The Manganese-Responsive Transcriptional Regulator MumR Protects Acinetobacter baumannii from Oxidative Stress

Acinetobacter baumannii is an emerging opportunistic pathogen that primarily infects critically ill patients in nosocomial settings. Because of its rapid acquisition of antibiotic resistance, infections caused by A. baumannii have become extremely difficult to treat, underlying the importance of identifying new antimicrobial targets for this pathogen. Manganese (Mn) is an essential nutrient metal required for a number of bacterial processes, including the response to oxidative stress.

ABSTRACT

Acinetobacter baumannii is an emerging opportunistic pathogen that primarily infects critically ill patients in nosocomial settings. Because of its rapid acquisition of antibiotic resistance, infections caused by A. baumannii have become extremely difficult to treat, underlying the importance of identifying new antimicrobial targets for this pathogen. Manganese (Mn) is an essential nutrient metal required for a number of bacterial processes, including the response to oxidative stress. Here, we show that exogenous Mn can restore A. baumannii viability in the presence of reactive oxygen species (ROS). This restoration is not dependent on the high-affinity Nramp family Mn transporter, MumT, as a ΔmumT mutant is no more sensitive to hydrogen peroxide (H₂O₂) killing than wild-type A. baumannii. However, mumR, which encodes the transcriptional regulator of mumT, is critical for growth and survival in the presence of H₂O₂, suggesting that MumR regulates additional genes that contribute to H₂O₂ resistance. RNA sequencing revealed a role for mumR in regulating the activity of a number of metabolic pathways, including two pathways, phenylacetate and gamma-aminobutyric acid catabolism, which were found to be important for resisting killing by H₂O₂. Finally, ΔmumR exhibited reduced fitness in a murine model of pneumonia, indicating that MumR-regulated gene products are crucial for protection against the host immune response. In summary, these results suggest that MumR facilitates resistance to the host immune response by activating a transcriptional program that is critical for surviving both Mn starvation and oxidative stress.

INTRODUCTION

Acinetobacter baumannii is a Gram-negative bacterial pathogen that has emerged as an increasingly important global public health threat (1). A. baumannii predominantly infects critically ill patients in hospital units, where infections manifest in a number of disease states, including ventilator-associated pneumonias and bloodstream infections (2, 3). Due to its rapid acquisition of antibiotic resistance, A. baumannii was recently classified by the World Health Organization as a “Priority 1” critical pathogen for the development of new antimicrobials (4). Therefore, it is of paramount importance to identify mechanisms by which A. baumannii causes disease within the mammalian host in order to develop novel therapeutics and treatment strategies.

During infection, A. baumannii must withstand a number of environmental stresses imposed by the host immune system, including the production of reactive oxygen species (ROS) by phagocytic cells and the withholding of essential nutrient metals, including iron (Fe), zinc (Zn), and manganese (Mn) by host immune proteins (5–12). Mn is a redox-active trace element that serves as an essential enzymatic cofactor in all domains of life (13). In bacteria, Mn is required for a number of essential processes, including DNA replication and intermediary metabolism, although one of the best-described functions of Mn is in mediating the response to oxidative stress (14–16). ROS are chemically reactive molecules derived from oxygen that can impart severe damage to bacterial cellular components, including proteins, lipids, and nucleic acids (17). In bacteria, Mn protects against toxicity caused by ROS through several mechanisms. Mn serves as a cofactor for a number of enzymes that assist in the detoxification of ROS, including superoxide dismutase (Mn-SOD), which converts superoxide to hydrogen peroxide (H₂O₂) and water, and DNA protection during starvation protein (Dps), which protects against oxidative stress by binding DNA and by sequestering Fe (18, 19). Additionally, Mn can offset ROS-induced toxicity by substituting for oxidized Fe that has been displaced from Fe-dependent mononuclear enzymes (20). Mn also possesses protein-independent antioxidant functions, which involve low-molecular-weight Mn-metabolite complexes that directly detoxify ROS (21–23).

A. baumannii must import Mn and respond to ROS in order to survive within the mammalian host; however, how A. baumannii coordinates these processes to resist the host immune response remains unclear. Previously, we determined that A. baumannii acquires Mn through the high-affinity Nramp family transporter MumT (9). MumT is critical for the growth of A. baumannii in the presence of the S100 family immune protein calprotectin, which binds and chelates Mn with high affinity (9, 24, 25). In addition, we determined that H₂O₂ is produced in the murine lung in response to A. baumannii infection and found that the regulatory protein OxyR facilitates a transcriptional response to this stressor (6). However, strains inactivated for oxyR remain capable of surviving high doses of H₂O₂, suggesting that A. baumannii may utilize additional mechanisms to regulate a transcriptional response to ROS (6). Given the integral nature of Mn in facilitating the defense against ROS in other bacterial species, we hypothesized that Mn-dependent processes may play a crucial role in the response of A. baumannii to oxidative stress.

In this work, we show that exogenous Mn can restore growth to A. baumannii in the presence of H₂O₂. We determined that resistance to ROS does not require the Mn transporter MumT but is dependent on its transcriptional regulator, MumR, suggesting that MumR might regulate other factors required for protection against ROS. In support of this hypothesis, we profiled transcriptional changes in a ΔmumR strain and identified two catabolic pathways, phenylacetate and gamma-aminobutyric acid (GABA) degradation, regulated by MumR that promote protection against H₂O₂ toxicity. Finally, we confirmed the importance of MumR during bacterial infection by assessing the fitness of a mutant lacking this gene in a murine model of pneumonia, suggesting that transcriptional regulation by MumR plays a multifactorial role in mediating protection against immune cell killing.

