• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of bmcgenoBioMed Centralsearchsubmit a manuscriptregisterthis articleBMC Genomics
BMC Genomics. 2009; 10: 470.
Published online Oct 12, 2009. doi:  10.1186/1471-2164-10-470
PMCID: PMC2770081

A subset of the diverse COG0523 family of putative metal chaperones is linked to zinc homeostasis in all kingdoms of life

Abstract

Background

COG0523 proteins are, like the nickel chaperones of the UreG family, part of the G3E family of GTPases linking them to metallocenter biosynthesis. Even though the first COG0523-encoding gene, cobW, was identified almost 20 years ago, little is known concerning the function of other members belonging to this ubiquitous family.

Results

Based on a combination of comparative genomics, literature and phylogenetic analyses and experimental validations, the COG0523 family can be separated into at least fifteen subgroups. The CobW subgroup involved in cobalamin synthesis represents only one small sub-fraction of the family. Another, larger subgroup, is suggested to play a predominant role in the response to zinc limitation based on the presence of the corresponding COG0523-encoding genes downstream from putative Zur binding sites in many bacterial genomes. Zur binding sites in these genomes are also associated with candidate zinc-independent paralogs of zinc-dependent enzymes. Finally, the potential role of COG0523 in zinc homeostasis is not limited to Bacteria. We have predicted a link between COG0523 and regulation by zinc in Archaea and show that two COG0523 genes are induced upon zinc depletion in a eukaryotic reference organism, Chlamydomonas reinhardtii.

Conclusion

This work lays the foundation for the pursuit by experimental methods of the specific role of COG0523 members in metal trafficking. Based on phylogeny and comparative genomics, both the metal specificity and the protein target(s) might vary from one COG0523 subgroup to another. Additionally, Zur-dependent expression of COG0523 and putative paralogs of zinc-dependent proteins may represent a mechanism for hierarchal zinc distribution and zinc sparing in the face of inadequate zinc nutrition.

Background

Transition metals perform vital roles in many chemical reactions essential for life. A recent bioinformatic approach suggests Zn-, non-heme Fe- and Cu-proteins constitute 10% of bacterial and eukaryotic proteomes and 13% of archaeal proteomes [1-3]. The roles of these metals can be varied. In some oxidoreductases, for instance, iron and copper are exploited for their ability to accept or donate electrons, while in hemoglobin and hemocyanin, these metals are used for oxygen transport [4-6]. Zinc, on the other hand, serves as an electrophile or Lewis acid in many protein-catalyzed reactions. The activity of metalloproteins (many of which are essential proteins) is, consequently, strictly dependent on the presence of a metal and in most cases of a specific metal. Ensuring proper metal allocation is therefore critical for survival.

It was initially assumed that free pools of metal ions were available within the cell, such that a nascent polypeptide would acquire its cofactor solely through the metal-affinity of the chelating ligands. As discussed in several recent reviews, this picture of metal metabolism was oversimplified [7,8], as: i) the ions are chelated intracellularly by proteins and small molecule ligands, and ii) metal-binding ligands are not sufficiently selective to ensure that the proper cofactor is loaded. Since the discovery of the first copper metallochaperone, Atx1[9], numerous protein factors involved in metallocenter biosynthesis have been characterized. The mechanisms by which the cell ensures the correct metal ions are loaded into metalloproteins are just beginning to be understood.

Studies involving the maturation of Ni-urease and Ni-Fe hydrogenase have provided the most extensive picture of metallocenter biosynthesis (for a review see [10] and [11]). These two nickel-containing proteins require a suite of accessory proteins to properly insert Ni into the catalytic site (only one exception has been found to date; Bacillus subtilis encodes a functional urease in the absence of the canonical accessory proteins [12]). In both cases, a GTPase (UreG for urease or HypB for hydrogenase) is involved in the incorporation of the Ni cofactor. These two proteins belong to the G3E family of P-loop GTPases as defined by Leipe and colleagues [13]. Other members of this family include MeaB (ArgK), required for the activation of methylmalonyl-CoA mutase (a B12-dependent enzyme) [14], and COG0523, a large and diverse subfamily of proteins with poorly defined functions.

COG0523 proteins occur in all kingdoms of life, and most sequenced genomes encode one or more homologs. The first member of the COG0523 family was identified as being involved in cobalamin biosynthesis in Pseudomonas denitrificans and hence named CobW [15]. Other members of COG0523 include the nitrile hydratase activator, which is required for Fe-type nitrile hydratase activity [16], and YciC of Bacillus subtilis. Due to repression by Zur, a zinc-responsive transcription factor, yciC was originally assumed to code for a low-affinity zinc transporter [17-19]. Without the means to automatically distinguish between these different functions, these annotations have been propagated amongst all members of the family in sequenced genomes. Therefore, as a result of these studies, genes encoding a COG0523 protein have been automatically and arbitrarily assigned either a function in cobalamin biosynthesis, in the activation of nitrile hydratase, or as a low-affinity zinc transporter. Nevertheless, we note that each of these functions is related in the general sense to intracellular metal delivery. The diversity of the metals putatively handled by COG0523, Ni, Fe, or Zn, suggests that there might be different sub-groups identifiable within the COG0523 superfamily.

The COG0523 family is a striking example of systematic, homology-based mis-annotation. Although members are frequently annotated as having specific functions, these 'functions' are based only on homology to a few family members and are therefore suspect. The simplest way to annotate a genome is based on sequence homology to characterized genes. Sequence homology does not necessarily equate to functional identity or even similarity. Therefore, this approach to annotation is frequently inadequate as exemplified in the literature [20-22] and illustrated by the development of alternative paradigms for functional annotation [23-26].

To provide an improved annotation for the various members of this family and gain insight into the role members of this protein family may perform, we conducted an extensive comparative genomic analysis of the G3E family of P-loop GTPases and more specifically of COG0523 members. By combining phylogenetic analysis, physical clustering analysis, and regulatory site detection, we predict that the COG0523 family comprises subfamilies that have specialized and distinct functions in metal metabolism. We also hypothesize that several, but not all, of these subfamilies have a role in survival under conditions of poor zinc nutrition in both prokaryotic and eukaryotic organisms.

Results and Discussion

Phylogenomic analysis of COG0523 proteins as members of the G3E family of P-loop GTPases

Phylogenetic analysis first performed by Leipe, et al. [13] and repeated here with a diverse set of family members (see Figure Figure1)1) shows that COG0523 belongs to the G3E family of P-loop GTPases (G3E family), which is separated from the rest of the SIMIBI class of GTPases (for SIgnal recognition GTPases, MInD superfamily, and BIoD superfamily) by a glutamate residue in the Walker B motif and an intact NKXD motif (Figure (Figure2A)2A) [13]. Characterized members of the G3E family perform two roles in metallocenter assembly: 1) facilitating incorporation of the cofactor in an energy- dependent manner into the target protein's catalytic site (the insertase role) and, 2) storage and delivery of a metal cofactor to a target metalloprotein (the metallochaperone role). G3E family proteins have been found to function as either metal-insertases or as a dual function metallochaperone/insertase. For example, MeaB appears to fulfill the role of an adenosylcobalamin (Co2+)-insertase, facilitating the insertion of B12 into methylmalonyl-CoA mutase (MCM) [27]. A large structural rearrangement occurs upon interaction between MeaB and its target [27], suggesting that MeaB may be responsible for the structural changes that must occur for B12 cofactor incorporation. In addition, MeaB also appears to protect radical intermediates that are essential for the activity of MCM [28,29]. In hydrogenase maturation, HypB is thought to carry out both the insertase and metallochaperone roles in most organisms due to the presence of a histidine stretch located at the N-terminus of these proteins. In Bradyrhizobium japonicum, this histidine stretch was found to bind 18-Ni ions per dimer [30]. In Escherichia coli, the common histidine stretch is missing and SlyD is presumed to be the metallochaperone component that delivers Ni to the assembly complex [31,32]. Most UreG proteins studied to date lack the histidine stretch that is found in most HypB proteins. Accordingly, in urease maturation, UreG appears to function as an insertase and another accessory protein, UreE, functions as the metallochaperone, delivering Ni to the maturation complex [33-35].

Figure 1
Summary of phylogenomic analysis of G3E family and COG0523 members in prokaryotes. A, Phylogeny of extracted GTPase domains from diverse members of the G3E family plus genome context for corresponding genes. B, COG0523 distance tree. Each subgroup (3,4 ...
Figure 2
COG0523 amino acid conservation plot. A, Plot of amino acid conservation. The conserved GTPase motifs are highlighted in red. The conserved GCXCC motif is highlighted in yellow. The most common positions of His-stretches are shown. B, Typical histidine-rich ...

To support the proposition that the presence of a histidine stretch in G3E family proteins could be indicative of metallochaperone activity, we analyzed the distribution of genes encoding UreG and UreE (see 'G3E' subsystem in the SEED database [36]). We found that several genomes containing ureG homologs lack any recognizable ureE homologs. In all but two of those genomes, UreG contains an added histidine-rich motif at the N- or C-terminus (Table (Table1).1). As has been suggested for the UreG of Mycobacterium tuberculosis, this His-stretch may be able to compensate for the absence of UreE [37]. In addition, the absence of ureE does not correlate with the presence of slyD as would be expected if SlyD performs the metallochaperone role in those organisms. Bradyrhizobium japonicum USDA 110 and Frankia sp. Ccl3 are two exceptions to this trend as they lack both ureE and a His-stretch extension in UreG. In these cases, the Ni-metallochaperone involved in urease maturation could be HypB, which is present in both of these organisms (see 'G3E' subsystem). Indeed, it has been shown in Helicobacter pylori that HypB is required for activity of both hydrogenase and urease [38], and a physical interaction between UreG and HypB has been verified [39]. Although ureE is present in the genome of H. pylori, the corresponding protein lacks a His-stretch. As expected, the addition of a His-stretch to UreE was found to eliminate the need for HypB in the maturation of urease [40].

Table 1
Co-occurrence profile between ureE, slyD, and a His-stretch in UreG.

We compared the amino acid sequence of 887 COG0523 proteins from all kingdoms (see 'G3E' subsystem for sequences). We observed that like UreG and HypB orthologs, COG0523 proteins are found with and without His-stretches, suggesting a distribution of insertase and metallochaperone activity among various members. Approximately 40% of the sequences analyzed contain a histidine-rich region, commonly found near the C-terminus (Figure (Figure2A2A and and2B);2B); 365 COG0523 proteins contain the minimal HxHxHxH motif, where x represents 0 - 4 residues. Some proteins contain a His-stretch with up to 29 histidines, such as Ava_3717 [Genbank:75703646] from Anabaena variabilis.

The region of highest similarity between COG0523 and the other members of the G3E family is the GTPase domain, defined by the canonical Walker A and Walker B motifs (Figure (Figure2)2) [13]. GTPase activity of HypB and UreG has been shown to be essential for the metallocenter biosynthesis of hydrogenase and urease, respectively [41,42], and postulated to be responsible for incorporation of B12 in MCM by MeaB [29]. GTPase activity has been verified for YjiA, a COG0523 homolog from Escherichia coli, for which the crystal structure was solved [43]. In addition to the GTPase domain, all members of COG0523 have a conserved, putative, metal-binding CXCC motif (Figure (Figure2A).2A). Analysis of the YjiA crystal structure reveals this motif is found in the Switch I region of the protein, suggesting that binding of GTP/GDP affects its conformation [43]. The same motif was found to be essential for the activity of the nitrile hydratase activator protein, a member of COG0523 assumed to be involved in the incorporation of iron into Fe-type nitrile hydratase [44,45].

