The Role of Electrostatics in Colicin Nuclease Domain Translocation into Bacterial Cells

The mechanism(s) by which nuclease colicins translocate distinct cytotoxic enzymes (DNases, rRNases, and tRNases) to the cytoplasm of Escherichia coli is unknown. Previous in vitro investigations on isolated colicin nuclease domains have shown that they have a strong propensity to associate with anionic phospholipid vesicles, implying that electrostatic interactions with biological membranes play a role in their import. In the present work we set out to test this hypothesis in vivo. We show that cell killing by the DNase toxin colicin E9 of E. coli HDL11, a strain in which the level of anionic phospholipid and hence inner membrane charge is regulated by isopropyl β-d-thiogalactopyranoside induction, is critically dependent on the level of inducer, whereas this is not the case for pore-forming colicins that take the same basic route into the periplasm. Moreover, there is a strong correlation between the level and rate of HDL11 cell killing and the net positive charge on a colicin DNase, with similar effects seen for wild type E. coli cells, data that are consistent with a direct, electrostatically mediated interaction between colicin nucleases and the bacterial inner membrane. We next sought to identify how membrane-associated colicin nucleases might be translocated into the cell. We show that neither the Sec or Tat systems are involved in nuclease colicin uptake but that nuclease colicin toxicity is instead dependent on functional FtsH, an inner membrane AAA+ ATPase and protease that dislocates misfolded membrane proteins to the cytoplasm for destruction. The mechanism(s) by which nuclease colicins translocate distinct cytotoxic enzymes (DNases, rRNases, and tRNases) to the cytoplasm of Escherichia coli is unknown. Previous in vitro investigations on isolated colicin nuclease domains have shown that they have a strong propensity to associate with anionic phospholipid vesicles, implying that electrostatic interactions with biological membranes play a role in their import. In the present work we set out to test this hypothesis in vivo. We show that cell killing by the DNase toxin colicin E9 of E. coli HDL11, a strain in which the level of anionic phospholipid and hence inner membrane charge is regulated by isopropyl β-d-thiogalactopyranoside induction, is critically dependent on the level of inducer, whereas this is not the case for pore-forming colicins that take the same basic route into the periplasm. Moreover, there is a strong correlation between the level and rate of HDL11 cell killing and the net positive charge on a colicin DNase, with similar effects seen for wild type E. coli cells, data that are consistent with a direct, electrostatically mediated interaction between colicin nucleases and the bacterial inner membrane. We next sought to identify how membrane-associated colicin nucleases might be translocated into the cell. We show that neither the Sec or Tat systems are involved in nuclease colicin uptake but that nuclease colicin toxicity is instead dependent on functional FtsH, an inner membrane AAA+ ATPase and protease that dislocates misfolded membrane proteins to the cytoplasm for destruction. Colicins are SOS-induced protein antibiotics that penetrate and kill cells in competing Escherichia coli populations (1Kirkup B.C. Kim M.A. Nature. 2004; 428: 412-414Crossref PubMed Scopus (352) Google Scholar, 2Czaran T.L. Kim R.F. Pagie L. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 786-790Crossref PubMed Scopus (371) Google Scholar). Colicin entry is known to begin with binding to an extracellular receptor followed by translocation across the outer membrane (OM) 3The abbreviations used are: OM, outer membrane; IM, inner membrane; IPTG, isopropyl β-d-thiogalactopyranoside; ER, endoplasmic reticulum; LPS, lipopolysaccharide.3The abbreviations used are: OM, outer membrane; IM, inner membrane; IPTG, isopropyl β-d-thiogalactopyranoside; ER, endoplasmic reticulum; LPS, lipopolysaccharide. by a mechanism that has similarities to that used by filamentous bacteriophages in the early stages of phage infection (3Cascales E. Kim S.K. Duché D. Kleanthous C. Lloubès R. Postle K. Riley M. Slatin S. Cavard D. Microbiol. Mol. Biol. Rev. 2007; 71: 158-229Crossref PubMed Scopus (756) Google Scholar). Once in the periplasm the cytotoxic domains of colicins target the IM or, in the case of the nuclease colicins, are translocated entirely across the IM to reach their cytoplasmic nucleic acid substrates (3Cascales E. Kim S.K. Duché D. Kleanthous C. Lloubès R. Postle K. Riley M. Slatin S. Cavard D. Microbiol. Mol. Biol. Rev. 2007; 71: 158-229Crossref PubMed Scopus (756) Google Scholar, 4James R. Kim C.N. Moore G.R. Kleanthous C. Biochimie (Paris). 