PQS and pyochelin in Pseudomonas aeruginosa share inner membrane transporters to mediate iron uptake

ABSTRACT Bacteria absorb different forms of iron through various channels to meet their needs. Our previous studies have shown that TseF, a type VI secretion system effector for Fe uptake, facilitates the delivery of outer membrane vesicle-associated Pseudomonas quinolone signal (PQS)-Fe3+ to bacterial cells by a process involving the Fe(III) pyochelin receptor FptA and the porin OprF. However, the form in which the PQS-Fe3+ complex enters the periplasm and how it is moved into the cytoplasm remain unclear. Here, we first demonstrate that the PQS-Fe3+ complex enters the cell directly through FptA or OprF. Next, we show that inner membrane transporters such as FptX, PchHI, and FepBCDG are not only necessary for Pseudomonas aeruginosa to absorb PQS-Fe3+ and pyochelin (PCH)-Fe3+ but are also necessary for the virulence of P. aeruginosa toward Galleria mellonella larvae. Furthermore, we suggest that the function of PQS-Fe3+ (but not PQS)-mediated quorum-sensing regulation is dependent on FptX, PchHI, and FepBCDG. Additionally, the findings indicate that unlike FptX, neither FepBCDG nor PchHI play roles in the autoregulatory loop involving PchR, but further deletion of fepBCDG and pchHI can reverse the inactive PchR phenotype caused by fptX deletion and reactivate the expression of the PCH pathway genes under iron-limited conditions. Finally, this work identifies the interaction between FptX, PchHI, and FepBCDG, indicating that a larger complex could be formed to mediate the uptake of PQS-Fe3+ and PCH-Fe3+. These results pave the way for a better understanding of the PQS and PCH iron absorption pathways and provide future directions for research on tackling P. aeruginosa infections. IMPORTANCE Pseudomonas aeruginosa has evolved a number of strategies to acquire the iron it needs from its host, with the most common being the synthesis, secretion, and uptake of siderophores such as pyoverdine, pyochelin, and the quorum-sensing signaling molecule Pseudomonas quinolone signal (PQS). However, despite intensive studies of the siderophore uptake pathways of P. aeruginosa, our understanding of how siderophores transport iron across the inner membrane into the cytoplasm is still incomplete. Herein, we reveal that PQS and pyochelin in P. aeruginosa share inner membrane transporters such as FptX, PchHI, and FepBCDG to mediate iron uptake. Meanwhile, PQS and pyochelin-mediated signaling operate to a large extent via these inner membrane transporters. Our study revealed the existence of shared uptake pathways between PQS and pyochelin, which could lead us to reexamine the role of these two molecules in the iron uptake and virulence of P. aeruginosa.

ATPase domain that is co-expressed with fptX in the biosynthetic gene cluster of PCH, is also involved in the siderophore-free iron uptake via PCH into the bacterial cytoplasm (5).However, despite intensive studies of the PCH pathway, our understanding of PCH-Fe 3+ uptake remains incomplete.
Herein, we reveal that PQS and PCH in P. aeruginosa share inner membrane trans porters such as FptX, PchHI, and FepBCDG to mediate iron uptake.Further investiga tion showed that FptX, PchHI, and FepBCDG were also involved in the virulence of P. aeruginosa.This study highlights the important roles of FptX, PchHI, and FepBCDG in the ability of P. aeruginosa to mediate iron uptake and virulence.These findings contribute to further understanding of the molecular mechanism through which P. aeruginosa absorbs iron ions.

FptX, PchHI, and FepBCDG are involved in iron uptake via PQS
In the previously proposed iron uptake mechanism mediated by P. aeruginosa PQS, PQS-Fe 3+ enters cells through the TseF-FptA/OprF pathway (4).However, our research group has discovered two possible ways in which PQS-Fe 3+ enters cells through the TseF-FptA/OprF pathway.One is that PQS-Fe 3+ first transmits iron to the receptors FptA or OprF before diffusing into cells in the form of PQS.The other is that PQS-Fe 3+ directly enters cells through the receptors FptA or OprF.To further determine which of these two possible pathways plays a primary role, in this study, we compared the differences in the regulation of the lectin gene lecA and pyocyanin synthesis operon phz1 (phzA1B1C1D1G1) with the exogenous addition of PQS and PQS-Fe 3+ in PAΔ3FeΔpqsA [a mutant defective in the PVD biosynthetic pathway (ΔpvdA), ferrous iron transport (ΔfeoB), PCH synthetase (ΔpchE), and the PQS biosynthetic pathway (ΔpqsA)] and PAΔ3FeΔpqsAΔfptAΔoprFΔtseF strains, in which the transport (TseF-FptA/ OprF) pathway was deleted in the background of strain PAΔ3FeΔpqsA (4).The results are shown in Fig. S1.The exogenous addition of PQS and PQS-Fe 3+ significantly induced the expression of phzA1 and lecA in strain PAΔ3FeΔpqsA, and adding PQS to the PAΔ3FeΔpqsAΔfptAΔoprFΔtseF strain also significantly induced the expression of phzA1 and lecA in iron-sufficient tryptone soya broth (TSB) medium.However, adding PQS-Fe 3+ did not effectively activate the expression of phzA1 or lecA in the PAΔ3FeΔpqsAΔfptAΔoprFΔtseF strain in the TSB medium.These results suggest that the TseF-FptA/OprF pathway does not affect the diffusion of PQS into the cell, but it is necessary for P. aeruginosa to absorb PQS-Fe 3+ and allow PQS to function as a signaling molecule, indicating that PQS-Fe 3+ enters the cell directly through the TseF-FptA/OprF pathway in the PAΔ3Fe strain.
Although the outer membrane receptor of PQS-Fe 3+ is known, the mechanism of PQS-Fe 3+ transport through the inner membrane has not yet been revealed.FptA is a known PCH-Fe 3+ outer membrane receptor protein that mediates the uptake of PCH-Fe 3+ (43).Because PQS-Fe 3+ and PCH-Fe 3+ share the same outer membrane receptor protein, we speculated that P. aeruginosa may also share the inner membrane transporter for the uptake of PQS-Fe 3+ and PCH-Fe 3+ .FptX is a known inner membrane transporter that plays an important role in PCH-Fe 3+ uptake (43).Therefore, this study examined the effect of fptX mutation on the uptake of PQS-Fe 3+ in P. aeruginosa.The results are shown in Fig. 1.The growth of all strains was normal, and there was no significant difference under iron-sufficient conditions, where the growth curves were made with 6.7 µM FeCl 3 (this concentration corresponds to the concentration of ferric ions in 20 µM PQS-Fe 3+ ) in the succinate minimal medium (MM) (Fig. 1A).For iron-limited growth, the MM was treated with the iron chelator 2, 2′-bipyridine (BP) at a concentration of 750 µM, a concentration that almost completely inhibited the growth of all strains (Fig. 1C).When 20 µM PQS-Fe 3+ and 750 µM BP was added to the MM, compared with PAΔ3Fe, further deleting fptX inhibited the uptake of PQS-Fe 3+ by P. aeruginosa.However, this inhibitory effect was weaker than that in the negative control strain PAΔ3FeΔfptAΔoprF (Fig. 1B), indicating that there were other inner membrane transporters mediating the All data represent the results of at least three independent experiments.The error bars represent the standard deviations.(D and E) Interactions between FptX, FepB, FepC, FepD, and FepG identified by bacterial two-hybrid assays.Images of colonies formed by co-transformants on MacConkey agar plates (red colonies indicate a positive interaction).The β-galactosidase activity of co-transformants was measured after plating on MacConkey agar plates (Fig. S2).Zip, leucine zipper domain of the yeast transcription factor GCN4 (positive control); T18, empty vector pUT18CM; and T25, empty vector pKT25M.(F-H) Growth curves for P. aeruginosa PAΔ3Fe and its derivative mutants in the MM.The experimental conditions were similar to those shown in panels A-C.All data represent the results of at least three independent experiments.The error bars represent the standard deviations.intracellular transport of PQS-Fe 3+ .Importantly, recent studies have found that the ABC family inner membrane transporter PchHI is also involved in the uptake of PCH-Fe 3+ by P. aeruginosa (5).Therefore, we speculated that the inner membrane transporter PchHI was also involved in the uptake of PQS-Fe 3+ by P. aeruginosa.The results of the growth curve analysis confirmed this assumption (Fig. 1A through C).After the deletion of pchHI based on PAΔ3Fe, the growth phenotype of PAΔ3FeΔpchHI was similar to that of the positive control strain PAΔ3Fe.However, in the further deletion of pchHI based on PAΔ3FeΔfptX, compared to PAΔ3FeΔfptX, strain PAΔ3FeΔfptXpchHI had a lower ability to absorb PQS-Fe 3+ in P. aeruginosa (Fig. 1B).The above results indicate that FptX and PchHI are jointly involved in the uptake of PQS-Fe 3+ in P. aeruginosa and that FptX plays a dominant role in the process.
Although the ability of strain PAΔ3FeΔfptXΔpchHI to absorb PQS-Fe 3+ was significantly reduced compared to PAΔ3Fe, it was still higher than that of the negative control strain PAΔ3FeΔfptAΔoprF (Fig. 1B), indicating that there were other inner membrane transport ers mediating the uptake of PQS-Fe 3+ in addition to FptX and PchHI.Previously, it was reported that PchH interacted with FptX, linking both the ABC transporter PchHI and the inner membrane transporter FptX (5), implying that other inner membrane transporters with similar functions may also interact with FptX.To identify other inner membrane transporters that may mediate the uptake of PQS-Fe 3+ , this study screened the proteins interacting with FptX from P. aeruginosa by constructing a bacterial two-hybrid screening library.The following four FptX-interacting proteins were identified: FepB (PA4159), FepC (PA4158), FepD (PA4160), and FepG (PA4161).These four proteins form the ABC family's inner membrane transporter complex, FepBCDG (45).Bioinformatic predictions indicate that FepB is a periplasmic substrate-binding protein; FepC is an ATPase that carries an ATP-binding domain, and FepD and FepG are both cytoplasmic membrane permea ses that may form heterodimers.Additionally, they may be involved in the uptake of enterobactin-Fe 3+ in P. aeruginosa (7) (Table S3).To rule out false positives, we further analyzed the interaction of full-length FepB, FepC, FepD, and FepG with FptX through bacterial two-hybrid assays.The results showed that FptX could interact with FepB, FepD, and FepG but could not interact with FepC (Fig. 1D; Fig. S2).In addition, we further examined the interactions between FepB, FepC, FepD, and FepG.The results showed that there was no interaction between FepC and FepB, while other proteins could interact with each other (Fig. 1E; Fig. S2), indicating that FepB, FepC, FepD, and FepG together constituted the FepBCDG protein complex, a finding that was consistent with the results of bioinformatic analysis.In conclusion, these results indicate that FptX interacts with the FepBCDG protein complex, implying that the FepBCDG protein complex may also participate in the uptake of PQS-Fe 3+ in P. aeruginosa.
To verify the above speculation, we further deleted fepBCDG based on PAΔ3FeΔfptXΔpchHI and detected its growth.The results showed that compared with strain PAΔ3FeΔfptXΔpchHI, the PAΔ3FeΔfptXΔpchHIΔfepBCDG strain basically lost the ability to absorb PQS-Fe 3+ to maintain normal growth (Fig. 1F through H).This phenotype was similar to that of the negative control strain PAΔ3FeΔfptAΔoprF.However, after the deletion of fepBCDG based on PAΔ3Fe, the growth phenotype of PAΔ3FeΔfepBCDG was similar to that of the positive control strain PAΔ3Fe.The above results indicate that FptX, PchHI, and FepBCDG are jointly involved in the uptake of PQS-Fe 3+ in P. aeruginosa and that FptX plays a dominant role.These results were also confirmed by genetic complementation (Fig. S3), which showed that the complementation of fptX, pchHI, and fepBCDG could promote the uptake of PQS-Fe 3+ by PAΔ3FeΔfptXΔpchHIΔfepBCDG to varying degrees.

