Rare earth element (REE)-dependent growth of Pseudomonas putida KT2440 depends on the ABC-transporter PedA1A2BC and is influenced by iron availability

In the soil-dwelling organism Pseudomonas putida KT2440, the rare earth element (REE)-utilizing and pyrroloquinoline quinone (PQQ)-dependent ethanol dehydrogenase PedH is part of a periplasmic oxidation system that is vital for growth on various alcoholic volatiles. Expression of PedH and its Ca2+-dependent counterpart PedE is inversely regulated in response to lanthanide (Ln3+) bioavailability, a mechanism termed the REE-switch. In the present study, we demonstrate that copper, zinc, and in particular, iron availability influences this regulation in a pyoverdine-independent manner by increasing the minimal Ln3+ concentration required for the REE-switch to occur by several orders of magnitude. A combined genetic- and physiological approach reveals that an ABC-type transporter system encoded by the gene cluster pedA1A2BC is essential for efficient growth with low (nanomolar) Ln3+ concentrations. In the absence of pedA1A2BC, a ~100-fold higher La3+ concentration is needed for PedH-dependent growth but not for the ability to repress growth based on PedE activity. From these results, we conclude that cytoplasmic uptake of lanthanides through PedA1A2BC is essential to facilitate REE-dependent growth under environmental conditions with poor REE bioavailability. Our data further suggest that the La3+/Fe3+ ratio impacts the REE-switch through the mismetallation of putative La3+-binding proteins, such as the sensor histidine kinase PedS2, in the presence of high iron concentrations. As such, this study provides an example for the complexity of bacteria-metal interactions and highlights the importance of medium compositions when studying physiological traits in vitro in particular in regards to REE-dependent phenomena.


Nancy, France
Metal ions are essential for all living organisms as they play important roles in stabilizing 4 1 macromolecular cellular structures, by catalyzing biochemical reactions or acting as cofactors for 4 2 enzymes (Gray, 2003;Merchant and Helmann, 2012). They can, however, also be toxic to cells at 4 3 elevated levels through the generation of reactive oxygen species or by aspecific interactions such 4 4 as mismetallation (Cornelis et al., 2011;Dixon and Stockwell, 2014;Foster et al., 2014). Bacteria have hence developed a sophisticated toolset to maintain cellular metal homeostasis (Andrews et 4 6 al., 2003;Chandrangsu et al., 2017;Schalk and Cunrath, 2016;Semrau et al., 2018). Common 4 7 mechanisms include release of metal-specific scavenger molecules, the activation of high-affinity 4 8 transport systems, the production of metal storage proteins, and the expression of specific efflux TES, the addition of ≥ 100 µM of La 3+ was required for growth inhibition in the Δ pedH strain it was the case upon the supplementation with TES. Since citrate is used as a metal chelator in FeSO 4 the addition of citrate had no effect, strongly indicating that FeSO 4 is predominantly PedH-dependent growth under any tested FeSO 4 concentration ( Figure 3B). Notably, strain condition, no growth was observed.

3 4
From these data, it can be speculated that beside the PedH-dependent growth also the inhibition concentrations. In addition, strain Δ pedH grew readily on 2-phenylethanol under all these 2 3 8 conditions even with no FeSO 4 supplementation ( Figure 4A). When 10 µM La 3+ was available, increased pedE promoter activity was detected.

4 3
Due to the very low concentrations of REEs (nM range) required for REE-dependent growth it is 2 4 4 commonly speculated that specific REE uptake systems must exist. From our previous results, we can conclude that pyoverdine is not such a system. A search of the genomic context of the ped 2 4 6 gene cluster identified a putative ABC transporter system located nearby the two PQQ-EDHs 2 4 7 encoding genes pedE and pedH ( Figure 5A). The ABC-transporter is predicted to be encoded as  Oldehinkel et al. , 2006). As the gene pedA1 is predicted to be such a substrate binding protein, it 2 5 5 is very likely that this transporter represents an import system. oxidized degradation intermediates of 2-phenylethanol, namely 2-phenylacetaldehyde and 2 6 0 phenylacetic acid, to a similar OD 600 within 48 h of incubation ( Figure 5B), indicating that the 2 6 1 transport system is not involved in carbon substrate uptake. When subsequently different La 3+ concentrations were tested in a similar setup, we found that of TES ( Figure 6A). This was in contrast to the Δ pedE deletion strain, which grew in the addition ( Figure 6B). Notably, beside the substantial difference in lag-phase, also the critical  of TES in the medium ( Figure 6C). Δ pedA1A2BC ( Figure 6D). This indicated that the impaired growth phenotype of the ABC- response to La 3+ supplementation (26-fold for KT2440*::Tn7-pedH-lux and 23-fold in 2 9 0 Δ pedA1A2BC::Tn7-pedH-lux).