2002; 84: 381-389Crossref PubMed Scopus (76) Google Scholar). There is currently no information available as to how this latter step is accomplished, although it is likely to require unfolding of the nuclease (5Mosbahi K. Kim C. Keeble A.H. Mobasheri H. Morel B. James R. Moore G.R. Lea E.J. Kleanthous C. Nat. Struct. Biol. 2002; 9: 476-484Crossref PubMed Scopus (44) Google Scholar). Passage across the OM for both pore-forming and nuclease type colicins is mediated by an OM receptor(s) and proteins of either the Tol (for Group A colicins such as A, N, and E1–E9) or Ton (for Group B colicins such as B, D, Ia, and Ib) complexes. Colicins are able to utilize a wide variety of OM proteins as their primary receptor, such as FepA, FhuA, Cir, BtuB, and OmpF, all involved in passive or active nutrient transport across the OM (3Cascales E. Kim S.K. Duché D. Kleanthous C. Lloubès R. Postle K. Riley M. Slatin S. Cavard D. Microbiol. Mol. Biol. Rev. 2007; 71: 158-229Crossref PubMed Scopus (756) Google Scholar). In contrast, the translocation step is restricted to either the Ton or Tol proteins, with movement across the OM likely by a common mechanism because colicins can be engineered to take either route by the simple exchange of colicin domains or indeed the requisite phage domains (6Jakes K.S. Kim N.G. Zinder N.D. J. Bacteriol. 1988; 170: 4231-4238Crossref PubMed Google Scholar, 7Braun V. FEMS Microbiol. Rev. 1995; 4: 295-307Crossref Scopus (281) Google Scholar). The physiological role of the Ton (TonB·ExbB·ExbD) complex is in the energy-dependent transport of nutrients across the OM. The complex is coupled to the proton motive force across the IM and is transduced to OM receptors via TonB through so-called “TonB box” sequences, which stimulate the passage of nutrients across the OM (8Wiener M.C. Curr. Opin. Struct. Biol. 2005; 15: 394-400Crossref PubMed Scopus (108) Google Scholar). The Tol·Pal system (TolA·TolQ·TolR·TolB·Pal) is also a transmembrane assembly responsive to the proton-motive force. In contrast to the Ton system, however, the Tol·Pal complex is required for stability of the OM, with recent data suggesting that its physiological role is that of an energized tether that maintains the appropriate juxtaposition between the inner and outer membranes and newly formed peptidoglycan during cell division (9Lloubes R. Kim E. Walburger A. Bouveret E. Lazdunski C. Bernadac A. Journet L. Res. Microbiol. 2001; 152: 523-529Crossref PubMed Scopus (136) Google Scholar, 10Gerding M.A. Kim Y. Pecora N.D. Niki H. de Boer P.A. Mol. Microbiol. 2007; 63: 1008-1025Crossref PubMed Scopus (259) Google Scholar). with energized systems the cytotoxic domains of pore-forming colicins to the of the IM into which they that the cytoplasmic membrane Kim Biochimie (Paris). 2002; 84: PubMed Scopus Google Scholar). In contrast, nuclease type colicins, or their cytotoxic into the cytoplasm they on and and or and R. Kim C.N. Moore G.R. Kleanthous C. Biochimie (Paris). 2002; 84: 381-389Crossref PubMed Scopus (76) Google Scholar, H. Kim Biochimie (Paris). 2002; 84: PubMed Scopus Google Scholar). work on colicins and has the that the nuclease domains are the of the toxin in the periplasm passage across the In the case of colicin D, the protease has as the or a by M. Kim L. A. V. Mol. 2001; PubMed Scopus Google Scholar, Kim J. Biol. 2005; PubMed Scopus Google Scholar). A for this of is the of or between the nuclease domains or indeed between nucleases that via the Tol or Ton that for entry to the they are to the of the we the in the of nuclease charge on colicin which that to to the cytoplasm there is an electrostatically with the E. coli We also identify a translocation route for membrane-associated colicin and cells in or on with or and are by Kim B. M. H. K. Mol. Microbiol. PubMed Scopus Google Scholar). and are by Kim K. B.C. J. Bacteriol. 