The FinO family of bacterial RNA chaperonesJ.N. Mark Glover a,*, Steven G. Chaulk a, Ross A. Edwards a, David Arthur a,Jun Lu a, Laura S. Frost b

a Department of Biochemistry, University of Alberta, Edmonton, AB T6G 2H7, Canadab Department of Biological Sciences, University of Alberta, Edmonton, AB T6G 2E9, Canada

A R T I C L E I N F O

Article history:Available online 4 August 2014Communicated by Ananda Chakrabarty

Keywords:FinOFinPAntisense RNARNA chaperonePlasmid conjugation

A B S T R A C T

Antisense RNAs have long been known to regulate diverse aspects of plasmid biology. Herewe review the FinOP system that modulates F plasmid gene expression through regula-tion of the F plasmid transcription factor, TraJ. FinOP is a two component system composedof an antisense RNA, FinP, which represses TraJ translation, and a protein, FinO, which isrequired to stabilize FinP and facilitate its interactions with its traJ mRNA target. We reviewthe evidence that FinO acts as an RNA chaperone to bind and destabilize internal stem-loop structures within the individual RNAs that would otherwise block intermolecular RNAduplexing. Recent structural studies have provided mechanistic insights into how FinO mayfacilitate interactions between FinP and traJ mRNA. We also review recent findings thattwo other proteins, Escherichia coli ProQ and Neisseria meningitidis NMB1681, may repre-sent FinO-like RNA chaperones.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

The importance of small, non-coding RNAs (ncRNAs)1

that regulate gene expression in all kingdoms of life isbecoming increasingly clear. ncRNA function involves basepairing with either RNA or DNA targets in an energeticallyfavorable interaction that in theory should require noenzyme co-factors. However, the interaction of two nucleicacid polymers often involves the destabilization of inter-nal structures (e.g., hairpins) within the individual nucleicacids. Although these internal structures are less energet-ically favorable than the final paired structure (e.g., duplex)between the two molecules, the bimolecular interactionmay be kinetically unlikely in the absence of factors thatcan destabilize the internal structures as a first step in theprocess. This paradox lead to the proposal of the RNAchaperone hypothesis by Herschlag (1995) and it is now

widely believed that RNA-based regulatory processes oftenrequire protein co-factors that may act as chaperonesto regulate RNA-target interactions. In bacteria, the bestcharacterized RNA chaperone is the Sm protein Hfq, whichforms hexameric ring structures that provide multiplebinding surfaces for RNAs (for recent reviews, see(Weichenrieder, 2014; Sauer, 2013; Wagner, 2013; Wiluszand Wilusz, 2013). Here we review the evidence that theregulatory protein FinO of F-like plasmids represents anotherclass of bacterial RNA chaperones. Whereas FinO is bestcharacterized as a critical inhibitor of the F plasmid traoperon, recent findings indicate that FinO-like proteins withRNA chaperone activity may be widespread throughoutbacterial species.

2. Discovery of an antisense RNA system that regulatesF plasmid gene expression

Antisense RNA, defined as a transcript that is expressedin cis to its target RNA, has a long and important associa-tion with plasmid biology. It has been the subject of intensestudy on the regulation of replication initiation and copynumber control, expression of antidotes to post-segregational

* Corresponding author.E-mail address: mark.glover@ualberta.ca (J.N. Mark Glover).

1 Abbreviations: ncRNA, non-coding RNA; HF, Thigh frequencies oftransfer; UTR, untranslated region; FRET, Förster resonance energy trans-fer; SAXS, small angle X-ray scattering.

http://dx.doi.org/10.1016/j.plasmid.2014.07.0030147-619X/© 2014 Elsevier Inc. All rights reserved.

Plasmid 78 (2015) 79–87

Contents lists available at ScienceDirect

Plasmid

journal homepage: www.elsevier.com/ locate /yplas

killing, control of the frequency of bacterial conjugation, andthe expression of traits carried as cargo by plasmids suchas Tn10 transposition and heavy metal resistance. One ofthe best-described systems is the fertility inhibition systemof IncF (F-like) plasmids that downregulates conjugative geneexpression.

Bacterial conjugation by transfer systems encoded onIncF plasmids requires the expression of transcripts for pilussynthesis, DNA processing, mating pair stabilization, surfaceand entry exclusion as well as regulation (reviewed inGubbins et al., 2005; Arutyunov and Frost, 2013). Positiveactivation of tra transcription requires the TraJ protein, whichde-silences the tra promoters and positively regulates themain transfer promoter PY in conjunction with host-encoded ArcA (Frost and Koraimann, 2010; Wagner et al.,2013). The control of traJ mRNA translation is central toregulating the levels of TraJ in the cell and the resultinglevel of transfer. This is accomplished by the “fertility in-hibition” system of FinOP.

