Computational and Structural Biotechnology Journal 20 (2022) 757–765 journal homepage: www.elsevier.com/locate/csbj Structural and biochemical characterization of the relaxosome auxiliary proteins encoded on the Bacillus subtilis plasmid pLS20 Isidro Crespo a,1, Nerea Bernardo a,1, Anna Cuppari a, Barbara M. Calisto a, Jorge Val-Calvo b, Andrés Miguel-Arribas b, Wilfried J.J. Meijer b, Xavi Carpena a, Fernando Gil-Ortiz a, Marc Malfois a, D. Roeland Boer a,⇑ a ALBA Synchrotron Light Source, Carrer de la Llum 2-26, Cerdanyola del Vallès, 08290 Barcelona, Spain b Centro de Biología Molecular "Severo Ochoa" (CSIC-UAM), C. Nicolás Cabrera 1, Universidad Autónoma, Canto Blanco, 28049 Madrid, Spain a r t i c l e i n f o a b s t r a c t Article history: Bacterial conjugation is an important route for horizontal gene transfer. The initial step in this process Received 14 March 2021 involves a macromolecular protein-DNA complex called the relaxosome, which in plasmids consists of Received in revised form 22 December 2021 the origin of transfer (oriT) and several proteins that prepare the transfer. The relaxosome protein named Accepted 30 December 2021 relaxase introduces a nick in one of the strands of the oriT to initiate the process. Additional relaxosome Available online 05 January 2022 proteins can exist. Recently, several relaxosome proteins encoded on the Bacillus subtilis plasmid pLS20 were identified, including the relaxase, named RelpLS20, and two auxiliary DNA-binding factors, named Keywords: Aux1pLS20 and Aux2pLS20. Here, we extend this characterization in order to define their function. We present Structural biology Bacterial conjugation the low-resolution SAXS envelope of the Aux1pLS20 and the atomic X-ray structure of the C-terminal domain Relaxosome of Aux2pLS20. We also study the interactions between the auxiliary proteins and the full-length RelpLS20, as Auxiliary protein well as its separate domains. The results show that the quaternary structure of the auxiliary protein DNA binding protein Aux1pLS20 involves a tetramer, as previously determined. The crystal structure of the C-terminal domain Ribbon-Helix-Helix of Aux2pLS20 shows that it forms a tetramer and suggests that it is an analog of TraMpF of plasmid F. This Antibiotic resistance is the first evidence of the existence of a TraMpF analog in gram positive conjugative systems, although, Firmicutes unlike other TraMpF analogs, Aux2pLS20 does not interact with the relaxase. Aux1pLS20 interacts with the Horizontal gene transfer C-terminal domain, but not the N-terminal domain, of the relaxase RelpLS20. Thus, the pLS20 relaxosome exhibits some unique features despite the apparent similarity to some well-studied G- conjugation systems. Ó 2022 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://creative- commons.org/licenses/by-nc-nd/4.0/). 1. Introduction tion for the transfer of a conjugative element from a donor to a recipient cell [7–11]. The relaxosome is another important compo- Horizontal gene transfer (HGT) is the exchange of genes nent of T4SS-mediated conjugation and prepares the DNA for between organisms not related to transmission of genes between transfer [5,12]. Conjugation also plays an important role in bacte- parents and offspring and has important implications on evolution. rial virulence, as the T4SS system is also used for the extrusion of Conjugation is one of the routes that allow HGT and is common in virulence factor into the host environment (eg [13,14]). Bacterial the bacterial realm [1–6]. A membrane-embedded cellular machin- T4SS systems occur both in Gram positive (G+) and Gram negative ery named the Type 4 Secretion System (T4SS) is used in conjuga- (G-) bacteria and may be encoded in Mobile Genetic Elements (MGEs) such as plasmids [6]. Conjugation requires the initial preparation of the DNA that is Abbreviations: HGT, Horizontal Gene Transfer; MGE, Mobile Genetic Element; to be transferred [5,10,11,15]. This