Norepinephrine represses the expression of toxA and the siderophore genes in Pseudomonas aeruginosa

Abstract

Among the different extracellular virulence factors produced by Pseudomonas aeruginosa are exotoxin A (ETA) and the pyoverdine and pyochelin siderophores. Production of ETA and the siderophores requires the function of the iron-starvation sigma factor PvdS, the transcriptional activator RegA, and the AraC-activator PchR. Iron represses production of ETA and the siderophores by repressing expression of pvdS, regA, and pchR. PvdS regulates the expression of the ETA gene, toxA, regA, and the pyoverdine synthesis genes. The catecholamine norepinephrine (NE) enhances the growth of pathogenic bacteria by transferring iron from host binding proteins. In this study, we elucidated the mechanism by which NE and other catecholamines induce P. aeruginosa growth. We also investigated if NE regulates expression of toxA and the siderophore genes; and the mechanism of this regulation. NE enhanced the growth of P. aeruginosa by supplying iron from transferrin. This provision of iron repressed expression of toxA, the pyoverdine genes pvdD and pvdE, and their regulators, pvdS, regA, and pchR, suggesting that NE accomplishes this repression through PvdS and PchR. Additionally, NE bypassed PvdS and supported the growth of a pvdS deletion mutant indicating that NE transfers iron to P. aeruginosa independently of pyoverdine. Thus, NE apparently influences the pathogenesis of P. aeruginosa by affecting its pattern of growth and the production of virulence factors.

Introduction

Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen that causes serious infections in immunocompromised patients including severely burned patients, HIV-infected individuals, cancer patients undergoing chemotherapy and patients suffering from cystic fibrosis (Lyczak et al., 2000; Pollack, 2000). The extensive damage caused by P. aeruginosa is due to the production of several cell-associated and extracellular virulence factors including alginate, pili, exotoxin A (ETA), elastases, alkaline protease, type III secretion effectors, and phospholipase C (Lyczak et al., 2000; Pollack 2000). Under iron-limited conditions within the host, P. aeruginosa can obtain iron through two iron scavenging systems (siderophores); pyoverdine, which has a high affinity for iron (FeIII), and pyochelin, which has a low affinity for iron (Vasil & Ochsner, 1999; Michel et al., 2005;Visca et al., 2007). The uptake of iron-complexed siderophores occurs through specific receptors (outer membrane proteins). These receptors depend on the cytoplasmic protein TonB for their function (TonB-dependent receptors). In P. aeruginosa strain PAO1, there are 34 well-characterized and putative TonB-dependent receptors (Cornelis & Bodilis, 2009), including the pyoverdine receptors FpvA and FpvB (Poole et al., 1993; Ghysels et al., 2004), the pyochelin receptor FptA (Ankenbaur & Quan, 1994), the enterobactin receptors PfeA and PirA (Dean & Poole, 1993; Ghysels et al., 2005), the haem uptake receptors HasR, PhuR, and HuxC (Ochsner et al., 2000), the ferric citrate receptor FecA (Banin et al., 2005), the aerobactin receptor ChtA (Cuiv et al., 2007), the ferrioxamine uptake receptor FoxA (Llamas et al., 2006) and the ferrichrome receptor FiuA (Llamas et al., 2006). The production of ETA and the siderophores by P. aeruginosa, like many bacterial virulence factors, is negatively regulated by iron availability (Vasil & Ochsner, 1999; Hamood et al., 2004). Under iron-replete conditions, expression of the ETA gene toxA, the siderophore genes, and the siderophore receptor genes is repressed through the ferric uptake regulator, Fur (Vasil & Ochsner, 1999, Hamood, et al., 2004). Expression of the main regulators of these genes, PvdS and PchR, is also negatively regulated by iron (Michel et al., 2005; Visca et al., 2007). Other environmental conditions within the host may also influence the production of ETA and the siderophores (Storey et al., 1998; Gallant et al., 2000; Xiong et al., 2000).