The concept of fertility inhibition (fin) goes back to theearliest days of research on plasmid conjugation. F+ donorsexhibited high frequencies of transfer (HFT) to newly in-troduced recipient (F− of plasmid-free) cells with the resultingdonors and transconjugants capable of the same high matingfrequencies. In contrast, other plasmids usually transferredat reduced frequencies when recipient (plasmid-free) cellswere introduced but new transconjugants exhibited dere-pressed (HFT) levels of transfer. In the absence of newrecipient cells, it took approximately seven generations tore-establish fertility inhibition (Stocker, 1963). This “epi-demic spread” and (Willetts, 1974; Cullum et al., 1978) wasattributed to relief from fertility inhibition by an unknownmechanism.

In the 60s, Japanese researchers identified plasmids car-rying antibiotic resistance (R factors) that, when introducedinto F+ donor cells, inhibited F conjugation (Egawa andHirota, 1962; Watanabe and Fukasawa, 1962). This wastermed i+ or fi+ by Hirota and Watanabe, respectively. TheseR factors transferred 50–1000-fold less well than F buttransfer-proficient or derepressed (drd) mutants of these Rfactors could be isolated. Meynell and Datta (1967) sur-mised that F and these R factors were closely related andthat the R factors encoded a repressor molecule, FinO, whichcould inhibit conjugative transfer by both F and the R factors(Meynell et al., 1968). Genetic studies further revealed thatthe transfer operons of F and F-related (IncF) R factors werepositively regulated by TraJ and that two mutations thatmapped either proximally or distally to traJ, finP and finO,respectively, resulted in derepressed levels of transfer(Timmis et al., 1978). Sequence analysis of F revealed an IS3insertion within finO, explaining why F was naturally de-repressed for transfer (Cheah and Skurray, 1986; Yoshiokaet al., 1987). The site of action of FinO was mapped to thefinP locus near traJ (Finnegan and Willetts, 1971, 1973;Grindley et al., 1973) but the nature of FinP remained puz-zling until promoter assessment studies suggested that atranscript encoding an antisense RNA in cis to traJ was FinP(Mullineaux and Willetts, 1985). Unlike FinO, FinP is plasmid-specific in that FinP RNAs from different plasmids only acton their cognate plasmid traJ. Sequencing finP from a numberof F-like plasmids demonstrated that this specificity resided

in different sequences within predicted loops of FinP RNA(Willetts and Maule, 1986; Finlay et al., 1986).

3. FinP and its target traJ mRNA

FinP is a 79 nucleotide (nt) RNA transcribed in the op-posite direction to its target traJ mRNA, which has a 105nt untranslated region (UTR) 5′ to its coding sequence(Mullineaux and Willetts, 1985) (Fig. 1). Similar results wereobtained for the F-like plasmids R100 (Dempsey, 1989) andR1 (Koraimann and Hogenauer, 1989). FinP was predictedto have two stem-loops named SLI and SLII whereas thetraJ UTR has three stem loops SLIc, SLIIc and SLIII (SLIc andSLIIc are complementary to SLI and SLII, respectively). Thesestructures were mapped using single- and double-strandednucleases and a preliminary rate constant of associationof the two RNAs was determined (kapp of 5 × 105 M−1 s−1)that agreed with other natural antisense RNA systems (vanBiesen et al., 1993). In vivo, FinP has a short half-life thatis extended by the presence of FinO. This was due to in-creased stabilization of the FinP transcript rather thanincreased activity of the finP promoter. Construction of anF plasmid derivative that lacked a finP promoter gave wildtype levels of transfer. This could be complemented by sup-plying finO and finP in trans from foreign promoterssuggesting that FinP required FinO for increased stabilitypossibly by protecting it from cellular nucleases such asribonuclease E (RNase E) Lee et al., 1992. Incidentally,overexpression of finP from foreign promoters can com-pensate for the absence of FinO leading to almost wild-type levels of repression (Lee et al., 1992; Koraimann etal., 1991, 1996).

Finnegan and Willetts (1971, 1973) reported two classesof fertility inhibition mutants in finP. One wascomplementable (finP mutants) in trans and the other wasonly partially complementable (fisO for “site of fertility in-hibition by FinO”) in the presence of FinO. The finP mutationswere thought to destabilize the stems in the stem-loops ofFinP whereas the fisO mutations in SLI affected FinO binding.Mutants in the spacer region between SLI and SLII were alsorelatively insensitive to the presence of FinO (Frost et al.,1989). This turned out to be due to an increased sensitiv-ity to RNase E (Jerome et al., 1999). Jerome and Frost (1999)determined that efficient FinO binding required an intactSLII and short (4–6 bases) single-stranded regions at the 5′and 3′ ends of FinP. FinO protected FinP from RNase E andpromoted FinP-traJ UTR interactions, blocking access to thetraJ mRNA ribosome binding site and leading to inhibitionof traJ mRNA translation. The resulting duplex RNA is de-graded by RNase III (Jerome et al., 1999).