step involves an initial cut or T4SS, Type IV secretion system; EM, Electron Micoscopy; RHH, Ribbon-Helix-Helix; EDTA, Ethylenediaminetetraacetic acid; IPTG, Isopropyl b-D-1- nick at a specific site of one of the DNA strands, which then allows thiogalactopyranoside; PMSF, phenylmethylsulfonyl fluoride; SEC, Size Exclusion for the unwinding of this strand and subsequent transfer [5]. This Chromatography; SDS-PAGE, sodium dodecyl sulphate–polyacrylamide gel elec- critical step is performed by the relaxase [16–18], which have in trophoresis; SAXS, Small-angle X-ray scattering; oriT, Origin of Transfer; AUC, common that they consist of an N-terminal endonuclease domain, Analytical Ultracentrifugation; ITC, Isothermal titration calorimetry. ⇑ Corresponding author. followed by additional domains with a variety of functions [18– E-mail address: rboer@cells.es (D.R. Boer). 21]. The endonuclease domain of the relaxase remains covalently 1 These authors contributed equally to this work. attached to the ssDNA strand and this complex is transferred to https://doi.org/10.1016/j.csbj.2021.12.041 2001-0370/Ó 2022 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 the recipient cell. The process is strictly regulated at various levels, struct (Mw 48.75 kDa) and the C-terminal (residues 180–410, Mw which includes the implication of auxiliary proteins that form part 28.27 kDa) and N-terminal domain (residues 1–232, Mw of the relaxosome complex [5,22]. The auxiliary factors can be plas- 26.92 kDa) of RelpLS20. The pET28-based clone of Aux1pLS20 with C- mid encoded or provided by the donor. In most cases, the auxiliary terminal His tag was obtained as described in [26]. All the other proteins are either indispensable or enhance conjugation fre- cDNAs were chemically synthesized, adding an N-terminal HRV-3C quency. The functions of the auxiliary proteins include specific protease cleavage site before the methionine codon of each protein DNA binding on the relaxosome, connecting the relaxosome with construct. The cDNA synthesis and subcloning into pHTP1 vectors the T4SS channel, and interacting with the relaxase. were purchased from NZYTech (Campus do Lumiar, Lisboa, Portu- Recently, the relaxase of the Bacillus subtilis plasmid pLS20 was gal). Escherichia coli BL21 (DE3) were transformed with pHTP1 carry- identified [23], which was named RelpLS20. Sequence comparison ing the insert of the different constructs of aux1pLS20, aux2pLS20, showed that the N-terminal domain of RelpLS20 contains sequence relpLS20, relC-terpLS20 and relN-terpLS20 and were inoculated in fresh motifs corresponding to relaxase domains and was marked as the Luria-Bertani broth (LB) media complemented with 50 mg/ml kana- founding member of a new class of relaxases, named MOBL [23]. mycin at 37 °C overnight. Then, the cells from the overnight culture However, other authors suggest that this relaxase could also be were collected by centrifugation (4000  g for 10 min) and sus- classified as belonging to the MOBP subfamily, MOBMG [19,24,25]. pended in expression media (typically 1 L of Terrific Broth (TB) with In addition to the relaxase, two auxiliary proteins of plasmid 50 mg/ml kanamycin), at a ratio of 15 ml of preculture per liter of pLS20 have been identified, which were named Aux1pLS20 and medium. Cells in expression media were grown at 37 °C until an Aux2pLS20 [26]. They are encoded immediately upstream of RelpLS20 OD600 = 0.8–1 was reached. After that, protein expression was and were predicted to be Ribbon-Helix-Helix (RHH) proteins based induced overnight at 20 °C by addition of 1 mM isopropyl-b-D-1-thio on the occurrence of a b-strand followed by two a-helices at their N galactopyranoside (IPTG, Omnipur). After overnight induction, cells termini. RHH proteins are transcription factors that have a range of were centrifuged at 4000g for 30 min, and pellets were frozen regulatory functions in prokaryotes and bacteriophages and have