Ir

PK

a

ARRAA

KCCP

1

w(mrtwciaaCca

sia

I

h0

International Journal of Biological Macromolecules 106 (2018) 763–767

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

j ourna l h o mepa ge: www.elsev ier .com/ locate / i jb iomac

dentification of functional interactome of a key cell divisionegulatory protein CedA of E.coli

ankaj Sharma, Anil Kumar Tomar, Bishwajit Kundu ∗

usuma School of Biological Sciences, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

r t i c l e i n f o

rticle history:eceived 21 July 2017eceived in revised form 9 August 2017ccepted 10 August 2017vailable online 14 August 2017

eywords:edA-binding proteinshromosomal DNA over-replication

a b s t r a c t

Cell division is compromised in DnaAcos mutant Escherichia coli cells that results in filamentous cellmorphology. This is countered by over-expression of CedA protein that induces cytokinesis and thus,regular cell morphology is regained; however via an unknown mechanism. To understand the processsystematically, exact role of CedA should be deciphered. Protein interactions are crucial for functionalorganization of a cell and their identification helps in revealing exact function(s) of a protein and its bind-ing partners. Thus, this study was intended to identify CedA binding proteins (CBPs) to gain more cluesof CedA function. We isolated CBPs by pull down assay using purified recombinant CedA and identifiednine CBPs by mass spectrometric analysis (MALDI-TOF MS and LC–MS/MS), viz. PDHA1, RL2, DNAK, LPP,

ull down assay RPOB, G6PD, GLMS, RL3 and YBCJ. Based on CBPs identified, we hypothesize that CedA plays a crucial andmultifaceted role in cell cycle regulation and specific pathways in which CedA participates may includetranscription and energy metabolism. However, further validation through in-vitro and in-vivo experi-ments is necessary. In conclusion, identification of CBPs may help us in deciphering mechanism of CedAmediated cell division during chromosomal DNA over-replication.

© 2017 Elsevier B.V. All rights reserved.

. Introduction

Cell division and chromosomal DNA replication in bacteria occurith high accuracy, co-ordination and regulation. In Escherichia coli

E. coli), DnaA protein initiates and regulates the process of chro-osomal DNA replication [1,2]. Basically, it binds to the origin of

eplication (oriC) and initiates formation of the DNA replication ini-iation complex. Previously, several studies have been performedith DnaA mutants in an attempt to decipher the mechanism(s) of

hromosome replication and cell division [3–5]. One such mutants DnaAcos, a cold sensitive DnaA mutant that carries four aminocid substitutions. It induces excessive initiation from E. coli oriCt 30◦C and thus, causes over-replication of chromosomal DNA.onsequently, the process of cell division is compromised in theseells which undergo excessive karyokinesis without cytokinesisnd attain a filamentous morphology [1,6,7].

A multi-copy suppressor gene cedA, that encodes a cell divi-

ion activator protein (CedA), is expressed in dnaAcos mutants andnterestingly, its expression controls the cell division. It is an 80mino acids long DNA-binding protein. Katayama et al. reported

∗ Corresponding author at: Room No. 204, Kusuma School of Biological Sciences,ndian Institute of Technology Delhi (IIT Delhi), Hauz Khas, New Delhi, 110016, India.

E-mail address: bkundu@bioschool.iitd.ac.in (B. Kundu).

ttp://dx.doi.org/10.1016/j.ijbiomac.2017.08.073141-8130/© 2017 Elsevier B.V. All rights reserved.

that CedA over-expression in DnaAcos mutants started septation,cell division and regular colony formation and thus, inhibited theformation of filamentous morphology at 30 ◦C; however, chro-mosomal DNA over-replication continued [1]. It is evident fromtheir study that though CedA is important for the initiation ofcell division, it does not inhibit dnaAcos to bind at oriC to initi-ate chromosomal replication. CedA was also identified as one of thecomponents of RNA polymerase complex in E. coli [8]. So, it becomesinteresting to decipher how CedA in case of DnaAcos mutants main-tains a balance between two entirely integrated events, cell divisionand chromosomal replication.