In vitro, FinO binds to FinP and the traJ UTR with similarkinetics (Jerome and Frost, 1999). However, in vivo, FinO doesnot affect the stability or translational activity of the traJmRNA prior to its role in promoting FinP-traJ UTR duplexformation. The function of a third stem-loop (SLIII) in thetraJ UTR, 25 bases from SLIIc, is not readily apparent.However, it is involved in Hfq-mediated degradation of thetraJ mRNA (Jerome and Frost, 1999) possibly in conjunc-tion with a small RNA utpR, found on the “dark side” of theF plasmid, opposite the tra operon (Will, Majdalani and Frost,unpublished observations (Gubbins et al., 2005). A schematic

80 J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

diagram that summarizes the role of the FinOP system inthe regulation of TraJ is shown in Fig. 2.

4. Early characterization of FinO

The finO gene is located at the distal end of the tra operon,approximately 33 kb away from finP. In F, it is interruptedby an IS3 element rendering F naturally derepressed. Thisis unusual since most IncF plasmids have wild-type finOgenes. finO encodes a 22.2 kDa protein that was shown tobe required for FinP repression of traJ translation (re-viewed in Frost et al., 1994). The promoter for finOtranscription has not been determined unequivocally. It mightbe transcribed from the main promoter PY 33 kb away or it

might also be expressed from secondary promoters at thedistal end of the operon (Fowler and Thompson, 1986; Hamet al., 1989; Nuk et al., 2011). Willetts and Maule (1986) iden-tified two variants of finO among F-like plasmids, types 1and 2 that repressed transfer 100–1000-fold or 20–50-fold. This difference was traced to the presence of orf286upstream of finO in plasmids capable of high levels of trans-fer repression rather than sequence differences in finO itself.Interestingly, orf286 extended the half-life of finO mRNA incis but not in trans. The orf286-finO mRNA transcript is ableto fold into a complex secondary structure that increasesthe stability of the finO transcript. The ribosome binding sitefor finO is exposed in a loop that allows access to ribosomesand translation of the mRNA (van Biesen and Frost, 1992).

Fig. 1. The components of the FinOP regulatory system. Left center is shown a region of the F plasmid 5′ to the tra operon which encodes traJ and itsantisense FinP RNA. The sequences and predicted secondary structures of these RNAs are shown, as well as the ribosomal binding site (RBS) and initiatingAUG codon in traJ mRNA. On the right is shown two views of the structure of FinO26-186 with secondary structure elements labeled (PDB accession number1DVO).

81J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

5. Towards a structural mechanism of FinO function

Amino acid sequence analysis of the FinO protein indi-cated that it likely adopted a novel structure and was highlypositively charged, consistent with its role in RNA pro-cesses. Proteolytic mapping of purified FinO both alone andin complex with its target FinP RNA indicated that N- andC-terminal regions of the protein are flexible in the absenceof RNA but become resistant to proteolysis in the presenceof RNA, suggesting that the terminal tails of FinO may contactRNA (Ghetu et al., 1999). The crystal structure of a frag-ment of FinO lacking the flexible N-terminal 25 amino acidsrevealed a largely helical structure resembling a right-handed fist with extended index finger and thumb,corresponding to N- and C-terminal helices (Ghetu et al.,2000) (Fig. 1). Positively charged surfaces near the N-terminusof α1, as well as a large positively charged patch on the bodyof the protein suggested two possible RNA contact sur-faces. This idea was tested through RNA–protein crosslinkingand FRET experiments utilizing a series of 12 single cyste-ine substituted FinO mutants (Ghetu et al., 2002). Thecrosslinking experiments demonstrated that cysteine resi-dues positioned near either positively charged patch couldefficiently crosslink to FinP SLII, however cysteine residuespositioned near a negatively charged patch (bottom of FinOas illustrated in Fig. 1) did not crosslink to SLII RNA. The FRETexperiments utilized a technique that enabled the analysisof FRET on distinct FinO-SLII complexes that had been sepa-rated by native gel electrophoresis, thereby reducingbackground from unbound protein or RNA. A two-strandedversion of SLII RNA was used that could be 5′ tagged withfluorescein at either the 5′ single stranded tail or at the topof the duplex where the loop would be in a natural SLIIhairpin. FRET was measured between the fluorescein linkedto the RNA and specific fluorophores covalently linked to