RESULTS

Manganese rescues A. baumannii from H₂O₂ toxicity.

Previous studies have demonstrated an important role for Mn uptake and homeostasis in the response of bacteria to oxidative stress. In order to investigate the role of Mn in the oxidative stress response of A. baumannii, cultures were grown to mid-exponential phase in the presence or absence of supplemental Mn and exposed to H₂O₂ challenge. While cultures grown in rich media were sensitive to H₂O₂ killing, A. baumannii grown in media supplemented with Mn exhibited enhanced resistance to this oxidant (Fig. 1A). Additionally, while micromolar concentrations of H₂O₂ were sufficient to inhibit A. baumannii growth in rich media, Mn supplementation to the growth medium partially rescued growth under this condition (Fig. 1B). Together, these data suggest that Mn utilization by A. baumannii is critical for defense against oxidative stress.

FIG 1.

Manganese promotes resistance to H₂O₂. (A) A. baumannii cultures were grown to the exponential phase in the presence or absence of 250 μM manganese (Mn) and pulsed with H₂O₂ at the indicated concentrations for 30 min. The means and standard errors of the means of recovered CFU are shown on a log₁₀ scale. The limit of detection was 2-log₁₀. *, P < 0.05; **, P < 0.01 by Student's t test. (B) Growth of A. baumannii in the presence and absence of 300 μM H₂O₂ and 250 μM Mn was monitored by OD₆₀₀, and means and standard errors of the means were graphed.

We previously identified a high-affinity Mn import system, MumT, which supports growth of A. baumannii under Mn-limiting conditions (9). mumT is encoded adjacent to the LysR family transcriptional regulator mumR (Fig. 2A), which acts as a positive regulator of mumT and induces its transcription during Mn starvation (9). We therefore hypothesized that Mn uptake by MumT is crucial for defense against oxidative killing. To directly assess the role of this system in the defense against oxidative stress, mutants lacking either mumT or mumR were grown to mid-exponential phase and exposed to a pulse of H₂O_2. Surprisingly, a ΔmumT strain exhibited slightly heightened resistance to H₂O₂ killing relative to wild-type (WT) A. baumannii, whereas a mutant lacking the regulatory protein MumR was sensitive to killing by H₂O₂ (Fig. 2B). The survival of ΔmumR was partially rescued when cultures were grown with supplemental Mn prior to H₂O₂ exposure (Fig. 2C), suggesting that MumR may regulate other Mn transporters or play additional roles in coordinating Mn homeostasis in response to oxidative stress. In addition, while ΔmumT exhibited a growth defect in rich media, the addition of H₂O₂ into the growth medium only moderately exacerbated this growth defect. By contrast, growth of ΔmumR was significantly more inhibited than a WT strain in the presence of H₂O₂ (Fig. 2D and E). Expression of mumR on a multicopy plasmid restored growth in the presence of H₂O_2, demonstrating that mumR is important for the response of A. baumannii to oxidative stress (Fig. S1 in the supplemental material).

FIG 2.

mumR inactivation sensitizes A. baumannii to H₂O₂. (A) Schematic of the mum locus, with gene names and predicted protein functions. (B) A. baumannii WT, ΔmumR, and ΔmumT cultures were grown to exponential phase and pulsed with H₂O₂ at the indicated concentrations for 30 min. The means and standard errors of the means of recovered CFU are shown on a log₁₀ scale. The limit of detection was 2-log₁₀. **, P < 0.01, ****; P < 0.0001 by one-way analysis of variance (ANOVA) with Dunnett’s multiple-comparison posttest. (C) A. baumannii WT, ΔmumR, and ΔmumT cultures were grown to exponential phase in the presence or absence of 250 μM Mn and pulsed with a dose of 20 mM H₂O₂ for 30 min. The means and standard errors of the means of recovered CFU are shown on a log₁₀ scale. The limit of detection was 2-log₁₀. ****, P < 0.0001 by Student's t test. (D and E) Growth of A. baumannii WT, ΔmumR, and ΔmumT in the absence (solid lines) and presence (dotted lines) of 100 μM H₂O₂ was monitored by OD₆₀₀, and (D) means and standard errors of the means were graphed. Time to mid-log phase (E) was recorded as the time at which individual cultures reached an OD₆₀₀ of 0.5 and was normalized to that of an untreated control group. ***, P < 0.001 by one-way ANOVA with Dunnett’s multiple-comparison posttest.

MumR and OxyR act independently to protect against oxidative stress.