FptX, PchHI, and FepBCDG are involved in iron uptake via PCH
FptX is a known PCH-Fe 3+ inner membrane transporter.However, it is only responsible for about 50% of the transport of PCH-Fe 3+ into cells (44).In addition, the heterodimeric ABC transporter PchHI has also been reported to be involved in PCH-Fe 3+ uptake by P. aeruginosa (5).As FptX, PchHI, and FepBCDG are jointly involved in the uptake of PQS-Fe 3+ in P. aeruginosa, they may also be jointly involved in the uptake of PCH-Fe 3+ .To verify this hypothesis, we took the double deletion mutant ΔpvdAΔpqsH as the starting strain and the strain ΔpvdAΔpqsHΔpchE, with the simultaneous deletion of PVD, PQS, and PCH, as the negative control strain and analyzed the effects of the deletion mutations fptX, pchHI, and fepBCDG on the uptake of PCH-Fe 3+ in P. aeruginosa.The results are shown in Fig. 2A and B. When 6.7 µM FeCl 3 was added to the MM, all strains grew normally and showed no significant differences (Fig. 2A).However, when 250 µM BP was added to the MM, compared with ΔpvdAΔpqsH, strain ΔpvdAΔpqsHΔfptX lost its partial ability to utilize PCH-Fe 3+ to sustain normal growth.Knocking out the pchHI gene based on the ΔpvdAΔpqsHΔfptX strain further reduced the ability of P. aeruginosa to absorb PCH-Fe 3+ .This result indicates that both FptX and PchHI play important roles in the uptake of PCH-Fe 3+ in P. aeruginosa.However, the growth status of strain ΔpvdAΔpqsHΔfptXΔpchHI was significantly better than that of the negative control strain ΔpvdAΔpqsHΔpchE.Therefore, based on ΔpvdAΔpqsHΔfptXΔpchHI, we further deleted the fepBCDG gene.Compared with ΔpvdAΔpqsHΔfptXΔpchHI, the growth phenotype of the ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG strain was further inhibited.Consistent with the negative control strain ΔpvdAΔpqsHΔpchE, the strain ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG completely lost the utilization of PCH-Fe 3+ (Fig. 2B).This finding suggests that FepBCDG also participates in the uptake of PCH-Fe 3+ in P. aeruginosa.These results were also confirmed by genetic complementation (Fig. S4).However, after the deletion of pchHI or fepBCDG based on ΔpvdAΔpqsH, the growth phenotypes of ΔpvdAΔpqsHΔpchHI and ΔpvdAΔpqsHΔfepBCDG were similar to that of the positive control strain ΔpvdAΔpqsH (Fig. 2B).The above results indicate that FptX, PchHI, and FepBCDG are jointly involved in the uptake of PCH-Fe 3+ in P. aeruginosa and that FptX plays a dominant role.In addition, the mutant ΔpvdAΔpqsHΔpchE was taken as the starting strain, and exogenous PCH or PCH extract (ΔpvdAΔpqsH extract or ΔpvdAΔpqsHΔpchE extract) was added to the medium to analyze the effects of the deletion mutations fptX, pchHI, and fepBCDG on the uptake of PCH-Fe 3+ in P. aeruginosa.The results are shown in Fig. 2C through G.When 6.7 µM FeCl 3 was added to the MM, all strains grew normally and showed no significant differences (Fig. 2C).However, whether or not the ΔpvdAΔpqsHΔpchE extract was added to the MM supplied with 250 µM BP, the growth of all strains was defective and was consistent with that of the negative control strain ΔpvdAΔpqsHΔpchEΔfptA (Fig. 2D and E).However, the exogenous addition of ΔpvdAΔpqsH extract promoted the growth of all strains except for ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG and the negative control strain ΔpvdAΔpqsHΔpchEΔfptA; among these strains, the positive control strain ΔpvdAΔpqsHΔpchE grew best (Fig. 2F).Compared with ΔpvdAΔpqsHΔpchE, ΔpvdAΔpqsHΔpchEΔfptX lost its partial ability to sustain normal growth.Knocking out the pchHI gene based on the ΔpvdAΔpqsHΔpchEΔfptX strain further reduced the ability of P. aeruginosa to absorb PCH-Fe 3+ .However, compared with the negative control strain ΔpvdAΔpqsHΔpchEΔfptA, growth was only partially affected in the strain ΔpvdAΔpqsHΔpchEΔfptXΔpchHI.The exogenous addition of PCH promoted the growth of all strains except for ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG and the negative control strain ΔpvdAΔpqsHΔpchEΔfptA (Fig. 2G).Compared with ΔpvdAΔpqsHΔpchE, the ΔpvdAΔpqsHΔpchEΔfptX strain lost its partial ability to utilize PCH-Fe 3+ to sustain normal growth.Knocking out the pchHI gene on the basis of the ΔpvdAΔpqsHΔpchEΔfptX strain further reduced the ability of P. aeruginosa to uptake PCH-Fe 3+ .However, com pared with the negative control strain ΔpvdAΔpqsHΔpchEΔfptA, growth was only partially affected in the strain ΔpvdAΔpqsHΔpchEΔfptXΔpchHI.Surprisingly, based on ΔpvdAΔpqsHΔpchEΔfptXΔpchHI, after further deletion of the fepBCDG gene, strain ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG was consistent with the negative control strain ΔpvdAΔpqsHΔpchEΔfptA and completely lost its growth promotion effect (Fig. 2F and  G).The results suggested that FepBCDG is also involved in the uptake of PCH-Fe 3+ by P. aeruginosa.These results were also confirmed by genetic complementation (Fig. S4).Overall, these data suggest that FptX, PchHI, and FepBCDG are jointly involved in the uptake of PCH-Fe 3+ in P. aeruginosa.