Protein–protein interactions are crucial for understandingstructural and functional organization of a cell. Their identifi-cation helps us in decoding exact physiological function(s) of aspecific protein and its binding partners. In addition, interactionnetwork analysis plays a decisive role in exposing underlying mech-anism(s) of various related biological processes. Thus, this studywas designed to identify functional partners of CedA as a first stepto reveal its putative role in cell division associated events. Here,we have cloned, expressed and purified E. coli CedA and then, iden-tified its binding proteins by pull down assay followed by mass

spectrometric analysis.

764 P. Sharma et al. / International Journal of Biolog

Fig. 1. Experimental set up for isolation and identification of CedA-binding proteins(

2

2

N

gFiwkwacfbsHNfw(

2

SMubgdasi

were identified by LC–MS/MS including the proteins identified

CBPs).

. Materials and methods

The experimental set up is shown in Fig. 1.

.1. Cloning, expression and purification of CedA

The cedA gene (NCBI Ref. Seq. NC 000913.3) was cloned withdeI and BamHI restriction sites using the following primer set:

Forward primer: 5′CTTATCATATGCGTTTAGTGAAGCC 3′

Reverse primer: 5′ATACTGGATCCTTACTCAGTCACTTCC3′

After PCR amplification and digestion with restriction enzymes,ene was ligated into pET28a vector using T4 DNA ligase (Thermoisher Scientific, United States). CedA clone was then transformednto BL21-Gold (DE3) E. coli expression host. The transformed cells

ere cultured in LB media (HiMedia, India) containing 50 �g/mlanamycin and incubated at 37 ◦C with continuous shaking. Cellsere induced with 1 mM IPTG (Sigma-Aldrich, United States) when

bsorbance at 595 nm was about 0.6. After four hours of induction,ells were harvested by centrifugation and cell lysis was per-ormed by sonication. CedA expression in cell lysate was checkedy 15% sodium dodecyl sulfate-polyacrylamide gel electrophore-is (SDS-PAGE) and it was further processed for its purification.is6-tagged CedA was isoalted by affinity chromatography usingi-NTA-Agarose beads under native conditions following manu-

acturer’s recommendations (Qiagen, Germany) and purified CedAas obtained after buffer exchange using PD-10 desalting column

GE Healthcare, Sweden).

.2. Electrophoresis and mass spectrometry

Purified CedA collected after buffer exchange was run on 15%DS-PAGE [9]. Its intact mass was determined by UltrafleXtremeALDI-TOF/TOF Mass Spectrometer (Bruker Daltonik, Germany)

sing sinapinic acid matrix. Further, its identity was confirmedy MALDI-TOF MS analysis as described earlier [10]. Briefly, sin-le band observed on gel was excised manually and in-gel trypsinigested. The digested peptides were mixed with matrix solutionnd spotted on to MALDI target plate. Generated mass spectra were

earched against Swissprot databases for peptide mass fingerprint-ng using Mascot search engine (Matrix Sciences, UK).

ical Macromolecules 106 (2018) 763–767

2.3. CedA-binding proteins

2.3.1. Isolation by pull down assayE. coli CedA-binding proteins (CBPs) were isolated by pull down

assay using CedA as a bait. His6-tagged CedA was mixed with Ni-NTA-Agarose beads and incubated at 4 ◦C for 2hr to allow it to bindwith the beads. These beads were then transferred into a columnand washed twice with assay buffer (100 mM sodium phosphate,50 mM sodium chloride and 10 mM imidazole buffer, pH 8.0). E.colicell lysate diluted with assay buffer was loaded onto the column,followed by washing with assay buffer containing 50 mM imida-zole. Finally, CBPs were eluted with assay buffer containing 250 mMimidazole.