In addition to the GTPase domain, MeaB and most COG0523 proteins contain an additional C-terminal domain. On average, COG0523 is 99 and 147 amino acids larger than HypB and UreG, respectively, and only 26 residues larger than MeaB. The smallest G3E protein, UreG, is the GTPase component of a complex composed of UreD and UreF, where the three proteins act together in the activation of urease [46]. Activation of MCM appears to only require delivery of the cofactor by adenosyltransferase and the activity of MeaB[47]. The size of G3E proteins could be indicative of the number of other accessory proteins required for activation of the target metalloprotein.

While the N-terminal GTPase domain is well conserved among COG0523 members, the C-terminal region is highly variable (Figure (Figure2A).2A). Indeed, COG0523 proteins fall under the category of "segmentally variable genes (SVGs)," as defined by Zheng et al [48]. SVG profiles for four members of the family (HP0312, NMB1263, VCA0527, yeiR) can be found at http://geneva.bu.edu. SVGs are genes that code for proteins that have highly variable regions interspersed with well-conserved regions. The authors observed that SVGs encode proteins that are involved in adaptation to environmental stresses and proposed that highly variable domains are an indication of protein-protein interaction specificity or specificity of small molecule binding.

Finally, as summarized in Figure Figure1,1, while hypB genes are consistently found in hydrogenase maturation gene clusters, ureG genes in the urease maturation clusters and meaB genes cluster with MCM-encoding genes, COG0523 genes are found in multiple gene clusters. Most genomes contain only one homolog of hypB, ureG, or meaB. Conversely, up to 11 COG0523 genes can be found in a single genome, as seen in Cyanothece sp. ATCC 51142. In addition, the available functional analyses of COG0523 members suggest varied functions and an interaction with various metals [15,18,45]. We predict that the different gene clusters involving COG0523 represent distinct subgroups. In contrast to the HypB, UreG, and MeaB subfamilies, which are composed of chaperones for a single protein, each COG0523 subgroup may perform a chaperone role in different metallocenter biosynthesis of various proteins.

In summary, this analysis suggests that like HypB or UreG, COG0523 proteins are most certainly metal insertase and/or metallochaperones. However, the metal substrate and the metalloprotein target(s) of most COG0523 family proteins is not obvious, and the initial analysis of the COG0523 gene neighborhoods suggests that there could be a greater diversity of targets than observed for the other G3E family subgroups. To investigate the presence of diverse COG0523 subfamilies, we combined literature analysis (Table (Table22 and Additional File 1) with predictions of Zur and B12 regulation (see below) and phylogenetic and gene neighborhood analyses. This approach led to the identification of fifteen subfamilies (summarized in Figure Figure1,1, detailed in Additional Files 2 and 3). Each subfamily is monophyletic and the corresponding genes belong to similar genomic neighborhoods and/or share conserved regulatory sites. Two exceptions are subfamilies 1 (Figure (Figure1C)1C) and 5 (Figure (Figure1B),1B), which appear to be paraphyletic; the clade composed of subfamily 1 also contains subfamily 2 and the clade that contains subfamily 5 also contains subfamily 4 (Additional File 2). Five subfamilies (1, 2, 5, 12 and 13) are analyzed in more detail below. The 10 others are detailed in Additional File 3.

Table 2
Literature reports of COG0523 mutant data.

The CobW subfamily involved in cobalamin biosynthesis

CobW was the first member of COG0523 to be described and so-named because the disruption of the corresponding gene in Pseudomonas denitrificans resulted in the inability to synthesize cobalamin (Table (Table2)2) [15]. Although "cobalamin biosynthesis protein" is the most highly propagated annotation for COG0523 members, our comparative genomic and phylogenetic analysis reveals that true CobW proteins (Subgroup 12, Figure Figure1)1) represent only 12.5% of the COG0523 family (for a list of putative cobW genes, see 'Coenzyme B12 biosynthesis' subsystem). In our previous genomic analysis, cobW genes were identified in Proteobacteria located within the cobalamin biosynthesis gene clusters under control of the B12 riboswitch, a regulatory RNA element modulating gene expression in response to changing B12 concentrations [49]. As many more genome sequences have become available, we updated this analysis and report that 54 out of 65 cobW orthologs analyzed belong to B12-regulated gene clusters in γ-, β-, and α-proteobacteria (Additional File 4). In three α-proteobacteria from the Rhodospirillaceae family, cobW genes belong to the cobalamin biosynthesis gene clusters that are not preceded by B12 riboswitches. Finally, cobW orthologs in cyanobacteria are neither clustered with B12 biosynthesis genes nor regulated by a B12 riboswitch. However, these orthologs are highly similar to other CobW proteins and the corresponding genes co-occur with the cobalamin biosynthesis genes of the aerobic pathway.

In the majority of cases, cobW is found adjacent to the cobalt chelatase component, cobN (Figure (Figure11 and Additional File 3) and all CobW proteins analyzed contain a His-stretch, which on average is composed of 7 histidines (the least being 4 histidines and the most being 15). The exact function of CobW is still not clear; it could be involved in the presentation of the cobalt ion to the cobalt chelatase, protection of the cofactor, or involved in inserting a metal in a metal-dependent enzyme of the pathway, such as Fe-dependent CobG [50].

The Nitrile hydratase activator subfamily

Based on our analysis, less then 0.7% of the COG0523 family is represented by the nitrile hydratase (NHase) activators (Subgroup 2, Figure Figure1).1). A complete list of identified Fe-type NHase activators from both Genbank and SEED databases is given in Additional File 3. In the literature, these proteins are referred to as Nha3, P44K, or P47K, depending on the organism in which the protein was identified (Additional File 3). Here we refer to this subgroup of COG0523 as Nha3. Nha3 is found clustered exclusively with the genes encoding the two subunits of the Fe-type NHase (Figure (Figure11 and Additional File 3) and has been found to be required for the in vivo activity of Fe-type NHase [16]. NHases are enzymes that use either a non-heme iron (III) or non-corrin cobalt (III) for the hydration of nitriles to amides [51]. NHase types can be differentiated by the strictly conserved metal binding motifs CSLCSCT for Fe(III) and CTLCSCY for Co(II) [52]. Although, the same coordination geometry has been determined for both Co(III)- and Fe(III)-binding sites [53], the two types of NHases specifically incorporate the correct metal. This specificity is thought to be due to activator proteins, which are required for full activity of their respective NHase. For Co-type NHases, metallocenter biosynthesis is thought to occur via subunit exchange, a mechanism called "self-subunit swapping" [54,55]. The accessory protein in this case, NhlE, is a self-subunit swapping chaperone and the corresponding gene is always found adjacent with the target NHase genes (Additional File 3). No sequence similarity is found between the Co-type accessory protein and the Fe-type accessory protein supporting the conclusion that Co- and Fe-type metallocenters are assembled by different mechanisms.

Even if the involvement of Nha3 in Fe-type NHase activation is documented, its exact role is not known. It has been postulated that it has an insertase role involved in incorporation of iron into the active site of the hydratase [45]. When the Fe-dependent NHase from Rhodococcus sp. N-771 was expressed in E. coli without Nha3 in Co-supplemented media, it incorporated Co instead of Fe [56]. Therefore, nitrile hydratase activator proteins may not only be involved in incorporating Fe, but also in ensuring that competing metal ions are excluded. In addition, the coexpression of Nha3 with NHase was found to be unnecessary with the coexpression of the GroESL chaperones [57]. This observation supports the hypothesis that COG0523 proteins like the rest of the G3E family could be involved in the structural rearrangements that must take place to ensure the metal cofactor is incorporated into the catalytic site.

Zur-regulated COG0523 proteins

Extensive analysis of the literature (Table (Table22 and Additional File 1) reveals that members of COG0523 have been implicated in the virulence of several pathogens whose hosts are known to induce Zn-limitation as a defense strategy. In 1973, Kochan introduced the concept of nutritional immunity as a defense strategy against invading pathogens [58]. The host organism actively deprives metals from the invaders inducing both hypoferremia and hypozincemia (deficiency of iron and zinc, respectively, in the blood) as part of the acute inflammatory response [59-61]. Therefore, the mechanisms that enable a pathogen to overcome this host-induced Zn-starvation are considered essential to a pathogen's ability to cause infection [62-64]. In Mycobacterium tuberculosis, a COG0523-like gene, RV0106, (shown also to be repressed by Zur [65]) is up-regulated during human macrophage infection [66] (although RV0106 shows homology to COG0523, it is missing both GTPase motifs, and the second cysteine of the CXCC motif is not conserved). In the closely related Mycobacterium avium subsp. paratuberculosis, this gene is found on a pathogenicity island [66,67] and the corresponding protein was the second strongest antigen consistently reactive with cattle sera infected with M. avium or Micobacterium bovis [68]. COG0523 is also found in a pathogenicity island from Enterococcus faecalis [69]. The loss of COG0523 in Brucella suis rendered this bacterium incapable of intramacrophagic replication [70], while the loss of COG0523 in Burkholderia pseudomallei results in the inability to infect Caenorhabditis elegans (Table (Table2)2) [71]. An ortholog from Francisella tularensis was expressed exclusively in bacteria separated from infected murine spleen tissue [72]. This gene is down-regulated in the Francisella novicida ΔpmrA mutant [73]. PmrA is a transcription factor found to be essential for survival/growth inside human and murine macrophage cell lines [73].

In plants, an opposing defense strategy may be employed, as repression of zinc uptake machinery is required for full virulence of the plant pathogens, Xanthomonas campestris and Xanthomonas oryzae [74-76]. In contrast to animal pathogens and further supporting a role for COG0523 in zinc homeostasis, two COG0523 homologs of Agrobacterium tumefaciens as well as the genes encoding the high-affinity zinc transporter, ZnuABC, are down-regulated in response to plant signals (Additional File 1) [77].

The most extensive analysis of the zinc-dependent regulation of a COG0523 gene has been performed on the COG0523 (yciC) in Bacillus subtilis encoding a member of subfamily 1 (Figure (Figure11 and Additional File 3). The expression of yciC is controlled by the Zn-dependent Zur repressor and is thus up-regulated under Zn-limiting conditions [17]. In addition to the work on yciC, an early comparative genomic analysis had identified Zur-regulated yciC orthologs in several Gram-positive bacteria (Bacillus, Staphylococcus, Enterococcus) [62]. As this initial analysis of putative Zur-binding sites had been done when a limited set of genomes was available and, as discussed above, scattered observation links this family to zinc limitation, we expanded the analysis to all currently complete bacterial genomes (see Methods).

Sixty-eight yciC/COG0523 genes were found to be downstream of a potential Zur-binding site mainly in Firmicutes and γ-, β-, and α-proteobacteria (Additional File 5). Two COG0523 genes were found downstream of a putative Zur site in the cyanobacteria, Prochlorococcus marinus, Nostoc sp. PCC 7120 and several Cyanothece species (Additional File 5). While most proteins encoded by Zur-regulated COG0523 members are found in subfamily 1 (75%), several paralogs are found in other subfamilies. For instance, in Pseudomonas entomophila, Pseudomonas fluorescens, and Pseudomonas putida there are two COG0523 homologs per genome predicted to be downstream of a Zur-binding site (Additional File 5). Our phylogenetic analysis reveals that one paralog belongs to subfamily 1 while the other belongs to subfamily 11 (Additional File 2). Zur-regulated COG0523 paralogs are also found in subfamily 5, 8, 10 and 14 (Additional File 2 and Additional File 5). One possibility is that the presence of several Zur-regulated COG0523 subfamilies could be indicative of more than one function of COG0523 under zinc limitation (as discussed below).