Stressful conditions, such as trauma, increase the level of the catecholamine hormone norepinephrine (NE) (Udupa, 1978; Crum et al., 1988). Several studies showed that, in minimal medium containing serum, stress-related neurohormones such as NE enhance the growth of different pathogenic bacteria including enteropathogenic Escherichia coli (Lyte & Ernst, 1992; Lyte et al., 1996; Freestone et al., 1999; Burton et al., 2002; Freestone et al., 2003), Salmonella enterica (Lyte & Ernst, 1992; Lyte et al., 1996; Freestone et al., 1999; Freestone et al., 2007; Karavolos et al., 2008), Campylobacter jejuni (Cogan et al., 2007), Bordetella bronchiseptica (Anderson & Armstrong, 2008), and P. aeruginosa (Lyte & Ernst, 1992; Freestone et al., 1999). One of the mechanisms identified through which NE and other catecholamine stress hormones stimulate the growth of pathogenic bacteria is by facilitating the acquisition of iron from host iron-binding proteins such as transferrin and lactoferrin (Freestone et al., 2000; Freestone et al., 2002; Freestone et al., 2003). Besides stimulating bacterial growth, NE also enhances the production of virulence factors. For example, Lyte et al. (1996) previously showed that NE increases the production of shiga toxin by enterohaemorrhagic E. coli O157:H7. Additionally, Cogan et al. (2007) recently showed that NE increases the growth rate, motility, and invasion of cultured epithelial cells by C. jejuni.

Because NE increases the availability of host iron to bacteria, we were interested in whether this provision might influence expression of genes regulated by environmental iron levels. In this study, we examined the effect of NE on P. aeruginosa growth, the ability of NE to modulate P. aeruginosa uptake of transferrin-complexed iron, and its effects on the expression of the P. aeruginosa iron-regulated toxA and siderophores genes.

Materials and methods

Bacterial strains, plasmids, media and growth conditions

The P. aeruginosa prototypic, genome-sequenced strain PAO1 (Holloway et al., 1979; Vasil & Ochsner, 1999) was utilized to examine the effects of NE on the growth of P. aeruginosa and the expression of virulence genes. The PAO1 pvdS deletion mutant, PAO∷pvdS (Cunliffe et al., 1995), was utilized to examine the relationship between NE and the pyoverdine system. Plasmids used in this study are listed in Table 1. For growth analysis, transferrin-iron uptake measurements and gene expression profiling, the strains were grown overnight in Luria Bertani (LB) broth (Miller, 1972) at 37 °C with shaking at 250 revolutions per minutes (rpm). Aliquots of the cultures were pelleted, washed, resuspended, and serially diluted (1:10) in the minimal salt medium SAPI containing 30% adult bovine serum (SAPIS) (Lyte & Ernst, 1992; Freestone et al., 2000). The subcultures were grown at 37 °C with shaking for various time periods, except for studies involving expression of the toxA and regA genes where the cultures were grown at 32 °C. Time courses of growth were conducted using a dilution that produced 10² or 10³ colony forming units (CFU) mL^-1 (Colmer & Hamood, 1999; Gaines et al., 2005). NE and other stress-related catecholamines were added to SAPIS at a concentration of 500 μM as previously described (Freestone et al., 1999; Freestone et al., 2003; Freestone et al., 2008). Iron was added to SAPIS as FeCl₃ at a concentration of 25 μg mL^-1 of Fe³⁺ (Colmer & Hamood, 1999).

Table 1. Plasmids used in this study.

Plasmids	Description	Reference
pSW228	pSW205 carrying toxA-lacZ fusion, cbr	(West et al., 1994)
pRL88	pSW205 carrying regA-lacZ fusion, cbr	(Storey et al., 1990)
ppvdD	pMP190 carrying pvdD-lacZ fusion, smr, cmr	(Rombel et al., 1995)
pPMP190∷PpvdE	pMP190 carrying pvdE-lacZ fusion, smr, cmr	(Rombel et al., 1995)
pMP220∷PpvdS	pMP220 carrying pvdS-lacZ fusion, tcr	(Ambrosi et al., 2002)
pDH10	pMP190 carrying pchR-lacZ fusion, smr, cmr	(Heinrichs & Poole, 1996)

R, resistance; cb, carbenicillin; sm, streptomycin; cm, chloramphenicol; tc, tetracycline.