2.3.2. Identification by mass spectrometryMass spectrometric identification of CBPs was done by two

methods, MALDI-TOF MS and LC–MS/MS. In first set up, CBPs iso-lated in previous step were separated on 15% SDS-PAGE, proteinbands were manually excised, trypsin digested and identified byMALDI-TOF MS analysis. While in second set up, CBPs were identi-fied by LC–MS/MS. Protein bands on gel were excised, cut into smallpieces, pooled and trypsin digested. Extraction buffer (1:2 (v/v) 5%formic acid/acetonitrile) was added to digested peptides, vortexedand incubated for 30 min at 37 ◦C. Extracted peptide samples weredesalted using C18 tips (Pierce, Thermo Fisher Scientific, UnitedStates) according to manufacturer’s protocol. LC–MS/MS analysiswas performed on Easy-nLC II (Thermo Scientific, United States)connected to an ESI mass spectrometer (amaZon SL, Bruker Dalton-ics). Released peptides were separated by a reversed-phase column(10 cm × 75 �m, 3 �m EASY-Column) using acetonitrile gradientcontaining 0.1% formic acid in the mobile phase at a flow rateof 0.3 �l/min for 55 min. Peptides generated were analyzed usingCompass HyStar 3.2, Data Analysis 4.1 and BioTools 3.2 software.For identification, MS/MS ion search was performed using Mas-cot search engine (parameters: monoisotopic mass values, masstolerance ±0.5 Da and maximum missed cleavage = 1).

3. Results

3.1. Purification and identification of CedA

CedA was successfully cloned, expressed and purified asdescribed in methodology section. Its clone was confirmed bynucleotide sequencing. The purification, at each step, was mon-itored by electrophoresis (Fig. 2). Highly purified CedA wasobtained through Ni-NTA-Agarose affinity chromatography andPD-10 buffer exchange, as evident from a single protein bandobserved on SDS-PAGE gel (Fig. 2D). The intact mass of purifiedCedA was determined 12.275 kDa by MALDI-TOF (Fig. 3). Further,its identity was confirmed by mass spectrometric analysis. Peptidespectra searched against Swiss-Prot matched to CedA from E.coliwith a significant mascot score, 69 (Fig. 4).

3.2. CedA-binding proteins

CBPs isolated by pull down assay using His6-tagged CedAas bait were separated by SDS-PAGE (Fig. 5) and identified byMALDI-TOF MS and LC–MS/MS. CedA and its three functionalpartners, viz. pyruvate dehydrogenase E1 component (PDHA1,band 2 on gel, Fig. 5), chaperone protein DnaK (DNAK, band 3)and 50S ribosomal protein L2 (RL2, band 6) were identified byMALDI-TOF MS (Table 1). On the other hand, a total of nine CBPs

by MALDI-TOF MS. Other proteins identified were outer mem-brane lipoprotein Lpp (LPP), DNA directed RNA polymerase subunitbeta (RPOB), glucose-6-phosphate 1-dehydrogenase (G6PD), 50S

P. Sharma et al. / International Journal of Biologi

Fig. 2. Cloning and purification of CedA. (A) CedA clone confirmation after colonyPCR, Lanes- 1: DNA ladder, 2–7: colony PCR products; (B) CedA expression in celllysate, Lanes- 1: Protein molecular weight markers, 3: supernatant; 5: Pellet; (C)Ni-NTA affinity chromatography, Lanes- 1: Protein molecular weight markers, 2:UcP

ra(

4

hSC

nbound fraction, 3: Washing, 4–8: Elution fractions; (D) Purified CedA fractionsollected after PD-10 desalting, Lanes- 1: Protein molecular weight markers, 2–4:urified CedA.

ibosomal protein L3 (RL3), glucoseamine-fructose-6-phosphatemino-transferase (GLMS) and an uncharacterized protein ybcJYBCJ) (Table 1).

