection.

5.2 The Nickel and Cobalt Exporter RcnA

The expression of a nickel and cobalt export system, RcnA, is a mechanism utilized by bacteria to counter metal toxicity and was first described by Rodrigue et al. in 2005. That study demonstrated that E. coli rcnA was induced upon addition of nickel or cobalt to the media; strains containing a deletion of rcnA were more sensitive to nickel and cobalt toxicity and contained more intracellular nickel and cobalt; and strains harboring a multicopy plasmid expressing RcnA contained less intracellular nickel and cobalt than strains containing an empty multicopy plasmid (Rodrigue et al. 2005). Brucella spp. Encode a putative nickel and cobalt export permease RcnA (Fig. 5.1). However, Brucella RcnA shows less than 40% amino acid similarity to that of E. coli RcnA and is missing the distinctive histidine cluster found within E. coli RcnA sequence. This begs the question whether the ortholog of RcnA in Brucella is truly a nickel and cobalt exporter, or are there other mechanisms to detoxify the cells of these metal cations?

5.3 Nickel-Responsive Regulators NikR and RcnR

While nickel is an essential cofactor for several proteins in bacteria, an excessive amount of intracellular nickel can be toxic by causing oxidative damage, by replacing essential metal ions in metalloenzymes, or by binding to non-metalloenzyme active sites or secondary sites leading to decreased enzyme activity (Macomber and Hausinger 2011). To combat this problem, bacteria possess mechanisms to regulate intracellular nickel concentrations. Escherichia coli and Helicobacter pylori encode the well characterized nickel responsive regulator, NikR, which is a nickel dependent ribbon-helix-helix transcriptional regulator. NikR regulates nik genes in response to intracellular metal concentrations and other stimuli, and E. coli also encodes a repressor of the nickel exporter RcnA, called RcnR (Iwig et al. 2006; Schreiter and Drennan 2007).

In E. coli, NikR solely regulates the nik operon (Li and Zamble 2009). In H. pylori, NikR is a repressor of the nickel uptake gene nixA and an activator of the ure operon (Ernst et al. 2005). A putative nickel responsive regulator is encoded adjacent to the nikABCDE locus in B. abortus 2308. The ability of NikR to regulate the nickel import systems and potentially the ure2 operon has not been experimentally characterized, and there is little data describing the role of this regulator in Brucella. While the aim of the study was not to characterize NikR, Jubier-Maurin et al. identified a sequence upstream of nikA that closely resembles the NikR binding site in E. coli (Jubier-Maurin et al. 2001). Rossetti et al. were the first to identify a potential role of NikR in virulence during B. melitensis infection of HeLa cells. Their data demonstrated that nikR expression increased 12 h post-infection of HeLa cells infected with B. melitensis (Table 5.1) (Rossetti et al. 2011). This study also revealed increased expression of the ure2 operon, but no differential expression of any of the nik genes during HeLa cell infection. Thus, the authors suggested that NikR could be both a repressor of the nik genes and an activator of the ure2 genes (Rossetti et al. 2011). However, since this was not a targeted experiment, rather a discovery tool for genes differentially regulated during infection of HeLa cells, the link between NikR and the regulation of the ure2 operon remains unknown. It is possible that another regulator is responsible for the induction of ure2, and it is coincidental that both ure2 and nikR are induced upon infection of HeLa cells. It is clear that further studies are necessary to deduce the function of NikR in Brucella spp.