For the analyses of the effects of stress hormones on low cell density cultures, P. aeruginosa was inoculated at approximately 50-100 CFU mL^-1 into SAPIS; the precise size of the bacterial inoculum was determined by pour-plate analysis using LB agar. Cultures were incubated statically at 37 °C in a 5% CO₂ humidified incubator for 18 h. At the end of the incubation, the cultures were thoroughly resuspended by vigorous pipetting and the numbers of bacteria were enumerated by pour-plate analysis using LB agar as previously described (Freestone et al., 2000).

Transferrin iron uptake analysis

⁵⁵Fe-labelled transferrin (⁵⁵Fe-Tf) was prepared as described previously (Freestone et al., 2000). Exponentially growing P. aeruginosa were inoculated at 1 × 10⁸ CFU mL^-1 into SAPIS supplemented with 1 × 10⁵ counts per minute (cpm) of ⁵⁵Fe-Tf with and without 500 μM concentrations of the stress hormones indicated in the text. Cultures were incubated at 37 °C in a 5% CO₂ humidified incubator for 6 h, harvested by centrifugation at 5000 g for 5 min, washed in PBS and assayed for cell numbers and ⁵⁵Fe incorporation using pour-plate analysis and scintillation counting as described previously (Freestone et al., 2000). Two independent assays were performed in triplicate; variation within individual assay sets was less than 8%, and between experiments less than 10%.

β-galactosidase assay

For each experiment, three separate flasks of each tested strain were utilized. Duplicates samples were obtained from each time point. The β-galactosidase assay was performed as previously described (Miller, 1972). Cells were grown in SAPIS for 14-16 h to an OD₆₀₀ of 2.5-3.0.

Statistical analysis

Statistics were calculated using InStat (Graph Pad Software). One-way analysis of variance with the Tukey-Kramer multiple comparisons test was used to determine significant differences among three or more growth conditions and across time. Unpaired t tests were used to compare pairs of conditions.

Results

NE enhances the growth of PAO1

Previous studies showed that the addition of NE to SAPIS enhances the growth of different Gram-negative bacteria including P. aeruginosa (Lyte & Ernst, 1992; Freestone et al., 1999; Belay & Sonnenfeld, 2002), a process thought to occur through the ability of NE to obtain iron from the serum iron-binding protein transferrin and transfer it to the bacteria (Freestone et al., 2000). However, the experimental conditions described in those P. aeruginosa growth-related studies differ from the ones that we previously used to examine the expression of iron-regulated genes (Colmer & Hamood, 1999; Gaines et al., 2005). To resolve these differences, we first examined the effect of NE throughout the growth cycle (24 h) of P. aeruginosa in SAPIS at 37 °C, with vigorous shaking at 250 rpm, a starting inoculum of 10³ CFU mL^-1, and the addition of 500 μM NE to the SAPIS. NE did not affect the growth of P. aeruginosa at 4 and 8 h post inoculation, but significantly (P<0.001) enhanced it at 12, 16, 20, and 24 h (Fig. 1).

Fig. 1.

The effect of either NE or iron (Fe) or NE plus Fe on the growth of the P. aeruginosa in SAPIS. An overnight culture of PAO1 was washed, subcultured into fresh LB broth and grown to an OD₆₀₀ at 0.02-0.03. The culture was then serially diluted (10-fold) and inoculated in either SAPIS, SAPIS-NE (500 μM NE), SAPIS-Fe (25 μg mL^-1 Fe³⁺) at a starting inoculum that corresponds to 10³ CFU mL^-1. Cells were grown at 37° C with vigorous shaking at 250 rpm for 16 h and aliquots were obtained every four hours to determine the OD₆₀₀. Values represent the means of three independent experiments ± SEM. One-way analysis of variance (ANOVA) with the Tukey multiple comparisons test was used to determine significance (see text).