. Discussion

CedA is a ∼10 kDa DNA-binding E. coli protein with reportedomologues in other Gram-negative enterobacteria includinghigella flexneri and Salmonella typhimurium. The NMR determinededA C-terminal domain structure is quite similar to that of other

Fig. 3. Intact mass of purified CedA (∼12

cal Macromolecules 106 (2018) 763–767 765

known DNA-binding proteins, such as N-terminal DNA-bindingdomains of �-integrase and Tn916 integrase proteins and it bindsto dsDNA through a �-sheet structure [7]. Previous studies havesuggested a putative role of CedA in the regulation of cell divi-sion; however, the mechanism via which it regulates the processis still unknown [1,6,7]. It was reported that CedA activated celldivision inhibited by chromosomal DNA over-replication in DnaA-cos cells [1]. Lee et al. identified RNA polymerase complex in E. coliusing affinity isolation and mass spectrometric analysis and CedAwas identified as one of the components of the complex alongwith several other proteins, such as RpoD, DnaK, NusA, RpoN, DnaJ,RpoS, RpoE, NusG, etc. [8]. Recently, Abe et al. have identified CedAdomains/residues crucial for its binding with DNA and RNA poly-merase [6]. These studies provide evidences supporting crucial, butunknown, role of CedA in cell cycle regulation. It is well knownthat protein interactions play crucial role in cell organization andit is essential to identify specific functional partners of a protein inorder to fully understand its role in a biological process. Thus, weperformed this study and identified nine CBPs in E. coli by affinityseparation followed by MS analysis, including PDHA1, RL2, DNAK,LPP, RPOB, G6PD, GLMS, RL3 and YBCJ (Table 1).

Binding of CedA with a diverse set of proteins in E. coli sug-gests that it has a multifaceted role. Manual clustering of CBPsinto functional categories reveals that seven of them are associ-ated with energy metabolic pathways (PDHA1, G6PD and GLMS)and transcription (RPOB, RL2, RL3 and YBCJ). Pyruvate dehydroge-nase complex converts pyruvate into acetyl-CoA and CO2 and linksglycolysis to TCA cycle. This nuclear-encoded multi-enzyme com-plex consists of multiple copies of three enzyme components (E1,E2 and E3). The E1 component (PDHA1), a hetero-tetramer hav-ing two � and two � subunits, plays the key role in the conversionof pyruvate [11,12]. G6PD is an important rate limiting enzyme ofpentose phosphate pathway, which is involved in the first step of d-ribulose 5-phosphate synthesis from d-glucose 6-phosphate whereit catalyzes the oxidation of glucose 6-phosphate. In addition, italso protects E. coli against the oxidative stress [13,14]. GLMS,also known as glutamine: fructose-6-phosphate aminotransferase,is the first enzyme of hexosamine biosynthesis pathway. It con-trols the flux of glucose and converts fructose 6-phosphate intoglucosamine 6-phosphate [15,16]. RL2, RL3 and YBCJ are RNA bind-ing proteins. RL2 and RL3 are known to participate in assembly

of bacterial ribosomal subunits, while no conclusive informationis available for YBCJ function [17–19]. RPOB is DNA dependentRNA polymerase. It is basically one of the five units that form

kDa) as determined by MALDI-TOF.

766 P. Sharma et al. / International Journal of Biological Macromolecules 106 (2018) 763–767

Fig. 4. Mass spectra of trypsin digested CedA obtained by MALDI-TOF MS analysis. (Inset) Confirmation of purified CedA by peptide mass fingerprinting followed by databasesearch.

Table 1List of CedA-binding proteins identified by mass spectrometry.

