OUP user menu

Characterization of the Acinetobacter baumannii Fur regulator: cloning and sequencing of the fur homolog gene

Catherine Daniel, Stéphanie Haentjens, Marie-Christine Bissinger, René J. Courcol
DOI: http://dx.doi.org/10.1111/j.1574-6968.1999.tb13375.x 199-209 First published online: 1 January 1999


Growth kinetics, siderophore activity and iron-regulated bacterial proteins of Acinetobacter baumannii BM2580 were studied in iron-restricted and iron-supplemented chemically defined media. Iron-regulated outer membrane proteins of 75 kDa and 80 kDa were expressed under iron-restricted conditions. Cloning and sequencing of the complete iron-uptake regulatory (fur) gene from A baumannii BM2580 is reported for the first time. This gene is preceded by a single autoregulated promoter whose −10 region overlaps the Fur binding site. The open reading frame identified encodes a polypeptide consisting of 145 amino acids. The fur gene is followed by a divergent open reading frame coding for the C-terminus of a putative PilU protein. Sequence analysis indicates that the Fur protein of A baumannii was 63% identical to the Escherichia coli Fur protein.

  • Iron
  • Acinetobacter baumannii
  • Ferric uptake regulator
  • Iron-regulated protein
  • Siderophore

1 Introduction

Acinetobacter baumannii, formerly known as Acinetobacter calcoaceticus subsp. anitratus, belongs to a bacterial species that is widely distributed throughout the environment. Lately, numerous outbreaks of nosocomial infections caused by A. baumannii have been reported, which are of particular concern because of the widespread and increasing antibiotic resistance of the isolated strains [1]. Acinetobacter species are able to survive under restricted nutrient conditions, such as those imposed by the host. Iron is one of the essential bacterial nutrients that is tightly controlled by the host, through its chelation by the high-affinity binding glycoproteins, transferrin and lactoferrin. Thus, bacteria have developed efficient iron-uptake mechanisms that allow them to scavenge iron from these host proteins, either by direct interaction with the host proteins, by direct interaction with the iron protein complexes, or by the synthesis and secretion of high-affinity extracellular siderophore compounds [2]. Only very few reports have shown the ability of clinical isolates of A. baumannii to grow and produce siderophore compounds under iron-deficient conditions [35]. More recently, a high-affinity siderophore-mediated system has been described in a clinical strain of A. baumannii [6].

The genes involved in iron uptake in several aerobic bacteria, such as Escherichia coli, are coordinately regulated by the ferric uptake regulator protein, Fur. The Fur repressor protein functions as a dimer with ferrous iron as a corepressor to bind the operator regions of genes under Fur control [7].

In the present work, we cloned and sequenced the A. baumannii fur gene. The Fur protein was highly related (>60%) to other Fur proteins. By sequencing the fur flanking regions, a pilU-related sequence was identified which suggests the presence of a type IV pilus in A. baumannii. This paper is also the first to report a kinetic study on siderophore production, bacterial growth, and iron-regulated proteins in these bacteria.

2 Materials and methods

2.1 Bacterial strain, plasmids, medium and growth conditions

A. baumannii BM2580 was used in this study [8]. Bacteria were grown at 37°C in 500-ml shake flask cultures in an orbital shaker (172 rpm). Chemically defined media containing ≤0.04 µM Fe (CDM-Fe) or 100 µM FeSO4 (CDM+Fe) were used as previously described [9]. Growth was monitored by colony counts using a pour-plate procedure. All experiments were performed in duplicate. The coefficients of variation (CV) in all experiments were equal to or less than 15%. Bacterial cells were harvested at various stages of growth.

2.2 Siderophore detection

Siderophore activity in culture supernatants was detected with the chrome azurol assay (CAS) prepared as described by Schwyn and Neilands [10]. Results were expressed as equivalents of Desferrioxamine activity (µM), with reference to a standard curve as previously described [9].

2.3 Outer membrane preparation

Outer-membrane proteins (OMPs) were prepared by a slight modification of the method reported by Sprott et al. [11]. Bacteria harvested by centrifugation were disrupted ultrasonically. Unbroken cells were discarded by centrifugation. The supernatant was separated into cytoplasmic and membrane fractions by further centrifugation (45 000×g, 45 min, 4°C). The pellet containing the cell envelope was treated with 2% Triton X-100 (Sigma) to solubilize the cytoplasmic membranes. The detergent-insoluble fraction in the pellet was treated with lysozyme (1 mg ml−1) to digest cell-wall peptidoglycan. The remaining OMPs in this fraction were concentrated by further centrifugation.