As lower starting inocula had been employed to determine the effect of NE on P. aeruginosa growth previously (Lyte et al., 1996), we also examined the effect of NE when cultures were inoculated with 50-100 CFU mL^-1. The low inoculum reflects the number of P. aeruginosa cells that are likely to be present at the initial stage of infection. The cultures were incubated at 37° C in a 5% CO₂, humidified atmosphere under static conditions. Rather than using relative growth index, the number of CFU mL^-1 at 18 h were determined by pour-plate analysis. To determine if the effect of NE is unique or common to other stress hormones, we also evaluated the response of P. aeruginosa to dopamine, epinephrine, and the synthetic inotrope dobutamine. A much greater induction of growth was seen with NE over the unsupplemented control culture (P<0.001) (Fig. 2a). Indeed, all the stress hormones were potent stimulators of growth, inducing nearly 3 log increases in growth over the control (Fig. 2a).

Fig. 2.

(a) Catecholamine stress hormones enhance P. aeruginosa growth. P. aeruginosa was inoculated at 67 CFU mL^-1 into duplicate 1-mL aliquots of SAPIS containing 500 μM concentrations of the catecholamines shown and incubated at 37° C in a 5% CO₂, humidified atmosphere under static conditions for 18 h and enumerated for growth (CFU mL^-1). The results shown are the means of two independent assays performed in triplicate ± SD. A much greater induction of growth was seen with all four catecholamines over the unsupplemented control culture. (b) Growth induction of P. aeruginosa by stress hormones is via provision of iron from transferrin. The data shows the uptake of ⁵⁵Fe from ⁵⁵Fe-transferrin by P. aeruginosa incubated at 1 × 10⁸ CFU mL^-1 for 6 h in the presence of 500 μM concentrations of the catecholamines shown. The values shown represent the means of bacterial ⁵⁵Fe incorporation from triplicate 1 ml assays; standard deviations were less than 10 %. No significant differences in growth levels between control (SAPIS) and catecholamine-supplemented cultures were observed (data not shown).

One-way ANOVA with the Tukey multiple comparisons test was used to determine significance; **P<0.01, ***P<0.001.

Freestone et al. (2000; 2002; 2007) showed that NE and other catecholamine stress hormones provide iron to E. coli, S. enterica, and Yersinia enterocolitica from the host iron-binding protein transferrin. Therefore, we examined whether NE affects the uptake of iron from transferrin by P. aeruginosa. ⁵⁵Fe uptake by P. aeruginosa was significantly greater in the presence of the different catecholamines (Fig. 2b). This indicates that at least part of the mechanistic basis for catecholamine-induced growth induction of P. aeruginosa is via provision of iron from transferrin.

To investigate this hypothesis further, we examined the effect of either iron alone or iron plus NE on the growth of PAO1. Similarly to the effect of NE, the addition of iron to SAPIS enhanced the growth of PAO1 at 12, 16, 20, and 24 h post inoculation (Fig. 1); yet, except for the 12-h time point, this enhancement in PAO1 growth by iron was not significantly different from that induced by NE (Fig. 1). However, upon the addition of both iron and NE, PAO1 growth was significantly higher throughout the growth cycle than that in SAPIS containing NE alone (Fig. 1). PAO1 growth in the presence of NE and Fe was also significantly higher than that in SAPIS containing iron alone, but at 4, 8, 20, and 24 hours. This enhancement indicates that, in addition to the free iron available in the medium, NE mobilizes even more iron to PAO1. These results confirm that NE enhances the growth of P. aeruginosa via provision of iron from host-sequestered sources and indicate that the magnitude of this enhancement is culture-density dependent.

NE represses the expression of toxA, the siderophores genes, and their regulators

Earlier studies revealed that NE enhanced the production of virulence factors in pathogenic bacteria (Lyte et al., 1996; Lyte et al., 1997; Cogan et al., 2007). In contrast, the provision of additional iron to P. aeruginosa by the catecholamines would most likely negatively impact the expression of the genes that are negatively regulated by iron. Among these iron-regulated genes, the most likely targets are toxA and the siderophore genes, whose expression is negatively regulated by iron (Vasil & Ochsner, 1999; Hamood et al., 2004). We, therefore, investigated the modulatory effects of NE on these genes. First, we examined the effect of NE on toxA expression using the toxA-lacZ fusion plasmid pSW228 (West et al., 1994). PAO1/pSW228 was grown in SAPIS and SAPIS-NE for 16 h at 32 °C (toxA is maximally expressed at 32 °C) and the level of β-galactosidase activity determined as previously described (Miller, 1972; Colmer & Hamood, 1999; Gaines et al., 2005). As shown in Fig. 3a, toxA expression in the presence of NE was significantly reduced (P<0.05).