Of the three COG0523 genes in Acinetobacter baylyi ADP1, only one is regulated by Zur

To test the predictive power of our COG0523 phylogenomic analysis, the regulation of the three COG0523 genes from Acinetobacter baylyi ADP1 was analyzed. The first, ACIAD1614 [Genbank: ACIAD1614, 49530751], is predicted to be a nha3 homolog (subfamily 2). The second, ACIAD1025 [Genbank: ACIAD1025, 49530203], is predicted to be most similar to subfamilies 10 and 11. The third, ACIAD1741 [Genbank: ACIAD1741, 49530869], is predicted to be regulated by Zur (subfamily 1). These groupings suggest that the expression of only the ACIAD1741-encoding gene should be under Zur control. Expression of the three A. baylyi COG0523 genes was analyzed by RT-PCR (see Material and Methods) in a WT strain and in a Δzur derivative (ΔACIAD0176). As shown in Figure Figure3,3, the presence or absence of Zur does not affect the expression of ACIAD1025 and ACIAD1614 under the conditions tested. Nevertheless, as we predicted, ACIAD1741 is only expressed in Δzur background.

Figure 3
Differential expression of the three COG0523 genes of Acinetobacter baylyi ADP1. Amplification of ACIAD1741 (lane 2, 3, 4 and 5), ACIAD1614 (lane 6, 7, 8 and 9) and ACIAD1025 (lane 10, 11, 12 and 13) transcripts from A. baylyi ADP1 (WT) and A. baylyi ...

Identification of putative Zur-regulated genes encoding paralogs of Zn-dependent enzymes

Genome context analysis revealed that a significant proportion of the yciC genes (Zur-regulated COG0523) are located within chromosomal gene clusters including genes for zinc transport (e.g., znuABC), the zur repressor, and various genes encoding paralogs of zinc-dependent proteins (Figure (Figure44 and and5).5). Nine families of Zn-dependent enzymes whose paralogs belong to Zur regulons in γ-, and β-proteobacteria (Figure (Figure5)5) were found. These Zn-dependent enzymes include phosphoribosyl-AMP cyclohydrolase (HisI), dihydroorotase (PyrC), γ-class carbonic anhydrase (Cam), porphobilinogen synthase (HemB), cysteinyl-tRNA synthetase (CysRS), threonyl-tRNA synthetase (ThrRS), N-acetylmuramoyl-L-alanine amidase, queuosine biosynthesis enzyme QueD, and C4-type zinc finger regulator DksA (see 'Zinc regulated enzymes' subsystem).

Figure 4
Representative gene clusters composed of Zur-regulated COG0523 members. Genes labeled yciC represent subfamily 1 COG0523 members. COG0523-11 and COG0523-8 refer to subfamilies 11 and 8, respectively. Abbreviations not found in text: DUF1826, Pfam protein ...
Figure 5
Genomic co-localization of genes belonging to putative Zur regulon. Genomic distribution, candidate Zur-dependent regulation and genomic co-localization of genes encoding Zur, COG0523 and Zn-dependent enzymes and their paralogs in γ- and β-proteobacteria. ...

Differential regulation of distinct isofunctional genes by co-factor availability is a known regulatory mechanism in bacteria (for a review, see [78]) and in eukaryotes (as discussed below). For instance, the coenzyme B12-independent isozymes of methionine synthase and ribonucleotide reductase are regulated by B12 riboswitches in the genomes that encode both B12-dependent and -independent isozymes [49]. Likewise, a similar regulatory strategy has been described for zinc availability. Zn-independent proteins are negatively regulated by Zur and expressed under Zn-limiting conditions to replace the Zn-dependent proteins. Examples include paralogs of ribosomal proteins [62] and alternative isozymes of GTP cyclohydrolase I (FolE1 and FolE2) [79]. In both cases, a Zn-dependent protein is functionally replaced by a Zn-independent counterpart during conditions of zinc deficiency.

Our comparative analysis of Zur regulons revealed co-regulation and frequent co-localization on the chromosome between COG0523 and paralogs of these Zn-dependent enzymes. For example, Cupriavidus metallidurans has a Zur-regulated gene cluster encoding YciC, FolE2, and paralogs of CysS, QueD, Cam, and PyrC, whereas the Zur regulon in Pseudomonas fluorescens includes two COG0523 proteins, FolE2, and paralogs of AmiA, DksA, HisI, Cam, and PyrC (Figure (Figure44 and and55).

We hypothesize that these alternative enzymes could require a metal other than Zn (or no metal) and are therefore expressed during Zn-limitation to replace or compensate for the decreased activity of their Zn-dependent counterparts. Indeed, the carbonic anhydrases found in our analysis are members of the γ-class. The γ-class carbonic anhydrase from Methanosarcina thermophila exhibited highest activity with Fe and, when purified under anaerobic conditions, contained Fe and not Zn [80,81]. The Zur-regulated cam we have identified could therefore encode an Fe-dependent carbonic anhydrase expressed to compensate for the Zn-dependent carbonic anhydrases.

The proteins of three other families downstream of putative Zur binding sites are missing the conserved zinc binding residues. As shown in Figure Figure55 and Additional File 6, some genomes encode three PyrC paralogs. One paralog is similar to the dihydroorotase from Escherichia coli. These proteins have a binuclear zinc center chelated by the conserved metal binding residues His 16, His 18, Lys 102, Asp 250, His 139 and His 177 [82]. The second PyrC paralog is an inactive dihydroorotase, which is referred to in the literature as PyrC' [83]. Similar to PyrC', the zinc-binding residues are not conserved in the PyrC paralog whose gene we predict to be regulated by Zur. However, unlike PyrC', this PyrC paralog has previously been shown to display dihydroorotase activity [84]. For porphobilinogen synthase, the existence of zinc binding and non-zinc binding variants is documented in the literature [85]. As expected, HemB1 contains the zinc chelating cysteine ligands while those cysteines are not conserved in the protein, HemB2, encoded by the gene putatively regulated by Zur in Pseudomonas putida (Additional File 6). HemB2 we would accordingly expect to be active with magnesium and/or potassium instead of zinc. In addition, the DksA paralogs downstream of putative Zur binding sites are missing the canonical C4-zinc finger motifs (Additional File 6).

Not all paralogs seem to have lost their zinc-binding sites as the zinc-binding residues are conserved in the HisI, CysRS, ThrRS, QueD, and AmiA paralogs encoded by genes predicted to be induced during zinc depletion. An alternative explanation for the existence of these paralogs could be to increase protein copy number during zinc deficiency. The analysis of the metal content of some paralogs identified in this study is currently underway.

COG0523 in eukaryotes: Two Chlamydomonas reinhardtii COG0523 homologs are induced under zinc limitation

COG0523 is widespread in eukaryotes, with most organisms containing one to four homologs (see 'G3E'subsystem), and have been associated with stress phenotypes (see Table Table22 and Additional File 1). In Arabidopsis thaliana, one of the three COG0523 genes (At1 g80480 [Genbank: AT1g80480, 51536562]), which was isolated as a member of the actively-transcribed plastid chromosome in mustard seed [86], is induced under heat-stress [87]. Deletion of COG0523 from Saccharomyces cerevisiae, YNR029c [Genbank: 6324356], confers sensitivity to the metal chelator, glycol-bis (2-aminoethylether)-N,N,N',N'-tetraacetic acid (EGTA) [88] as well as salt-sensitive and heat-sensitive phenotypes [89] (Table (Table22).

Gene clustering is not very informative in eukaryotes but most eukaryotic COG0523 homologs including Homo sapiens belong to subfamily 5 (Figure (Figure6).6). The prokaryotic members of subfamily 5 cluster on the genome with genes that encode WD40-repeat proteins, which form a β-propeller structure thought to mediate protein-protein interactions [90], and with znuABC and creatinase encoding genes (Additional File 3). Several prokaryotic members of subfamily 5 are also predicted to be downstream of a Zur-binding site (Additional File 5), suggesting a role for bacterial members of subfamily 5 in the response to zinc limitation.

Figure 6
Phylogeny of eukaryotic COG0523 members. Lightly shaded tree represents collapsed COG0523 tree. Branches are labeled with corresponding subfamily number. Subfamily 5 Tree: branches representing eukaryotic homologs are colored. Blue diamonds indicate putative ...

Little if any work has been performed on the role of COG0523 in eukaryotes, which do not encode a Zur homolog. Therefore, we sought to investigate the regulation of COG0523 during metal depletion in a eukaryotic reference organism. Previous studies have established the alga Chlamydomonas reinhardtii as a choice reference organism for the study of trace metal homeostasis because it is straightforward to deplete the medium of zinc, copper, iron or manganese (as seen in [91-93]). Sentinel genes for each of these deficiencies are known, such as CYC6 for copper deficiency, FOX1 for iron-deficiency, NRAMP1 for manganese deficiency and ZRT3 for Zn-deficiency [92,94,95]. Furthermore, Chlamydomonas has retained many pathways present in the common ancestor to the plant and animal lineages and displays the metabolic flexibility of "back up" or alternate systems [96,97]. For instance, the replacement of B12-independent methionine synthase with a B12-dependent form when this cofactor is available, the use of Mn-dependent superoxide dismutase (SOD) in place of Fe-SOD in iron-limitation and the replacement of plastocyanin with cytochrome c6 in copper-deficiency [93,98,99].

We identified 15 genes encoding proteins with COG0523 domains in versions 3.1 and 4.0 of the C. reinhardtii draft genome [96] (see Additional File 7 for protein IDs). Of these 15 gene models, only 10 gene models encoded full-length COG0523 GTPase domains. Therefore, we tested the expression of these 10 genes as a function of Zn, Cu, Fe and Mn nutrition. Transcripts for two of these, encoding proteins 123019 and 117458 (version 3.1 protein IDs), are increased in abundance by several orders of magnitude when cells are grown in zinc-limiting conditions as opposed to zinc-replete conditions, only slightly induced under copper limitation relative to copper-replete conditions and unaffected by iron or manganese nutrition (Figure (Figure7).7). The zinc sensors and regulatory factors responsible for mediating this response to zinc depletion ARE yet unknown in C. reinhardtii.

Figure 7
Differential expression of the COG0523 genes of Chlamydomonas reinhardtii. Chlamydomonas strain 2137 was grown under various metal-deficient or replete conditions in triplicate experiments, represented by squares, circles and triangles. RNA was isolated ...

Our phylogenetic analysis reveals that protein 123019 belongs to subfamily 1, while protein 117458 belongs to subfamily 5 (Figure (Figure66 and Additional File 2). We therefore substantiate the role of COG0523 family members in Zn homeostasis in eukaryotes as well as in bacteria. Several COG0523 proteins encoded by eukaryotic genomes belong to subfamily 5 (Figure (Figure6).6). We predict that the expression of some of these other eukaryotic COG0523 proteins may also be regulated by zinc.

In addition, supporting the functional diversity revealed by our gene neighborhood analysis, the expression of the eight other COG0523 family members from C. reinhardtii are not significantly affected by the deficiency of metals tested.