OxyR was recently described as critical for the growth of A. baumannii in the presence of H₂O₂ (6). We therefore hypothesized that MumR may defend against oxidative stress by modulating the expression or activity of OxyR. To test this possibility, a ΔmumRΔoxyR double mutant was constructed, and the growth of this mutant was monitored in the presence or absence of 100 μM H₂O₂. Whereas single mutants lacking mumR or oxyR were partially restricted for growth in the presence of 100 μM H₂O₂, a double mutant lacking both transcriptional regulators was unable to grow under this condition (Fig. 3A and B). While the ΔoxyR mutant is sensitive to growth inhibition by H₂O₂, we have previously shown that this strain exhibits resistance to a single pulse of H₂O₂ at high concentrations (6). In contrast to this phenotype, ΔmumRΔoxyR grown to mid-exponential phase exhibited enhanced sensitivity to H₂O₂ challenge relative to the WT, suggesting that mumR is critical for resisting H₂O₂ killing, even in the absence of oxyR (Fig. 3C). Additionally, ΔmumRΔoxyR was more sensitive to growth restriction by H₂O₂ spotted onto a solid medium than either ΔmumR or ΔoxyR (Fig. 3D). These data reveal that MumR and OxyR do not function epistatically and suggest that these proteins act independently to promote survival and growth in the presence of H₂O₂.

FIG 3.

MumR and OxyR act independently to resist H₂O₂ stress. (A and B) A. baumannii WT, ΔmumR, ΔoxyR, and ΔmumRΔoxyR cultures were grown in the absence (A) or presence (B) of 100 μM H₂O₂, and growth was monitored by OD₆₀₀, and means and standard errors of the means were graphed. Error bars are included in panels A and B, but in most cases are too small to be seen. (C) A. baumannii WT, ΔmumR, ΔoxyR, and ΔmumRΔoxyR cultures were grown to exponential phase and incubated with H₂O₂ at the indicated concentrations for 30 min. The means and standard errors of the means of recovered CFU are shown on a log₁₀ scale. The limit of detection was 2-log₁₀. *, P < 0.05 by one-way ANOVA with Dunnett’s multiple-comparison posttest. (D) Quantification of the diameter of the zone of inhibition by H₂O₂ spotted onto discs placed on a solid medium. *, P < 0.05; ****, P < 0.0001 by one-way ANOVA with Dunnett’s multiple-comparison posttest.

The transcriptome of A. baumannii is altered in the absence of mumR.

To further delineate the mechanism by which MumR protects against oxidative stress, RNA was isolated from WT and ΔmumR cultures grown to mid-exponential phase, and RNA sequencing (RNA-Seq) was performed. Relative to a WT strain, the transcription of 59 genes was altered in ΔmumR (Table 1 and 2 ). Of these genes, 43 were downregulated in ΔmumR (Table 1), while 16 were upregulated (Table 2), suggesting that MumR functions primarily as a transcriptional activator. The repertoire of MumR-regulated genes fell into several broad functional classes, with many predicted to contribute to carbon and nitrogen metabolism (Fig. 4A). Consistent with our previous study, transcripts from mumT and the remaining mum operon genes, including the urea decarboxylase-encoding mumC, were the most strongly downregulated in the ΔmumR mutant (Table 1). The next most highly downregulated transcripts in the ΔmumR mutant belonged to genes involved in a number of catabolic pathways, including the catabolism of phenylacetate, GABA, and acetoin (Table 1). Surprisingly, in the absence of mumR, altered transcription of other genes predicted to contribute to Mn import was not observed, suggesting that outside of regulating mumT, MumR may play additional roles in regulating Mn homeostasis, potentially through the coordination of metabolic pathways.

TABLE 1.

Genes with decreased expression in the ΔmumR strain relative to WT

Function and locus	Annotation (KEGG no.)	Fold change
Amino acid transport and metabolism
A1S_3281	gabT	−42.0
A1S_1349	paaI	−18.2
A1S_3283	GABA permease (K11735)	−17.3
A1S_1348	paaY	−8.7
A1S_1346	paaK2	−7.9
A1S_1345	paaK1	−7.4
A1S_1347	paaX	−6.6
A1S_1344	paaJ	−4.6
General metabolism
A1S_1703	Dihydrolipoamide dehydrogenase	−17.9
A1S_1702	Dihydrolipoamide dehydrogenase (K00382)	−17
A1S_1701	Dihydrolipoamide acetyltransferase (K00627)	−11.9
A1S_1700	Acetoin:2,6-dichlorophenolindophenol oxidoreductase subunit beta (K21417)	−8.5
A1S_1704	Meso-butanediol dehydrogenase/(S,S)-butanediol dehydrogenase/diacetyl reductase (K03366)	−7.0
A1S_1699	Acetoin:2,6-dichlorophenolindophenol oxidoreductase subunit alpha (K21416)	−6.4
A1S_1777	Methylenetetrahydrofolate reductase (NADPH) (K00297)	−4.1
Urea metabolism
A1S_1270	mumC	−25.6
A1S_1267	mumL	−23.3
A1S_1269	mumH	−22.3
A1S_1268	mumU	−19.4
Transport
A1S_1266	mumT	−43.7
A1S_3272	Putative MFS membrane transport protein (K08224)	−4.5
A1S_0474	Catecholate siderophore receptor (K16090)	−4.2
Hypothetical
A1S_3900	Hypothetical protein	−18.6
A1S_3867	Hypothetical protein	−12.1
A1S_1854	Hypothetical protein	−6.9
A1S_3693	Hypothetical protein	−6.4
A1S_3899	Hypothetical protein	−6.4
A1S_3699	Hypothetical protein	−4.8
A1S_3742	Hypothetical protein	−4.5
A1S_3868	Hypothetical protein	−4.4
A1S_3694	Hypothetical protein	−4.0
Other
A1S_1853	Cu2+-containing amine oxidase (K00276)	−11.5
A1S_2459	Putative oxidoreductase	−10.4
A1S_3225	Carbonic anhydrase (K01673)	−9.9
A1S_1591	Phage major capsid protein HK97	−7.4
A1S_1591	Phage major capsid protein HK97	−7.4
A1S_1588	Phage terminase-like protein large subunit	−5.2
A1S_1588	Phage terminase-like protein large subunit	−5.2
A1S_2458	Linoleoyl-CoA desaturase (K00508)	−4.9
A1S_3104	Putative ATP-dependent RNA helicase	−4.8
A1S_1590	Peptidase U35 phage prohead HK97	−4.2
A1S_1590	Peptidase U35 phage prohead HK97	−4.2
A1S_2216	csuB	−4.1
A1S_0109	Homoserine lactone synthase	−4.0