9 1
In contrast to PedE, the signal peptide of PedH contains two adjacent arginine residues, which is 2 9 2 an indication that it might be transported to the periplasm in a folded state via the Tat (twin-2 9 3 arginine translocation) protein translocation system (Berks, 2015). Therefore, one could argue 2 9 4 that the transport of lanthanides into the cytoplasm might be beneficial as the incorporation into substrate, impaired growth on 2-phenylethanol in the presence of La 3+ should be observable. In the present study, we reveal that iron availability severely affects the REE-switch in 3 0 9 Pseudomonas putida KT2440. This is evidenced by the reduction of the critical concentration of La 3+ that is required both to promote PedH-dependent growth and for the repression of growth 3 1 1 based on PedE activity. By using a Δ pvdD deletion strain, we demonstrate that the production of 3 1 2 the iron chelating siderophore pyoverdine is not required for PedH-dependent growth under low 3 1 3 La 3+ conditions. Our data suggest that the observed effects during high Fe 2+/3+ /La 3+ ratios are by Fe 2+/3+ ions that are in excess in the medium, and can also be present in the same 3+ oxidation state (Foster et al., 2014;Tottey et al., 2008;Tripathi and Srivastava, 2006;Webb, 1970 pyoverdine deficient strain would be unable to take up enough of this essential element that is, most likely, present at trace levels in the medium even without additional supplementation.

0
It is further interesting to point out that also micromolar Cu 2+ and Zn 2+ inhibited growth in involved in this process, too.

9
We provide compelling evidence that the predicted ABC-transporter PedA1A2BC is essential for and >50% for pedA1, pedB and pedC) and that all bacterial strains that have been reported to Since the PedH enzyme, like all currently known Ln 3+ -dependent enzymes, resides in the 3 5 7 periplasm and since the purified apoenzyme of PedH can be converted into the catalytically represents a so-far uncharacterized transcriptional regulator or another REE-dependent enzyme.

6 4
It has been demonstrated that the location of protein folding can regulate metal binding (Tottey et periplasm. However, we could not find evidence that PedH is a Tat substrate and consequently both still showed PedH-dependent growth. However, it cannot be excluded that the two Tat   3  7  1 systems are functionally redundant since attempts to generate a tatC1/C2 double mutant strain 3 7 2 proved unsuccessful.

7 3
We can further conclude that the putative La 3+ transport into the cytoplasm is not required to activate pedH transcription. It is however possible that additional genes/proteins required for PedH-dependent growth rely on, or are regulated by, the cytoplasmic presence of REEs. In this context it is interesting to note that in a recent proteomic approach, we found that besides PedE 3 7 7 and PedH, additional proteins of unknown function show differential abundance in response to whether any of these proteins is required for PedH function. In P. putida KT2440, no close homolog to any of the aforementioned TonB-dependent receptors Knief et al., 2010). One could therefore speculate that a specific Ln 3+ -chelator system is perhaps homologous TonB-dependent receptor could also be explained by structural differences in the 3 9 6 REE-specific chelator system that might be employed by P. putida compared to that of the 3 9 7 methylotrophic bacteria. Lastly, it is also possible that in P. putida the REE uptake across the 3 9 8 outer membrane proceeds via the same chelator systems that are used for pyoverdin independent 3 9 9 Fe-acquisition. This could further provide another explanation for the impact of the Fe 2+/3+ to 4 0 0 La 3+ ratio on the REE-switch.

9 3
The two-step electrochemical oxidation of alcohols using a novel recombinant PQQ alcohol 5 9 4 dehydrogenase as a catalyst for a bioanode. Bioelectrochemistry 94, 75-78. folding location can regulate manganese-binding versus copper-or zinc-binding. Nature