FptX, PchHI, and FepBCDG are necessary for PQS-Fe 3+ and PCH-Fe 3+ uptake
To analyze the roles of FptX, PchHI, and FepBCDG in the uptake of PQS-Fe 3+ and PCH-Fe 3+ in P. aeruginosa, the growth differences between  To further analyze the roles of FptX, PchHI, and FepBCDG in the uptake of PQS-Fe 3+ and PCH-Fe 3+ , we first cultured strains PAΔ3FeΔpqsH and PAΔ3FeΔpqsHΔfptXΔpchHIΔfepBCDG (for the analysis of PQS-Fe 3+ uptake) as well as strains ΔpvdAΔpqsHΔpchE, ΔpvdAΔpqsHΔpchEΔfptX, and ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG (for the analysis of PCH-Fe 3+ uptake) in the MM to the mid-log phase, and then subcultured these strains in phos phate-buffered saline (PBS) (containing 0.4% glucose) supplemented with PQS-Fe 3+ , PCH, or hemin.Cell samples were collected at different time points to measure the intracellular metal ion content.The results are shown in Fig. 3E through H.Under the treatment conditions of the exogenous addition of PQS-Fe 3+ or PCH, the intracellular iron content of strains PAΔ3FeΔpqsH and ΔpvdAΔpqsHΔpchE increased rapidly with the extension of culture time; meanwhile, the intracellular iron content of strains PAΔ3FeΔpqsHΔfptXΔpchHIΔfepBCDG and ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG changed little and was consistent with that of the negative control strains PAΔ3FeΔpqsHΔfptAΔoprF and ΔpvdAΔpqsHΔpchEΔfptA, and significantly lower than that of strains PAΔ3FeΔpqsH and ΔpvdAΔpqsHΔpchE (Fig. 3F and H).However, under the same conditions, there were no significant differences in the intracellular zinc or manganese ion contents in these strains (Fig. S5).Additionally, under the exogenous addition of hemin, the intracellular iron, zinc, and manganese ion contents exhibited no significant differences among these strains (Fig. 3E and G; Fig. S5).In addition, complementary fptX, pchHI, and fepBCDG genes could promote the increase of intracellular iron content in the corresponding mutant cells to varying degrees under the exogenous addition of PQS-Fe 3+ or PCH (Fig. S5).However, complementary fptX, pchHI, or fepBCDG genes had no effect on the contents of intracellular zinc or manganese ions under the same conditions (Fig. S5).These results suggested that the mutation of fptX, pchHI, and fepBCDG significantly reduced the uptake of PQS-Fe 3+ and PCH-Fe 3+ in the PAΔ3FeΔpqsH and ΔpvdAΔpqsHΔpchE strains, respectively.In summary, FptX, PchHI, and FepBCDG are necessary for P. aeruginosa to absorb PQS-Fe 3+ and PCH-Fe 3+ .

Lack of an energy source impairs the uptake of PQS-Fe 3+ and PCH-Fe 3+ by P. aeruginosa inner membrane transporters
We have demonstrated that FptX, PchHI, and FepBCDG are involved in the uptake of PQS-Fe 3+ and PCH-Fe 3+ .Interestingly, FptX is a proton motive-dependent permease (7,44).The PchH and PchI proteins carry an ATP-binding domain and a transmem brane domain, respectively, on the same polypeptide (46,47), and the two proteins may function together to form a complete heterodimeric ABC transporter (5,46).Therefore, this study analyzed the effects of lacking an energy source on the func tions of PQS-Fe 3+ and PCH-Fe 3+ inner membrane transporters.For FptX, which uses proton motive force as its energy source, carbonyl cyanide-m-chlorophenylhydrazone (CCCP) was used to inhibit the cell's proton motive force and cause it to lose its energy source.Under the iron-limited culture conditions of adding PQS-Fe 3+ , the growth ability of PAΔ3FeΔpchHIΔfepBCDG was close to that of PAΔ3Fe and was significantly higher than that of PAΔ3FeΔpchHIΔfepBCDGΔfptX. However, when 10 µM CCCP was added to the medium, the growth of both the PAΔ3FeΔpchHIΔfepBCDG and PAΔ3FeΔpchHIΔfepBCDGΔfptX strains was completely inhibited, while PAΔ3Fe still grew effectively (Fig. 4A).In addition, in the ΔpvdAΔpqsH mutant and its deriv ative strains (for analyzing the role of endogenous PCH), the growth ability of ΔpvdAΔpqsHΔpchHIΔfepBCDG was close to that of ΔpvdAΔpqsH and was significantly higher than that of ΔpvdAΔpqsHΔpchHIΔfepBCDGΔfptX in the MM supplied with 250 µM BP.However, when 250 µM CCCP was added to the medium, the growth of both the ΔpvdAΔpqsHΔpchHIΔfepBCDG and ΔpvdAΔpqsHΔpchHIΔfepBCDGΔfptX strains was completely inhibited, while ΔpvdAΔpqsH still grew effectively (Fig. 4B), indicat ing that CCCP inhibited the energy source of FptX, resulting in its loss of the ability to transport PQS-Fe 3+ and PCH-Fe 3+ .For PchH and PchI, which use ATP as an energy source, the site-directed mutation of the conserved amino acid resi dues on the Walker B motif of their ATPase domain was constructed, and their mutation sites were both E490A, so that they both lost ATPase activity and their energy source (Fig. 4C).As shown in Fig. 4D, compared with the complementary wild-type gene pchHI, complementary pchH*I (i.e., PchH E490A I) could not restore the growth of PAΔ3FeΔfptXΔpchHIΔfepBCDG in the MM supplied with 750 µM BP.However, the growth phenotypes of the PAΔ3FeΔfptXΔpchHIΔfepBCDG strains with complementary pchHI* (i.e., PchHI E490A ) and complementary wild-type pchHI were similar.In addition, a similar situation was observed in the genetic complementary strains to ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG.Compared with the complementary wild-type pchHI, complementary pchH*I (i.e., PchH E490A I) did not restore the growth of ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG in the MM supplied with 250 µM BP.However, the growth phenotypes of ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG strains with complementary pchHI* (i.e., PchHI E490A ) and complementary wild-type pchHI were similar (Fig. 4E).These results indicated that the energy of the PchHI transporter complex was derived from the ATPase activity of PchH, but not PchI.FepC provides energy for the ABC-type trans porter complex FepBCDG.The energy source of FepC was lost through the site-directed mutagenesis of the conserved amino acid residue on the Walker B motif of the FepC ATPase domain (the mutation site was E166A) (Fig. 4C).The results are shown in Fig. 4F and G. Compared with the complementary wild-type gene fepC, the complementary fepC* (i.e., FepC E166A ) did not restore the growth of the PAΔ3FeΔfptXΔpchHIΔfepC strain in the MM supplied with 750 µM BP, suggesting that the energy of the FepBCDG trans porter complex was derived from the ATPase activity of FepC.To sum up, these results indicate that the lack of an energy source impairs the capacity of the P. aeruginosa inner membrane transporters to transport PQS-Fe 3+ and PCH-Fe 3+ .aeruginosa mutants PAΔ3FeΔpqsH and PAΔ3FeΔpqsHΔfptXΔpchHIΔfepBCDG were cultured in the MM to the mid-log phase.
Cells were collected and treated as described in Materials and Methods.The intracellular metal ion content was determined using inductively coupled plasma mass spectrometry (ICP-MS) at different time points.(E) Supplied with 20 µM hemin.
(G and H) P. aeruginosa mutants ΔpvdAΔpqsHΔpchE, ΔpvdAΔpqsHΔpchEΔfptX, and ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG were cultured in the MM to the mid-log phase.Cells were collected and treated as described in Materials and Methods.
The intracellular metal ion content was determined using ICP-MS at different time points.(G) Supplied with 20 µM hemin.