COG0523 in Archaea

Although COG0523 was previously assumed to be missing from Archaea [13], the availability of recently sequenced genomes reveals that out of 44 archaeal genomes in the SEED database, eight genomes contain at least one COG0523 homolog, with Methanosarcina acetivorans C2A containing eight homologs. Most Archaeal members belong to subfamily 13, members of which co-localize with corrinoid-dependent methyltransferases (Figure (Figure11 and Additional File 3). In Methanosarcina barkeri, Methanosarcina acetivorans, Methanosarcina mazei, and Methanococcus maripaludis S2, COG0523 clusters with genes involved in methanol:CoM methylation: mtaA, mtaB, (both are Zn-dependent [100]), mtaC (corrinoid protein [101]) and ramM (iron-sulfur protein [102]) (Figure (Figure11 and Additional File 3). Clustering between COG0523 and methanogenesis genes is not limited to Archaea but also found in Clostridium botulinum (Additional File 3). Another clostridium, Desulfitobacterium hafniense DCB-2, encodes a COG0523 that clusters with a MeTr homolog (methyltetrahydrofolate:corrinoid/iron-sulfur protein methyltransferase) (Additional File 3). Also, proteome analysis of acetate- and methanol-grown M. acetivoran scells revealed the presence of MA4382 (COG0523) in methanol-grown cells [103]. Finally, MM1072 (M. mazei COG0523) is induced to the same extent as its neighboring ramM homolog, MM1071, during growth in high salt conditions (2.38 and 2.21 fold, respectively) [104].

Archaeal genomes sequenced to date lack any recognizable homolog of the Fur (Fe) or Zur (Zn) transcriptional regulators. Alternatively, there is a large group of MntR/DtxR-like regulators in Archaea (Methanosarcina, Pyrococci, Archaeaglobus, Methanobacterium) that regulate iron homeostasis, whereas another small group of MntR/DtxR-like repressors in Methanosarcina spp. named ZntR (e.g. MA0022 in M. acetivorans), is predicted to regulate the zinc uptake operon, znuABC (D.A.R., unpublished observation, see Additional File 8). Comparative genomic analysis of this novel zinc regulon in M. acetivorans reveals that a COG0523 homolog, MA4381, is co-regulated with znuABC based on the presence of candidate-binding sites of the ZntR repressor (Additional File 8).

Conclusion

COG0523 is a diverse family of metal chaperones

Based on relatedness to the G3E family of GTPases, we expect COG0523 to also be involved in metallocenter biosynthesis of target metalloproteins. The diversity of genomic co-localization suggests that COG0523 is more diverse than the other subfamilies of G3E. Both the metal specificity and the protein target(s) might vary from one subgroup to another.

While the known roles in cobalamin biosynthesis and response to zinc limitation predominate, our analysis implies members of COG0523 are not limited to those roles. Based on genome context (co-localization and/or presence of a B12 riboswitch) and protein similarity analyses, only 12.5% of sequenced COG0523 from the SEED are true CobW proteins and assigned to the cobalamin biosynthesis pathway. Only ~30% of COG0523 members analyzed are linked to the zinc homeostasis either through putative Zur sites (~8%) or co-localization with genes involved in the response to zinc starvation (~20%). In addition the third known role, NHase activator, only applies to less than 1% of sequenced COG0523 genes. Over half of COG0523 may perform a role in the activity of unknown proteins.

A ubiquitous subset of COG0523 is linked to zinc

Although involvement in the response to zinc deficiency applies to only a subset of COG0523, we show that this function is not limited to Bacteria but also present in Archaea and Eukaryota. Two C. reinhardtii COG0523 homologs, which belong to separate phylogenetic subgroups, are induced under zinc-deficient conditions compared to zinc-replete conditions. In addition, the expression of the eight other homologs was not significantly affected by metal deprivation confirming the diversity of COG0523.

Identification of novel zinc homeostasis mechanisms

The comparative genomic analysis of the zinc repressor Zur regulons in Bacteria has revealed insights into previously unknown zinc homeostasis mechanisms. At least nine protein families that are homologs or isozymes of known Zn-dependent proteins were identified as candidate members of the reconstructed Zur regulons in γ- and β-proteobacteria, suggesting their up-regulation during zinc limitation. Based on sequence analysis, four of these protein families do not contain the canonical zinc binding residues. We propose, therefore, that these paralogs may require a metal other than Zn for catalysis and are involved in the adaptation to poor zinc nutrition. The presence of these paralogs could aid in compensating for the loss in activity of the Zn-dependent protein analogs and reducing in the total amount of Zn required by the cell.

Putative roles of COG0523 in response to zinc limitation

At this stage, if the exact role of COG0523 members in survival in low zinc conditions is still to be determined, several hypotheses can be proposed. Our comparative genomic analysis suggests that COG0523 may be a metal chaperone for a protein that is also part of the Zur regulon. The 'alternative enzymes' of Zn-dependent proteins may require a metal other than zinc for catalysis and may also require a metal chaperone for efficient cofactor acquisition. However, in about half of the genomes analyzed, yciC (Zur-regulated COG0523) appears to not belong to any operon. For instance, although adjacent to folE2, yciC is usually regulated by Zur independently (Figure (Figure55).

Another possibility would be that COG0523 is involved in the allocation and reallocation of zinc. Zinc is not an essential cofactor for metabolic steps where zinc-independent back-up proteins can substitute. Accordingly, in conditions of poor zinc nutrition, we expect that zinc delivery is prioritized to proteins that do not have zinc-independent substitutes (and where zinc function is hence essential). Induction of the C. reinhardtii genes, 123019 and 117458, containing the putative metal delivery COG0523 domain, may affect prioritized delivery to a subset of zinc targets. These delivery factors might be particularly important in a compartmentalized eukaryotic cell. In bacteria, COG0523 may also function as either a zinc chaperone as proposed recently by Gabriel et al. [105] or as a molecular chaperone that aids in the folding of essential zinc metalloproteins ensuring that essential Zn-proteins acquire zinc while nonessential proteins are excluded (the possible existence of a zinc metallochaperone is discussed in the recent review [106]).

As a third hypothesis, some members of COG0523 may be a chaperone involved in incorporating a metal other than zinc into Zn-dependent enzyme(s) based on zinc availability. In vitro, the activity of several Zn-dependent enzymes is slightly less, the same, or in some cases higher with a metal cofactor other than zinc (for recent examples see [107-109]). Under zinc limitation and supplementation with cobalt, the zinc in carbonic anhydrase of the marine diatom, Thalassiosira weissflogii, is substituted with cobalt in vivo [110]. The genome of the closely related Thalassiosira pseudonana encodes seven COG0523 proteins. Interestingly, the genomes of cyanobacteria and algae tend to encode relatively high numbers of COG0523 proteins. Zinc-containing carbonic anhydrases are important for assimilation of CO2, and algae tend to express multiple isoforms in various organelles [111], which might require mechanisms for preferential metal delivery. The symbiotic alga Chlorella sp. NC64A has twelve homologs and the free-living Chlorella vulgaris C-169 has seven. The genome of Micromonas sp. RCC299 encodes ten COG0523 homologs and the smallest known free-living eukaryote, Ostreococcus tauri, has four homologs. As stated above, the cyanobacterium Cyanothece sp. ATCC 51142 has eleven homologs, while Anabaena variabilis has five and Nostoc sp. PCC7120 and Prochlorococcus marinus susp. marinus both have four homologs. This high number of paralogs might reflect their particular lifestyles.

Lastly, there is some evidence from Magnetospirillum magneticum AMB-1 that a MeaB homolog may function as a cytoplasmic ATPase required for energizing Fe uptake [112]. Indeed, neither MeaB homolog encoded in the M. magneticum genome appears to be co-transcribed with a gene encoding methylmalonyl-CoA mutase (also the case for ~40% of meaB homologs; see 'G3E' subsystem). Therefore, a role for some COG0523 members in affecting metal transport cannot be ruled out at this point.

Further experimental work is now required to discriminate between these different potential roles. To complicate the problem, there are up to three Zur-regulated COG0523 paralogs in some genomes, therefore, a combination of the above functional hypotheses may prove to be operational.

Methods

Comparative genomic analysis of COG0523 gene family and G3E family

Analysis of 'COG0523,' 'G3E,' and 'Zinc regulated enzymes' subsystems were performed in the SEED database [23,36]. COG0523 gene sequences in the SEED database were identified by homology to known COG0523 members and the presence of the conserved CXCC motif and P-loop GTPase domain in the corresponding protein sequences. cobW gene sequences were identified based on homology to cobW from Pseudomonas denitrificans [15] and occurrence within cobalamin biosynthesis operons and/or downstream of a putative B12 riboswitch. Genomic search for candidate B12 riboswitches was performed as previously described [49].

Sequence Analysis

All COG0523, HypB, UreG, and MeaB sequences presented here were downloaded from the SEED or Genbank [113] databases. The fig numbers (internal identifications in the SEED) and Genbank accession numbers can be found in Additional File 7. Identification of histidine motifs was performed with Fuzzpro from the EMBOSS software package [114]. Amino acid sequences were aligned using the ClustalW2 algorithm with default parameters [115]. For alignments of PyrC, DksA and HemB, ESPript 2.2 was also used [116]. Phylogenetic analyses were carried out by employing the Phylip 3.67 program package [117]. Distance-based matrices were generated between all pairs of sequences using the Jones-Taylor-Thornton matrix as employed in Protdist (Phylip). Phylogenetic trees were generated from these matrices using the neighbor-joining method as implemented in Neighbor (Phylip). Reliability of branches was determined with the bootstrap method of 1000 replicates using Bootseq (Phylip).

For the G3E family distance tree, the GTPase domain was extracted and aligned. The GTPase domain of CooC was used as an outgroup. Although it has been previously assumed that CooC is a member of the G3E family [14,29], GTPase sequence motifs suggest that it is actually a member of the closely related MinD/BioB family. The COG0523 distance tree was built with 177 full-length COG0523 sequences. Significant gene clusters between prokaryotic COG0523 genes and neighboring genes were identified in the SEED database. Members of COG0523 with the highest functional coupling score for each significant gene cluster were chosen for inclusion in the phylogenetic analysis. Functional coupling scores and significant gene clusters were computed by the SEED database. For an explanation of functional coupling scores refer to [118]. In addition, COG0523 proteins whose genes were identified through our analysis of the Zur regulon were also included. The COG0523 proteins from six eukaryotes were also used including the 10 COG0523 homologs from C. reinhardtii whose transcript levels were investigated by real-time PCR. RV0106 from Mycobacterium tuberculosis CDC1551 was used as an outgroup in this analysis. This protein, while having similarity to COG0523, does not contain the canonical CXCC motif (CXSC). In addition, it is missing the canonical Walker A motif of the GTPase domain, suggesting that these COG0523-like proteins do not have GTPase activity. Subfamilies were defined based on the following criteria. Each subfamily had to be monophyletic. The exceptions are subfamilies 1 and 5, which are paraphyletic. Subfamily 1 becomes monophyletic with the subtraction of the nitrile hydratase activators. Subfamily 5 becomes monophyletic with the subtraction of subfamily 4. The genes encoding proteins in each subfamily belong to similar gene clusters and/or have shared regulatory sites. Bootstrap values were all above 900. Tree illustration was performed with Treedyn [119].

Plot of amino acid conservation

Thirty-two protein sequences representing the 15 COG0523 subfamilies were aligned using ClustalW2 and default parameters. Accession numbers for sequences used can be found in Additional File 7. Columns containing gaps in eight or more sequences were removed. Residue conservation at each position as determined by Jalview was plotted [120].

Comparative genomic analysis of Zur regulons

Complete bacterial genomes were downloaded from GenBank [113]. The taxon-specific training sets for identification of Zur-binding motifs were composed of the previously identified Zur-binding sites in Firmicutes, α-, β-, and γ-proteobacteria [62] and the DNA motif search profiles were constructed using the SignalX program. Analyzed genomes encoding both an ortholog of Zur regulator and COG0523 proteins were scanned with the constructed taxon-specific Zur motif profiles (see Additional File 9 for sequence logo motifs) using the Genome Explorer software and the identified genes with candidate Zur-binding sites were analyzed by the consistency check comparative procedure as previously described [78]. Candidate ZntR-binding motif was obtained by applying the SignalX program to the training set of the znuACB regulatory regions from methonogenic Archaea that have an ortholog of the DtxR-like regulator ZntR (MA0022). Positional nucleotide weights in the recognition profile and Z scores of candidate sites were calculated as the sum of the respective positional nucleotide weights as described in [121]. The threshold for the site search was defined as the lowest score observed in the training set. Sequence logos for DNA-binding sites were constructed using WebLogo 2.0 [122].