TABLE 2.

Genes with increased expression in ΔmumR strain relative to WT

Function and locus	Annotation (KEGG no.)	Fold change
Transcription
A1S_1746	Putative transcriptional regulator	5.6
A1S_1855	Putative transcriptional regulator	5.1
Transport
A1S_1755	adeT	6.7
A1S_0010	RND type efflux pump	4.8
Hypothetical
A1S_0946	Hypothetical protein	5.6
A1S_0517	Hypothetical protein	5.1
A1S_3780	Hypothetical protein	4.5
A1S_3121	Hypothetical protein	4.1
Other
A1S_3122	Putative membrane protein	8.3
A1S_0945	Putative ferredoxin	7.4
A1S_0359	Putative beta-lactamase	4.9
A1S_0566	NAD(P) transhydrogenase subunit alpha (K00324)	4.6
A1S_0567	NAD(P) transhydrogenase subunit alpha (K00324)	4.4
A1S_3050	Putative membrane protein	4.2
A1S_2850	Putative acyl-CoA transferase/carnitine dehydratase	4.2
A1S_0568	NAD(P) transhydrogenase subunit beta (K00325)	4.0

FIG 4.

The transcriptome of A. baumannii is altered in the absence of mumR. RNA sequencing was performed on RNA extracted from WT A. baumannii and ΔmumR cells grown to mid-exponential phase and treated with 5 mM H₂O₂ or a vehicle control for 10 min. (A) Classification of transcript IDs significantly up- or downregulated in ΔmumR A. baumannii treated with vehicle control compared to WT cells (>2 log₂ fold change [FC]; P < 0.05, corrected for FDR). (B and C) Venn diagrams comparing common upregulated (B) and downregulated (C) transcripts in WT and ΔmumR following a 10-min exposure to 5 mM H₂O₂ (>2 log₂ FC; P < 0.05 corrected for FDR).

We have previously defined the response of A. baumannii to H₂O₂ stress and found that transcription of nearly 300 genes is altered under this condition (6). To determine whether MumR is important for inducing these transcriptional changes, the transcriptional profile of ΔmumR in the presence of H₂O₂ was determined (Table S2 and S3). Under this condition, the majority of transcriptional changes occurring in WT A. baumannii in the presence of H₂O₂ were not dependent on MumR, as transcriptional profiles of ΔmumR and WT were generally similar following exposure to H₂O₂ (Fig. 4B and C). In addition, we did not observe significant changes in mumR transcript levels in a WT strain following H₂O₂ treatment, suggesting that transcription of mumR is not being induced in this condition (6). Together, these data suggest that genes under MumR transcriptional control during growth in rich media may play a role in defending against ROS, perhaps by allowing A. baumannii to bypass pathways or processes that are damaged by oxidative stress.

MumR-regulated catabolic pathways promote defense against H₂O₂ killing.

Based on our RNA sequencing data, we hypothesized that genes regulated by MumR during growth in rich media may be protective against oxidative stress. To test this hypothesis, genes in two MumR-regulated pathways, phenylacetate catabolism and GABA catabolism, were deleted, and the resistance of the resulting mutants to H₂O₂ was assessed. These pathways were prioritized for future study, as transcripts from phenylacetate and GABA degradation genes were among the most strongly downregulated in the ΔmumR strain (Table 1). Consistent with our hypothesis, a mutant lacking the GABA transaminase-encoding gabT exhibited reduced growth in the presence of H₂O₂ and was more sensitive to killing by H₂O₂ (Fig. 5A to C). In addition, a ΔpaaJKXYI mutant, which lacks part of the phenylacetate degradation pathway (26, 27), was more susceptible to H₂O₂ killing than a WT strain (Fig. 5C). However, this mutant was no more restricted for growth in the presence of H₂O₂ than WT (Fig. 5B), suggesting that phenylacetate degradation may only be required to withstand H₂O₂ encountered at high concentrations.

FIG 5.

MumR-regulated pathways promote defense against H₂O₂ killing. (A and B) A. baumannii WT, ΔmumR, ΔgabT, and ΔpaaJKXYI cultures were grown in the absence (A) or presence (B) of 200 μM H₂O₂; growth was monitored by OD₆₀₀, and means and standard errors of the means were graphed. (C) A. baumannii WT, ΔmumR, ΔgabT, and ΔpaaJKXYI cultures were grown to exponential phase and incubated with H₂O₂ at the indicated concentrations for 30 min. The means and standard errors of the means of recovered CFU are shown on a log₁₀ scale. The limit of detection was 2-log₁₀. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 by one-way ANOVA with Dunnett’s multiple-comparison posttest.