2.4 SDS-PAGE and Western blot analysis

Proteins were separated on sodium dodecyl sulfate-polyacrylamide gel slabs. Gels were fixed and stained with Coomassie blue or were transferred by electroblotting onto nitrocellulose membrane (Amersham). After transfer, membranes were blocked with 5% skim milk–0.3% Tween-20, then probed with rabbit polyclonal anti-E. coli Fur serum [12], diluted 1:500 for 1 h. Proteins were detected using anti-rabbit IgG horseradish peroxidase conjugate (Bio-Rad) diluted 1:3000. Reactive bands were visualized with 3,3′-diaminobenzidine (Sigma).

2.5 DNA manipulation

Genomic DNA was purified with the Wizard Genomic DNA Purification kit (Promega) and digested with various restriction endonucleases as instructed by the manufacturer (Boehringer Mannheim). Plasmid DNA was purified with spin columns (Qiagen).

The PCR products were separated by electrophoresis and bands of interest were cloned into the plasmid T/A cloning vector pMosBlue (Amersham) as described previously [13].

2.6 Cloning strategy, sequencing and analysis of fur homolog

Amino acid sequences of Fur proteins of E. coli (X02589), Yersinia pestis (Z12101), Vibrio anguillarum (L19717), Vibrio vulnificus (L06428), Vibrio cholerae (M85154), Pseudomonas aeruginosa (L00604), and Legionella pneumophila (U06072) were aligned. The following consensus sequences were deduced: sequence 1 (AGLK(V/I)TLPR), where the A residue is positioned as A-11 in the E. coli Fur protein; sequence 2 (HDH(L/M)(I/V)C(L/V)(D/K)CG), where the first H residue is positioned as H-88 in the E. coli Fur protein.

We designed the 32-fold sense degenerate oligonucleotide primer FUR3 (5′-GCIGGIYTIAARRTIACIYTICCIMG-3′; I=inosine; Y=C or T; R=A or G; M=C or A) based on the first amino acid consensus sequence and the 1152-fold antisense degenerate oligonucleotide primer FUR5 (5′-CCRCAITYIAVRCAIAYIADRTGRTCRTG-3′; V=G, C, or A; D=G, T, or A) based on the second amino acid consensus sequence. The DNA template was mixed with 15 pmol of each primer, 2 units of Taq DNA polymerase, 0.3 mM dNTP, and 1.5 mM MgCl2 in a final volume of 50 µl, with the buffer recommended by the manufacturer (Boehringer Mannheim). PCR parameters were 94°C for 4 min followed by 35 cycles at 94°C for 4 min, followed by 35 cycles at 94°C for 20 s, at 57°C for 40 s, at 47°C for 30 s, at 37°C for 20 s, and at 72°C for 2 min, followed by a final extension at 72°C for 20 min.

A. baumannii fur gene probe was prepared by PCR amplification of A. baumannii BM2580 using the internal primers to the putative fur gene FUR6 (5′-CAAAACAACATCATCTTAGCGCCG-3′) and FUR7 (5′-TGATGATCTTCTTGCATGATTTCG-3′) (Fig. 3) resulting in a single band of the expected size (178 bp, data not shown). The PCR product was labeled by random-primed incorporation of digoxigenin-labeled dUTP with the DNA labeling kit from Boehringer Mannheim.

Figure 3

Strategy for cloning the complete fur gene. (A) FUR3 and FUR5 were designed on the basis of sequence data from other Fur proteins. (B) FUR6 and FUR7 were designed on the basis of the nucleotide sequence of the DNA fragment obtained from A. baumannii by PCR using the FUR3 and FUR5 primers. (C) The fragment FUR6–FUR7 was used as a probe on Southern transfers of A. baumannii BM2580, digested with various restriction enzymes, and hybridized to a single HindIII–ClaI band. (D) FUR8 (5′-GCAGATGGGGTCCTGATCAAT-3′) and FUR11 (5′-ATACACCAAAGATTGAACAACC-3′) were designed on the basis of the nucleotide sequence of the two overlapping PCR products. Right arrows represent sense primers and the left arrows represent antisense primers. The shaded box represents the ORF.