2001; PubMed Scopus Google Scholar). and are by Kim H. H. K. PubMed Scopus Google Scholar). HDL11 is by R. Kim de B. J. Biol. PubMed Google and and which are wild type for phospholipid are by Kim C. A. J. Bacteriol. PubMed Google and Kim M. Y. Y. M. M. H. Mol. Biol. Scopus Google The and are by J. Microbiol. Google and the strain wild type and the and have K. Kim J. Biol. PubMed Scopus Google Scholar, K. Kim Mol. Microbiol. 2001; PubMed Scopus Google Scholar, Kim K. K. J. Biol. PubMed Scopus Google Scholar). in the DNase of colicin E9 and DNase by the the as the The of the into the same of the colicins used either as with their proteins C. Kim D. Sci. Scopus Google or as isolated cell killing are by the of the protein R. Kim A. A. Moore G.R. James R. Kleanthous C. J. PubMed Scopus Google Scholar, M. Kim B. Moore G.R. Kleanthous C. C.N. James R. J. Bacteriol. 2005; PubMed Scopus Google Scholar). The colicin and colicin and and colicin E9 as D. Kim G.R. James R. Kleanthous C. PubMed Scopus Google Scholar). Colicin and and The cells in to an to a of and the cells for a all The cells in of and by passage through a of the cell by to a of and for this The by and the and the in a of into of the same The proteins to a and with a cell in an of E. coli HDL11 into of with or and with the appropriate colicin The of the test of with of to which of a of the appropriate strain with an of of because this strain on of of of colicin on the which test the of colicins of the colicin into and The to for to the cells and with of the strain as The and for a of the test of the as the a strain an system M. Kim B. Moore G.R. Kleanthous C. C.N. James R. J. Bacteriol. 2005; PubMed Scopus Google Scholar). cells to which they into with an and wild type colicins or proteins to a of between and colicin of followed a of to with M. Kim B. Moore G.R. Kleanthous C. C.N. James R. J. Bacteriol. 2005; PubMed Scopus Google for the by colicin and the of by by the by colicin in the and of of the and the for in to in the or of of Colicin a of to of the cells with the of the to for a which cells in and with the appropriate the cells and the of to a to which no colicin Colicin DNase in E. coli of have that the colicin and E9 DNase domains and with anionic but not phospholipid vesicles, with electrostatically the domains to K. Kim D. Lea E. Moore G.R. James R. Kleanthous C. J. Biol. 2004; PubMed Scopus Google Scholar, K. Kim D. James R. Moore G.R. Kleanthous C. Sci. 15: PubMed Scopus Google Scholar). Colicin also have the to in (5Mosbahi K. Kim C. Keeble A.H. Mobasheri H. Morel B. James R. Moore G.R. Lea E.J. Kleanthous C. Nat. Struct. Biol. 2002; 9: 476-484Crossref PubMed Scopus (44) Google Scholar). that the E. coli IM is of and anionic R. B. Riley M. M. Escherichia coli and and to the that colicin which are might with or both of the membrane systems of E. coli route to the The present work set out to test this In the of electrostatically interactions in the translocation of the E9 DNase we used an E. coli strain in anionic The strain HDL11 a of the the of the and is R. Kim de B. J. Biol. PubMed Google Scholar). The of the is for the of the anionic phospholipid In the of IPTG, HDL11 or but acid and has a of anionic of R. Kim de B. J. Biol. PubMed Google Scholar). is to in the of IPTG, with for the as is the case in wild type R. Kim de B. J. Biol. PubMed Google Scholar). The of of in of this not the of anionic phospholipid in the We that killing of strain HDL11 by an of colicin E9 cell by of the of the dependent on In the of IPTG, cell for the of the although cell In the of IPTG, however, the rate of cell killing to a level similar to that in wild type not The of or not the level of cell killing with or IPTG, the level between and The data show that although HDL11 is not to colicin the of cell killing are dependent on the level of anionic phospholipid by the The phospholipid of the IM and the inner of the OM is in the HDL11 the of on colicin E9 toxicity either The outer of the OM is of and is to be in the of the OM be to and this early in colicin receptor binding and translocation across the OM. this we the of HDL11 to the A pore-forming colicins and A, which translocate