MumR promotes the fitness of A. baumannii in a murine model of pneumonia.

Because A. baumannii experiences both oxidative stress and Mn restriction during infection of mammalian hosts (6, 9), we hypothesized that MumR might contribute to the pathogenesis of this organism during infection. Mice were infected intranasally with WT, ΔmumR, or an inoculum containing an equal mixture of both strains, and bacterial burdens were enumerated after 36 h of infection. Consistent with this hypothesis, the ΔmumR strain was recovered from the lungs of mice at lower levels than WT A. baumannii, regardless of the presence of a WT strain in the same inoculum (Fig. 6A and B). Interestingly, WT A. baumannii was recovered from the lungs and livers of mice at lower levels following coinfection with ΔmumR than when mice were inoculated with WT alone (Fig. 6A and B), highlighting the possibility that MumR may regulate genes that possess immunomodulatory functions. Together, these data support a model in which MumR promotes fitness during bacterial pneumonia by regulating genes critical for defense against Mn restriction, oxidative killing, and immunomodulation of the host.

FIG 6.

MumR promotes the fitness of A. baumannii in a murine model of pneumonia. Bacterial burdens were enumerated from the lungs (A) and livers (B) of mice 36 h following intranasal inoculation of WT, ΔmumR, or a 1:1 mixture of both strains. Mono- and coinfection data are each combined from two independent experiments. The limit of detection was 2-log₁₀. *, P < 0.05; **, P < 0.01 by Kruskal-Wallis with Dunn’s multiple-comparison posttest.

DISCUSSION

During infection, microbes must compete with the host immune system to obtain nutrients, including the essential transition metal Mn. As a result of this selective pressure, bacterial pathogens have evolved complex regulatory mechanisms to sense and respond to Mn limitation within tissue sites (13, 28). These regulatory mechanisms can play multifaceted roles during infection; for example, Mn supports bacterial growth, but it can also aid in the detoxification of ROS produced by host immune cells (18–23). As a result, many bacterial pathogens coordinate their transcriptional responses to Mn limitation and oxidative stress. In Escherichia coli and Salmonella, OxyR serves as a transcriptional activator of mntH, which encodes an Nramp family Mn transporter (29, 30). Additionally, the metalloregulator MntR of Streptococcus oligofermentans functions dually as both an Mn and redox sensor (31). We have previously shown that the pathogen A. baumannii induces a transcriptional response to H₂O₂ stress that is largely dependent on OxyR (6). However, mutants inactivated for oxyR remain resistant to H₂O₂ killing, suggesting that A. baumannii encodes other regulatory mechanisms required to withstand oxidative stress (6).

In this work, we characterize the interplay between the Mn starvation and oxidative stress responses in A. baumannii. We previously determined that the transcriptional regulator MumR induces the expression of an Mn import system during conditions of Mn starvation (9). Here, we show that MumR also plays a role in promoting resistance to H₂O₂ stress during conditions where Mn is replete (Fig. 2). While assays were performed under Mn-replete conditions, we observed a rescue in A. baumannii survival when additional Mn was supplemented into the growth media prior to H₂O₂ exposure, suggesting that acquisition or utilization of Mn might be critical for defense against ROS (Fig. 2). However, this protection is not mediated through the regulation of the Nramp family Mn transporter mumT, as a ΔmumT strain is more resistant to H₂O₂ killing than WT A. baumannii (Fig. 2). To determine whether MumR might regulate other factors important for protecting against ROS in Mn-replete conditions, we profiled transcriptional changes in a ΔmumR strain and determined that at least two MumR-regulated catabolic pathways, phenylacetate and GABA metabolism, promote resistance to H₂O₂ stress (Fig. 5). Finally, we evaluated the fitness of a ΔmumR strain in vivo and determined that this factor is required for full virulence in a murine pneumonia model, suggesting that MumR-regulated gene products are critical for protecting against restriction by the host immune response (Fig. 6).

MumR shares homology to the LysR family of proteins, which represent a conserved class of transcriptional regulators containing an N-terminal DNA-binding helix-turn-helix motif and a C-terminal coinducer-binding domain (32). The LysR family of transcriptional regulators includes the redox-sensitive protein OxyR, which contains two conserved redox-active cysteine residues that modulate its activity (33, 34). In this study, we found that MumR does not modulate the activity of OxyR, as a double mutant lacking both genes was more sensitive to oxidative stress than either single mutant alone (Fig. 3). Additionally, while the coding sequence of MumR contains two cysteine residues in its predicted substrate-binding domain, the transcriptional signature observed in the ΔmumR strain following H₂O₂ treatment was similar to WT A. baumannii, suggesting that MumR does not specifically sense alterations in the redox environment of the bacterial cell (Fig. 4).

