Investigation of the human pathogen Acinetobacter baumannii under iron limiting conditions

Abstract

Background: Iron acquisition systems are important virulence factors in pathogenic bacteria. To identify these systems in Acinetobacter baumannii, the transcriptomic response of the completely sequenced strain ATCC 17978 under iron limiting conditions was investigated using a genomic microarray that contained probes for all annotated open reading frames.

Results: Under low iron conditions, transcription levels were more than 2-fold up-regulated for 463 genes, including 95 genes that were up-regulated more than 4-fold. Of particular significance, three siderophore biosynthesis gene clusters, including one novel cluster, were highly up-regulated. Binding sites for the ferric uptake regulator were identified in the promoter regions of many up-regulated genes, suggesting a prominent role for this regulator in the Acinetobacter iron acquisition response. Down-regulation under iron limitation was less dramatic as the transcription of only 202 genes varied more than 2-fold. Various genes involved in motility featured prominently amongst the genes down-regulated when iron was less readily available. Motility assays confirmed that these transcriptional changes are manifested at the phenotypic level. The siderophore biosynthesis gene clusters were further investigated by means of comparative genomic analysis of 10 sequenced Acinetobacter isolates. These analyses revealed important roles for mobile genetic elements in shaping the siderophore meditated iron acquisition mechanisms between different Acinetobacter strains.

Conclusions: A. baumannii grown under iron limited conditions resulted in major transcriptional changes of not only many iron acquisition related genes, but also genes involved in other processes such as motility. Overall, this study showed that A. baumannii is well adaptable to growth in an environment which has limiting iron availability.

Figure 1

Growth curves of A. baumannii with varying iron concentrations. Growth under different iron concentrations was tested in Mueller-Hinton (MH) broth and MH supplemented with 2,2'-dipyridyl (DIP) to final concentrations of 100, 200 and 300 μM. Absorbance was measured every hour at OD₆₀₀ for 6 hours; the data represent the average of three separate experiments and the error bars show the standard deviation.

Figure 2

Overview of transcriptional responses to iron starvation. The A. baumannii ATCC 17978 transcriptome was compared under iron replete and iron limiting conditions (200 μM DIP). All 3367 genes of the A. baumannii ATCC 17978 genome are represented on the X-axis, ordered according locus tag. Differential expression levels between iron replete and iron limiting conditions are displayed in Log₂-values on the Y-axis. Up- and down-regulated genes under iron limiting conditions are displayed in the green and red sections, respectively. Gene names or functions have been provided for various highly differentially expressed genes, such as siderophore biosynthesis genes and the fumarases. The putative FUR binding sites displayed in the top section were identified as described in Methods and the nucleotide sequences are listed in Additional file 2.

Figure 3

Validation of the microarray results. The transcriptomic results obtained by microarray hybridisation were validated by quantitive RT-PCR (qRT-PCR) analysis. The level of differential expression of 12 genes was compared and showed a correlation between microarray (Y-axis) and qRT-PCR analysis (X-axis). The level of differential expression between iron replete and iron limitation is given in Log₂-values.

Figure 4

Microarray results displayed by COG-function. Depiction of cluster of orthologous groups (COG) and the percentage of up-regulated (green) and down-regulated genes (red) within such group determined by microarray. The total number of genes per COG is shown in parentheses.

Figure 5

The optimized A. baumannii FUR motif. Per position, the size of the nucleotide (T in red, A in green, C in blue and G in black) indicates its prevalence in the 22 included sequences from Table 1. The motif shows a palindrome with a central non-conserved nucleotide in position 13 which is indicated by the star. The figure of the A. baumannii ATCC 17978 FUR motif was created using WebLogo 3.0 [64].

Figure 6

Transcriptional profiling of three siderophore gene clusters identified in A. baumannii ATCC 17978. Transcriptional alteration of the three siderophore gene clusters to low iron conditions are shown, (A) siderophore gene cluster 1 (A1S_1647-1657), (B) siderophore cluster 2 (A1S_2562-2581) and (C) the acinetobactin gene cluster (A1S_2372-2392). The top arrows show predicted gene function; siderophore biosynthesis in purple, receptors and uptake mechanisms in light orange, efflux pumps in orange and genes of unknown function in white. The relative transcriptional differences between A. baumannii grown under iron replete and iron limiting conditions are depicted in the bottom set of arrows according to the green color scale bar, all values are in times-fold difference. Genes depicted in white were not differentially expressed and those in dark green were overexpressed more than 25-fold. No significant down-regulation was observed within the siderophore gene clusters. Putative FUR boxes are shown as black arrows.

Figure 7

Gene clusters with a putative role in motility. Many genes in the motility COG were found down-regulated including genes of (A) type I and (B) type IV pili. The top set of arrows show predicted gene function; pilus proteins in orange, pilus assembly proteins in purple, and regulatory proteins in light blue. The relative transcriptional differences between A. baumannii grown under iron rich and iron limiting conditions is depicted in the bottom set of arrows, all values are in times-fold. Genes depicted in white were not differentially expressed, those shaded in light red were down-regulated less than 2-fold, whereas those in dark red were down-regulated 2-fold or more. No significant up-regulation was observed within gene clusters related to motility.

Figure 8

Swarming of A. baumannii ATCC 17978. A. baumannii colony material was spotted on Luria-Bertani medium containing 0.25% agar. Swarming motility (A) is visible as the channel-like growth around the dense white colony material. The absence of halo growth around the colony (B) indicates lack of swarming motility of A. baumannii when available iron is limited.

Figure 9

Comparison of siderophore cluster 1 in sequenced Acinetobacter isolates. The alignment of siderophore cluster 1 between strain ATCC 17978, ADP1 and SDF. Arrows indicate open reading frames, in black genes with high homology (> 50% identity) and in grey genes with low homology (< 50% identity). White genes represent a 3.5 Kb transposon insertion, which replaced the A1S_1648-1656 orthologs in strain SDF.