FptX, PchHI, and FepBCDG affect the expression of the lectin gene lecA and the PCH biosynthetic operon
The PQS signaling molecule in P. aeruginosa regulates the expression of many virulence genes through the transcription regulator PqsR, including the pqsABCDE synthesis operon, the pyocyanin phzA1B1C1D1G1 synthesis operon, and the lectin lecA gene (22).In P. aeruginosa PQS-deficient mutant ΔpqsA, the exogenous addition of PQS significantly induced the expression of pqsA and lecA genes in iron-sufficient Luria-Bertani (LB) medium, while the exogenous addition of PQS or PQS-Fe 3+ (PQS: Fe 3+ = 3:1) could also effectively induce the expression of the pqsA gene in iron-deficient casamino acid (CAA) medium, indicating that the iron-chelating activity of PQS did not affect the function of PQS signal molecules under this specific condition (38).Because the three inner membrane transporters FptX, PchHI, and FepBCDG participate in the uptake of PQS-Fe 3+ in P. aeruginosa, it is possible that they also affect the function of PQS as a quorum-sensing signal molecule.Using the lecA gene promoter, which is regulated by PQS, as a probe, we compared the differences in the regulation of the lectin gene lecA by the exogenous addition of PQS and PQS-Fe 3+ in the PAΔ3FeΔpqsA and PAΔ3FeΔpqsAΔfptXΔpchHIΔfepBCDG strains.The results are shown in Fig. 5A.The exogenous addition of PQS did not significantly activate lecA in the iron-limited medium (the TSB medium containing 300 µM BP) but significantly activated lecA in the TSB medium (Fig. S1).Additionally, the growth of P. aeruginosa was not significantly different in the TSB medium supplemented with or without 300 µM BP (data not shown).These results indicated that under iron-limited conditions, PQS was ineffective for uptake by P. aeruginosa.In contrast, when PQS-Fe 3+ was added, the expression of the lecA gene was only significantly activated in the PAΔ3FeΔpqsA strain, while it was not activated in the PAΔ3FeΔpqsAΔfptXΔpchHIΔfepBCDG strain, indicating that under iron-limited culture conditions, PQS mainly entered cells in the form of PQS-Fe 3+ , and this process depended on FptX, PchHI, and FepBCDG.The siderophore PCH of P. aeruginosa regulates the expression of many genes via combining the transcription regulator PchR, including the PCH biosynthesis operons pchDCBA and pchEFGHI, and the PCH-Fe 3+ uptake operon fptABCX (44,48).Because FptX, PchHI, and FepBCDG participate in the uptake of PCH-Fe 3+ in P. aeruginosa, and both FptA and FptX are involved in the positive autoregulatory loop through import ing the PCH-Fe 3+ complex interacting with PchR into the bacteria (5,44), we asked whether PchHI and FepBCDG also play important roles in PCH-mediated signaling.To answer this question, we used the fptA, pchD, and pchE gene promoters as probes to analyze the role of PCH-mediated signaling by FptX, PchHI, and FepBCDG through lacZ transcription fusion of the promoter.P. aeruginosa PAO1 was used as the starting strain to analyze the regulation of these genes by endogenous PCH.Consistent with previous reports (44), when the fptA or fptX genes were deleted, the expression levels of fptA, pchD, and pchE were significantly downregulated under iron-limited conditions.However, when further deletion of fepBCDG was performed based on PAO1ΔfptX, the expression of the fptA, pchD, and pchE genes was unaffected.In contrast to the single PAO1ΔfptX mutant, the PAO1ΔfptXΔpchHI and PAO1ΔfptXΔpchHIΔfepBCDG mutants had the phenotype that can activate the expression of the fptA, pchD, and pchE genes.As expected, the deletion of pqsR based on PAO1ΔfptXΔpchHIΔfepBCDG had a strong inhibitory effect on the expression of the genes of the PCH locus, indicating that the positive autoregulatory loop involving PchR was no longer active (Fig. 5B).In addition, ΔpvdAΔpchE and its derivative mutants were used to study how exogenous PCH regulated the expression of the PCH genes.The results showed that when no exogenous PCH was added to the medium, there were only background gene expres sion levels and no significant differences in the expression of the fptA, pchD, or pchE genes between ΔpvdAΔpchE and its derivative mutants (Fig. 5C).In contrast, when PCH was added, the expression levels of the fptA, pchD, and pchE genes were significantly upregulated in the ΔpvdAΔpchE mutant.As expected, after the deletion of the fptA or fptX genes based on the ΔpvdAΔpchE mutant, the exogenous addition of PCH failed to activate the expression of fptA, pchD, and pchE.Similarly, the deletion of pqsR based on the ΔpvdAΔpchEΔfptXΔpchHIΔfepBCDG mutant made the expression of the PCH were cultured in the MM and the levels of fptABCX, pchDCBA, and pchEFGHI transcription in P. aeruginosa PAO1 and its derivative mutant cells were monitored using the fptA-lacZ, pchD-lacZ, and pchE-lacZ transcriptional fusions, respectively.(C and D) Cells cultured in the MM were supplied with or without 20 µM pyochelin and the transcription levels of fptABCX, pchDCBA, and pchEFGHI in P. aeruginosa ΔpvdApchE and its derivative mutant cells were monitored using the fptA-lacZ, pchD-lacZ, and pchE-lacZ transcriptional fusions, respectively.(C) No addition of pyochelin.(D) Addition of 20 µM pyochelin.The graphs show the mean and standard deviation of three experiments performed using five replicates each time.n.s., not significant; **P < 0.01; and ***P < 0.001.genes no longer responsive to exogenously added PCH.However, in contrast to the ΔpvdAΔpchEΔfptX mutant, the ΔpvdAΔpchEΔfptXΔfepBCDG, ΔpvdAΔpchEΔfptXΔpchHI, and ΔpvdAΔpchEΔfptXΔpchHIΔfepBCDG mutants exhibited the phenotype that can activate the expression of the PCH genes under the same culture conditions (Fig. 5D).These results demonstrated that, unlike FptX, both FepBCDG and PchHI were not involved in the autoregulatory loop involving PchR, but further deletion of fepBCDG and pchHI could reverse the inactive PchR phenotype caused by fptX deletion and reactivate the expression of the genes of the PCH pathway under iron-limited conditions.
FptX, PchHI, and FepBCDG are necessary for the virulence of P. aeruginosa in Galleria mellonella larvae PQS and PCH play important roles in the virulence of P. aeruginosa (49,50), and FptX, PchHI, and FepBCDG, which are all inner membrane transporters, jointly partic ipate in the uptake of PQS-Fe 3+ and PCH-Fe 3+ in P. aeruginosa.Thus, we speculated that FptX, PchHI, and FepBCDG may affect the virulence of P. aeruginosa.To test this hypothesis, we analyzed the effect of these inner membrane transporter deletions on the toxicity of P. aeruginosa to G. mellonella larvae as well as the viability of the bacteria in the larvae.G. mellonella is a host that can be used to examine the viru lence of this pathogen (4).The ΔpvdA mutant strain was used as the starting strain to avoid iron uptake by the siderophore PVD.The results showed that the toxicity of strain ΔpvdAΔfptXΔpchHIΔfepBCDG to the G. mellonella larvae was significantly lower than that of the ΔpvdA strain (Fig. 6A), while the complementary fptX, pchHI, and fepBCDG recovered the toxicity of P. aeruginosa to varying degrees (Fig. S6).We further evaluated the role of PQS-mediated or PCH-mediated iron uptake in the biology of P. aeruginosa by examining the interactions of relevant mutants with G. mellonella.Mutants defective in fptX, pchHI, and fepBCDG were able to effectively compete with the parental strain PAΔ3Fe or with ΔpvdAΔpqsH in LB media.The deletion of fptX, pchHI, or fepBCDG from mutant PAΔ3Fe or ΔpvdAΔpqsH resulted in mutants that could not effectively compete against PAΔ3Fe or ΔpvdAΔpqsH when co-inoculated in the host (Fig. 6B and C).However, complementing fptX, pchHI, and fepBCDG could restore the survival of the competition-deficient phenotype of PAΔ3FeΔfptXΔpchHIΔfepBCDG or ΔpvdAΔpqsHΔfptXΔpchHIΔfepBCDG in the G. mellonella larvae to varying degrees (Fig. S6).These results indicate that FptX, PchHI, and FepBCDG are necessary for P. aeruginosa virulence in this host.