across the OM by the same basic mechanism as nuclease both colicins as their primary receptor and OM and OmpF, followed by translocation to the periplasm through binding of the Tol·Pal We that both pore-forming colicins HDL11 with killing of to the suggesting that the for colicins across the OM has We that the of anionic phospholipid on colicin E9 toxicity likely effects the We also DNase type colicins, and which have Kim A. Moore G.R. James R. Kleanthous C. J. Biol. PubMed Scopus Google Scholar, Kim A.H. A. James R. Moore G.R. Kleanthous C. Sci. 2004; PubMed Scopus Google Scholar). DNase colicins are that a of in their and translocation domains but are in their and although all are basic domains they in the and type of In the net positive charge for all DNase domains with and E9 and We that in contrast to colicin killing of E. coli HDL11 dependent on the colicins or cytotoxic HDL11 of to the data in the of are shown in for the the of not the level of cell colicin not the of on cell but not to the with colicin E9 DNase colicins are in the of their and translocation and are to the of on effects for the of colicin DNase charge and on cell The and hence of the and are of and can be because a of the of all DNase colicins no such colicin has the in vitro in both and with and E9 Kim A.H. A. James R. Moore G.R. Kleanthous C. Sci. 2004; PubMed Scopus Google Scholar). The net positive charge of the their to associate with the IM of this is we it be to or cell killing by or the positive charge on a colicin nuclease also the that the of HDL11 to between the enzymes of the DNase colicin and to the of positive test this hypothesis we engineered charge in the DNase of colicin both and the net positive of the active that are not involved in or binding Kim C. J. Biol. 2004; PubMed Scopus Google Scholar). and and for the isolated engineered to the of positive charge to and killing HDL11 to the wild type colicin with the In contrast, the of positive into the DNase of colicin E9 and cell killing of HDL11 to wild type colicin E9 We the of all the engineered charge and no correlation with their biological both of to wild type colicin E9 not data that the cell killing of a colicin DNase E. coli HDL11 with anionic phospholipid can be or by the of net positive and that this is not to in between the colicin there is a between cell killing and colicin DNase we all of the HDL11 cell killing data as the in the of as a of the positive charge of the the it is that there is a strong correlation between the net positive charge on a colicin DNase and its HDL11 We also for in toxicity HDL11 with and this to a wild type strain of E. coli In both a correlation with charge not but the in with the correlation between charge and as by in is the IM wild type charge and in E. coli that is of anionic of DNase Colicins into E. coli the of electrostatic interactions in colicin in wild type and to on cell entry we to of colicin colicins, and has used to colicin entry V. Kim J. K. K. J. Bacteriol. PubMed Google Scholar, H. Kim R. Lazdunski C. L. J. PubMed Scopus Google Scholar). of cells used by H. Kim R. Lazdunski C. L. J. PubMed Scopus Google to show that the pore-forming toxin colicin A is and to the extracellular its cytotoxic is the have by Duché D. J. Bacteriol. 2007; PubMed Scopus Google for the DNase colicin that the toxin to its OM receptor the Tol proteins in the periplasm. Colicins to or proteins in the OM and and this likely colicins such as and have with their cytotoxic and translocation domains M. Kim D. Nature. PubMed Scopus Google Scholar, S. Kim K. C. M. Mol. 2001; PubMed Scopus Google Scholar). is the or a during and this is consistent with the of a engineered across the with on of the with C.N. Kim B. N.G. R. M. Moore G.R. Kleanthous C. James R. J. Bacteriol. 2004; PubMed Scopus Google Scholar, N.G. Kim Moore G.R. James R. Kleanthous C. Proc. Natl. Acad. Sci. U. S. A. 2005; PubMed Scopus Google Scholar). In the of on HDL11 and work on colicin translocation V. Kim J. K. K. J. Bacteriol. PubMed Google Scholar, H. Kim R. Lazdunski C. L. J. PubMed Scopus Google Scholar, D. J. Bacteriol. 2007; PubMed Scopus Google Scholar, M. Kim D. Nature. PubMed Scopus Google Scholar, S. Kim K. C. M. Mol. 2001; PubMed Scopus Google Scholar, C.N. Kim B. N.G. R. M. Moore G.R. Kleanthous C. James R. J. Bacteriol. 