an and stored at 80 °C. Before purification, the pellets were thawed interlaced, inseparable dimeric arrangement [27]. The well-studied and resuspended in lysis buffer at a ratio of 5 ml/g of cells. The lysis Salmonella phage protein Arc is the founding member of this family. buffer contained 0.5 M NaCl, 50 mM Tris pH 7.5, 5 mM imidazole, It was shown that the pLS20 auxiliary RHH proteins are essential 1 mM EDTA, 5% glycerol, and 0.2 mM of PMSF (phenylmethylsul- for pLS20 conjugation and that they bind to specific sites of the fonyl fluoride) as protease inhibitor. The cell suspension was lysed oriTpLS20. Aux1pLS20 binds to a 25 bp region, which contained an by sonication, adding DNase I to a final concentration of 200 mg/ml inverted repeat with sequence 50 -TGGTACCA-30 , which was pro- and lysozyme to a final concentration of 100 mg/ml during sonica- posed to be its binding site. Aux2pLS20 is able to bind to a much lar- tion. Insoluble matter was precipitated by centrifugation (18 ger region of the oriT spanning several hundreds of bases. To explain 000g, 30 min), and the supernatant was filtered through a this, it was proposed that Aux2pLS20 binds the TGTGCAT sequence 0.22 mm filter and applied to a nickel-charged His-TrapTM HP chelat- which was the only sequence to be present three times in the oriT ing column 5 ml (GE Healthcare Life Sciences). The column was [26]. The oligomerization state of these proteins was studied by an- washed with 10 column volumes (50 ml) of binding buffer alytical ultracentrifugation (AUC) and was proposed to be tetrameric (20 mM Tris pH 8.0, 500 mM NaCl, 5 mM Imidazole) to elute for Aux1pLS20 and hexameric for Aux2pLS20, although tetrameric unspecific bound proteins. Bound proteins were eluted using forms were compatible with the sedimentation profile as well sequential steps of 4, 10 and 25% of elution buffer (0.5 M NaCl, [26]. This study also used AUC to probe for interactions between 50 mM Tris pH 7.5, 0.5 M imidazole). Fractions were analyzed on the different relaxosome proteins, but none could be identified. SDS-PAGE and fractions containing pure protein were combined The exact function of the auxiliary proteins in the conjugative and concentrated over an Amicon ultra 10 kDa MWCO (Millipore). process of pLS20 is unknown. To gain knowledge on the function The buffer was exchanged using three HiTrap 5 ml Desalting col- of these proteins, we characterize Aux1pLS20 and Aux2pLS20 using umns (Cytiva) connected in series to 20 mM Tris pH 8.0, 500 mM various biophysical techniques and have investigated the structures NaCl. After elution, proteins were incubated overnight at 4 °C with of these proteins. We present the low-resolution, SAXS envelope of 1 mg of HRV-3C protease (in-house preparation). After incubation, the Aux1pLS20 and the crystal structure of the C-terminal domain of HRV-3C, tags and uncut protein were removed by application of Aux2pLS20 at atomic resolution. In addition, we also study the inter- the incubation solution through a His-TrapTM HP chelating column. actions between the auxiliary proteins and the relaxase RelpLS20. Our The flow-through was concentrated and further purified by Size results gave new insights into the relaxosome of pLS20, which has Exclusion Chromatography to remove aggregates and other contam- implications on similar G+ conjugative systems. Thus, we show that inants. The SEC column (Generon ProSEC 16/60 3–70 HR) was equi- the C-terminal domain of Aux2pLS20 forms a four-helix bundle and librated in 20 mM Tris pH 8.0, 500 mM NaCl. Fractions containing with structural homology to TraM of plasmid F, thereby providing the target protein were concentrated over an Amicon ultra 10 kDa the first evidence of the existence of a TraMpF analog in G+ conjuga- MWCO (Millipore). Typically, a yield of 20 mg of Aux1pLS20, tive systems. Unlike other TraMpF analogs, Aux2pLS20 does not inter- 10 mg for Aux2pLS20, 100 mg for full length RelpLS20, 