DISCUSSION
P. aeruginosa is an opportunistic human pathogen that is listed by the World Health Organization as one of the pathogens for which the development of new antimicrobial treatments is urgently needed (51,52).During infection, P. aeruginosa faces stressful environments and must overcome the host's immune reactions.In order to survive under these stressful conditions, P. aeruginosa secretes a large number of virulence factors, including siderophores (53).Siderophores are small organic compounds produced and secreted by bacteria to absorb iron (13), an essential nutrient for bacterial growth and virulence.In the iron uptake system of P. aeruginosa, there are three main types of molecules that can be used to capture extracellular Fe 3+ : two types of siderophores (PCH and PVD) and the quorum-sensing signaling molecule PQS.Previous studies have shown that the PQS-and PCH-mediated iron uptake pathways share the same outer membrane transporter, FptA (4,14), while FptX and PchHI are the inner membrane transporters involved in PCH-mediated iron uptake (5,14).The results of the present study demon strate that FptX and PchHI are not only the inner membrane transporters of iron uptake mediated by PCH but are also the inner membrane transporters of the PQS-mediated iron uptake pathway.In addition, this study has demonstrated that another inner membrane transporter, also plays an important role in the uptake of PQS-Fe 3+ and PCH-Fe 3+ in P. aeruginosa.
During the uptake of PCH-Fe 3+ in P. aeruginosa, FptX transports a portion of the PCH-Fe 3+ into the cytoplasm.FptX is a proton motive-dependent permease that can use proton motive force momentum as its energy source (43).PchH and PchI proteins contain an ATP-binding domain and a transmembrane domain on the same polypeptide (5).This feature corresponds to the YbtPQ and IrtAB ABC transporters of Yersinia pestis and Mycobacterium tuberculosis, respectively, and both YbtPQ and IrtAB play important roles in the process of iron uptake (54,55).FepC contains an ATP-binding domain, and this domain provides energy for FepBCDG (7).In the present study, different methods were used to deprive these proteins of their energy source.The results showed that the loss of the energy sources of FptX, PchHI, and FepBCDG led to the inability of P. aeruginosa to utilize PQS-Fe 3+ and PCH-Fe 3+ (Fig. 4).Interestingly, we found that the energy of the PchHI transporter complex was derived from the ATPase activity of PchH but not PchI (Fig. 4D and E).
A promoter-lacZ transcriptional fusion assay showed that unlike FptX, FepBCDG and PchHI were not involved in the autoregulatory loop involving PchR, but further deletion of fepBCDG and pchHI could reverse the inactive PchR phenotype caused by fptX deletion and reactivate the expression of the genes of the PCH pathway under iron-limited conditions (Fig. 5B and D).It has previously been shown that PchR-mediated transcrip tional activation of the PCH genes does not require interaction with PCH-Fe 3+ under very strong iron-limited conditions (56).The deletion of pchHI and fepBCDG resulted in a similar phenotype in which PchR also became active in the absence of the apparent transport of PCH-Fe 3+ into the bacteria.Recent studies have reported that during the uptake of PCH-Fe 3+ in P. aeruginosa, a fraction of the PCH-Fe 3+ complexes is transpor ted across the inner membrane into the cytoplasm by FptX to interact with PchR in the auto-regulatory loop, while another fraction of the PCH-Fe 3+ complexes undergoes dissociation in the bacterial periplasm via an unknown mechanism, and the free iron is transported further across the inner membrane into the bacterial cytoplasm by PchHI (5).Because FepBCDG and PchHI exhibit similar patterns of effect on the PqsR-mediated autoregulatory loop, it has been suggested that FepBCDG, like PchHI, may act as an ABC transporter to translocate siderophore-free iron into the cytoplasm.
The present study has improved our understanding of the molecular mechanisms of PCH-mediated iron uptake systems.However, there is little information concerning the molecular mechanisms through which iron is released from the PCH-Fe 3+ complexes in P. aeruginosa cells.It has been reported that FadD1, the fatty acid coenzyme-A ligase, is an interacting partner of the inner membrane transporter FptX, implying that it may play a role in modifying PCH (57).Therefore, we speculate that FadD1 may be involved in the release of iron from the PCH-Fe 3+ complexes.
The results showed that in the TSB medium, the exogenous addition of PQS significantly induced the expression of phzA1 and lecA in P. aeruginosa strains PAΔ3FeΔpqsA and PAΔ3FeΔpqsAΔfptAΔoprFΔtseF (Fig. S1).However, in the iron-limited medium (TSB medium containing 300 µM BP), the exogenous addition of PQS only weakly activated the expression of lecA in P. aeruginosa strain PAΔ3FeΔpqsA (Fig. 5A), indicating that PQS may not be efficiently absorbed under iron-limited conditions.Surprisingly, this result was opposite to the results reported by Diggle et al. (38).This difference in the regulatory phenotype may be the result of different culture conditions.An iron-limited CAA medium was used by Diggle et al., and the addition of PQS to this medium could significantly inhibit the growth of P. aeruginosa [see Fig. 6 in this reference (38)].Additionally, in the present study, it was found that whether in iron-rich or iron-limited media, the exogenous addition of PQS-Fe 3+ could only activate the expression of phzA1 and/or lecA in P. aeruginosa strain PAΔ3FeΔpqsA, while it could not activate the expression of these two genes in strains PAΔ3FeΔpqsAΔfptAΔoprFΔtseF or PAΔ3FeΔpqsAΔfptXΔpchHIΔfepBCDG (Fig. S1; Fig. 5A).These results suggest that the function of PQS-Fe 3+ -mediated quorum-sensing regulation is dependent on the TseF-FptA/OprF pathway and three inner membrane transporters, namely FptX, PchHI, and FepBCDG.
Based on the results of this study, we proposed a model for P. aeruginosa to transport PQS-Fe 3+ and PCH-Fe 3+ across the inner membrane into the cytoplasm (Fig. 8).This model also prompted us to suggest the following new hypothesis: although the secretion pathways of PQS and PCH remain unknown, they may share the same secretion pathway, and they may function synergistically.Further work is required to verify this hypothesis.
Previous studies have shown that pchHI is an important virulence gene in P. aeru ginosa.When pchHI is mutated, the virulence of P. aeruginosa toward Dictyostelium, Drosophila, and mice was significantly decreased (58), indicating that PchH and PchI are necessary for the virulence of P. aeruginosa (59).The results of the present study are consistent with these previous reports.Here, we found that when fptX, pchHI, and fepBCDG were deleted, the virulence of P. aeruginosa toward G. mellonella larvae was significantly decreased.However, the virulence of P. aeruginosa was restored to varying degrees after complementing these genes (Fig. 6; Fig. S6), suggesting that FptX, PchHI, and FepBCDG are not only involved in the uptake of PQS-Fe 3+ and PCH-Fe 3+ but that they also play crucial roles in the virulence of P. aeruginosa.
In conclusion, PQS and pyochelin in P. aeruginosa share inner membrane transporters, including FptX, PchHI, and FepBCDG, to mediate iron uptake.These findings provide a special perspective for the prevention and treatment of P. aeruginosa infection, and the results have greatly expanded current understandings of bacterial adaptation to complex environments.