Acinetobacter RT-PCR

Acinetobacter baylyi ADP1 (ADP1) Δzur:kanR was a generous gift from Véronique de Bernardinis (Genoscope, Institut de Génomique (CEA), Evry, France) [123]. Overnight cultures of ADP1 and Δzur:kanR cultured in Luria Broth was used to inoculate 5 ml culture of Luria Broth supplemented with 50 μM ZnSO4 to ensure repression of transcription by Zur. Samples (1 ml) were harvested in early stationary phase, RNAprotect Bacteria Reagent (Qiagen) was added and cells were frozen at negative 80°C overnight. Pellets were thawed and RNA was extracted using TRIzol LS reagent (Invitrogen) followed by RNeasy mini kit (Qiagen). Contaminating DNA was removed using DNase I (RNA-free) (Ambion). RT-PCR reactions were carried out with Superscript™ III One-Step RT-PCR System with Platinum®Taq High Fidelity (Invitrogen). Reactions were composed of 7.5 μl 2× reaction mix, 1 μl RNA (200 pg RNA), 0.3 μl Forward Primer (10 μM), 0.3 μl Reverse Primer (10 μM), 0.3 μl Superscript™ III RT/Platinum® Taq High Fidelity enzyme mix, and water to a final volume of 15 μl. Reverse transcriptase minus controls were performed using 5 PRIME Taq Master Mix (Fisher). Reactions were composed of 6 μl 5 PRIME Master Mix, 1 μl RNA (200 pg RNA), 0.3 μl Forward Primer (10 μM), 0.3 μl Reverse Primer (10 μM), and water to a final volume of 15 μl. Growth of strains and RT-PCR were performed in experimental triplicate. Primer sequences used in this analysis are available in Additional File 10.

Chlamydomonas RNA analysis

Cultures of Chlamydomonas reinhardtii wild-type strain 2137 were maintained in aerated Tris-Acetate-Phosphate (TAP) medium with shaking in the light (60-100 μmol m-2 s-1). To characterize Zn-responsive gene expression, cells were initially grown to late exponential phase in TAP supplemented with 2.5 μM zinc, followed by a round of growth with no supplemental zinc, before they were inoculated into the experimental conditions at a density of 105 cells/mL. Characterization of the effects of copper, iron, and manganese was performed similarly with the exception that cultures were grown under a second round of metal deficiency prior to the experiment. The iron deficient concentration used was 1 μM [95]. When cultures reached mid- to late-exponential phase, total RNA was prepared as described in [124]. cDNA preparation and real-time PCR were performed as described in [92]with CBLP used as the reference gene. All experiments were performed in experimental triplicate. Additionally, all RT-PCR analysis was performed in technical triplicate. Primer sequences are available in Additional File 10. MIQE checklist is available in Additional File 11.

List of abbreviations

MCM: methylmalonyl-CoAmutase; SVG: segmentally variable gene; NHase: nitrile hydratase.

Authors' contributions

CEH carried out and designed the phylogenetic, comparative genomic, and sequence analysis of COG0523 and the G3E family, the RT-PCR analysis in A. baylyi, the sequence analysis of the Zur-regulated back-up proteins, the literature search and drafted the manuscript. DAR carried out the analysis of putative Zur and ZntR binding sites and B12 riboswitches, analysis of putative back-up enzyme paralogs, and assisted in drafting the paper. JK prepared C. reinhardtii RNA from Cu, Fe, Mn and Zn-deficient cells, designed and tested primers for qRT-PCR, performed the qRT- PCR and assisted in drafting the paper. JK and DM established conditions for zinc-deficiency in Chlamydomonas, isolated RNA, designed primers for qRT-PCR, and assisted in drafting the paper. SSM and VDC designed experiments, analyzed data and assisted in drafting the paper. All authors read and approved the final manuscript.

Supplementary Material

Additional file 1:

Literature reports of COG0523 expression data.

Additional file 2:

Phylogeny of COG0523 subgroups. A, Each identified subgroup is shaded and labeled. The branches representing proteins encoded by putative Zur-regulated genes are marked with a black square. The branches representing C. reinhardtii COG0523 homologs encoded by the genes induced by zinc deficiency are marked with a green square. Branches representing the Pseudomonas paralogs discussed in the text are labeled. Protein IDs for each branch can be found in Additional File 7.

Additional file 3:

Detailed description of each COG0523 subfamily. For subfamilies 2-15, representative genomic context figures are given as is a list of the locus-tags for members of each subfamily identified by physical clustering to common genes.

Additional file 4:

cobW genes downstream of a putative B12 riboswitch. Presence/absence of a putative B12riboswitch upstream of cobW is shown as well as the first gene in each putative cobalamin-regulated biosynthesis operon. CobW locus tags are in bold. B12riboswitches previously reported by [49] are marked with an asterisk.

Additional file 5:

Putative Zur-binding sites in bacterial genomes. Candidate Zur-binding sites upstream of genes encoding COG0523 family proteins and paralogs of Zn-dependent enzymes in Bacilli, α-, β- and γ-proteobacteria and Cyanobacteria. Distance refers to the location of the putative Zur-binding site upstream from the first gene in each operon.

Additional file 6:

Sequence analysis of PyrC, DksA and HemB paralogs. Alignments of protein sequences encoded by genomes from Figure Figure4.4. A, Alignment of the three PyrC paralogs. The residues that chelate the α-metal are highlighted in blue while the β-metal ligands are highlighted in green. Lys102 serves as a ligand for both metals ions. B, DksA paralogs. C, HemB paralogs. Secondary structure as determined from the crystal structure of the Escherichia coli homolog in each case is given (PDB identifiers: 1J79 (PyrC), 1TJL (DksA), 1L6S (HemB)). For PyrC and DksA, the alignments show only the portion of the alignment containing the zinc binding residues. Columns containing the zinc chelating residues as determined from the crystal structure are highlighted in yellow. Genome abbreviations: EC, Escherichia coli; PE, Pseudomonas entomophila L48; PF, Pseudomonas fluorescens Pf-5; PA, Pseudomonas aeruginosa PAO1; ABa, Acinetobacter baylyi ADP1; VA, Vibrio alginolyticus 12G01; HC, Hahella chejuensis KCTC 2396; BCe, Burkholderia cenocepacia AU 1054; BC, Burkholderia cepacia R18194; CM, Cupriavidus metallidurans CH34l; AB, Alcanivorax borkumensis SK2; KP, Klebsiella pneumoniae MGH 78578 (on virulence plasmid pLVPK); AV, Azotobacter vinelandii; BPe, Bordetella pertussis Tohama I; BF, Burkholderia fungorum; BM, Burkholderia mallei ATCC 23344; BX, Burkholderia xenovorans LB400; BP, Burkholderia pseudomallei 1710b; RE, Ralstonia eutropha JMP134; RS, Ralstonia solanacearum GMI1000; MF, Methylobacillus flagelatus KT.

Additional file 7:

Protein IDs used in sequence analyses. Protein IDs used for various sequence analyses are organized by figure and correspond to the indicated database.

Additional file 8:

Putative ZntR-regulated genes in archaeal genomes. Candidate ZntR-binding sites upstream of genes encoding COG0523, ZntR and ZnuABC in Archaea.

Additional file 9:

Sequence logos for DNA-binding motifs for candidate Zinc regulators. The taxonomy-specific DNA motif logos were constructed using Zur- and ZntR-binding sites indentified for COG0523 and other zinc-responsive genes described in the Additional Files 5 and 7.

Additional file 10:

Primers used in transcription analyses. Sequences for primers used in the transcription analysis of A. baylyi and C. reinhardtii COG0523 homologs.

Additional file 11:

MIQE checklist for C. reinhardtii qRT-PCR.

Acknowledgements

We thank John Helmann for sparking our interest in the COG0523 family, Ross Overbeek for generating the initial clustering analysis data in SEED, Dirk Iwata-Reuyl for critical reading of the manuscript, Véronique de Bernardinis for sending A. baylyi mutants. This work was supported by the U.S. Department of Energy Grant (grant no. DE-FG02-07ER64498, to V. de C.-L), National Institutes of Health (GM42143, to S.M. and 1F32GM083562 to D.M.), the Sol Leshin Program for BGU-UCLA Academic Cooperation (to S.M.), and by the grant from the Russian Academy of Science (program "Molecular and Cellular Biology", to D.A.R.).