S. No. Protein Match Mascot Score Protein Name Molecular weight (kDa)

MALDI-TOF MS1 ODP1 ECO57 113 Pyruvate Dehydrogenase E1 component 99.662 DNAK ECOHS 61 Chaperone protein DnaK 69.133 RL2 ECO24 85 50S ribosomal protein L2 29.864 CEDA ECOBW 156 Cell division activator CedA 9.37

LC–MS/MS1 CEDA ECOBW 143 Cell division activator CedA 9.372 ODP1 ECO57 113 Pyruvate Dehydrogenase E1 component 99.663 LPP ECOLI 80 Major outer memberan lipoprotein Lpp 8.324 RPOB ECOBW 73 DNA directed RNA polymerase subunit beta 150.635 DNAK ECOHS 67 Chaperone protein DnaK 69.136 G6PD ECO57 47 Glucose-6-phosphate 1-dehydrogenase 55.737 RL2 ECO24 45 50S ribosomal protein L2 29.86

bosomseamiracter

RtimiasmatadaotfcoDbp

8 RL3 ECO24 40 50S ri9 GLMS ECOLI 31 Gluco10 YBCJ ECOLI 30 Uncha

NA polymerase enzyme, which regulates the bacterial transcrip-ional process [20,21]. CedA interaction with these CBPs hints aboutts distinctive role(s) in cell cycle regulation as well as in energy

etabolism, though unknown to date. DNAK and LPP were othermportant CBPs identified and we consider that their binding isssociated with the septum formation and cell division in E. colipecifically during chromosomal DNA over-replication. DNAK is aultifunctional chaperone of highly conserved HSP70 family which

ssists in protein folding, disaggregation and remodeling of pro-ein complexes. DNAK interacts with a large number of proteins tossist them and mutations induced in DNAK cause several cellularefects in E. coli [22]. Here, we guess that DNAK binds with CedA andssists in the regulation of cell division during chromosomal DNAver-replication. However, it’s a mere assumption and experimen-al evidences are required to confirm it. LPP is a protein requiredor maintaining structural and functional integrity of bacterialell envelope [23,24]. As mentioned earlier that over-expressionf CedA in E. coli ceases the filamentous structure formation in

naAcos cells, its interaction with LPP is not surprising. It is quiteelievable that CedA interacts with LPP to regain regular cell mor-hology during chromosomal DNA over-replication.

al protein L3 22.24ne-fructose-6-phosphate aminotransferase 66.89ised Protein ybcJ 7.39

A restricted search to experimentally validated interactions overstring database version 10.5 [25] retrieved only two functional part-ners of CedA- RNA polymerase subunits rpoC and rpoZ (Fig. 6A).Next, we performed network analysis of CedA and CBPs- includingthose identified in our study as well as rpoC and rpoZ (Fig. 6B). Aweak network was observed as only 6 out of 12 proteins given asinput were connected through the constructed network. This wasexpected as very limited experimental evidences are available forCedA and its interactome. The major nodes were RPOB, rpoC, rpoZand DNAK and each of them had four connections. Our findings willadd new nodes to the network and hence, a strong network can beexpected that, in turn, will be crucial for better understanding ofCedA associated processes.

5. Conclusion

Here, we have expressed and purified E.coli CedA and then, usedit as a bait to isolate and identify its functional partners. Overall

nine CBPs were identified. As inferred from this interactome, weconsider that CedA is primarily associated with regulation of cellcycle pathways, specifically transcription and energy metabolism.However, these probable roles require further confirmation and

P. Sharma et al. / International Journal of Biologi

Fig. 5. 15% SDS-PAGE profile of CedA binding proteins, Lanes- L1: Protein molecularweight markers; L2: Eluted CedA binding proteins (1–7 protein bands); L3: PurifiedHis6-tagged CedA.