Bacterial strains and growth conditions
The bacterial strains and plasmids used in this study are listed in Table S1.The Escherichia coli strains were grown at 37°C in either LB or TSB medium.The P. aeruginosa strains were grown at 37°C in either LB, TSB, or MM (60).The P. aeruginosa PAO1 strain was the parent strain of all of the derivatives used in this study.To construct in-frame deletion mutants, the pK18mobsacB derivatives were transformed into relevant P. aeruginosa strains through E. coli S17-1-mediated conjugation and were screened as described by Lin et al. (4,61).For overexpression or complementation in the various P. aeruginosa strains, the pME6032 derivatives were transformed into the relevant P. aeruginosa strains and induced by the addition of 1 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG).Antibiotics were used at the following concentrations for P. aeruginosa: kanamycin (30 µg/mL), chloramphenicol (30 µg/mL), gentamicin (200 µg/mL), and tetracycline (160 µg/mL for liquid growth or 200 µg/mL for solid growth).Antibiotics were used at the following concentrations for E. coli: kanamycin (30 µg/mL), gentamicin (10 µg/mL), ampicillin (100 µg/mL), chloramphenicol (30 µg/mL), and tetracycline (20 µg/mL).

Plasmid construction
The construction of the knockout plasmid was modified from a previously reported study (4,61,62).Briefly, to construct the recombinant suicide plasmids for deletion, for the fptX gene, the 803-bp upstream and 783-bp downstream fragments flanking the fptX gene were amplified with the primer pairs fptX Up F/fptX Up R and fptX Low F/fptX Low R, respectively (Table S2).The upstream and downstream polymerase chain reaction (PCR) fragments were ligated using overlapping PCR, and the resulting PCR products were inserted into the XbaI/HindIII sites of the suicide vector pK18mobsacB to yield the plasmid p-fptX.The gentamicin resistance cassette from p34s-Gm was subsequently inserted into the same HindIII site of p-fptX to yield the recombinant suicide plas mid pK18-ΔfptX.The recombinant suicide plasmids pK18-ΔpchHI, pK18-ΔfepBCDG, and pK18-ΔfepC were constructed in a similar manner using primers listed in Table S2.
To construct the complementation plasmid pME6032-fptX, PCR-amplified fptX was inserted into the EcoRI/BglII sites of the pME6032, giving rise to the recombinant plasmid pME6032-fptX.The recombinant plasmids pME6032-pvdA, pME6032-pchHI, pME6032-fepBCDG, and pME6032-fepC were constructed using the same method.3+ and PCH-Fe 3+ by P. aeruginosa.Extracellular PQS-Fe 3+ is transported through OMVs.Under the mediation of the T6SS effector protein TseF, PQS-Fe 3+ on OMV enters the periplasm through the outer membrane receptors OprF and FptA.Similarly, extracellular PCH-Fe 3+ also enters the periplasm through FptA.Then, a fraction of the PQS-Fe 3+ and PCH-Fe 3+ complexes enters the cytoplasm directly through FptX; a fraction of the PQS-Fe 3+ and PCH-Fe 3+ complexes undergoes dissociation in the bacterial periplasm via an unknown mechanism, and the free iron is transported further across the inner membrane into the cytoplasm by PchHI and FepBCDG.The entry of PQS-Fe 3+ and PCH-Fe 3+ into the cytoplasm activates the expression of the PQS and PCH genes to facilitate their synthesis.
The lecA-lacZ transcriptional fusions were constructed via the PCR amplification of the 1,036-bp upstream DNA region from the lecA gene using the primer pairs lecA F/lecA R (Table S2).The PCR amplification products were cloned directly into the pMini-CTX::lacZ vector (62), yielding lacZ reporter constructs.The recombinant plasmids fptA-lacZ were constructed using the same method (Table S1).

Site-directed mutagenesis
To construct the site-directed mutagenesis complementary vector pME6032-pchH*I, the conservative amino acid residue glutamate on the Walker B motif of the ATPase domain of PchH was replaced with alanine.The 1,563-bp front section and the 2,003-bp rear section DNA segments of the pchHI gene were amplified using the primer pairs SD-pchH*I Up F/SD-pchH*I Up R and SD-pchH*I Low F/SD-pchH*I Low R, respectively, and the upstream and downstream segments were connected to form a gene segment using overlapping PCR.The product [containing the native Shine-Dalgarno (SD) sequence] of overlapping PCR was inserted into the KpnI/BglII sites of the pME6032 plasmid to yield the site-directed mutagenesis recombinant plasmid pME6032-pchH*I.Using the same method, this study constructed pME6032-phHI* and pME6032-fepC* (Table S1).

Growth assay
The growth assay protocol was as described previously with some modifications (17).P. aeruginosa strains were grown overnight in TSB; the overnight cultures were harvested, and the cells were washed with the MM twice to remove nutrient-rich substances prior to subculture.Subculture proceeded in the MM with BP (250 or 750 µM) with or without IPTG (1 mM), PQS-Fe 3+ (20 µM), FeCl 3 (6.7 µM), or PCH (20 µM) to a final OD 600 of ~0.01.The cultures were incubated at 37°C, and OD 600 readings were taken every 12 h for 72-120 h.

Construction and screening of the bacterial two-hybrid library
The protocol used to construct genome fragment libraries was as described previously with some modifications (63,64).The genomic DNA of P. aeruginosa strain PAO1 was prepared using a PureLink Genomic DNA kit (Invitrogen, Carlsbad, CA, USA) and partially digested by Sau3AI.The randomly digested DNA was separated on 0.8% agarose gels, and fragments ranging in size from 1,000 to 3,000 bp were gel-purified using a Qia gen Gel Extraction Kit (Qiagen, Valencia, CA, USA).The genomic DNA libraries were constructed using the pKT25M vectors.The pKT25M vectors were digested with BamHI and dephosphorylated with phosphatase.The pools of DNA fragments were ligated overnight at 16°C into the different pKT25M linearized vectors using T4 ligase (NEB, Ipswich, MA, USA).The resulting ligation mixture was transformed into E. coli TG1 competent cells.The libraries were collected and pooled as prey libraries and stored in a freezer at −80°C.
The plasmid pUT18CM-fptX was used as bait to probe the genomic DNA libraries.Basically, 25-50 ng of each pKT25M-derived library was transformed into 100 µL of electrocompetent BTH101 cells carrying the pUT18CM-fptX bait vector and plated on MacConkey agar medium containing 0.5 mM IPTG.Bacteria expressing interacting hybrid proteins will form red colonies on MacConkey agar medium, while cells expressing non-interacting proteins will remain white.IPTG was used to increase β-galactosidase expression.A co-transformant containing pKT25-zip and pUT18-zip was used as a positive control for expected growth on the screening medium.A co-transformant containing empty vectors pKT25 and pUT18 was used as a negative control.The red colonies were picked up and recultivated in the liquid medium, and plasmids were isolated and further analyzed using DNA sequencing.