Alternatively, we hypothesize that pathways regulated by MumR under normal growth conditions might play a critical role in the defense against ROS. Transcription of a number of catabolic pathways, including phenylacetate and GABA metabolism, was downregulated in the ΔmumR strain during growth in rich media. Mutants lacking the GABA transaminase gabT or the phenylacetate degradation genes paaJKXYI exhibited heightened sensitivity to ROS, suggesting that the functions of these pathways play an important role in the response of A. baumannii to oxidative stress (Fig. 5). In E. coli, the transaminase GabT functions as the first of two metabolic steps in the GABA shunt, a pathway that produces succinate from the amino acid GABA (35, 36). In A. baumannii and other bacteria, the paaJKXYI gene cluster encodes part of an operon that converts the aromatic compound phenylacetate, a breakdown product of phenylalanine, into succinyl coenzyme A (succinyl-CoA) and acetyl-CoA (26, 27, 37). While the specific mechanisms by which these pathways protect from oxidative stress are still unknown, it is possible that the products of GABA and phenylacetate catabolism might be required to repair or recover from oxidative damage. For example, the utilization of these pathways might allow A. baumannii to bypass other metabolic pathways impaired by oxidative stress, such as those requiring iron-sulfur cluster-containing enzymes critical for carrying out a multitude of metabolic functions (38, 39).

It is also possible that the induction of GABA and phenylacetate catabolism is linked to Mn homeostasis within the bacterial cell. We have previously shown that the amino acid histidine binds Zn in A. baumannii and that histidine catabolism is upregulated during Zn limitation. Together, these data support a model in which histidine serves as a component of the labile Zn pool in A. baumannii (10). Similarly, it is possible that GABA, phenylacetate, or intermediates within their catabolic pathways might function as intracellular reservoirs of Mn. In this case, activation of these pathways by MumR would liberate Mn, which could then be bioavailable to serve as an enzymatic cofactor or to detoxify ROS directly. Within bacterial cells, the labile Mn pool is believed to consist of Mn in complex with low-molecular-weight metabolites (40, 41); however, the specific intracellular reservoir of Mn has yet to be defined. The potential role phenylacetate and GABA catabolism might play in Mn homeostasis thus opens up an intriguing avenue of future research.

Finally, we show in this work that MumR is critical for the full virulence of A. baumannii in a murine pneumonia model of infection. While we have previously demonstrated the importance of the MumR-regulated Mn transporter MumT during infection (9), our study supports a model in which MumR functions in response to both Mn limitation and oxidative stress during times of infection. Additionally, we observed that coinfection with a ΔmumR strain reduces the virulence of WT A. baumannii during infection, suggesting that MumR-regulated genes may have immunomodulatory functions (Fig. 6). Intermediates of the phenylacetate degradation pathway modulate neutrophil recruitment to A. baumannii infection sites (37). Thus, we hypothesize that in addition to regulating responses to Mn limitation and ROS, MumR may also function to limit immune cell recruitment during A. baumannii infection.

Together, this work contributes to our understanding of the emerging multidrug-resistant pathogen A. baumannii and highlights the relationship between Mn homeostasis, the oxidative stress response, and bacterial virulence in this organism. Due to the interplay of these processes in A. baumannii, targeting these responses could serve as a promising therapeutic strategy for the treatment of this emerging global health threat.

MATERIALS AND METHODS

Bacterial strains and reagents.

The strains used in this study are described in Table 3. All experiments were performed using the clinical isolate A. baumannii ATCC 17978 or a mutant derivative. Cloning was performed using E. coli DH5α. Strains were cultured in Luria-Bertani broth (LB) at 37°C. Kanamycin (Km) (Sigma) was used at 40 μg/ml concentration, carbenicillin (Carb) (Fisher) was used at 75 μg/ml, chloramphenicol (Cm) (Fisher) was used at 15 μg/ml, and tetracycline (Tet) (Alfa Aesar) was used at 10 μg/ml for selections.

TABLE 3.

Bacterial strains used in this study

Strain	Relevant characteristics	Reference or source
17978	Wild type	ATCC
ΔoxyR	In-frame ΔoxyR::aphA	6
ΔmumR	In-frame ΔmumR::aphA	9
ΔmumRΔoxyR	In-frame ΔoxyR::tetA ΔmumR::aphA	This study
ΔgabT	In-frame ΔgabT::aphA	This study
ΔpaaJKXYI	In-frame ΔpaaJKXYI::aphA	This study
17978 pPr01MU368tet	Empty vector control	This study
ΔmumR pPr01MU368tet	Empty vector control	This study
ΔmumR pPr01.mumR.MU368tet	Complementation strain containing plasmid p.Pr01.mumR.MU368tet	This study
Escherichia coli DH5α p.oxyR::tetA.FLP2	Cloning strain containing oxyR knockout vector	This study
Escherichia coli DH5α p.gabT::aphA.FLP2	Cloning strain containing gabT knockout vector	This study
Escherichia coli DH5α p.paaJKXYI::aphA.FLP2	Cloning strain containing paaJKXYI knockout vector	This study

Strain generation.