the single cysteine positions on FinO. The results indicatedthat efficient FRET occurred between the fluorescein linkedto the 5′ single stranded tail of SLII and any position on FinO,however, lower levels of FRET were observed between RNAmolecules labeled at the opposite end of the molecule atthe position of the hairpin loop. This suggested that FinOspecifically contacts the single stranded regions and perhapsthe base of the SLII stem. The regions of the SLII RNA thatcontact FinO were probed in more detail by RNase diges-tion experiments (Arthur et al., 2011). These results showedthat FinO strongly protected the 3′ single stranded tail fromRNases, as well as the duplex stem extending ∼1/2 turn fromthe single stranded tails. Interestingly, the results sug-gested that, unlike the 3′ tail, the susceptibility of the 5′ singlestranded tail to RNase degradation was actually enhancedby FinO binding. Not only was the 3′ tail highly protectedagainst RNase degradation by FinO, but it was also foundthat modification of the 3′ nucleotide, either by phosphory-lation or oxidation, markedly reduced FinO binding to SLIIRNA, suggesting an intimate recognition of the 3′ nucleo-tide. Intriguingly, both Hfq (Wilusz and Wilusz, 2013) as wellas RNA polymerase III regulatory factor La (Dong et al., 2004)specifically recognize 3′ nucleotides of their substrate RNAs.

To date, FinO-RNA complexes have eluded high-resolutionstructural analyses, however structural models for FinO andits interactions with SLII have been developed using an ap-proach that integrates the crosslinking, FRET and footprintingresults described above, with small angle X-ray scattering(SAXS) data obtained from purified FinO–SLII complexes(Arthur et al., 2011). SAXS can yield moderate resolutionstructural information for flexible macromolecular com-plexes in solution, and, together with constraints providedby other experimental methods as well as models of the in-dividual molecular components, it can be a powerful toolto develop molecular models for such complexes (Rambo

Fig. 2. Schematic overview of the regulation of traJ by the FinOP system. In the absence of FinO (i), FinP is degraded by RNase E, allowing translation oftraJ mRNA (ii), which in turn drives tra operon transcription, resulting in a transfer-competent cell (iii). As FinO protein accumulates, FinO binds FinP (iv),stabilizing it against RNase E degradation and facilitating initial interactions with traJ mRNA (v), leading to duplexing and inhibition of traJ translation(vi).

82 J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

and Tainer, 2013). In this method, a molecular dockingprogram, HADDOCK, was used that allows the definition ofprobable contact residues in both the RNA and protein tolimit the number of possible docked complexes. Starting with2500 docked FinO-SLII complexes, these were ranked basedon their agreement with the experimental SAXS data andthe top 10% were then clustered into five groups based onstructural similarity. The top group in terms of agreementwith the SAXS data has the RNA oriented against FinO suchthat the large, positively charged patch on the body of theprotein contacts the 3′ single-stranded tail of SLII as wellas the base of the duplex (Fig. 3). The duplex portion of SLIIis directed toward α1 (the “index finger”). Interestingly, whilemodels were produced that satisfied the biochemical re-straints within the initial 2500 docked complexes in whichthe SLII duplex was directed away from α1, these modelsdid not provide good fits to the SAXS data. Thus, SAXS wasable to provide key information that likely has defined theoverall orientation of FinO interaction with a minimal RNAsubstrate.

6. Mechanism of FinO RNA chaperone activity

Early work using full length FinP and traJ mRNA sug-gested that FinO could enhance pairing between these RNAs∼5-fold (van Biesen and Frost, 1994). Pairing between thesehighly structured RNAs is thought to initiate with interactionsbetween complementary hairpin loops via “kissing”

interactions (Gubbins et al., 2003) however the highly stableduplex regions within each RNA were predicted to block theattainment of the complete FinP-traJ mRNA duplex. To testthe idea that FinO might function to overcome energy bar-riers associated with the internal structures within FinP andthe complementary region of traJ mRNA, Arthur et al. usedan assay which tested the ability of FinO to catalyze RNAstrand exchange between a two-stranded version of SLII, anda complementary single strand (Arthur et al., 2003). Theresults indicated that FinO could dramatically enhance strandexchange activity, consistent with the notion that FinO func-tions to destabilize the internal hairpins within target RNAs.Deletions of the flexible N-terminal regions progressivelyreduced strand exchange activity so that FinO45-186, whichlacks the flexible N-terminus as well as the N-terminal portionof α1, exhibited no strand exchange activity. Point muta-tions in FinO suggested that the N-terminus of α1 is especiallycritical for strand exchange activity. This region is highly pos-itively charged but also contains conserved hydrophobicresidues that appear to be particularly important for strandexchange. The behavior of the same mutants in assays thatmeasured the ability of FinO to catalyze duplexing betweencomplementary RNAs derived from FinP and traJ mRNAstrongly suggested that the ability of FinO to catalyze RNAstrand exchange is linked to its ability to facilitate sense-antisense RNA interactions. Interestingly, an inversecorrelation was observed between either strand exchangeor RNA duplexing activity and the SLII binding affinity of

Fig. 3. Structural model of FinO and its recognition of SLII (Arthur et al., 2011). Regions of the RNA in red are protected against RNase degradation byFinO. Residues shown in sticks on α1, α2, α4 and α6 were found to strongly crosslink to SLII. An electrostatic surface is displayed for FinO in the rightpanel. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