References

  • Andreini C, Banci L, Bertini I, Rosato A. Zinc through the three domains of life. J Proteome Res. 2006;5:3173–3178. doi: 10.1021/pr0603699. [PubMed] [Cross Ref]
  • Andreini C, Banci L, Bertini I, Elmi S, Rosato A. Non-heme iron through the three domains of life. Proteins. 2007;67:317–324. doi: 10.1002/prot.21324. [PubMed] [Cross Ref]
  • Andreini C, Banci L, Bertini I, Rosato A. Occurrence of copper proteins through the three domains of life: a bioinformatic approach. J Proteome Res. 2008;7:209–216. doi: 10.1021/pr070480u. [PubMed] [Cross Ref]
  • Johnson DC, Dean DR, Smith AD, Johnson MK. Structure, function, and formation of biological iron-sulfur clusters. Annu Rev Biochem. 2005;74:247–281. doi: 10.1146/annurev.biochem.74.082803.133518. [PubMed] [Cross Ref]
  • De Rienzo F, Gabdoulline RR, Menziani MC, Wade RC. Blue copper proteins: a comparative analysis of their molecular interaction properties. Protein Sci. 2000;9:1439–1454. doi: 10.1110/ps.9.8.1439. [PMC free article] [PubMed] [Cross Ref]
  • Kurtz DM. Oxygen-carrying proteins: three solutions to a common problem. Essays Biochem. 1999;34:85–100. [PubMed]
  • Finney L, O'Halloran T. Transition metal speciation in the cell: insights from the chemistry of metal ion receptors. Science. 2003;300:931–936. doi: 10.1126/science.1085049. [PubMed] [Cross Ref]
  • Waldron KJ, Robinson NJ. How do bacterial cells ensure that metalloproteins get the correct metal? Nat Rev Microbiol. 2009;7:25–35. doi: 10.1038/nrmicro2057. [PubMed] [Cross Ref]
  • Pufahl RA, Singer CP, Peariso KL, Lin SJ, Schmidt PJ, Fahrni CJ, Culotta VC, Penner-Hahn JE, O'Halloran TV. Metal ion chaperone function of the soluble Cu(I) receptor Atx1. Science. 1997;278:853–856. doi: 10.1126/science.278.5339.853. [PubMed] [Cross Ref]
  • Kuchar J, Hausinger RP. Biosynthesis of metal sites. Chem Rev. 2004;104:509–525. doi: 10.1021/cr020613p. [PubMed] [Cross Ref]
  • Leach M, Zamble D. Metallocenter assembly of the hydrogenase enzymes. Curr Opin Chem Biol. 2007;11:159–165. doi: 10.1016/j.cbpa.2007.01.011. [PubMed] [Cross Ref]
  • Kim JK, Mulrooney SB, Hausinger RP. Biosynthesis of active Bacillus subtilis urease in the absence of known urease accessory proteins. J Bacteriol. 2005;187:7150–7154. doi: 10.1128/JB.187.20.7150-7154.2005. [PMC free article] [PubMed] [Cross Ref]
  • Leipe DD, Wolf YI, Koonin EV, Aravind L. Classification and evolution of P-loop GTPases and related ATPases. J Mol Biol. 2002;317:41–72. doi: 10.1006/jmbi.2001.5378. [PubMed] [Cross Ref]
  • Hubbard PA, Padovani D, Labunska T, Mahlstedt SA, Banerjee R, Drennan CL. Crystal structure and mutagenesis of the metallochaperone MeaB: insight into the causes of methylmalonic aciduria. J Biol Chem. 2007;282:31308–31316. doi: 10.1074/jbc.M704850200. [PubMed] [Cross Ref]
  • Crouzet J, Levy-Schil S, Cameron B, Cauchois L, Rigault S, Rouyez MC, Blanche F, Debussche L, Thibaut D. Nucleotide sequence and genetic analysis of a 13.1-kilobase-pair Pseudomonas denitrificans DNA fragment containing five cob genes and identification of structural genes encoding Cob(I)alamin adenosyltransferase, cobyric acid synthase, and bifunctional cobinamide kinase-cobinamide phosphate guanylyltransferase. J Bacteriol. 1991;173:6074–6087. [PMC free article] [PubMed]
  • Nojiri M, Yohda M, Odaka M, Matsushita Y, Tsujimura M, Yoshida T, Dohmae N, Takio K, Endo I. Functional expression of nitrile hydratase in Escherichia coli: requirement of a nitrile hydratase activator and post-translational modification of a ligand cysteine. J Biochem. 1999;125:696–704. [PubMed]
  • Gaballa A, Helmann JD. Identification of a zinc-specific metalloregulatory protein, Zur, controlling zinc transport operons in Bacillus subtilis. J Bacteriol. 1998;180:5815–5821. [PMC free article] [PubMed]
  • Gaballa A, Wang T, Ye RW, Helmann JD. Functional analysis of the Bacillus subtilis Zur regulon. J Bacteriol. 2002;184:6508–6514. doi: 10.1128/JB.184.23.6508-6514.2002. [PMC free article] [PubMed] [Cross Ref]
  • Smith KF, Bibb LA, Schmitt MP, Oram DM. Regulation and activity of a zinc uptake regulator, Zur, in Corynebacterium diphtheriae. J Bacteriol. 2009;191:1595–1603. doi: 10.1128/JB.01392-08. [PMC free article] [PubMed] [Cross Ref]
  • Galperin MY, Koonin EV. Sources of systematic error in functional annotation of genomes: domain rearrangement, non-orthologous gene displacement and operon disruption. In Silico Biol. 1998;1:55–67. [PubMed]
  • Brenner SE. Errors in genome annotation. Trends Genet. 1999;15:132–133. doi: 10.1016/S0168-9525(99)01706-0. [PubMed] [Cross Ref]
  • Devos D, Valencia A. Intrinsic errors in genome annotation. Trends Genet. 2001;17:429–431. doi: 10.1016/S0168-9525(01)02348-4. [PubMed] [Cross Ref]
  • Overbeek R, Begley T, Butler RM, Choudhuri JV, Chuang HY, Cohoon M, de Crécy-Lagard V, Diaz N, Disz T, Edwards R, et al. The subsystems approach to genome annotation and its use in the project to annotate 1000 genomes. Nucleic Acids Res. 2005;33:5691–5702. doi: 10.1093/nar/gki866. [PMC free article] [PubMed] [Cross Ref]
  • Glasner JD, Liss P, Plunkett G, Darling A, Prasad T, Rusch M, Byrnes A, Gilson M, Biehl B, Blattner FR, et al. ASAP, a systematic annotation package for community analysis of genomes. Nucleic Acids Res. 2003;31:147–151. doi: 10.1093/nar/gkg125. [PMC free article] [PubMed] [Cross Ref]
  • Mao X, Cai T, Olyarchuk JG, Wei L. Automated genome annotation and pathway identification using the KEGG Orthology (KO) as a controlled vocabulary. Bioinformatics. 2005;21:3787–3793. doi: 10.1093/bioinformatics/bti430. [PubMed] [Cross Ref]
  • Vallenet D, Labarre L, Rouy Z, Barbe V, Bocs S, Cruveiller S, Lajus A, Pascal G, Scarpelli C, Médigue C. MaGe: a microbial genome annotation system supported by synteny results. Nucleic Acids Res. 2006;34:53–65. doi: 10.1093/nar/gkj406. [PMC free article] [PubMed] [Cross Ref]
  • Padovani D, Labunska T, Banerjee R. Energetics of interaction between the G-protein chaperone, MeaB, and B12-dependent methylmalonyl-CoA mutase. J Biol Chem. 2006;281:17838–17844. doi: 10.1074/jbc.M600047200. [PubMed] [Cross Ref]
  • Korotkova N, Lidstrom ME. MeaB is a component of the methylmalonyl-CoA mutase complex required for protection of the enzyme from inactivation. J Biol Chem. 2004;279:13652–13658. doi: 10.1074/jbc.M312852200. [PubMed] [Cross Ref]
  • Padovani D, Banerjee R. Assembly and protection of the radical enzyme, methylmalonyl-CoA mutase, by its chaperone. Biochemistry. 2006;45:9300–9306. doi: 10.1021/bi0604532. [PubMed] [Cross Ref]
  • Fu C, Olson JW, Maier RJ. HypB protein of Bradyrhizobium japonicum is a metal-binding GTPase capable of binding 18 divalent nickel ions per dimer. Proc Natl Acad Sci USA. 1995;92:2333–2337. doi: 10.1073/pnas.92.6.2333. [PMC free article] [PubMed] [Cross Ref]
  • Zhang JW, Butland G, Greenblatt JF, Emili A, Zamble DB. A role for SlyD in the Escherichia coli hydrogenase biosynthetic pathway. J Biol Chem. 2005;280:4360–4366. doi: 10.1074/jbc.M411799200. [PubMed] [Cross Ref]
  • Leach MR, Zhang JW, Zamble DB. The role of complex formation between the Escherichia coli hydrogenase accessory factors HypB and SlyD. J Biol Chem. 2007;282:16177–16186. doi: 10.1074/jbc.M610834200. [PubMed] [Cross Ref]
  • Soriano A, Colpas GJ, Hausinger RP. UreE stimulation of GTP-dependent urease activation in the UreD-UreF-UreG-urease apoprotein complex. Biochemistry. 2000;39:12435–12440. doi: 10.1021/bi001296o. [PubMed] [Cross Ref]
  • Remaut H, Safarov N, Ciurli S, Van Beeumen J. Structural basis for Ni(2+) transport and assembly of the urease active site by the metallochaperone UreE from Bacillus pasteurii. J Biol Chem. 2001;276:49365–49370. doi: 10.1074/jbc.M108304200. [PubMed] [Cross Ref]
  • Song HK, Mulrooney SB, Huber R, Hausinger RP. Crystal structure of Klebsiella aerogenes UreE, a nickel-binding metallochaperone for urease activation. J Biol Chem. 2001;276:49359–49364. doi: 10.1074/jbc.M108619200. [PubMed] [Cross Ref]
  • The SEED: an Annotation/Analysis Tool Provided by FIG http://theseed.uchicago.edu/FIG/index.cgi
  • Zambelli B, Musiani F, Savini M, Tucker P, Ciurli S. Biochemical studies on Mycobacterium tuberculosis UreG and comparative modeling reveal structural and functional conservation among the bacterial UreG family. Biochemistry. 2007;46:3171–3182. doi: 10.1021/bi6024676. [PubMed] [Cross Ref]
  • Olson JW, Mehta NS, Maier RJ. Requirement of nickel metabolism proteins HypA and HypB for full activity of both hydrogenase and urease in Helicobacter pylori. Mol Microbiol. 2001;39:176–182. doi: 10.1046/j.1365-2958.2001.02244.x. [PubMed] [Cross Ref]
  • Stingl K, Schauer K, Ecobichon C, Labigne A, Lenormand P, Rousselle JC, Namane A, de Reuse H. In vivo interactome of Helicobacter pylori urease revealed by tandem affinity purification. Mol Cell Proteomics. 2008;7:2429–2441. doi: 10.1074/mcp.M800160-MCP200. [PubMed] [Cross Ref]
  • Benoit S, Maier RJ. Dependence of Helicobacter pylori urease activity on the nickel-sequestering ability of the UreE accessory protein. J Bacteriol. 2003;185:4787–4795. doi: 10.1128/JB.185.16.4787-4795.2003. [PMC free article] [PubMed] [Cross Ref]
  • Mehta N, Benoit S, Maier RJ. Roles of conserved nucleotide-binding domains in accessory proteins, HypB and UreG, in the maturation of nickel-enzymes required for efficient Helicobacter pylori colonization. Microb Pathog. 2003;35:229–234. doi: 10.1016/S0882-4010(03)00151-7. [PubMed] [Cross Ref]
  • Olson JW, Maier RJ. Dual roles of Bradyrhizobium japonicum nickelin protein in nickel storage and GTP-dependent Ni mobilization. J Bacteriol. 2000;182:1702–1705. doi: 10.1128/JB.182.6.1702-1705.2000. [PMC free article] [PubMed] [Cross Ref]
  • Khil PP, Obmolova G, Teplyakov A, Howard AJ, Gilliland GL, Camerini-Otero RD. Crystal structure of the Escherichia coli YjiA protein suggests a GTP-dependent regulatory function. Proteins. 2004;54:371–374. doi: 10.1002/prot.10430. [PubMed] [Cross Ref]
  • Hashimoto Y, Nishiyama M, Horinouchi S, Beppu T. Nitrile hydratase gene from Rhodococcus sp. N-774 requirement for its downstream region for efficient expression. Biosci Biotechnol Biochem. 1994;58:1859–1865. doi: 10.1271/bbb.58.1859. [PubMed] [Cross Ref]