2004; PubMed Scopus Google Scholar, N.G. Kim Moore G.R. James R. Kleanthous C. Proc. Natl. Acad. Sci. U. S. A. 2005; PubMed Scopus Google we that electrostatic interactions between the domains and the anionic phospholipid of the IM might entry of the nuclease to the cytoplasm and the rate of cell test this hypothesis we on wild type E. coli cells DNase colicins of The rate of cell entry by the of on the cell by the translocated DNase, the latter by a Previous work has shown that DNase colicins are strong of the D. Kim M. A. Moore G.R. James R. Kleanthous C. J. Bacteriol. 2004; PubMed Scopus Google and that of an to the a of by a DNase colicin that is of cell killing M. Kim B. Moore G.R. Kleanthous C. C.N. James R. J. Bacteriol. 2005; PubMed Scopus Google Scholar). the for of for colicins with DNase domains of positive charge to cells, which are wild type for anionic phospholipid DNase and The data show a in the of with the colicin a the consistent with electrostatic entry to the of cell and The data are as the in of cell or in the of the colicins and are wild type for phospholipid HDL11 in the of has wild type of anionic and in the of has of anionic The for and for and HDL11 cell type in a We also E. coli HDL11 cell by the of as a of the for of HDL11 cells with colicin E9 in the and of in the of colicin E9 for HDL11 cells is the same as that for the strain E. coli consistent with the anionic in the cell to wild type the to kill E. coli HDL11 cells that have not by for the same consistent with the that IM charge entry of the nuclease to the of cell for the of DNase colicins in this and with the of data The data the between nuclease charge and the rate of cell killing and show how this the charge on the IM of the or is The of this are colicin E9 of the with for cell killing of HDL11 is but that to in the of IPTG, and colicin for cell entry is and is to IM can of the colicin that are on the of the OM, data are consistent with recent suggesting that DNase colicins to their OM receptor during passage to the periplasm D. J. Bacteriol. 2007; PubMed Scopus Google Scholar). Moreover, we show that the IM translocation step is by the of positive charge on the DNase with enzymes of cell of Colicin across the the AAA+ ATPase of DNase colicin cell entry and on its net positive charge that colicin nuclease domains into the the of how they to the In the case of bacterial such as that on cells, it has that the endoplasmic via the that in protein is J. Kim A. S. J. Biol. PubMed Scopus Google Scholar, A. Kim H. A. V. J. Biol. PubMed Scopus Google Scholar). We the that colicin nuclease domains be by the known protein of the E. coli cytoplasmic the Sec and Tat with in or are in the of substrates and cell Kim K. B.C. J. Bacteriol. 2001; PubMed Scopus Google Scholar). the of to colicin E9 to the wild type that the Tat is not involved in colicin not The of the Sec transport system be but in and that this are known Kim J. PubMed Google Scholar). an all which such and by Kim H. H. K. PubMed Scopus Google no colicin or to the not suggesting that the Sec complex is also not required for colicin although we this out as a protein because of the be E. coli a in the IM to the protein of the the AAA+ protease has shown to be for the and of misfolded or membrane proteins K. Kim Y. Rev. Microbiol. 2005; PubMed Scopus Google Scholar). membrane with AAA+ ATPase and domains and a in the E. coli cytoplasmic membrane K. Kim Y. Rev. Microbiol. 2005; PubMed Scopus Google Scholar). The of in colicin translocation has because of early the of by of both nuclease and pore-forming colicins H. Kim S. Y. J. Bacteriol. PubMed Google Scholar, Kim H. Y. Y. K. H. J. Bacteriol. PubMed Scopus Google Scholar). it is likely that the in which the In the work of H. Kim S. Y. J. Bacteriol. PubMed Google the strain which isolated as a to colicin to colicins D, Ia, and in this strain shown to be to a acid in FtsH, which Kim H. Y. Y. K. H. J. Bacteriol. PubMed Scopus Google Scholar). there are between and a strain has because the of is to be to a of a of and an in the strain of an that is by the that the not show in a Kim K. K. K. J. H. Mol. Microbiol. PubMed Scopus Google Scholar). the with the that E. coli of both pore-forming and nuclease