Ff(

vdor

A

SeIf

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

Chem. 280 (41) (2005) 34409–34419.[25] D. Szklarczyk, J.H. Morris, H. Cook, M. Kuhn, S. Wyder, M. Simonovic, A.

ig. 6. String database based interaction analysis of CedA. (A) Experimentally knownunctional partners of CedA; (B) Network of CedA and CBPs- including rpoC and rpoZ.Note: rif → RPOB; Non-connected nodes were removed).

alidation through in-vitro and in-vivo experiments. Once vali-ated, these findings would be crucial to decipher exact mechanismf CedA mediated cell division during chromosomal DNA over-eplication.

cknowledgements

This study was supported by Research Grant received fromcience and Engineering Research Board (SERB), Department of Sci-nce & Technology, Govt. of India, New Delhi. PS and AKT also thankndian Institute of Technology Delhi and SERB respectively for theirellowships.

eferences

[1] T. Katayama, M. Takata, K. Sekimizu, CedA is a novel Escherichia coli proteinthat activates the cell division inhibited by chromosomal DNAover-replication, Mol. Microbiol. 26 (4) (1997) 687–697.

cal Macromolecules 106 (2018) 763–767 767

[2] K. Schenk, A.B. Hervas, T.C. Rosch, M. Eisemann, B.A. Schmitt, S. Dahlke, L.Kleine-Borgmann, S.M. Murray, P.L. Graumann, Rapid turnover of DnaA atreplication origin regions contributes to initiation control of DNA replication,PLoS Genet. 13 (2) (2017) e1006561.

[3] K. Fujimitsu, M. Su’etsugu, Y. Yamaguchi, K. Mazda, N. Fu, H. Kawakami, T.Katayama, Modes of overinitiation, dnaA gene expression, and inhibition ofcell division in a novel cold-sensitive hda mutant of Escherichia coli, J.Bacteriol. 190 (15) (2008) 5368–5381.

[4] H. Kawakami, T. Iwura, M. Takata, K. Sekimizu, S. Hiraga, T. Katayama, Arrestof cell division and nucleoid partition by genetic alterations in the slidingclamp of the replicase and in DnaA, Mol. Genet. Genomics: MGG 266 (2)(2001) 167–179.

[5] S. Nishida, K. Fujimitsu, K. Sekimizu, T. Ohmura, T. Ueda, T. Katayama, Anucleotide switch in the Escherichia coli DnaA protein initiates chromosomalreplication: evidnece from a mutant DnaA protein defective in regulatory ATPhydrolysis in vitro and in vivo, J. Biol. Chem. 277 (17) (2002) 14986–14995.

[6] Y. Abe, N. Fujisaki, T. Miyoshi, N. Watanabe, T. Katayama, T. Ueda, Functionalanalysis of CedA based on its structure: residues important in binding of DNAand RNA polymerase and in the cell division regulation, J. Biochem. (Tokyo)159 (2) (2016) 217–223.

[7] H.A. Chen, P. Simpson, T. Huyton, D. Roper, S. Matthews, Solution structureand interactions of the Escherichia coli cell division activator protein CedA,Biochemistry 44 (18) (2005) 6738–6744.

[8] D.J. Lee, S.J. Busby, L.F. Westblade, B.T. Chait, Affinity isolation and I-DIRT massspectrometric analysis of the Escherichia coli O157:H7 Sakai RNA polymerasecomplex, J. Bacteriol. 190 (4) (2008) 1284–1289.

[9] U.K. Laemmli, Cleavage of structural proteins during the assembly of the headof bacteriophage T4, Nature 227 (5259) (1970) 680–685.

10] A.K. Tomar, B.S. Sooch, I. Raj, S. Singh, S. Yadav, Interaction analysis identifiessemenogelin I fragments as new binding partners of PIP in human seminalplasma, Int. J. Biol. Macromol. 52 (2013) 296–299.

11] S. Kale, P. Arjunan, W. Furey, F. Jordan, A dynamic loop at the active center ofthe Escherichia coli pyruvate dehydrogenase complex E1 componentmodulates substrate utilization and chemical communication with the E2component, J. Biol. Chem. 282 (38) (2007) 28106–28116.

12] P.E. Stephens, M.G. Darlison, H.M. Lewis, J.R. Guest, The pyruvatedehydrogenase complex of Escherichia coli K12. Nucleotide sequenceencoding the dihydrolipoamide acetyltransferase component, Eur. J. Biochem.133 (3) (1983) 481–489.