Total genomic DNA (3 µg) from A. baumannii BM2580 was digested with various restriction enzymes and electrophoresed on 1% agarose gel prior to transfer to nylon membranes (Amersham) by capillary blotting overnight. Hybridization to the blot was performed at 42°C in DIG Easy Hyb (Boehringer Mannheim). Hybrids were detected by an enzyme-linked immunoassay with an antibody conjugate (anti-digoxigenin alkaline phosphatase conjugate) and the chemiluminescent substrate CSPD (disodium-3-(−4-methoxyspiro{1,2-dioxetane-3-2′-(5′-chloro)tricyclo[]decan}-4-yl)phenyl phosphate; Boehringer Mannheim). For the detection of the chemiluminescent signal, the membranes were exposed to a Hyperfilm ECL (Amersham).

To construct a HindIII–ClaI minilibrary, genomic DNA from A. baumannii BM2580 was digested with HindIII and ClaI and was size-fractionated. DNA bands of appropriate size were extracted from the gel and ligated into the plasmid pBluescript KS(+) vector (Stratagene). The ligated DNA was used as template for PCR amplification of the fur gene of A. baumannii with two primer sets hybridizing to the vector pBluescript KS(+) and to the fur gene of A. baumannii, respectively.

Nucleotide sequence of both strands of plasmid DNA were determined by the dideoxy method using the ABI Prism Ready Reaction AmpliTaq FS, DyeDeoxy Terminator Cycle Sequencing kit (Applied Biosystems) and an automated DNA sequencer (model ABI 373A, Applied Biosystems).

3 Results and discussion

3.1 Effect of iron depletion on growth, siderophore activity, and protein composition

Little is known about the iron metabolism in A. baumannii, thus prompting our investigation. It is important to understand the potential of Acinetobacter to respond to iron-stressed conditions as it may reflect their ability to survive and cause infections in the human host. Thus, for this bacteria, this paper is the first to report a kinetic study on siderophore production and bacterial growth, data on iron-regulated proteins, and identification of a fur gene homolog.

In a preliminary step, bacterial growth was studied. As shown on Fig. 1, A. baumannii BM2580 cells grew well under iron-restricted conditions. Iron-restriction did not show any initial delay in the growth of the strain compared to the cells grown under iron-supplemented conditions. Growth kinetics of A. baumannii BM2580 were almost identical in both media until 18 h of growth. However, after 18 h of growth, cell lysis was observed in CDM-Fe followed by a steady-state at 24 h, resulting in a final reduced growth yield compared to the bacteria cultured in CDM+Fe (Fig. 1).

Figure 1

Kinetics of growth (bacterial counts expressed in number of viable cells per ml: ×, CDM+Fe; ◯, CDM-Fe and □, siderophore production (siderophore expressed in µM Desferal equivalents) of A. baumannii BM2580 in CDM-Fe. Values are means of two experiments. The coefficients of variation in all experiments were equal to or less than 15%.

As there was no difference between the growth curves of bacteria with the two kinds of chemically defined medium during the exponential phase, iron might act as a growth factor rather late, especially during the stationary phase, and therefore delaying the cell lysis observed early on with the iron-restricted medium. Our results suggest that A. baumannii required a low concentration of iron which was illustrated by good growth in the CDM-Fe, and that an active iron-uptake mechanism was operational. These results are consistent with those obtained by Echenique et al. [6] who showed that A. baumannii grew well under iron-restricted conditions and also reported an increased generation time compared with bacteria cultured in iron-supplemented conditions.

Few studies have been done on kinetics of siderophore secretion by A. baumannii [35]. Siderophore activity was investigated at each sampling time (Fig. 1). No siderophore activity was detected in CDM+Fe (data not shown) indicating that the siderophore production was tightly regulated by the availability of iron in the growth medium. The siderophore production described in our study seemed to be population-dependent as siderophore activity could not be detected before 6 h of culture when the number of bacteria was under 108 ml−1 (data not shown). As it started to be detected from 6 h of culture onwards, these results suggest that a high-affinity siderophore-dependent system for iron uptake, as described by Echenique et al. [6], was rapidly set up only a few hours after incubation. This is supported by the good growth observed during the first hours of culture. However, during the exponential phase, siderophore-secretion increased weakly during the first 9 h of incubation, suggesting a certain bacterial adaptation to growth conditions. To precede or accompany exponential growth, an active iron-uptake system was presumably required within the lag phase and the first hours of the exponential phase during which siderophore secretion was undetectable. This implies that the CAS assay used for detection was not sufficiently sensitive to detect the early secretion or that bacteria in the lag phase used a siderophore-independent pathway for iron-uptake. During cell lysis period, siderophore activity increased, which might be due to a cytoplasmic release of siderophore from dead cells. Then, from 24 h onwards, as growth curve and siderophore kinetics evolved similarly, a metabolic cycle implying secretion and reabsorption of the siderophore by the bacteria might be suggested [14].