The primers used in this study are listed in Table S1 in the supplemental material. ΔmumRΔoxyR, ΔpaaJKXYI, and ΔgabT mutants were created via allelic exchange as follows. For generation of the ΔpaaJKXYI and ΔgabT mutants, approximately 1,000 bp of DNA in both the 5´ and 3´ flanking regions surrounding gabT and 1,000 bp of DNA in the 5´ flanking region of paaJ and 1,000 bp of DNA in the 3´ flanking region of paaI were amplified using A. baumannii genomic DNA as a PCR template. The kanamycin resistance gene aphA was amplified by PCR from the vector pUCK1. These products were cloned into the pFLp2 vector using NEBuilder HiFi Assembly (New England Biolabs). The resulting pFLp2 constructs were then introduced into WT A. baumannii by triparental conjugation using an E. coli HB101 strain containing the helper plasmid pRK2013. Matings were plated onto LB Km40 Cm15 agar to select for strains containing the integrated plasmid. Strains were then plated onto agar containing 10% sucrose to select for clones that had resolved the integrated plasmid, and resulting sucrose-resistant colonies were patched onto LB Kan40 to screen for the loss of gabT or paaJKXYI and replacement with aphA. Deletion of loci was confirmed by multiple PCRs by use of both A. baumannii and aphA-specific primers. For construction of the ΔmumRΔoxyR mutant, 1,000 bp of DNA in both the 5´ and 3´ flanking regions surrounding oxyR was amplified by PCR. The tetracycline resistance gene tetA was amplified using genomic DNA isolated from the tetracycline-resistant A. baumannii isolate AB0057 as a PCR template. Products were cloned into pFLp2, and the resulting plasmid was introduced into ΔmumR by conjugation as described above. Trans-conjugates were selected for by plating matings onto LB Tet10 Cm15 agar, and the replacement of oxyR with tetA was selected for and confirmed as described above. The mumR complementation vector was constructed in pMU368 under the control of the 16S rRNA promoter (r01) by PCR amplification of the open reading frame (ORF), double digestion of the vector and PCR product with BamHI-HF and SacI-HF (New England Biolabs), and ligation with T4 DNA ligase (Promega).

H₂O₂ growth assays.

Bacterial strains were freshly streaked onto LB agar, and single colonies were inoculated into an LB growth medium. Cultures were grown overnight at 37°C with shaking at 180 rpm, subcultured at 1:50 into fresh LB medium for 1 h, and then diluted at 1:100 into LB containing H₂O₂ (30%; EMD Millipore), MnCl₂ (Sigma), or a combination of both at concentrations indicated in figure legends. Growth assays were performed using 96-well plates in 100 μl volumes, and growth was measured by optical density at 600 nm (OD₆₀₀) at 30-min intervals.

H₂O₂ killing assays.

Bacterial cultures were inoculated from freshly streaked colonies and grown overnight in LB medium at 37°C with shaking at 180 rpm. Cultures were then diluted 1:1,000 into LB or LB supplemented with 250 μM MnCl₂ (Sigma) and grown to mid-exponential phase. Bacteria were subsequently incubated with H₂O_2. Due to experimental variability in A. baumannii recovery after H₂O₂ treatment, incubations were performed using a range of H₂O₂ concentrations, as indicated in figure legends. Following 30 min of incubation, bacterial cultures were serially diluted in phosphate-buffered saline (PBS) containing catalase (2,000 U/ml, catalase from bovine liver; Sigma), and dilutions were spot plated onto LB agar for CFU enumeration.

H₂O₂ disc diffusion assays.

Bacterial cultures were grown overnight from freshly streaked colonies and then diluted 1:10 in LB. One hundred microliters of each diluted culture was added to 4 ml of melted soft agar (0.75%) and immediately poured onto LB agar plates and dried for 5 min. Sterile 6-mm paper discs were placed onto the centers of the plates and loaded with 10 μl of 30% H₂O₂ (EMD Millipore). After overnight incubation at 37°C, the diameter of the zone of growth inhibition surrounding the disc was measured.

Mouse infections.

WT A. baumannii and a kanamycin-resistant ΔmumR mutant were freshly streaked onto LB agar. One day prior to infection, single colonies of each strain were inoculated into LB medium and grown overnight at 37°C with shaking at 180 rpm. The next day, stationary-phase bacteria were subcultured at 1:1,000 and grown to mid-exponential phase. Bacteria were then centrifuged, washed twice in PBS, and resuspended in PBS at a concentration equivalent to 1 × 10¹⁰ CFU/ml. For competitive infections, suspensions of WT and ΔmumR A. baumannii were combined at a 1:1 ratio and mixed thoroughly. Prior to infection, mice were anesthetized by intraperitoneal injection of 2,2,2 tribromoethanol diluted in PBS. Anesthetized mice were infected intranasally with 4 × 10⁸ CFU of the inoculum in a 40-μl volume. Infections proceeded for 36 h. Mice were then euthanized by forced CO₂ inhalation, and lungs and livers were sterilely harvested and placed on ice. Organs were homogenized, and serial dilutions of homogenized tissues were spot plated onto LB agar for enumeration. For competitive infections, dilutions were plated onto both LB agar and LB agar containing 40 μg/ml kanamycin. All animal experiments were approved by the Vanderbilt University Medical Center (VUMC) Institutional Care and Use Committee and conformed to policies and guidelines established by VUMC, the Animal Welfare Act, the National Institutes of Health, and the American Veterinary Medical Association.

Growth for RNA-Seq and RNA isolation.