20 mg for act with the relaxase. Surprisingly, Aux1pLS20 did show an RelN-terpLS20 and RelC-terpLS20 were obtained from 10 g of pellet. interaction with the C-terminal domain of RelpLS20. We discuss the Purity was assessed to be >95% by SDS–PAGE, followed by BlueSafe implications of the results for the mechanism of relaxosome prepa- staining (NZYTech, Campus do Lumiar, Lisboa, Portugal). The protein ration and presentation to the T4SS channel. concentration was determined by nanodrop, using the theoretical extinction coefficient to calculate the concentrations, immediately prior to usage where possible. Surplus aliquots were stored at 2. Methods 80 °C until usage. 2.1. Protein expression and purification 2.2. Analytical size exclusion chromatography on Aux1pLS20, Aux2pLS20 and RelpLS20 mixtures All protein constructs were expressed and purified using standard protocols. The cDNAs expressing the protein constructs were To determine the elution volumes of the separate proteins, Aux1pLS20 without tags (Mw 9.02 kDa), Aux1pLS20 with C-terminal 1.2 nmol of either full length RelpLS20, RelC-terpLS20 and RelN- His tag (Mw 10.6 kDa), Aux2pLS20 (Mw 17.03 kDa), full length con- terpLS20 were used for all the interaction assays. For complex- 758 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 binding stoichiometry assays, molar ratios of 1:1, 5:1 and 10:1 of 172.0  172.0 lm2). The distance sample/detector was were prepared for Aux1pLS20:RelpLS20, Aux1pLS20:RelN-terpLS20, 2556.00 mm. 40 images were collected for each concentration Aux1pLS20:RelC-terpLS20 and Aux2pLS20:Aux1pLS20. All samples were (1.0 mg/ml, 2.5 mg/ml, 5.0 mg/ml, 7.5 mg/ml and 10 mg/ml) with incubated for 30 min on ice before injection. Thus, the maximum an exposure time of 0.1 s. The q-axis calibration was obtained by amount of Aux1pLS20 and Aux2pLS20 injected was 12 nmol in the measuring silver behenate [29]. The program pyFAI [30] was used 1:10 M ratio. Samples of the proteins alone and in complex were to integrate the 2D data into 1D data. The 1D data has been aver- prepared in a final volume of 30 mL in 20 mM Tris pH 8, 300 mM aged, subtracted, normalized by the concentration, extrapolated NaCl and 25 mL were injected on a Superdex 200 increase 5/150 to zero concentration and merged with Primus [31] from the GL equilibrated with 20 mM Tris pH 8, 300 mM NaCl. The elution ATSAS package. The radius of gyration Rg and the maximum dis- was run at a flow rate of 0.3 ml/min and the absorption of the elute tance Dmax has been determined with GNOM [32]. The low- was monitored at 260 and 280 nm. To estimate the molecular resolution envelop has been restored with DAMMIF [33]. weights (MW) of the homo- and heterocomplexes, a calibration of the Superdex 200 increase 5/150 GL column was performed using 2.5. Aux2pLS20 crystallization and structure solution proteins with known MW (i.e. BLC (239.64 KDa), RCO (81.28 KDa), BSA (69.29 KDa), MpARF3 (49.52 KDa), MpARF2 (44.63 KDa), Purified Aux2pLS20 was concentrated to a final protein concen- AtARF5 (44.29 KDa), AtARF1 (41.13 KDa) and p69 (24.1 KDa), tration of 10 mg/ml in 20 mM Tris (pH 8.0), 300 mM NaCl. Crystal- which were eluted using the same elution buffer used above. The lization experiments of Aux2pLS20 were performed using the derived relation between the elution volume (Vel) and MW was sitting-drop vapor-diffusion method at 18 °C, by equilibration of Vel = –0.6815  log(Mw) + 5.1906, with an R2 = 0.933. drops of 1 lL protein + 1 lL crystallization buffer against 100 lL of the crystallization buffer. Crystals were harvested from two dif- 2.3. Isothermal titration calorimetry (ITC) ferent crystallization conditions. Cryo-cooling in liquid nitrogen was performed by soaking crystals on a cryo-protecting solution Calorimetric measurements were carried out using a VP-ITC consisting in reservoir buffer complemented with 10% glycerol, fol- instrument from MicroCal Inc. (Northampton, USA), in the