The outer membrane proteins of A. baumannii, collected at different times, were examined to detect differences between bacterial cells cultured in iron-restricted vs. iron-supplemented medium. Bacteria growing in CDM-Fe expressed several new outer membrane proteins distributed in the range 70–80 kDa from 6 h of culture onwards as it is shown on Fig. 2 (lanes 2, 4, 6, 8, 10, 12, 14). These proteins were completely iron-repressed. Two prominent proteins of 80 and 75 kDa were more strongly expressed in the outer membrane of A. baumannii cultured in CDM-Fe and were expressed after only 6 h of culture (lane 2). These two proteins of 80 and 75 kDa were produced more abundantly after 12 h of culture, during the stationary phase (lanes 6, 8, 10, 12, 14). The expression of these proteins appeared and increased simultaneously with the siderophore synthesis (Fig. 2). In CDM+Fe, proteins appeared either during the log phase (19.5 kDa, 84 kDa) or the stationary phase (22 kDa, 24 kDa) (Fig. 2; lanes 5, 7, 9, 11, 13). Expression of these four proteins was time-dependent and iron-inducible. In each respective medium, the protein profiles obtained at 48 and 60 h were identical to those obtained at 36 h (data not shown).

Figure 2

SDS-PAGE profiles of outer membrane proteins from A. baumannii BM2580. Lanes 1, 3, 5, 7, 9, 11 and 13, CDM+Fe; lanes 2, 4, 6, 8, 10, 12 and 14, CDM-Fe; lane 15: molecular weight markers (kDa). Arrows indicate the 75- and 80-kDa iron-regulated proteins.

Some of the outer membrane proteins had been already identified by Coomassie blue-staining by Echenique et al. [6], or by immunoblotting with serum from septicemic patients [15] although no iron-inducible outer membrane proteins have been previously described in Acinetobacter species.

This expression of such a siderophore-mediated iron-uptake system represents a benefit for the invading bacteria by allowing it to compete with host-binding proteins for essential iron and may also represent a detrimental factor for the host since such systems were demonstrated to be important virulence factors in the establishments of bacterial infections.

The genes encoding the structural and regulatory components of this siderophore-mediated iron uptake system have not yet been identified in A. baumannii. The genes involved in iron uptake in E. coli and other aerobic bacteria are coordinately regulated by the ferric uptake regulator protein Fur.

3.2 Cloning of the A. baumannii fur homolog

Two degenerate primers FUR3 and FUR5 were designed. From the alignment of the E. coli, Y. pestis, V. anguillarum, V. vulnificus, V. cholerae, P. aeruginosa, and L. pneumophila sequenced Fur proteins, a 260-bp fragment was amplified from A. baumannii BM2580 with these primers and was subsequently cloned. Nucleotide-sequencing confirmed the identity of a fur homolog gene from A. baumannii. Using this sequence, two internal primers of the putative fur gene, FUR6 and FUR7 (Fig. 3) were designed to amplify, as predicted, a specific 178-bp DNA fragment from A. baumannii BM2580. This specific 178-bp DNA fragment was labeled, hybridized, under high-stringency conditions, with endonuclease digested DNA from strain BM2580. The probe hybridized strongly to a single HindIII–ClaI band of approximately 1.7-kb in size (data not shown). A single fragment of 1.7-kb was hypothesized to be large enough to contain the fur homolog structural gene and flanking regions at both ends. The DNA band containing the 1.7-kb fragment was cloned to construct a HindIII–ClaI minilibrary.

The ligated DNA was used subsequently as a template for PCR amplification of the fur region with two primer sets, Reverse Primer and T7 primers hybridizing to the vector plasmid pBluescript KS(+) and the internal primers FUR6 and FUR7, hybridizing to the fur gene of A. baumannii DNA, respectively. Two bands of approximately 700-and 1300-bp were amplified. When cloned and sequenced, both of these PCR products were found to represent a fragment of a putative fur gene. The two overlapping products had overlapping sequences of 178 nucleotides (equivalent to the FUR6–FUR7 fragment). To verify that these two overlapping amplified fragments belonged to the same fur gene, two novel oligonucleotide primers flanking the entire fur gene, FUR8 and FUR11, were synthetized using the nucleotide sequences. Total genomic DNA of A. baumannii BM2580 was amplified by PCR with the primers FUR8 and FUR11 giving the expected fragment of 747-bp in length which was cloned.