  • Lu J, Zheng Y, Yamagishi H, Odaka M, Tsujimura M, Maeda M, Endo I. Motif CXCC in nitrile hydratase activator is critical for NHase biogenesis in vivo. FEBS Lett. 2003;553:391–396. doi: 10.1016/S0014-5793(03)01070-6. [PubMed] [Cross Ref]
  • Soriano A, Hausinger RP. GTP-dependent activation of urease apoprotein in complex with the UreD, UreF, and UreG accessory proteins. Proc Natl Acad Sci USA. 1999;96:11140–11144. doi: 10.1073/pnas.96.20.11140. [PMC free article] [PubMed] [Cross Ref]
  • Padovani D, Labunska T, Palfey BA, Ballou DP, Banerjee R. Adenosyltransferase tailors and delivers coenzyme B12. Nat Chem Biol. 2008;4:194–196. doi: 10.1038/nchembio.67. [PubMed] [Cross Ref]
  • Zheng Y, Roberts RJ, Kasif S. Segmentally variable genes: a new perspective on adaptation. PLoS Biol. 2004;2:E81. doi: 10.1371/journal.pbio.0020081. [PMC free article] [PubMed] [Cross Ref]
  • Rodionov DA, Vitreschak AG, Mironov AA, Gelfand MS. Comparative genomics of the vitamin B12 metabolism and regulation in prokaryotes. J Biol Chem. 2003;278:41148–41159. doi: 10.1074/jbc.M305837200. [PubMed] [Cross Ref]
  • Schroeder S, Lawrence AD, Biedendieck R, Rose RS, Deery E, Graham RM, McLean KJ, Munro AW, Rigby SE, Warren MJ. Demonstration that CobG, the monooxygenase associated with the ring contraction process of the aerobic cobalamin (vitamin B12) biosynthetic pathway, contains an Fe-S center and a mononuclear non-heme iron center. J Biol Chem. 2009;284:4796–4805. doi: 10.1074/jbc.M807184200. [PubMed] [Cross Ref]
  • Mascharak PK. Structural and functional models of nitrile hydratase. Coord Chem Rev. 2002;225:201–214. doi: 10.1016/S0010-8545(01)00413-1. [Cross Ref]
  • Banerjee A, Sharma R, Banerjee UC. The nitrile-degrading enzymes: current status and future prospects. Appl Microbiol Biotechnol. 2002;60:33–44. doi: 10.1007/s00253-002-1062-0. [PubMed] [Cross Ref]
  • Endo I, Nojiri M, Tsujimura M, Nakasako M, Nagashima S, Yohda M, Odaka M. Fe-type nitrile hydratase. J Inorg Biochem. 2001;83:247–253. doi: 10.1016/S0162-0134(00)00171-9. [PubMed] [Cross Ref]
  • Zhou Z, Hashimoto Y, Shiraki K, Kobayashi M. Discovery of posttranslational maturation by self-subunit swapping. Proc Natl Acad Sci USA. 2008;105:14849–14854. doi: 10.1073/pnas.0803428105. [PMC free article] [PubMed] [Cross Ref]
  • Zhou Z, Hashimoto Y, Kobayashi M. Self-subunit swapping chaperone needed for the maturation of multimeric metalloenzyme nitrile hydratase by a subunit exchange mechanism also carries out the oxidation of the metal ligand cysteine residues and insertion of cobalt. J Biol Chem. 2009;284:14930–14938. doi: 10.1074/jbc.M808464200. [PMC free article] [PubMed] [Cross Ref]
  • Nojiri M, Nakayama H, Odaka M, Yohda M, Takio K, Endo I. Cobalt-substituted Fe-type nitrile hydratase of Rhodococcus sp. N-771. FEBS Lett. 2000;465:173–177. doi: 10.1016/S0014-5793(99)01746-9. [PubMed] [Cross Ref]
  • Stevens JM, Rao Saroja N, Jaouen M, Belghazi M, Schmitter JM, Mansuy D, Artaud I, Sari MA. Chaperone-assisted expression, purification, and characterization of recombinant nitrile hydratase NI1 from Comamonas testosteroni. Protein Expr Purif. 2003;29:70–76. doi: 10.1016/S1046-5928(03)00008-1. [PubMed] [Cross Ref]
  • Kochan I. The role of iron in bacterial infections, with special consideration of host-tubercle bacillus interaction. Curr Top Microbiol Immunol. 1973;60:1–30. [PubMed]
  • Weinberg E. Metal starvation of pathogens by hosts. Bioscience. 1975;25:314–318. doi: 10.2307/1297129. [Cross Ref]
  • Liuzzi JP, Lichten LA, Rivera S, Blanchard RK, Aydemir TB, Knutson MD, Ganz T, Cousins RJ. Interleukin-6 regulates the zinc transporter Zip14 in liver and contributes to the hypozincemia of the acute-phase response. Proc Natl Acad Sci USA. 2005;102:6843–6848. doi: 10.1073/pnas.0502257102. [PMC free article] [PubMed] [Cross Ref]
  • Motley ST, Morrow BJ, Liu X, Dodge IL, Vitiello A, Ward CK, Shaw KJ. Simultaneous analysis of host and pathogen interactions during an in vivo infection reveals local induction of host acute phase response proteins, a novel bacterial stress response, and evidence of a host-imposed metal ion limited environment. Cell Microbiol. 2004;6:849–865. doi: 10.1111/j.1462-5822.2004.00407.x. [PubMed] [Cross Ref]
  • Panina EM, Mironov AA, Gelfand MS. Comparative genomics of bacterial zinc regulons: enhanced ion transport, pathogenesis, and rearrangement of ribosomal proteins. Proc Natl Acad Sci USA. 2003;100:9912–9917. doi: 10.1073/pnas.1733691100. [PMC free article] [PubMed] [Cross Ref]
  • Kim S, Watanabe K, Shirahata T, Watarai M. Zinc uptake system (znuA locus) of Brucella abortus is essential for intracellular survival and virulence in mice. J Vet Med Sci. 2004;66:1059–1063. doi: 10.1292/jvms.66.1059. [PubMed] [Cross Ref]
  • Pasquali P, Ammendola S, Pistoia C, Petrucci P, Tarantino M, Valente C, Marenzoni ML, Rotilio G, Battistoni A. Attenuated Salmonella enterica serovar Typhimurium lacking the ZnuABC transporter confers immune-based protection against challenge infections in mice. Vaccine. 2008;26:3421–3426. doi: 10.1016/j.vaccine.2008.04.036. [PubMed] [Cross Ref]
  • Maciag A, Dainese E, Rodriguez GM, Milano A, Provvedi R, Pasca MR, Smith I, Palù G, Riccardi G, Manganelli R. Global analysis of the Mycobacterium tuberculosis Zur (FurB) regulon. J Bacteriol. 2007;189:730–740. doi: 10.1128/JB.01190-06. [PMC free article] [PubMed] [Cross Ref]
  • Cappelli G, Volpe E, Grassi M, Liseo B, Colizzi V, Mariani F. Profiling of Mycobacterium tuberculosis gene expression during human macrophage infection: upregulation of the alternative sigma factor G, a group of transcriptional regulators, and proteins with unknown function. Res Microbiol. 2006;157:445–455. doi: 10.1016/j.resmic.2005.10.007. [PubMed] [Cross Ref]
  • Stratmann J, Strommenger B, Goethe R, Dohmann K, Gerlach GF, Stevenson K, Li LL, Zhang Q, Kapur V, Bull TJ. A 38-kilobase pathogenicity island specific for Mycobacterium avium subsp. paratuberculosis encodes cell surface proteins expressed in the host. Infect Immun. 2004;72:1265–1274. doi: 10.1128/IAI.72.3.1265-1274.2004. [PMC free article] [PubMed] [Cross Ref]
  • Bannantine JP, Paustian ML, Waters WR, Stabel JR, Palmer MV, Li L, Kapur V. Profiling bovine antibody responses to Mycobacterium avium subsp. paratuberculosis infection by using protein arrays. Infect Immun. 2008;76:739–749. doi: 10.1128/IAI.00915-07. [PMC free article] [PubMed] [Cross Ref]
  • McBride SM, Coburn PS, Baghdayan AS, Willems RJ, Grande MJ, Shankar N, Gilmore MS. Genetic variation and evolution of the pathogenicity island of Enterococcus faecalis. J Bacteriol. 2009;191:3392–3402. doi: 10.1128/JB.00031-09. [PMC free article] [PubMed] [Cross Ref]
  • Köhler S, Foulongne V, Ouahrani-Bettache S, Bourg G, Teyssier J, Ramuz M, Liautard JP. The analysis of the intramacrophagic virulome of Brucella suis deciphers the environment encountered by the pathogen inside the macrophage host cell. Proc Natl Acad Sci USA. 2002;99:15711–15716. doi: 10.1073/pnas.232454299. [PMC free article] [PubMed] [Cross Ref]
  • Gan YH, Chua KL, Chua HH, Liu B, Hii CS, Chong HL, Tan P. Characterization of Burkholderia pseudomallei infection and identification of novel virulence factors using a Caenorhabditis elegans host system. Mol Microbiol. 2002;44:1185–1197. doi: 10.1046/j.1365-2958.2002.02957.x. [PubMed] [Cross Ref]
  • Twine SM, Mykytczuk NC, Petit MD, Shen H, Sjöstedt A, Wayne Conlan J, Kelly JF. In vivo proteomic analysis of the intracellular bacterial pathogen, Francisella tularensis, isolated from mouse spleen. Biochem Biophys Res Commun. 2006;345:1621–1633. doi: 10.1016/j.bbrc.2006.05.070. [PubMed] [Cross Ref]
  • Mohapatra NP, Soni S, Bell BL, Warren R, Ernst RK, Muszynski A, Carlson RW, Gunn JS. Identification of an orphan response regulator required for the virulence of Francisella spp. and transcription of pathogenicity island genes. Infect Immun. 2007;75:3305–3314. doi: 10.1128/IAI.00351-07. [PMC free article] [PubMed] [Cross Ref]
  • Tang DJ, Li XJ, He YQ, Feng JX, Chen B, Tang JL. The zinc uptake regulator Zur is essential for the full virulence of Xanthomonas campestris pv. campestris. Mol Plant Microbe Interact. 2005;18:652–658. doi: 10.1094/MPMI-18-0652. [PubMed] [Cross Ref]
  • Huang DL, Tang DJ, Liao Q, Li HC, Chen Q, He YQ, Feng JX, Jiang BL, Lu GT, Chen B, et al. The Zur of Xanthomonas campestris functions as a repressor and an activator of putative zinc homeostasis genes via recognizing two distinct sequences within its target promoters. Nucleic Acids Res. 2008;36:4295–4309. doi: 10.1093/nar/gkn328. [PMC free article] [PubMed] [Cross Ref]
  • Yang W, Liu Y, Chen L, Gao T, Hu B, Zhang D, Liu F. Zinc uptake regulator (zur) gene involved in zinc homeostasis and virulence of Xanthomonas oryzae pv. oryzae in rice. Curr Microbiol. 2007;54:307–314. doi: 10.1007/s00284-006-0485-8. [PubMed] [Cross Ref]
  • Yuan ZC, Haudecoeur E, Faure D, Kerr KF, Nester EW. Comparative transcriptome analysis of Agrobacterium tumefaciens in response to plant signal salicylic acid, indole-3-acetic acid and gamma-amino butyric acid reveals signalling cross-talk and Agrobacterium--plant co-evolution. Cell Microbiol. 2008;10:2339–2354. doi: 10.1111/j.1462-5822.2008.01215.x. [PubMed] [Cross Ref]
  • Rodionov DA. Comparative genomic reconstruction of transcriptional regulatory networks in bacteria. Chem Rev. 2007;107:3467–3497. doi: 10.1021/cr068309+. [PMC free article] [PubMed] [Cross Ref]
  • Sankara SB, Shah K, Gabriel S, Reddy R, Schimmel P, Rodionov DA, de V, Crécy-Lagard , Helmann JD, Iwata-Reuyl D, Swairjo MA. Zinc-independent folate biosynthesis: Genetic, biochemical, and structural investigations reveal new metal dependence for GTP Cyclohydrolase IB. Journal of Bacteriology. [PMC free article] [PubMed]
  • Tripp BC, Bell CB, Cruz F, Krebs C, Ferry JG. A role for iron in an ancient carbonic anhydrase. J Biol Chem. 2004;279:6683–6687. doi: 10.1074/jbc.M311648200. [PubMed] [Cross Ref]