Poly- lowed by direct plunge-freezing in liquid nitrogen. Data was col- morphism and Calorimetry Unit of the Scientific and Technological lected on two crystal systems on 21/09/2021 at the BL13-XALOC Centers of the University of Barcelona. In the experiment, 1.4 ml of beamline of the ALBA synchrotron Light Source [34,35], using RelpLS20 solution at 87.2 lM (4.25 mg/ml) were titrated with 300 mL MXCuBE software for data collection [36]. Crystals of one system of Aux1pLS20 at 2770 lM (25 mg/ml) at 298 K. The reference cell belonged to the space group P4212 and were grown using 0.2 M was filled with double deionized water (ddH2O). All measurements Magnesium chloride hexahydrate and 20% w/v Polyethylene glycol were carried out in 20 mM Tris-HCl pH 8, 300 mM NaCl. The buffer 3,350 as crystallization buffer. Data for this system was processed solution and ddH2O were degassed at room temperature with stir- to a resolution of 1.76 Å, with two protein monomers in the asym- ring under vacuum for 30 min. Upon experimental setup, the metric unit. We will refer to this crystal structure as Aux2-S. Crys- RelpLS20 solution present in the sample cell was stirred at tals of the other system belonged to space group P43212 and were 300 rpm. The titration was initiated at 298 K after a stable baseline obtained using 0.06 M D-Glucose, 0.06 M D-Mannose, 0.06 M D- was achieved, with an initial injection of 2 mL of Aux1pLS20 during Galactose, 0.06 M L-Fucose, 0.06 M D-Xylose, 0.06 M N-Acetyl-D- 4 s. The initial injection was followed by 29 injections of 10 mL dur- Glucosamine, 0.06 M Tris-Bicine pH 8.5, 20% v/v Ethylene glycol, ing 20 s, each spaced by 300 s. 10 % w/v PEG 8000 as crystallization buffer. The data of this system The calorimetric signal was integrated to obtain the enthalpy was processed to a resolution of 1.89 Å with two protein tetramers changes caused by complexation of RelpLS20 with Aux1pLS20. Data in the asymmetric unit. We will refer to this crystal structure as were analyzed using the software Origin 7.0 (January 2004) and fit- Aux2-L. Data were processed with Autoproc v1.0.5 [37–42]. See ted to one single-site binding model, subtracting the average heat Table 1 for further statistics. of the last three measurements after saturation of the RelLS20 bind- To solve the structure of Aux2pLS20, a search for four alpha ing sites to correct experimental heats of the RelpLS20-Aux1pLS20 helices was performed using the program Arcimboldo lite [43]. 1 dilution. The enthalpy (DH ¼ 5722  120:4 kJ mol ), entropy The size of the helices was adjusted according to the secondary 1 (DS ¼ 1:54 kJ mol ), binding constant structure predictions computed with PSIPRED [44]. The initial 1 model was built by auto-tracing in the output map with Buccaneer (K b ¼ 3:40  104  2:64  103 mol ) and number of binding sites (N ¼ 2:55  0:0375) resulted from nonlinear least square data fit- [45]. The structure was completed through alternate manual ting. Dissociation constant (K d ¼ 29:41 lM) was calculated as the model building with Coot v0.8.9 [46] and refinement with PHENIX inverse of Kb and the standard Gibbs free energy value v1.9.2–4158 [47]. The model was validated and further adjusted 1 and refined using MolProbity [48]. The crystallographic and refine- (DG0 ¼ 6180:92 kJ mol ) was calculated using the equation: ment parameters are given in Table 1. Figures were prepared using DG0 ¼ DH0  T DS0 PyMOL (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC.) where T = 298 K and DH0 and DS0 are the thermodynamic val- ues resulting from experimental data fitting at the same tempera- ture. The c value of the assay (c = 7.56) was calculated as 3. Results c ¼ n  P t  K b , indicating a good shape of the binding isotherm to calculate accurate thermodynamic parameters from the calorime- The relaxosome of pLS20 was shown to involve two RHH pro- try data [28]. teins [26], in addition to the recently identified relaxase, RelpLS20 and the