3.3 DNA sequence of the A. baumannii fur gene

The PCR amplification of total genomic DNA of A. baumannii BM2580 with the two primers FUR8 and FUR11 and nucleotide-sequencing of several clones from each amplification containing the 747-bp fragment were repeated several times to confirm our results. Nucleotide-sequencing of the 747-bp fragment confirmed that the two overlapping fragments described above belong to the same fur gene of A. baumannii. The A. baumannii fur gene sequence revealed an open reading frame of 435-bp. The open reading frame coded for a 145-amino acid protein, rich in histidine residues (7.5%), and having a calculated molecular mass of 16 060. This value was in good agreement with the estimated molecular weight of approximately 16 kDa which was observed when Western blots of cytoplasmic proteins from A. baumannii grown in iron-restricted media and iron-supplemented media were probed with rabbit polyclonal anti-E. coli Fur serum with no major effect on the expression of the Fur protein (Fig. 4). The fur ORF was preceded by a putative ribosome binding site (Fig. 5). Upstream of the AUG start codon, there were sequences similar to the −35 and −10 hexamers of promoters used by the major form of RNA polymerase in E. coli. An imperfect inverted repeat that overlaps the −10 region was identified. This region whose sequence was 5′-AATTATGATGCATTTGATC-3′, possessed a 11-of-19-base match (identical bases to the consensus are underlined) with the E. coli Fur binding site (Fur box, GATAATGATAATCATTATC), indicating that the A. baumannii fur gene was probably autoregulated in a manner similar to E. coli [7]. A palindromic region, indicated on Fig. 5 by arrows in the 3′-untranslated region, capable of forming a stem-loop structure reminiscent of a putative transcriptional terminator was located downstream from the fur coding sequence. Loci homologous to the A. baumannii BM2580 fur gene were identified by PCR amplification of total genomic DNA with the primers FUR8 and FUR11 from 4 other different clinical strains of A. baumannii (including A. baumannii BM2686 previously described [16]. In the nucleotide sequence, only one codon differs, since the alanine in position 337 in our sequence (Fig. 5) was coded by GCG instead of GCA in the sequence of the fur gene of three strains out of the five strains studied (data not shown). Consequently, these results confirmed our nucleotide sequence of the fur gene of A. baumannii BM2580 and show that the fur gene was highly conserved within clinical strains of A. baumannii.

Figure 4

Detection of A. baumannii BM2580 Fur by immunoblotting with rabbit polyclonal antiserum to E. coli Fur. Lane 1, iron-supplemented medium; lane 2, iron-restricted medium; lane 3, molecular weight markers in kDa. Arrow indicates the Fur protein. Other bands are non-specific.

Figure 5

DNA sequence of the A. baumannii fur gene and flanking regions. Amino acids are indicated by single letter designations and numbered 1–145 in the left margin. Putative −35 and −10 regions are overlined, as are the putative ribosomal binding sites (RBS). A sequence similar to the consensus Fur-binding site (Fur box) is highlighted by shading and identical bases to the consensus are underlined. Asterisks indicate stop codons. Arrows indicate a transcription termination signal (palindromic sequence). Downstream of the transcription termination signal is the 3′-end of a divergent open reading frame (ORF2) with homology to the 3′-region of the pilU gene of Pseudomonas aeruginosa. The sequence reported in this article has been assigned the GenBank/EBI Data Bank accession number Y14980.

3.4 Homology with the PilU protein of P. aeruginosa

A search of GenBank/EBI Data Bank was performed on the 3′-flanking region of the fur gene of A. baumannii BM2580 and revealed that the deduced amino acid sequence of the 3′-flanking region exhibited high similarity to the PilU protein of P. aeruginosa. We thus identified, downstream of the putative transcriptional terminator, the 3′-end of a divergent partial open reading frame (ORF2) with homology to the PilU protein (42% identity over a 60-residue overlap) of P. aeruginosa (Fig. 5). Evidently, P. aeruginosa and A. baumannii do not share a common gene arrangement downstream of fur, since in P. aeruginosa fur and pilU were not linked [17].