13] J.M. Sandoval, F.A. Arenas, C.C. Vasquez, Glucose-6-phosphate dehydrogenaseprotects Escherichia coli from tellurite-mediated oxidative stress, PLoS One 6(9) (2011) e25573.

14] J. Zhao, T. Baba, H. Mori, K. Shimizu, Effect of zwf gene knockout on themetabolism of Escherichia coli grown on glucose or acetate, Metab. Eng. 6 (2)(2004) 164–174.

15] S.L. Bearne, C. Blouin, Inhibition of Escherichia coli glucosamine-6-phosphatesynthase by reactive intermediate analogues. The role of the 2-aminofunction in catalysis, J. Biol. Chem. 275 (1) (2000) 135–140.

16] T. Jaeger, C. Mayer, N-Acetylmuramic acid 6-phosphate lyases (MurNAcetherases): role in cell wall metabolism, distribution, structure, andmechanism, Cell. Mol. Life Sci.: CMLS 65 (6) (2008) 928–939.

17] P. Hu, S.C. Janga, M. Babu, J.J. Diaz-Mejia, G. Butland, W. Yang, O. Pogoutse, X.Guo, S. Phanse, P. Wong, S. Chandran, C. Christopoulos, A. Nazarians-Armavil,N.K. Nasseri, G. Musso, M. Ali, N. Nazemof, V. Eroukova, A. Golshani, A.Paccanaro, J.F. Greenblatt, G. Moreno-Hagelsieb, A. Emili, Global functionalatlas of Escherichia coli encompassing previously uncharacterized proteins,PLoS Biol. 7 (4) (2009) e96.

18] M. Jiang, S.M. Sullivan, A.K. Walker, J.R. Strahler, P.C. Andrews, J.R. Maddock,Identification of novel Escherichia coli ribosome-associated proteins usingisobaric tags and multidimensional protein identification techniques, J.Bacteriol. 189 (9) (2007) 3434–3444.

19] M. Shasmal, S. Dey, T.R. Shaikh, S. Bhakta, J. Sengupta, E. coli metabolic proteinaldehyde-alcohol dehydrogenase-E binds to the ribosome: a uniquemoonlighting action revealed, Sci. Rep. 6 (2016) 19936.

20] R. Glyde, F. Ye, V.C. Darbari, N. Zhang, M. Buck, X. Zhang, Structures of RNApolymerase closed and intermediate complexes reveal mechanisms of DNAopening and transcription initiation, Mol. Cell. 67 (1) (2017), 106.e4–116.e4.

21] K.S. Murakami, S. Masuda, E.A. Campbell, O. Muzzin, S.A. Darst, Structuralbasis of transcription initiation: an RNA polymerase holoenzyme-DNAcomplex, Science 296 (5571) (2002) 1285–1290.

22] F. Angles, M.P. Castanie-Cornet, N. Slama, M. Dinclaux, A.M. Cirinesi, J.C.Portais, F. Letisse, P. Genevaux, Multilevel interaction of theDnaK/DnaJ(HSP70/HSP40) stress-responsive chaperone machine with thecentral metabolism, Sci. Rep. 7 (2017) 41341.

23] W. Shu, J. Liu, H. Ji, M. Lu, Core structure of the outer membrane lipoproteinfrom Escherichia coli at 1.9 A resolution, J. Mol. Biol. 299 (4) (2000)1101–1112.

24] F. Stenberg, P. Chovanec, S.L. Maslen, C.V. Robinson, L.L. Ilag, G. von Heijne,D.O. Daley, Protein complexes of the Escherichia coli cell envelope, J. Biol.

Santos, N.T. Doncheva, A. Roth, P. Bork, L.J. Jensen, C. von Mering, The STRINGdatabase in 2017: quality-controlled protein–protein association networks,made broadly accessible, Nucl. Acids Res. 45 (D1) (2017) D362–D368.