RcnR is a repressor of rcnA and rcnR in a nickel and cobalt responsive manner in E. coli (Blaha et al. 2011; Iwig et al. 2006). The two genes are transcribed divergently from one another in E. coli, and Blaha et al. identified a specific palindromic RcnR binding box, TACT-N7-AGTA, in the intergenic region of the two genes. Upon binding either of the metals, RcnR dissociated from the RcnR binding box, allowing for the expression of rcnA and rcnR (Blaha et al. 2011). Deletion of rcnR showed constitutively expressed rcnA (Iwig et al. 2006). Brucella spp. also encode a putative RcnR protein that is 40% identical and over 60% similar to the RcnR protein of E. coli. Contrary to the situation in E. coli, rcnR in Brucella strains is not located divergently to rcnA, but rather is located on a different chromosome from rcnA. Interestingly, an identical sequence to that of the E. coli RcnR binding box is located upstream of rcnR, but this putative RcnR-binding sequence is not observed upstream of rcnA in Brucella. It should be noted that Brucella abortus RcnR is 54% identical to the formaldehyde stress response regulator FrmR in E. coli, and the genomic organization of RcnR and surrounding genes in B. abortus is similar to that of the E. coli frmR and N. gonorrhoeae nmlR (an fmrR ortholog) loci (Chen et al. 2016). To date, no studies have characterized RcnR in Brucella strains, and empirical evidence will be needed to support the hypothesis that RcnR is a transcriptional regulator of rcnR and/or rcnA in Brucella spp. and not, in fact, an ortholog of FrmR related to formaldehyde resistance.

5.4 Nickel-Dependent Proteins in Brucella

Bacterial proteins that require nickel include urease, NiFe-hydrogenase, carbon monoxide dehydrogenase, acetyl-coenzyme A decarbonylase, methyl-coenzyme M reductase, nickel dependent superoxide dismutases and glyoxylases (Mulrooney and Hausinger 2003; Li and Zamble 2009). Of the above proteins, urease is the only protein directly linked to virulence in Brucella spp.

5.4.1 Urease

Urease enzymes hydrolyze urea into carbon dioxide and ammonia and thus, play a key role in nitrogen metabolism, as well as acid resistance as microbes pass through the acidic environment of the stomach (Li and Zamble 2009). Urease was one of the first proteins demonstrated to require nickel for catalysis (Alagna et al. 1984). Since then, extensive biochemical analyses of this protein have exposed much about specific binding sites for nickel within the protein (Mulrooney and Hausinger 2003). The Brucella ure operons can be split into two groups: structural proteins (encoded by ureA, ureB, and ureC) and accessory proteins (ureD, ureE, ureF, and ureG) (Sangari et al. 2007). This genetic organization is very similar to that of the genes encoding the trimeric urease of Klebsiella aerogenes (Mulrooney and Hausinger 1990; Sangari et al. 2007). It was shown that K. aerogenes UreE directly binds to nickel and is thought to function as a nickel carrier for the urease enzyme (Mulrooney et al. 2005). The Brucella genome contains two urease operons, ure1 and ure2. While both have been shown to contribute to urease activity and are activated in acidic conditions, only ure1 has been shown to be necessary for the virulence of B. suis or B. abortus in mice infected via the oral route (Bandara et al. 2007; Sangari et al. 2007). Upon deletion of either ureC1 (ureC from operon 1) or ureC2 (ureC from ure operon 2), ureC1 was deemed less fit for oral infection of a mouse model compared to B. abortus 2308 (Sangari et al. 2007). This has been predicted to be due to the dissimilar genetic identities of the ure operons. In B. suis 1330, the amino acid similarity between ure1 and ure2 is about 70%, and most of the urease activity in this strain is predicted to be due to ure1 (Bandara et al. 2007). Mutations in the ure genes does not affect the ability of Brucella spp. to survive and replicate in cell lines in vitro (Sangari et al. 2007; Bandara et al. 2007). This evidence supports the claim that Brucella is most likely utilizing urease to combat a low pH environment during the biologically relevant oral route of infection. Altogether, it has been shown that while ure2 contributes to urease activity, it is not necessary for infection of a host via the oral route. Therefore, it is possible that Brucella spp. have maintained the ure2 operon due to necessity of other genes (i.e., ureT, nikK, nikM, nikL, nikQ, nikO) located downstream of the ure2 genes in the operon for infecting the host.

5.4.2 Other Potential Nickel-Containing Proteins in Brucella

As stated above, several proteins other than urease have been identified as nickel-binding metalloproteins in microbes. However, Brucella urease is the only nickel binding protein that has been extensively studied in Brucella biology.