colicins, we this is an in E. with the of and a in the of the is a of the of of a in The can be in a the acid is and to the Kim K. K. K. J. H. Mol. Microbiol. PubMed Scopus Google Scholar). We the strain and the strain for to the nuclease type colicins D, and E9 and the pore-forming colicins and in colicins a and type colicins to kill strain whereas we no in the of and to colicins and The that the of the nuclease colicin translocation in can be out because and the strain show to colicin E9 Colicins and are pore-forming hence we also pore-forming colicins, and A, and no in their to kill or We and for to a of pore-forming and nuclease type colicins in The pore-forming colicins and B active both and whereas the nuclease type colicins and active not to the pore-forming colicins, colicin which cells through the of able to kill both and not has shown that the of colicin is to its of peptidoglycan M. Kim A. H. D. D. J. Biol. PubMed Scopus Google which that the pore-forming colicins, this toxin not have to into the We that of cells to all known of nuclease colicins and tRNases) but not pore-forming colicins or colicin M. to nuclease colicins is of the translocation route across the OM because is to both and all and pore-forming colicins are able to kill this We that is the route by which colicin nuclease domains their to the cytoplasm their electrostatically with the We the that the of is with the either by the of a protein or proteins the IM that nuclease colicin translocation or by a membrane K. Kim Y. Rev. Microbiol. 2005; PubMed Scopus Google Scholar). for colicins are which that the IM has not to colicins has to outer but is with in proteins that play a role in their import. in between nuclease colicins and by have not a that is that colicins kill cells the of with the wild type to colicin E9 with of cell killing not on of with and but with have by and K. Kim J. Biol. PubMed Scopus Google Scholar, K. Kim Mol. Microbiol. 2001; PubMed Scopus Google Scholar, Kim K. K. J. Biol. PubMed Scopus Google that with the of are the cytoplasmic the of to nuclease type colicin The of of the of with and in the ATPase is to the to the is in a of the protein of and and are active of the protease We DNase, and colicins and that all required the ATPase and protease of not E. coli is the membrane-associated protease that has in nuclease colicin but this is to a colicin D, with an E. coli not nuclease colicins such as M. Kim L. A. V. Mol. 2001; PubMed Scopus Google Scholar). that the of is required for both and nuclease colicin for this step in contrast to is of the of the nuclease or the of entry to the periplasm. The present work for the the of electrostatic charge in colicin nuclease cell of a direct, electrostatically mediated between the nuclease and the bacterial across the IM to the cytoplasm likely the of an system, in the same that OM and proteins are by colicin to translocation across the OM. We show that this latter step not transport through the Sec or Tat but is instead dependent on FtsH, an IM protease that misfolded membrane the that nuclease domains of colicins are for misfolded membrane proteins for in the data not this out of in colicin for through the of an IM protein in the strain that translocation to the The however, is that is the IM which have with toxin into is that bacterial that target cells to reach their target in the for a to reach the and Kim B. Sci. PubMed Scopus Google transport via the protein for the and of proteins, the toxin to the cytoplasm K. Kim 2002; PubMed Scopus Google Scholar). such as are able to the stages of and by the We that by with colicins by active active protease for for by might the colicin nuclease to the is that nuclease colicins are by the of FtsH, but a is to the we that is involved in colicin this be consistent with its C. Kim S. A. Mol. PubMed Scopus Google a for the that colicin nuclease with anionic K. Kim D. Lea E. Moore G.R. James R. Kleanthous C. J. Biol. 2004; PubMed Scopus Google Scholar, K. Kim D. James R. Moore G.R. Kleanthous C. Sci. 15: PubMed Scopus Google Scholar). with with this also to the protein to translocation into cells Kim 2002; PubMed Scopus Google Scholar). We for bacterial in and wild type and and of for in Tat proteins, of for strain HDL11, and and for in the Sec