Fig. 3.

NE represses the expression of toxA, pvdD, pvdS, pchR and regA in PAO1. PAO1 carrying plasmids that contain a lacZ fusion in each gene was grown in SAPIS or SAPIS-NE as described in Fig. 1 (except for the expression of toxA and regA, which was done at 32° C) for 16 h and the level of β-galactosidase activity was determined. Values represent the means of three independent experiments ± SEM. (a) toxA expression (pSW228); (b) pvdD expression (ppvdD); (c) pvdS expression (pMP220∷PpvdS); (d) pchR expression (pDH10); and (e) regA expression (pRL88). As a control, we also examined the effect of excess iron (SAPIS-Fe) on regA expression. Excess iron produced similar effects on the expression of toxA, pvdD, pvdS, and pchR (data not shown). Unpaired t tests and one-way ANOVA were used to determine significance; *P<0.05, **P<0.01, ***P<0.001.

NE, by providing iron, may also repress the production of P. aeruginosa siderophores. One of these siderophores is the yellow-green fluorescent pigment, pyoverdine (Visca et al., 2002). Increased levels of iron in the growth medium repress P. aeruginosa pyoverdine production by repressing expression of the pyoverdine synthesis genes (Vasil & Ochsner, 1999; Visca et al., 2002). The coloration of SAPIS-NE precluded using the regular pyoverdine fluorescence assay to examine the effect of NE on pyoverdine production (data not shown). Therefore, we examined the effect of NE on the expression of the pyoverdine synthesis genes pvdD and pvdE using the lacZ fusion reporter constructs ppvdD and pMP190:PpvdE (Rombel et al., 1995), respectively. Cells were grown in SAPIS and SAPIS-NE for 14-16 h to OD₆₀₀ of 2.5-3.0. When P. aeruginosa was cultured in SAPIS-NE, pvdD expression was significantly (P<0.001) repressed (Fig. 3b). NE also significantly reduced pvdE expression (data not shown).

The P. aeruginosa iron-starvation sigma factor PvdS enhances the transcription of toxA and the pyoverdine synthesis genes pvdD and pvdE (Cunliffe et al., 1995; Wilson et al., 2001); but iron represses the expression of the PvdS gene, pvdS (Ochsner et al., 1995). As evidence suggests that iron negatively regulates the expression of toxA and the pyoverdine genes through pvdS (Barton et al., 1996; Vasil & Ochsner, 1999), we determined if NE represses pvdS expression in PAO1 using the pvdS-lacZ fusion plasmid pMP220∷PpvdS (Ambrosi et al., 2005). As shown in Fig. 3c, the addition of NE to SAPIS repressed pvdS expression significantly (P<0.01).

Similarly to pyoverdine synthesis, pyochelin synthesis in P. aeruginosa is a complicated process that involves several genes including the pyochelin biosynthesis operons pchDCBA and pchEFGHI (Serino et al., 1995; Reimmann et al., 1998). The expression of these genes is activated by pyochelin through the AraC-type transcriptional activator PchR (Heinrichs & Poole, 1996; Michel et al., 2005). Additionally, the expression of pchR and the pyochelin synthesis genes is negatively regulated by iron through Fur (Vasil & Ochsner, 1999; Michel et al., 2005). We examined the effect of NE on pchR expression using the pchR-lacZ transcriptional fusion plasmid pDH10 (Heinrichs & Poole, 1996). Similarly to its effect on pvdS expression, NE significantly repressed pyochelin expression in PAO1 (P<0.05) (Fig. 3d). These results suggest that NE represses the expression of toxA, the pyoverdine, and pyochelin genes by repressing their main regulatory genes, pvdS and pchR.