83J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

the N-terminal mutants such that mutants with the loweststrand exchange activities actually exhibited significantlygreater RNA binding. It was suggested that FinO might utilizethe free energy of RNA binding to offset the unfavorable de-stabilization of hairpin base pairing that would be expectedto be part of the RNA strand exchange activity. The findingthat FinO RNA chaperone activity is dependent on a highlyflexible region of polypeptide chain is a common theme inboth RNA and protein chaperones (Tompa and Csermely,2004). While this disordered region is required for chap-erone activity, it appears to be dispensable for FinPstabilization in cells (Arthur et al., 2003). Interestinglyhowever, N-terminal deletions strongly abrogate the abilityof FinO to repress conjugation, such that FinO45-186 exhibitsno inhibition of conjugation, even though this mutant is ableto protect FinP against degradation similarly to the full-length protein (Arthur et al., 2003).

While more detailed structural studies are required toclearly elucidate the molecular mechanisms underlying FinOactivity, the results to date provide a basis to begin toimagine how FinO acts to regulate traJ expression (Fig. 4).Upon initial entry and replication of F plasmid in a newlyinfected cell, traJ is transcribed and translated, in turn ac-tivating the tra operon that leads to the accumulation ofFinO. FinO, utilizing its RNA binding core (amino acids 45–186), binds FinP SLII, thereby protecting it from degradationand allowing FinP accumulation. We suggest FinP initiallyrecognizes the complementary structure in traJ mRNA viakissing interactions between the complementary loops inthe two RNAs, perhaps facilitated by the N-terminal regionof FinO (Ghetu et al., 2000). The current model for FinO-SLII binding would direct the N-terminus toward the regionof the loop, providing further support for this suggestion.The N-terminus of α1 appears to be most critical for chap-erone activity, and we suggest that the highly positivelycharged nature of this region brings it into contact withthe RNA near or at the loops. The conserved hydrophobicresidues in this region, in particular Trp36, could interactwith the bases, potentially via stacking interactions to de-stabilize the internal stem-loop base pairs and thereby allowpropagation of intermolecular pairing between the twoRNAs. The ability of FinO to not only bind FinP but alsotraJ mRNA presents the possibility that FinO could eitheract by binding only one RNA of the pair, or in a pseudo-symmetric arrangement in which both FinP and traJ mRNAare bound by FinO. Eventually, the progression of the re-action would unwind the stem-loop structures, leading todisengagement of FinO45-186, however it is possible that theN-terminal region of FinO might still be associated withthe two RNAs as they complete the annealing process.

7. The FinO family of RNA chaperones – NMB1681and ProQ

In a class of its own for two decades, recent work hasestablished a FinO family of RNA chaperones consisting ofat least three proteins to date based on amino acid se-quence as well as structural similarity. The second memberof this RNA chaperone family is the Neisseria meningitidisprotein NMB1681. While mechanisms of gene regulation inN. meningitidis remain poorly understood, it is becoming clear

that RNA chaperones, in particular the N. meningitidis Hfqortholog, are associated with stress and virulence re-sponses (Fantappie et al., 2009; Hey et al., 2013). The crystalstructure of protein NMB1681 revealed a highly alpha-helical C-terminal core domain, highly similar to the foldedcore domain of FinO (Chaulk et al., 2010). NMB1681 alsocontains flexible N- and C-terminal tails although the lengthof these tails are shorter than the N-terminal flexible regionof FinO. The RNA chaperone capacity of NMB1681 was dem-onstrated by RNA binding, RNA strand-exchange and RNAduplexing assays developed for FinO (Chaulk et al., 2010).Like FinO, NMB1681 binds RNA tightly with a KD in thenanomolar range and prefers double stranded over single-stranded RNA. Consistent with these in vitro results, NMB1681was able to repress conjugation as well as stabilize FinP mRNAin a finO deficient Escherichia coli strain. While there is con-vincing evidence that NMB1681 can act as a RNA chaperoneboth in vitro and in vivo, it remains to be determined whatrole this RNA chaperone plays in N. meningitidis biology.