Bacterial two-hybrid assay
Bacterial two-hybrid assays were performed using previously described methods (65,66).In brief, pUT18CM and pKT25M carrying different genes were used in various combina tions to co-transform E. coli BTH101 cells, and the plates were cultured at 30°C for 24 h.Five independent colonies were selected and inoculated into LB liquid culture medium supplemented with 100 µg/mL ampicillin, 30 µg/mL kanamycin, and 0.5 mM IPTG.After overnight growth at 30°C, 3 µL of each culture was spotted onto MacConkey plates supplemented with 100 µg/mL ampicillin, 30 µg/mL kanamycin, 0.5 mM IPTG, and 1% maltose, then cultured for 20 h at 30°C.The formation of red colonies on MacConkey agar plates indicated an interaction between the two proteins, and white colonies indicated negative results.

β-galactosidase assay
The β-galactosidase assays were modified from a previously reported study (4,61).A total of 100 µL of bacterial culture was added to 900 µL of Z Buffer (40 mM NaH 2 PO 4 , 10 mM KCl, 60 mM Na 2 HPO 4 , 1 mM MgSO 4 , and 0.2% β-mercaptoethanol).A total of 1 µL of 0.1% sodium dodecyl sulfate and 50 µL of chloroform were added to the suspension, which was then mixed vigorously for 20 s.The suspension was then incubated for 1 h at 30°C.A total of 100 µL of 4 mg/mL 2-nitrophenyl β-D-galactopyranoside (Sigma, St. Louis, MO, USA) was added to the cells.The reaction was stopped by adding 500 µL of 1 M Na 2 CO 3 .The suspension was centrifuged at 10,000 × g for 3 min, and the absorbance of the supernatant was read at 420 and 550 nm using a microplate reader.The β-galactosidase activity was then calculated in Miller units (MUs) according to the following equation: MU = 1, 000 × OD 420 − 1.75 × OD 550 Time min × Volume mL × OD 600 .

Extraction of P. aeruginosa PCH
The extraction of P. aeruginosa PCH was modified from a previously reported study (67).Briefly, P. aeruginosa was grown in 1-L volumes of the MM at 37°C, 200 rpm for 48 h.The bacterial cells were removed by centrifugation (6,000 × g for 15 min at 23°C), and the supernatant fluid was brought to pH 1-2 with 1 M HCl.Ethyl acetate was added in a 1:5 ratio, and after vigorous shaking in separatory funnels, the ethyl acetate layers were collected and concentrated using rotary evaporation.The residue remaining in the flask after evaporation was dissolved in 1 mL methanol.When PCH extracts were used, they were added to the culture medium at a ratio of 1: 1,000.

In vitro co-culture assay
In vitro co-culture assays were modified from a previously reported study (68).P. aeruginosa strains ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG (pBBR1MCS-5), PAΔ3FeΔpqsAΔfptXΔpchHIΔfepBCDG (pBBR1MCS-5), PAΔ3FeΔpqsA (pME6032), and ΔpvdAΔpqsHΔpchE (pME6032) were cultured in 5 mL of TSB liquid medium (37°C, 200 rpm, and 20 h).One milliliter samples of different strain cultures with the same OD 600 were centrifuged at 4°C and 3,000 × g for 10 min.As much supernatant was removed as possible, and the bacterial cells were retained.One milliliter of fresh MM medium containing kanamycin was added, and the bacterial cells were resuspen ded.The cells were centrifuged at 4°C, 3,000 × g for 10 min to remove as much supernatant as possible, and the bacterial cells were retained.The cells were resuspen ded in 1 mL of fresh MM medium containing kanamycin, and the OD 600 values of different cultures were adjusted to the same level in order to start with the same number of cells.PAΔ3FeΔpqsA (pME6032) and ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG (pBBR1MCS-5) were mixed in a 1:1 ratio, as well as ΔpvdAΔpqsHΔpchE (pME6032) and ΔpvdAΔpqsHΔpchEΔfptXΔpchHIΔfepBCDG (pBBR1MCS-5), to produce a mixed bacterial suspension.The mixed bacterial suspension was diluted at a ratio of 1:1,000 in fresh MM medium containing kanamycin (if needed, appropriate amounts of BP, PQS-Fe 3+ , PCH, or hemin were included in the medium), and culturing was conducted at 37°C and 200 rpm.Samples were collected at regular intervals and diluted to 10 −7 using a 10-fold dilution method.Three microliters of each dilution was collected and spotted onto LB plates supplemented with kanamycin and tetracycline or kanamycin and gentamicin.
Incubation was conducted at 37°C for 2 days.The colonies were counted, and the number of colony-forming units (CFUs) in 1 mL was calculated.Finally, the growth curve was plotted using log 10 (CFU/mL).All assays were performed in triplicate.

Determination of intracellular metal ion contents
The intracellular metal ion content determination method was modified from previous studies (69,70).Briefly, P. aeruginosa strains were cultured in 5 mL of TSB liquid medium at 37°C and 200 rpm for 20 h.After 1 mL culture solutions were collected and washed twice with MM, the cells were subcultured in MM medium at a ratio of 1:100 until the exponential phase, centrifuged at 4°C and 2,000 × g for 10 min, and the bacterial cells were collected.PBS buffer containing 1 mM ethylene diamine tetra acetic acid (EDTA) was used to suspend the bacterial cells.The suspensions were centrifuged at 2,000 × g for 10 min at 4°C; the bacterial cells were collected, and the process was repeated once.The bacterial cells were suspended using PBS buffer, centrifuged at 4°C and 2,000 × g for 10 min, and then collected.The bacterial cells were resuspended in PBS and divided into four equal parts.Then, 0.4% glucose and 20 µM PQS-Fe 3+ (20 µM PCH or 20 µM hemin) were added to each part.Each portion was incubated at 37°C and 200 rpm for 0, 1, 2, and 4 h and centrifuged at 4°C and 2,000 × g for 20 min to collect the bacterial cells.PBS buffer containing 1 mM EDTA was used to suspend the bacterial cells; the suspensions were centrifuged at 4°C and 2,000 × g for 20 min; the bacterial cells were collected, and the process was repeated once.The cells were washed again with PBS buffer, centrifuged at 4°C and 2,000 × g for 20 min, and the cells were collected.The wet cell pellet weight was determined, and bacteria were chemically lysed using 5 mL Bugbuster (Novagen, Madison, WI, USA) (gram wet pellet cell paste) −1 according to the manufacturer's instructions.Bacterial cells were resuspended in Bugbuster solution by pipetting and then incubated on a rotating mixer at a slow setting for 20 min.The total protein content of each sample was measured using a BioRad protein assay (BioRad, Hercules, CA, USA) according to the manufacturer's instructions.The wet pellet weight and total protein content for each sample were noted.Each sample was diluted 100-fold in 3% molecular-grade nitric acid to a total volume of 10 mL.Samples were analyzed using inductively coupled plasma mass spectrometry (Varian 802-MS; Varian, Palo Alto, CA, USA), and the results were corrected using the appropriate buffers for reference and dilution factors.Triplicate cultures of each strain were analyzed during a single experiment, and the experiment was repeated at least three times.

G. mellonella-killing assay
The protocol for the G. mellonella-killing assay was as described previously with some modifications (71).P. aeruginosa strains were cultured in 5 mL of TSB liquid medium at 37°C and 200 rpm overnight.Subculture was conducted in 5 mL of fresh TSB contain ing kanamycin at a 1:100 ratio until the OD 600 value reached 0.5.Bacterial cells were collected via centrifugation at 4°C and 3,000 × g for 5 min.The bacterial cells were suspended in 0.85% NaCl solution, centrifuged at 4°C and 3,000 × g for 5 min, and collected, and the procedure was repeated twice.The bacterial cells were suspended and diluted with 0.85% NaCl solution to a cell count of 2 × 10 7 CFU/mL.The G. mellonella larvae were placed on ice for 5 min to put them under anesthesia.A microsyringe was used to inject 10 5 cells into the hemocoel of 3-day-old, fifth-instar G. mellonella larvae, and 0.85% NaCl solution was injected as a control.In each group, 50 G.mellonella larvae were injected and cultured at 25°C in the dark, and the procedure was repea ted with three groups for each strain.Data were recorded every 12 hours.Data were analyzed using Kaplan-Meier survival curves.Statistical significance was assessed using the Mantel-Cox log-rank test, applying Bonferroni's correction for multiple comparisons.