  • Macauley SR, Zimmerman SA, Apolinario EE, Evilia C, Hou YM, Ferry JG, Sowers KR. The archetype gamma-class carbonic anhydrase (Cam) contains iron when synthesized in vivo. Biochemistry. 2009;48:817–819. doi: 10.1021/bi802246s. [PubMed] [Cross Ref]
  • Thoden J, Phillips GJ, Neal T, Raushel F, Holden H. Molecular structure of dihydroorotase: a paradigm for catalysis through the use of a binuclear metal center. Biochemistry. 2001;40:6989–6997. doi: 10.1021/bi010682i. [PubMed] [Cross Ref]
  • Schurr M, Vickrey J, Kumar A, Campbell A, Cunin R, Benjamin R, Shanley M, O'Donovan G. Aspartate transcarbamoylase genes of Pseudomonas putida: requirement for an inactive dihydroorotase for assembly into the dodecameric holoenzyme. J Bacteriol. 1995;177:1751–1759. [PMC free article] [PubMed]
  • Brichta DM, Azad KN, Ralli P, O'Donovan GA. Pseudomonas aeruginosa dihydroorotases: a tale of three pyrCs. Arch Microbiol. 2004;182:7–17. doi: 10.1007/s00203-004-0687-z. [PubMed] [Cross Ref]
  • Jaffe E. An unusual phylogenetic variation in the metal ion binding sites of porphobilinogen synthase. Chem Biol. 2003;10:25–34. doi: 10.1016/S1074-5521(02)00296-X. [PubMed] [Cross Ref]
  • Pfalz J, Liere K, Kandlbinder A, Dietz KJ, Oelmüller R. pTAC2, -6, and -12 are components of the transcriptionally active plastid chromosome that are required for plastid gene expression. Plant Cell. 2006;18:176–197. doi: 10.1105/tpc.105.036392. [PMC free article] [PubMed] [Cross Ref]
  • TAIR The Arabidopsis Information Resource http://www.arabidopsis.org
  • Kumar A, Cheung KH, Tosches N, Masiar P, Liu Y, Miller P, Snyder M. The TRIPLES database: a community resource for yeast molecular biology. Nucleic Acids Res. 2002;30:73–75. doi: 10.1093/nar/30.1.73. [PMC free article] [PubMed] [Cross Ref]
  • de Jesus Ferreira MC, Bao X, Laizé V, Hohmann S. Transposon mutagenesis reveals novel loci affecting tolerance to salt stress and growth at low temperature. Curr Genet. 2001;40:27–39. doi: 10.1007/s002940100237. [PubMed] [Cross Ref]
  • Smith TF, Gaitatzes C, Saxena K, Neer EJ. The WD repeat: a common architecture for diverse functions. Trends Biochem Sci. 1999;24:181–185. doi: 10.1016/S0968-0004(99)01384-5. [PubMed] [Cross Ref]
  • Merchant S, Bogorad L. Rapid degradation of apoplastocyanin in Cu(II)-deficient cells of Chlamydomonas reinhardtii. J Biol Chem. 1986;261:15850–15853. [PubMed]
  • Allen MD, Kropat J, Tottey S, Del Campo JA, Merchant SS. Manganese deficiency in Chlamydomonas results in loss of photosystem II and MnSOD function, sensitivity to peroxides, and secondary phosphorus and iron deficiency. Plant Physiol. 2007;143:263–277. doi: 10.1104/pp.106.088609. [PMC free article] [PubMed] [Cross Ref]
  • Merchant SS, Allen MD, Kropat J, Moseley JL, Long JC, Tottey S, Terauchi AM. Between a rock and a hard place: trace element nutrition in Chlamydomonas. Biochim Biophys Acta. 2006;1763:578–594. doi: 10.1016/j.bbamcr.2006.04.007. [PubMed] [Cross Ref]
  • Quinn JM, Merchant S. Two copper-responsive elements associated with the Chlamydomonas Cyc6 gene function as targets for transcriptional activators. Plant Cell. 1995;7:623–628. doi: 10.1105/tpc.7.5.623. [PMC free article] [PubMed] [Cross Ref]
  • Allen MD, del Campo JA, Kropat J, Merchant SS. FEA1, FEA2, and FRE1, encoding two homologous secreted proteins and a candidate ferrireductase, are expressed coordinately with FOX1 and FTR1 in iron-deficient Chlamydomonas reinhardtii. Eukaryot Cell. 2007;6:1841–1852. doi: 10.1128/EC.00205-07. [PMC free article] [PubMed] [Cross Ref]
  • Merchant SS, Prochnik SE, Vallon O, Harris EH, Karpowicz SJ, Witman GB, Terry A, Salamov A, Fritz-Laylin LK, Maréchal-Drouard L, et al. The Chlamydomonas genome reveals the evolution of key animal and plant functions. Science. 2007;318:245–250. doi: 10.1126/science.1143609. [PMC free article] [PubMed] [Cross Ref]
  • Grossman AR, Croft M, Gladyshev VN, Merchant SS, Posewitz MC, Prochnik S, Spalding MH. Novel metabolism in Chlamydomonas through the lens of genomics. Curr Opin Plant Biol. 2007;10:190–198. doi: 10.1016/j.pbi.2007.01.012. [PubMed] [Cross Ref]
  • Croft MT, Warren MJ, Smith AG. Algae need their vitamins. Eukaryot Cell. 2006;5:1175–1183. doi: 10.1128/EC.00097-06. [PMC free article] [PubMed] [Cross Ref]
  • Merchant S, Bogorad L. Regulation by copper of the expression of plastocyanin and cytochrome c552 in Chlamydomonas reinhardi. Mol Cell Biol. 1986;6:462–469. [PMC free article] [PubMed]
  • Sauer K, Thauer R. Methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Zinc dependence and thermodynamics of the methanol:cob(I)alamin methyltransferase reaction. Eur J Biochem. 1997;249:280–285. doi: 10.1111/j.1432-1033.1997.t01-1-00280.x. [PubMed] [Cross Ref]
  • Sauer K, Harms U, Thauer RK. Methanol:coenzyme M methyltransferase from Methanosarcina barkeri. Purification, properties and encoding genes of the corrinoid protein MT1. Eur J Biochem. 1997;243:670–677. doi: 10.1111/j.1432-1033.1997.t01-1-00670.x. [PubMed] [Cross Ref]
  • Ferguson T, Soares JA, Lienard T, Gottschalk G, Krzycki JA. RamA, a protein required for reductive activation of corrinoid-dependent methylamine methyltransferase reactions in methanogenic archaea. J Biol Chem. 2009;284:2285–2295. doi: 10.1074/jbc.M807392200. [PMC free article] [PubMed] [Cross Ref]
  • Li Q, Li L, Rejtar T, Karger BL, Ferry JG. Proteome of Methanosarcina acetivorans Part I: an expanded view of the biology of the cell. J Proteome Res. 2005;4:112–128. doi: 10.1021/pr049832c. [PubMed] [Cross Ref]
  • Pflüger K, Ehrenreich A, Salmon K, Gunsalus RP, Deppenmeier U, Gottschalk G, Müller V. Identification of genes involved in salt adaptation in the archaeon Methanosarcina mazei Gö1 using genome-wide gene expression profiling. FEMS Microbiol Lett. 2007;277:79–89. doi: 10.1111/j.1574-6968.2007.00941.x. [PubMed] [Cross Ref]
  • Gabriel SE, Miyagi F, Gaballa A, Helmann JD. Regulation of the Bacillus subtilis yciC gene and insights into the DNA-binding specificity of the zinc-sensing metalloregulator Zur. J Bacteriol. 2008;190:3482–3488. doi: 10.1128/JB.01978-07. [PMC free article] [PubMed] [Cross Ref]
  • Eide D. Zinc transporters and the cellular trafficking of zinc. Biochim Biophys Acta. 2006;1763:711–722. doi: 10.1016/j.bbamcr.2006.03.005. [PubMed] [Cross Ref]
  • Campos-Bermudez VA, Leite NR, Krog R, Costa-Filho AJ, Soncini FC, Oliva G, Vila AJ. Biochemical and structural characterization of Salmonella typhimurium glyoxalase II: new insights into metal ion selectivity. Biochemistry. 2007;46:11069–11079. doi: 10.1021/bi7007245. [PubMed] [Cross Ref]
  • Cámara B, Marín M, Schlömann M, Hecht HJ, Junca H, Pieper DH. trans-Dienelactone hydrolase from Pseudomonas reinekei MT1, a novel zinc-dependent hydrolase. Biochem Biophys Res Commun. 2008;376:423–428. doi: 10.1016/j.bbrc.2008.09.006. [PubMed] [Cross Ref]
  • Hall RS, Xiang DF, Xu C, Raushel FM. N-Acetyl-D-glucosamine-6-phosphate deacetylase: substrate activation via a single divalent metal ion. Biochemistry. 2007;46:7942–7952. doi: 10.1021/bi700543x. [PMC free article] [PubMed] [Cross Ref]
  • Yee D, Morel FMM. In vivo substitution of zinc by cobalt in carbonic anhydrase of a marine diatom. Limnol Oceanogr. 1996;41:573–577.
  • Moroney J, Ynalvez R. Proposed carbon dioxide concentrating mechanism in Chlamydomonas reinhardtii. Eukaryot Cell. 2007;6:1251–1259. doi: 10.1128/EC.00064-07. [PMC free article] [PubMed] [Cross Ref]
  • Suzuki T, Okamura Y, Arakaki A, Takeyama H, Matsunaga T. Cytoplasmic ATPase involved in ferrous ion uptake from magnetotactic bacterium Magnetospirillum magneticum AMB-1. FEBS Lett. 2007;581:3443–3448. doi: 10.1016/j.febslet.2007.06.047. [PubMed] [Cross Ref]
  • Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW. GenBank. Nucleic Acids Res. 2009:D26–31. doi: 10.1093/nar/gkn723. [PMC free article] [PubMed] [Cross Ref]
  • Rice P, Longden I, Bleasby A. EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 2000;16:276–277. doi: 10.1016/S0168-9525(00)02024-2. [PubMed] [Cross Ref]
  • Thompson JD, Gibson TJ, Higgins DG. Multiple sequence alignment using ClustalW and ClustalX. Curr Protoc Bioinformatics. 2002;Chapter 2 [PubMed]
  • Gouet P, Courcelle E, Stuart D, Métoz F. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics. 1999;15:305–308. doi: 10.1093/bioinformatics/15.4.305. [PubMed] [Cross Ref]
  • PHYLIP (Phylogeny Inference Package) version 3.67 http://evolution.genetics.washington.edu/phylip.html
  • Overbeek R, Fonstein M, D'Souza M, Pusch G, Maltsev N. The use of gene clusters to infer functional coupling. Proc Natl Acad Sci USA. 1999;96:2896–2901. doi: 10.1073/pnas.96.6.2896. [PMC free article] [PubMed] [Cross Ref]
  • Chevenet F, Brun C, Bañuls A, Jacq B, Christen R. TreeDyn: towards dynamic graphics and annotations for analyses of trees. BMC Bioinformatics. 2006;7:439. doi: 10.1186/1471-2105-7-439. [PMC free article] [PubMed] [Cross Ref]
  • Waterhouse AM, Procter JB, Martin DM, Clamp M, Barton GJ. Jalview Version 2 - a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;25:1189–1191. doi: 10.1093/bioinformatics/btp033. [PMC free article] [PubMed] [Cross Ref]
  • Mironov AA, Koonin EV, Roytberg MA, Gelfand MS. Computer analysis of transcription regulatory patterns in completely sequenced bacterial genomes. Nucleic Acids Res. 1999;27:2981–2989. doi: 10.1093/nar/27.14.2981. [PMC free article] [PubMed] [Cross Ref]
  • Crooks G, Hon G, Chandonia J, Brenner S. WebLogo: a sequence logo generator. Genome Res. 2004;14:1188–1190. doi: 10.1101/gr.849004. [PMC free article] [PubMed] [Cross Ref]
  • de Berardinis V, Vallenet D, Castelli V, Besnard M, Pinet A, Cruaud C, Samair S, Lechaplais C, Gyapay G, Richez C, et al. A complete collection of single-gene deletion mutants of Acinetobacter baylyi ADP1. Mol Syst Biol. 2008;4:174. doi: 10.1038/msb.2008.10. [PMC free article] [PubMed] [Cross Ref]
  • Quinn JM, Merchant S. Copper-responsive gene expression during adaptation to copper deficiency. Methods Enzymol. 1998;297:263–279. full_text. [PubMed]

Articles from BMC Genomics are provided here courtesy of BioMed Central

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...