It is possible that this repression of toxA expression by NE may not be directly related to iron availability, and could occur through another toxA regulatory mechanism, such as the main toxA transcriptional activator regA (Frank & Iglewski, 1988). Previous studies showed that in the absence of functional RegA, P. aeruginosa produces neither ETA nor toxA mRNA (Frank & Iglewski, 1988; Vasil & Ochsner, 1999). We examined the effect of NE on regA expression using the regA translational fusion plasmid pRL88 (Storey et al., 1990). Similarly to its effect on toxA expression, NE repressed regA expression in PAO1 significantly (Fig. 3e). As a control, we also examined the effect of excess iron on regA expression. As shown in Fig. 3e, the addition of excess iron repressed regA expression to a greater degree than the addition of NE alone. Similar results were obtained with respect to the expression of toxA, pvdS, and pchR (data not shown).

NE enhances the growth of PAO∷pvdS in SAPIS

Freestone et al. (2003) previously showed that NE depends on the enterobactin siderophore to stimulate the growth of the enterohaemorrhagic E. coli O157:H7. NE enhanced the growth of O157:H7 but not its entA mutant (Freestone et al., 2003). This defect was complemented by the introduction of a plasmid carrying the entA gene (Freestone et al., 2003). It has also been demonstrated that pyoverdine is more critical than pyochelin in supporting the growth of P. aeruginosa in serum (Ankenbauer et al., 1985). In a medium containing 20% human serum, a pyoverdine-deficient mutant grew less efficiently than its parent strain PAO1, while the growth of the pyochelin-deficient mutant in the same medium was similar to that of PAO1 (Ankenbauer et al., 1985). Thus, to enhance the growth of PAO1 through iron, NE may either transfer iron to pyoverdine extracellularly, or transfer iron directly to P. aeruginosa independently of the pyoverdine system. We examined these two possibilities using PAO∷pvdS since; compared with PAO1, pyoverdine production by PAO∷pvdS is significantly reduced (Cunliffe et al., 1995). We compared the growth of PAO1 and PAO∷pvdS in SAPI, SAPIS, and SAPIS-NE. PAO∷pvdS failed to grow in SAPI alone (Fig. 4). The presence of serum in the medium rescued PAO∷pvdS growth, although the growth was markedly less than that of PAO1 (Fig 4). However, the addition of NE significantly (P<0.001) enhanced the growth of PAO∷pvdS to a level even greater than that attained by PAO1 (Fig. 4). These results suggest that NE supports the growth of PAO1 by a pyoverdine-independent mechanism, most likely by acting as xenosiderophores.

Fig. 4.

NE rescues the growth of PAO∷pvdS in SAPIS. PAO1 (white bars) and PAO∷pvdS (black bars) were grown either in SAPI, SAPIS, or SAPIS-NE. Cells were grown as described in Fig. 1 for 16 h, serially diluted in the respective medium, and 10 μL aliquots of each dilution were spotted on LB agar plates. The plates were incubated at 37 °C overnight and the numbers of colonies for each dilution were determined. The number of microorganisms (CFU mL^-1) in each condition was then calculated. Values represent the means of three independent experiments ± SEM. One-way ANOVA was used to determine significance; ***P<0.001. NG, no growth

Discussion

Our results suggest that, while NE is able to supply iron for P. aeruginosa growth, this provision also represses the expression of toxA and the pyoverdine genes by modulating the expression of their main regulator PvdS, and the pyochelin genes through PchR. In our analysis of the effects of host catecholamine stress hormones on P. aeruginosa, we utilized conditions that closely resemble the in vivo environment within the human bloodstream. The growth medium used contained factors such as complement, antibodies, and transferrin that the bacteria would experience when inside the host. We also utilized a low initial inoculum of P. aeruginosa to reflect the numbers of bacteria likely to be present at the beginning of an infection. Furthermore, the amount of NE that was added to SAPIS-NE was within the physiological limits (500 μM) and is comparable to that utilized in a wide diversity of cellular assays, both eukaryotic and prokaryotic, that examine the ability of catecholamines to affect cellular function (Freestone & Lyte, 2008).