G. mellonella co-infection experiments
A lacZ reporter gene was transferred to the neutral phage attachment site (attB) of the P. aeruginosa chromosome as follows: the recombinant plasmid pMini-CTX-Ptac::lacZ (4) was transformed into E. coli S17-1, and the resulting plasmid was then transferred to the P. aeruginosa chromosome (attB site) by mating and selection for tetracycline resistance.The selectable marker was removed by the transient expression of the Flp recombinase from plasmid pFLP2, which was then cured by counter-selection on sucrose plates (72).The resulting P. aeruginosa strain ΔpvdAΔpqsH attB∷Ptac-lacZ and existing PAΔ3Fe attB::Ptac-lacZ (4) were confirmed to grow as well as ΔpvdAΔpqsH and PAΔ3Fe in both in vitro and in-vivo competition assays.The G. mellonella co-infection experiment was performed as previously described with minor modifications (4,73).P. aeruginosa strains were cultured in 5 mL of TSB liquid medium at 37°C and 200 rpm for 20 h.Cells were subcultured in 5 mL of fresh TSB medium at a ratio of 1% until the exponential phase, centrifuged at 4°C and 3,000 × g for 10 min, and the cells were collected.The bacterial cells were suspended in PBS buffer, centrifuged at 4°C and 3,000 × g for 10 min, and the cells were collected.This step was repeated twice.The bacteria were suspended and diluted with PBS buffer to a cell count of 2 × 10 7 CFU/mL.In each experiment, a P. aeruginosa strain that contained lacZ reporter inserted at the neutral phage attachment site (producing blue colonies on X-Gal plates; 40 µg/mL) was mixed at a ratio of 1:1 with a P. aeruginosa strain without lacZ (producing white colonies on X-Gal plates).For G. mellonella co-infection, the G. mellonella larvae were placed on ice for 5 min to put them under anesthesia.Five microliters of a mixed bacterial suspension with a CFU of 1 × 10 5 was injected into the hemocoel of 3-day-old, fifth-instar G. mellonella larvae, and 0.85% NaCl solution was injected as a control.After 24 h, the hemolymph of eight infected larvae from each group was collected in 1.5 mL Eppendorf tubes containing 2 µL of 1% phenylthiourea on ice.The samples were serially diluted with sterile PBS and spread on LB agar plates containing kanamycin and 40 µg/mL X-Gal.Culturing was conducted overnight at 37°C, and the cultures were left at 22°C-25°C until blue colonies appeared.For in vitro co-culture, the mixed bacterial suspension was subcultured in fresh LB liquid medium containing kanamycin at a ratio of 1:1,000 at 37°C and 200 rpm for 12 h.The samples were serially diluted with sterile PBS and spread on LB agar plates containing kanamycin and 40 µg/mL X-Gal.Culturing was conducted overnight at 37°C, and the cultures were left at room temperature until blue colonies appeared.Both the total CFU and the ratio of blue-to-white bacteria were determined.For the competitive index (CI) calculation: CI = CFU ratio of the sample after treatment (white colonies/blue colonies)/CFU ratio of the sample before treatment (white colonies/blue colonies).The larvae were selected randomly for each test group.

Statistical analysis
All of the experiments were performed in triplicate and repeated on two different occasions.The data are expressed as the mean ± S.D. The differences between the frequencies were assessed using Student's t-test (bilateral and unpaired), and a P-value of 0.05 was considered to be statistically significant.The Shapiro-Wilk test and one-way analysis of variance were performed using the GraphPad Prism version 7.00 software (GraphPad Software Inc., San Diego, CA, USA) to examine the normality of the data and the homogeneity of the variances, respectively.GraphPad Prism 7 and Adobe Illustrator 2020 (CS6; Adobe, Mountain View, CA, USA) were used to create all of the figures.

MM+250FIG 3
FIG 3 Effects of the deletion of fptX, pchHI, and fepBCDG on the utilization of PQS-Fe 3+ and PCH-Fe 3+ by P. aeruginosa.(A and B) Co-culture growth curves of P. aeruginosa mutants PAΔ3FeΔpqsH and PAΔ3FeΔpqsHΔfptXΔpchHIΔfepBCDG in the (Continued on next page)

FIG 4
FIG 4 Effects of energy source deficiency on the uptake of PQS-Fe 3+ and PCH-Fe 3+ in P. aeruginosa.(A) Growth curves of P. aeruginosa PAΔ3Fe and its derivative mutants in the MM supplemented with 750 µM BP and 20 µM PQS-Fe 3+ (PQS:Fe 3+ = 3:1) and the presence or absence of 10 µM CCCP.(B) Growth curves of P. aeruginosa ΔpvdAΔpqsH and its derivative mutants in the MM supplemented with 250 µM BP and the presence or absence of 250 µM CCCP.(C) Schematic diagram of the PchH, PchI, and FepC ATPase domains.(D) Growth curve of complementation strains of P. aeruginosa PAΔ3FeΔfptXΔpchHIΔfepBCDG in the MM supplemented with 750 µM BP and 20 µM PQS-Fe 3+ (PQS:Fe 3+ = 3:1).(E) Growth curve of complementation strains of P. aeruginosa PAΔ3FeΔfptXΔpchHIΔfepBCDG in the MM supplemented with 250 µM BP. (F and G) Growth curve of complementation strains of P. aeruginosa PAΔ3FeΔfptXΔpchHIΔfepC in the MM.(F) Supplied with 6.7 µM FeCl 3 .(G) Supplied with 750 µM BP and 20 µM PQS-Fe 3+ .All data represent the results of at least three independent experiments.Error bars represent standard deviations.

FIG 5
FIG5 The effects of the deletion of fptX, pchHI, and fepBCDG on the expression of the lectin lecA gene and PCH genes in P. aeruginosa.(A) Cells cultured in the iron-limited medium (TSB medium containing 300 µM BP) were supplied with or without 40 µM PQS or PQS-Fe 3+ (PQS:Fe 3+ = 3:1), and the transcription levels of lecA in P. aeruginosa PAΔ3FeΔpqsA and PAΔ3FeΔpqsAΔfptXΔpchHIΔfepBCDG mutant cells were monitored using the lecA-lacZ transcriptional fusions.(B) Cells

FIG 6
FIG6 Deletion of fptX, pchHI, and fepBCDG reduces the virulence of P. aeruginosa to G. mellonella larvae.(A) Survival of relative P. aeruginosa strains in G. mellonella larvae.The ordinate represents the percentage of survival rate of G. mellonella infected with different strains after different lengths of time.(B and C) P. aeruginosa mutant strains (white colonies formed on X-Gal plates) and the PAΔ3Fe strain or ΔpvdAΔpqsH strain carrying the lacZ gene at the neutral phage attachment site (blue colonies formed on X-Gal plates) were mixed in the ratio of 1:1.The bacterial mixture was injected into the hemocoel of G. mellonella larvae, and the hemolymph of G. mellonella larvae was collected 24 h later.Competitive index (CI) = colony-forming unit (CFU) ratio (white colonies/blue colonies) of the samples after treatment divided by the CFU ratio (white colonies/blue colonies) of the samples before treatment.In vitro samples were cultured overnight and diluted 1:1,000 in LB medium.All data represent the results of at least three independent experiments.Error bars represent standard deviations.**P < 0.01 and ***P < 0.001.

FIG 8
FIG8 Schematic diagram of the uptake of PQS-Fe 3+ and PCH-Fe 3+ by P. aeruginosa.Extracellular PQS-Fe 3+ is transported through OMVs.Under the mediation