3.5 Similarity of A. baumannii Fur with other bacterial Fur

The deduced amino acid sequence of the A. baumannii Fur protein was aligned with the other 22 Fur amino acid sequences currently released in the GenBank/EBI Data Bank. It was 63% identical to the E. coli Fur sequence. A comparable level of similarity was shared between the A. baumannii Fur protein and the Fur proteins of other Gram-negative pathogens listed on Fig. 6 except for the Campylobacter strains and Helicobacter pylori from which the A. baumannii Fur protein highly diverged.

Figure 6

Comparison of Fur proteins sequences from the following GenBank accession numbers: EC, E. coli (X02589); YP, Y. pestis (Z12101); VA, V. anguillarum (L19717); VV, V. vulnificus (L06428); VC, V. cholerae (M86629); PA, P. aeruginosa (L00604); LP, L. pneumophila (U06072); PP, Pseudomonas putida (X82037); BP, B. pertussis (U11699); KP, Klebsiella pneumoniae (L23871); NG, Neisseria gonorrhoeae (L11361); NM, Neisseria meningitidis (L19777); ST, Salmonella typhimurium [19]; HI, Haemophilus influenzae (HI0190); HD, Haemophilus ducreyi (U37224); CU, C. upsaliensis (L77075); CJ, Campylobacter jejuni (Z35165); HP, H. pylori (HP1027); SY, Synechococcus sp. (L41065); SE, S. epidermidis (X97011); SP, Streptococcus pyogenes (U76538); Synechocystis sp (D90903); AB, A. baumannii. Amino acids are indicated by a single letter designation and numbered 1–148 for the E. coli Fur protein. Dashes indicate gaps introduced in the sequence for alignment purposes. Identical amino acids in at least 19 Fur proteins are shaded. Shown in bold letters are the residues Cys and His that are thought to be critical to metal binding.

Given the sufficiently large database of Fur amino acid sequences available from the GenBank/EBI Data Bank, it was possible to identify, with some reliability, the regions which were highly conserved and the potentially required for Fur functions. Three highly conserved motifs have been identified from the Fur amino acid sequences alignment: the first domain in the N-terminal region GLATVYRVL (positions 51–59 of the E. coli Fur, bold and underlined letters represent residues that were found in all the Fur protein sequences currently released in the GenBank/EBI Data Bank); the second domain in the C-terminal region HHDHX2CX2CGXVIEF (X represents intervening residues, positions 87–102 of the E. coli Fur); and the third domain in the N-terminal region LX2GLKVTLPR (positions 8–19 of the E. coli Fur, X represents intervening residues). The three highly conserved motifs identified in our study (GLATVYRVL, HHDHX2CX2CGXVIEF, LX2GLKVTLPR; X represent intervening residues) approximated to the three motifs noted by Achenbach and Yang [18] when reporting the sequence of the Klebsiella pneumoniae fur gene. The A. baumannii Fur contains the three highly conserved motifs LRKAGLKVTLPR, GLATVYRVL, HHDHLVCQNCNKVIEF (where G was substituted by the second N in the case of A. baumannii).

Whereas A. baumannii is an opportunistic human pathogen that presents a potential risk of severe infections for patients, the nature of the factors affecting the virulence of this bacterium is not yet fully understood. While considerable attention has focused on the ferric uptake regulator (fur) locus and the iron metabolism of important pathogenic bacteria, only very few reports have been published on iron metabolism in A. baumannii. Therefore, this study reports that A. baumannii is able to respond to the availability of iron in its environment and codes for a protein homolog to E. coli Fur protein. The expression of a high affinity-mediated iron uptake system and the expression of several proteins under iron limitation as well as the existence of a fur gene could be important virulence factors in the establishment of bacterial infections. Since Fur and Fur-like repressors are known to regulate some virulence-determinant genes in other bacteria [2], the A. baumannii Fur-like repressor protein may also regulate a subset of genes with a role in pathogenesis.


This work was supported by the Conseil Régional Nord-Pas-de-Calais, Centre Hospitalier Régional et Universitaire de Lille, and Direction de la Recherche et des Etudes Doctorales. We are extremely grateful to Dr. Michael Vasil for the gift of the rabbit polyclonal anti-E. coli serum. We are also grateful to Professor Michel Simonet for critical review of the manuscript and to Claude Vandeperre for photography.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
View Abstract