One of the major findings of our study is that NE enhanced the growth of PAO1. In the presence of NE (and other catecholamines), the growth of PAO1 was significantly enhanced (Figs. 1 and 2a). Furthermore, we showed that at least part of the mechanistic basis of this growth induction involves provision of iron from the serum iron-binding protein transferrin (Fig. 2b). Iron is essential for in vivo growth of all human pathogens; and, for P. aeruginosa, this element is also an important overall regulator of virulence. The environment within the mammalian body is typically iron limited due to the presence of the host iron-binding proteins transferrin and lactoferrin. Iron restriction by these proteins reduces P. aeruginosa biofilm formation in vitro (Singh, 2004), demonstrating the importance of this host defense. As found with other bacteria (Freestone et al., 2000; Freestone et al., 2002; Freestone et al., 2003), P. aeruginosa grew with much greater ease due to the ability of stress hormones to reduce the iron-binding affinity of transferrin, to the extent that the bacteriostatic nature of serum instead became a highly supportive culture medium.

Our second major finding is that NE represses the production of the P. aeruginosa virulence factors ETA and the siderophores, rather than enhancing virulence production as seen with other bacteria (Lyte et al., 1996; Lyte et al., 1997; Cogan et al., 2007). In the presence of NE, the expression of toxA and the pyoverdine gene pvdD was significantly inhibited (Fig. 3). Additionally, NE also represses the expression of the two main regulatory genes, pvdS and pchR, which are essential for the expression of toxA and the pyoverdine genes (pvdS), as well as the pyochelin operons (pchR) (Fig. 3). The expression of all of these genes is negatively regulated by iron. We showed that the NE-mediated transfer of iron from transferrin to P. aeruginosa increases the intracellular level of iron (Fig. 2b). Within P. aeruginosa, a single mechanism may repress the expression of all of these genes through a mechanism that involves Fur (Vasil & Ochsner, 1999). Upon its activation by iron, Fur specifically binds to the upstream regions of pvdS and regA thereby repressing the expression of toxA and the pyoverdine genes (Vasil & Ochsner, 1999; Visca et al., 2007). Additionally, iron-activated Fur represses the expression of the two pyochelin operons and pchR (Michel et al., 2005). However, several aspects of the complete mechanism through which NE transfers iron into P. aeruginosa are not known at this time. Previous studies indicated that in E. coli O157:H7, the NE-induced transfer of iron depends on the enterobactin siderophore (Freestone et al., 2003). In P. aeruginosa, there are two siderophore systems, of which pyoverdine is the most critical for its survival within host fluid such as blood or serum (Ankenbauer et al., 1985). Using PAO∷pvdS, we have ruled out the possibility that NE-induced transfer of iron involves pyoverdine. While PAO∷pvdS was defective in its growth in SAPIS, this defect was complemented in the presence of NE (Fig. 4). Thus, in our assays, NE may function as a xenosiderophore and internalize iron into P. aeruginosa through one of the putative TonB-dependent receptors (Cornelis & Bodilis, 2009). We also examined the effect of NE on the expression of other virulence genes. Preliminary results showed that NE had no significant effect on the expression of either lasB or the pyocyanin synthesis operon phzA1-G1 (data not shown). Whether NE influences the expression of additional P. aeruginosa virulence genes is not known at this time, but is a future investigative objective of our laboratories.

This study suggests the presence of a trade-off between the growth induction of P. aeruginosa through iron provision by NE in serum, and the iron-induced reduction in virulence factor production. At this time, the in vivo conditions during which NE represses the production of ETA and the siderophores are not understood. During injury such as severe thermal injury, the level of circulating NE increases due to the extensive destruction of the non-adrenergic nerve terminals within the injured tissues (Woolf et al., 1992; Gosain et al., 2009). Yet, previous studies demonstrated the importance of ETA and pyoverdine in the pathogenesis of P. aeruginosa infection of thermally injured wounds; the lethality rate among thermally injured mice that were infected with PAO1 mutants defective in either toxA or the pyoverdine genes was significantly less than that in thermally injured mice infected with PAO1 (Meyer et al., 1996; Fogle et al., 2002). Additionally, high titers of exotoxin A antibody were detected in serum samples obtained from burn patients (Hamood et al., 1996). NE may influence the production of these factors during sepsis (systemic infection) but not at the local site of infection (the thermally injured tissues). It is also possible that by the time P. aeruginosa establishes an infection within the thermally injured wound and invades the blood stream, the trauma-induced increase in the level of NE has subsided. Currently, we are trying to correlate the blood level of NE with the local and systemic spread of P. aeruginosa during different stages of infection of thermally injured wounds.