The third member of the FinO family of RNA chaper-ones is ProQ, best known as a regulator of the E. coli prolinetransporter ProP. ProQ not only regulates ProP as part of theosmotic response enabling E. coli growth in high osmolal-ity media, but also has been shown to play a role in biofilmformation (Sheidy and Zielke, 2013). Though the genetic con-nection between ProP and ProQ has been established, themechanism of regulation of ProP by ProQ remains elusive.ProQ is comprised of three domains: an N-terminal FinO-like domain (similar to FinO45-186); a linker region; and anHfq-like C-terminal domain. Using a full-length constructalong with N-terminal and C-terminal domain constructsChaulk et. al. assessed the RNA chaperone activities of thefull length protein as well as the individual domains.(Chaulket al., 2011) While the nanomolar scale RNA binding affin-ity of ProQ resides in the N-terminal FinO-like domain, boththe N-terminal and C-terminal domains display RNA strandexchange and RNA duplexing activities on FinP/TraJ basedsubstrates. The combination of weaker RNA binding affin-ity with retention of strand exchange and duplexing activityin the Hfq domain of ProQ is reminiscent of the N-terminalresidues 1–61 of FinO, which harbours the region neces-sary for FinO chaperone activity. In this manner, the FinOdomain may serve as an RNA binding anchor for the chap-erone activity within the Hfq domain (Chaulk et al., 2011).Although the effect of ProQ is thought to extend beyond reg-ulation of ProP, identification of RNAs that bind ProQ hasremained elusive. Recent work has shown that ProQ asso-ciates with ribosomes (Jiang et al., 2007) dependent on thepresence of mRNA (Sheidy and Zielke, 2013). While ProQcan tightly bind mRNA, it has not been shown to exhibitspecificity for ProP mRNA (Sheidy and Zielke, 2013). Iden-tification of the direct targets of ProQ is necessary to trulydetermine the mechanism of action of this RNA chaperone.

8. Concluding remarks

The FinO RNA chaperone provides an elegant and simplemechanism that allows an initial burst of tra expression,coupled to a gradual, FinOP-dependent repression of thesystem. The subsequent repression is likely necessary toreduce the metabolic burden of the plasmid on the host, as

84 J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

Fig. 4. A potential mechanism for FinO RNA chaperone function in the regulation of FinP-traM mRNA duplexing. (i) Plasmid transcription leads to pro-duction of traJ mRNA and antisense FinP. (ii) FinO binds to stem-loop structures in the RNAs utilizing the folded core of the protein which recognizes thebase of the stem-loop duplex as well as the single stranded tails. (iii–iv) FinO facilitates interactions between the complementary RNAs, likely involving“kissing” interactions between the RNA loops in a manner that may be stimulated by the N-terminal regions of FinO (red). Particularly critical is the regionat the N-terminus of α1 containing a cluster of lysine residues as well as Pro34 and Trp36 which are critical for RNA strand exchange and RNA duplexing.(v) The RNA chaperone activity of the N-terminal region of FinO facilitates the propagation of intermolecular base pairing spreading from the initial loop-loop interactions. (vi) Disengagement of the body of FinO as stem-loop structures are incorporated into the growing regions of intermolecular duplex.Final duplexing could be facilitated by the N-terminal regions of FinO that do not absolutely require the main body of FinO for chaperone function. (vii)Formation of the inhibitory FinP-traJ mRNA duplex. (For interpretation of the references to colour in this figure legend, the reader is referred to the webversion of this article.)

85J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

well as to reduce the possibility of pilus-dependent phageinfection. While FinO appears to only function to regulatethe expression of genes within its own plasmid, the recentfinding of structural and functional FinO orthologs in E. coliand N. meningitidis hint that this class of chaperones mayplay more diverse roles in gene regulation in prokaryotes.

Acknowledgments

This work was carried out with the support of a Grantfrom the Canadian Institutes of Health Research (CIHR).

References

Arthur, D.C., et al., 2011. Mapping interactions between the RNA chaperoneFinO and its RNA targets. Nucleic Acids Res. doi:10.1093/nar/gkr025.

Arthur, D., et al., 2003. FinO is an RNA chaperone that facilitates sense-antisense RNA interactions. EMBO J. 22, 6346–6355.

Arutyunov, D., Frost, L.S., 2013. F conjugation: back to the beginning.Plasmid 70, 18–32. doi:10.1016/j.plasmid.2013.03.010.

Chaulk, S., et al., 2010. N. meningitidis 1681 is a member of the FinO familyof RNA chaperones. RNA Biol. 7, 112–119, doi: 13688 [pii].

Chaulk, S.G., et al., 2011. ProQ is an RNA chaperone that controls ProP levelsin Escherichia coli. Biochemistry 50, 3095–3106. doi:10.1021/bi101683a.

Cheah, K.C., Skurray, R., 1986. The F plasmid carries an IS3 insertion withinfinO. J. Gen. Microbiol. 132, 3269–3275.

Cullum, J., Collins, J.F., Broda, P., 1978. The spread of plasmids in modelpopulations of Escherichia coli K12. Plasmid 1, 545–556.

Dempsey, W.B., 1989. Derepression of conjugal transfer of the antibioticresistance plasmid R100 by antisense RNA. J. Bacteriol. 171, 2886–2888.

Dong, G., Chakshusmathi, G., Wolin, S.L., Reinisch, K.M., 2004. Structureof the La motif: a winged helix domain mediates RNA binding via aconserved aromatic patch. EMBO J. 23, 1000–1007. doi:10.1038/sj.emboj.7600115.