Cultures of WT and ΔmumR A. baumannii were inoculated from freshly streaked colonies and grown overnight in biological triplicate at 37°C with shaking at 180 rpm. Following overnight growth, cultures were subcultured at 1:1,000 into 10 ml fresh LB medium and grown to mid-exponential phase. Cultures were then treated with 5.65 μl of 30% H₂O₂ (EMD Millipore) at a final concentration of 5 mM or with a vehicle control. After addition of H₂O₂ or vehicle control treatment, cultures were incubated for 10 min at 37°C with shaking at 180 rpm. Following treatment, 10 μl of bacterial cultures were saved for CFU enumeration, while the remaining cultures were immediately mixed with 10 ml of a 1:1 mixture of ice-cold acetone (J.T.Baker) and 70% ethanol (Sigma) and frozen at –70°C. CFU were enumerated by serial dilution in PBS containing catalase (2,000 U/ml, catalase from bovine liver; Sigma) and spot plating onto LB agar. For RNA purification, frozen cell suspensions were thawed and centrifuged at 7,000 rpm for 10 min. Pellets were dried and resuspended in LETS buffer (0.1 M LiCl, 10 mM EDTA, 10 mM Tris HCl [pH 7.4%], 1% SDS), and homogenized in a bead beater with Lysing Matrix B beads (MP Biomedicals) at a speed of 6 m/s for 45 s. Homogenized suspensions were then heated at 55°C for 5 min and centrifuged for 10 min at 15,000 rpm. The upper phase was collected and mixed with 1 ml of TRI Reagent (Sigma) and incubated for 5 min at room temperature. Two hundred microliters of chloroform (Acros Organics) were then added to samples and mixed by vigorous shaking for 15 s. Samples were incubated at room temperature for 2 min and centrifuged at 4°C for 15 min, and 600 μl of the upper aqueous phase was collected. RNA was precipitated from the upper aqueous phase by mixing with 1 ml isopropanol (Sigma) and washing with 70% ethanol (Sigma). Pellets containing purified RNA were then resuspended in 100 μl DNase/RNase-Free water (Thermo Fisher). DNA contamination was removed by treating with 8 μl RQ1 enzyme (Promega), 12 μl 10× RQ1 buffer, and 2 μl Riboblock RNase inhibitor (Thermo Fisher) for 2 h at 37°C. DNase was removed by purifying samples with the RNeasy kit (Qiagen) according to the manufacturer’s instructions. RNA was stored at –80°C.

RNA-Seq library preparation and sequencing.

RNA-seq library construction and sequencing were performed by HudsonAlpha Institute for Biotechnology (Huntsville, AL). The concentration and integrity of extracted total RNA were estimated by a Qubit 2.0 Fluorometer (Invitrogen) and an Agilent 2100 Bioanalyzer (Applied Biosystems), and 500 ng of RNA was utilized for downstream applications. rRNA was removed using the Ribo-Zero Gold (Epidemiology) kit (Illumina) according to the manufacturer’s instructions. Following rRNA removal, RNA was fragmented and primed for first-strand synthesis using the NEBNext first strand synthesis module (New England BioLabs). Directional second-strand synthesis was performed using the NEBNext Ultra Directional second strand synthesis kit. Libraries were then prepared from samples using the NEBNext DNA Library Prep master mix set (Illumina) with the following slight modifications. End repair was performed and followed by polyadenylic acid (poly[A]) addition and custom adapter ligation. Ligated samples were individually barcoded with unique in-house Genomic Services Lab (GSL) primers and amplified through 12 cycles of PCR. Library quantity was assessed by a Qubit 2.0 Fluorometer, and the library quality was assessed by utilizing a DNA High Sensitivity Chip on a Caliper GX (PerkinElmer). The quantitative PCR (qPCR)-based Kapa Biosystems library quantification kit (Kapa Biosystems) was used for final accurate quantification of libraries prior to sequencing. Each library was diluted to a final concentration of 12.5 nM and pooled equimolar prior to clustering. Paired-end sequencing was performed on an Illumina HiSeq2500 sequencer (Illumina). Raw RNA sequencing data and processed data are deposited in the NCBI Gene Expression Omnibus (GEO) under accession number GSE114130.

Processing of RNA-Seq reads.

RNA-Seq analysis was performed by the HudsonAlpha Institute for Biotechnology (Huntsville, AL). Approximately 25 million, 100-bp paired-end reads were generated from each sample. Further downstream analysis of the sequenced reads from each sample was performed as per the HudsonAlpha unique in-house pipeline. Briefly, quality control checks on raw sequence data from each sample were performed using FastQC (Babraham Bioinformatics, London, UK). Raw reads were imported into the commercial data analysis platform, Avadis NGS (Strand Scientifics, CA, USA) and mapped to the reference genome A. baumannii ATCC17978. After quality inspection, the aligned reads were filtered on the basis of read-quality metrics in which reads with a base quality score less than 30, alignment score less than 95, and mapping quality less than 40 were removed. Remaining reads were then filtered on the basis of their read statistics, and missing mates, translocated, unaligned, and flipped reads were removed. The reads list was then filtered to remove duplicates. Samples were grouped, and quantification of transcript abundance was done on this final read list by use of Trimmed Means of M-values (TMM) (42) as the normalization method. Differential expression of genes was calculated on the basis of fold change (using default cutoff greater than or equal to ±2.0) observed between defined conditions, and the P value of the differentially expressed gene list was estimated by Z-score calculations determined by Benjamini Hochberg false-discovery rate (FDR) correction of 0.05 (43).

Statistical analysis.

Raw data were recorded in Microsoft Excel and imported into GraphPad Prism for statistical analysis. Specific tests used to test for statistical significance for each experiment are indicated in figure legends.

Data availability.

Any materials and data will be made available to members of the scientific community upon communication with the corresponding author.