Egawa, R., Hirota, Y., 1962. Inhibition of fertility by multiple drug-resistancefactors in Escherichia coli K12, Japan. J. Genet. 37, 66–69.

Fantappie, L., et al., 2009. The RNA chaperone Hfq is involved in stressresponse and virulence in Neisseria meningitidis and is a pleiotropicregulator of protein expression. Infect. Immun. 77, 1842–1853.doi:10.1128/IAI.01216-08.

Finlay, B.B., Frost, L.S., Paranchych, W., Willetts, N.S., 1986. Nucleotidesequences of five IncF plasmid finP alleles. J. Bacteriol. 167, 754–757.

Finnegan, D.J., Willetts, N.S., 1971. Two classes of Flac mutants insensitiveto transfer inhibition by an F-like R factor. Mol. Gen. Genet. 111,256–264.

Finnegan, D., Willetts, N., 1973. The site of action of the F transfer inhibitor.Mol. Gen. Genet. 127, 307–316.

Fowler, T., Thompson, R., 1986. Shadow promoters in the F plasmid transferoperon. Mol. Gen. Genet. 202, 509–511.

Frost, L.S., Koraimann, G., 2010. Regulation of bacterial conjugation:balancing opportunity with adversity. Future Microbiol. 5, 1057–1071.doi:10.2217/fmb.10.70.

Frost, L., Lee, S., Yanchar, N., Paranchych, W., 1989. FinP and fisO mutationsin FinP anti-sense RNA suggest a model for FinOP action in therepression of bacterial conjugation by the Flac plasmid JCFL0. Mol. Gen.Genet. 218, 152–160.

Frost, L.S., Ippen-Ihler, K., Skurray, R.A., 1994. Analysis of the sequence andgene products of the transfer region of the F sex factor. Microbiol. Rev.58, 162–210.

Ghetu, A.F., et al., 1999. The FinO repressor of bacterial conjugation containstwo RNA binding regions. Biochemistry 38, 14036–14044.

Ghetu, A., Gubbins, M., Frost, L., Glover, J., 2000. Crystal structure of thebacterial conjugation repressor finO. Nat. Struct. Biol. 7, 565–569.

Ghetu, A., Arthur, D., Kerppola, T., Glover, J., 2002. Probing FinO-FinP RNAinteractions by site-directed protein-RNA crosslinking and gelFRET. RNA8, 816–823.

Grindley, N.D., Grindley, J.N., Smith, H.R., Anderson, E.S., 1973.Characterisation of derepressed mutants of an F-like R factor. Mol. Gen.Genet. 120, 27–34.

Gubbins, M., Arthur, D., Ghetu, A., Glover, J., Frost, L., 2003. Characterizingthe structural features of RNA/RNA interactions of the F-plasmid FinOPfertility inhibition system. J. Biol. Chem. 278, 27663–27671.

Gubbins, M.J., Will, W.R., Frost, L.S., 2005. Mullany, P. (Ed.), The DynamicBacterial Genome. Cambridge University Press, pp. 151–206.

Ham, L.M., Cram, D., Skurray, R., 1989. Transcriptional analysis of the Fplasmid surface exclusion region: mapping of traS, traT, and traDtranscripts. Plasmid 21, 1–8.

Herschlag, D., 1995. RNA chaperones and the RNA folding problem. J. Biol.Chem. 270, 20871–20874.

Hey, A., Li, M.S., Hudson, M.J., Langford, P.R., Kroll, J.S., 2013. Transcriptionalprofiling of Neisseria meningitidis interacting with human epithelialcells in a long-term in vitro colonization model. Infect. Immun. 81,4149–4159. doi:10.1128/IAI.00397-13.

Jerome, L., Frost, L., 1999. In vitro analysis of the interaction between theFinO protein and FinP antisense RNA of F-like conjugative plasmids.J. Biol. Chem. 274, 10356–10362.

Jerome, L., van Biesen, T., Frost, L., 1999. Degradation of FinP antisense RNAfrom F-like plasmids: the RNA-binding protein, FinO, protects FinP fromribonuclease E. J. Mol. Biol. 285, 1457–1473.

Jiang, M., et al., 2007. Identification of novel Escherichia coli ribosome-associated proteins using isobaric tags and multidimensional proteinidentification techniques. J. Bacteriol. 189, 3434–3444. doi:10.1128/JB.00090-07.

Koraimann, G., Hogenauer, G., 1989. A stable core region of the tra operonmRNA of plasmid R1–19. Nucleic Acids Res. 17, 1283–1298.

Koraimann, G., Koraimann, C., Koronakis, V., Schlager, S., Hogenauer, G.,1991. Repression and derepression of conjugation of plasmid R1 bywild-type and mutated finP antisense RNA. Mol. Microbiol. 5, 77–87.

Koraimann, G., Teferle, K., Markolin, G., Woger, W., Hogenauer, G., 1996.The FinOP repressor system of plasmid R1: analysis of the antisenseRNA control of traJ expression and conjugative DNA transfer. Mol.Microbiol. 21, 811–821.

Lee, S.H., Frost, L.S., Paranchych, W., 1992. FinOP repression of the F plasmidinvolves extension of the half-life of FinP antisense RNA by FinO. Mol.Gen. Genet. 235, 131–139.

Meynell, E., Datta, N., 1967. Mutant drug resistant factors of hightransmissibility. Nature 214, 885–887.

Meynell, E., Meynell, G.G., Datta, N., 1968. Phylogenetic relationships ofdrug-resistance factors and other transmissible bacterial plasmids.Bacteriol. Rev. 32, 55–83.

Mullineaux, P., Willetts, N., 1985. Promoters in the transfer region ofplasmid F. Basic Life Sci. 30, 605–614.

Nuk, M.R., Reisner, A., Zechner, E.L., 2011. The transfer operon of plasmidR1 extends beyond finO into the downstream replication genes. Plasmid65, 150–158. doi:10.1016/j.plasmid.2010.12.003.

Rambo, R.P., Tainer, J.A., 2013. Super-resolution in solution X-ray scatteringand its applications to structural systems biology. Ann. Rev. Biophys.42, 415–441. doi:10.1146/annurev-biophys-083012-130301.

Sauer, E., 2013. Structure and RNA-binding properties of the bacterial LSmprotein Hfq. RNA Biol. 10, 610–618. doi:10.4161/rna.24201.

Sheidy, D.T., Zielke, R.A., 2013. Analysis and expansion of the role of theEscherichia coli protein ProQ. PLoS One 8, e79656. doi:10.1371/journal.pone.0079656.

Stocker, B.A.D., Smith, S.M., Ozeki, H., 1963. High infectivity of Salmonellatyphimurium newly infected by the coli factor. J. Gen. Microbiol. 30,201–221.

Timmis, K.N., Andres, I., Achtman, M., 1978. Fertility repression of F-likeconjugative plasmids: physical mapping of the R6–5 finO and finPcistrons and identification of the finO protein. Proc. Natl. Acad. Sci. U.S.A.75, 5836–5840.

Tompa, P., Csermely, P., 2004. The role of structural disorder in the functionof RNA and protein chaperones. FASEB J. 18, 1169–1175. doi:10.1096/fj.04-1584rev.

van Biesen, T., Frost, L.S., 1992. Differential levels of fertility inhibitionamong F-like plasmids are related to the cellular concentration of finOmRNA. Mol. Microbiol. 6, 771–780.

van Biesen, T., Frost, L.S., 1994. The FinO protein of IncF plasmids bindsFinP antisense RNA and its target, traJ mRNA, and promotes duplexformation. Mol. Microbiol. 14, 427–436.

van Biesen, T., Soderbom, F., Wagner, E.G., Frost, L.S., 1993. Structural andfunctional analyses of the FinP antisense RNA regulatory system of theF conjugative plasmid. Mol. Microbiol. 10, 35–43.

Wagner, E.G., 2013. Cycling of RNAs on Hfq. RNA Biol. 10, 619–626.doi:10.4161/rna.24044.

Wagner, M.A., Bischof, K., Kati, D., Koraimann, G., 2013. Silencing andactivating type IV secretion genes of the F-like conjugative resistanceplasmid R1. Microbiology 159, 2481–2491. doi:10.1099/mic.0.071738-0.

Watanabe, T., Fukasawa, T., 1962. Episome-mediated transfer of drugresistance in Enterobacteriaceae IV. Interactions betweenresistancetransfer factor and F-factor in Escherichia coli K-12. J.Bacteriol. 83, 727–735.

86 J.N. Mark Glover et al./Plasmid 78 (2015) 79–87

Weichenrieder, O., 2014. RNA binding by Hfq and ring-forming (L)Smproteins: A trade-off between optimal sequence readout and RNAbackbone conformation. RNA Biol. 11.

Willetts, N.S., 1974. The kinetics of inhibition of Flac transfer by R100 inE. coli. Mol. Gen. Genet. 129, 123–130.

Willetts, N., Maule, J., 1986. Specificities of IncF plasmid conjugation genes.Genet. Res. 47, 1–11.

Wilusz, C.J., Wilusz, J., 2013. Lsm proteins and Hfq: Life at the 3′ end. RNABiol. 10, 592–601. doi:10.4161/rna.23695.

Yoshioka, Y., Ohtsubo, H., Ohtsubo, E., 1987. Repressor gene finO in plasmidsR100 and F: constitutive transfer of plasmid F is caused by insertionof IS3 into F finO. J. Bacteriol. 169, 619–623.

87J.N. Mark Glover et al./Plasmid 78 (2015) 79–87