oriT region [23]. To gain more insight into the role of these 2.4. SAXS experiments on Aux1pLS20 proteins in the relaxosome formation and their structures, we have performed biophysical and structural studies of the relaxosome SAXS experiments have been performed on the NCD-SWEET proteins. beamline at the synchrotron ALBA (Cerdanyola del Vallès, Barce- As a first step, we estimated the molecular weight (MW) of the lona, Spain) at 12.4 keV on Aux1pLS20 with C-terminal His tag. pLS20 relaxosome proteins in solution using Size Exclusion Chro- The data were collected on a Pilatus 1 M detector (with a pixel size matography (SEC). The elution volumes of the SEC profiles of the 759 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 Table 1 mixture and with a Vel of 2.00 ml for the Aux1pLS20/RelC-terpLS20 Summary of the data processing and refinement statistics of the crystallographic mixture. In contrast, no additional peak appeared in mixtures of analysis of the Aux2pLS20 structures. Aux1pLS20 and RelN-terpLS20 (Suppl. Fig. 2). No other interactions Aux2-S Aux2-L were found between the different constructs of these three pro- Data collection teins (Fig. 1C and Suppl. Fig. 2). To confirm the Aux1pLS20/RelpLS20 Beamline XALOC (ALBA) XALOC (ALBA) interaction and to determine its strength, we measured the ther- PDB code 7NUV 7QNQ modynamic parameters of the interaction in solution using k (Å) 0.9793 0.9793 Space group P 4 21 2 P 43 21 2 isothermal titration calorimetry (ITC, Fig. 1D). The Kd determined Unit cell parameters a = 44.840, b = 44.840, a = 37.854, b = 137.854, by the ITC measurements was 29.4 lM. The ITC results indicate (Å) c = 141.153 c = 121.709 that the interaction was favorable in terms of enthalpy and entropy, with a calculated DG0 of 6180.92 kJ mol1, mainly dri- a Resolution range (Å) 47.05–1.759 (1.874 – 97.48–1.892 (1.964 – 1.759) 1.892) ven by the exothermic component of the binding, with a DH0 of # of reflections: total 147,161 (5833) 2,316,344 (116090) 5722 ± 120.4 kJ mol1 and a DS0 of 1.54 kJ mol1. Both parame- unique 12,790 (641) 87,560 (4373) ters suggest that the interaction between Aux1pLS20 and RelpLS20 is Completeness 90.2 (39.8) 96.1 (57.4) strongly affected by hydrogen bond formation and van der Waals ellipsoidal (%) interactions [49]. The entropy component, although small, may <Ι / r(Ι)> 10.0 14.3 (1.5) be related to the burial of water-accessible surface area upon bind- Average multiplicity 11.5 26.5 (26.5) Rmerge (%) b 36.5 0.19 (3.15) ing, resulting in the release of interfacial water molecules to the Rmeas (%) c 38.3 0.20 (3.1) solvent, contributing favorably to the total entropy of interaction CC(1/2) (%) 96.8 0.998 (0.522) [50]. Structure Refinement Rcrystd /Rfreee (%) 20.61 / 24.17 18.0 / 20.7 r.m.s.deviation from 3.1. SAXS on Aux1pLS20 target values: Bond lengths (Å) 0.007 0.005 Aux1pLS20 is an RHH protein and is predicted to consist of 1 b- Bond angle distances 0.912 0.662 strand followed by three a-helices (Fig. 2A). The occurrence of (Å) the first sheet and two consecutive a-helices is consistent with Molprobity scores: Clashscore (‰) 1.92 0.82 the RHH motif. However, the third a-helix is an addition to RHH Poor rotamers (%) 0.85 0.36 proteins, as it is predicted to span 50 amino acids. To probe the Ramachandran Outliers 0.00 0 effect of this helix on the overall structure of Aux1pLS20, we per- (%) formed SAXS measurements at different concentrations (Suppl Ramachandran 99.21 100 Favoured (%) Fig. 3). The data are consistent with a MW of 38.9 kDa, which cor- Overall score (Å) 0.96 0.75 responds well with the expected MW of 44.0 kDa for a tetramer, Isotropic B factor and is in accordance with AUC data [26]. The radius of gyration analysis (Rg) was determined to be 34 Å, the maximum dimension (Dmax) Average model B- 27.0 44.16 as 130 Å and the Porod volume (VPorod) as 89 Å3. Reconstruction factors (Å2) B-factor from Wilson 18.0 48.00 of the envelop using DAMMIN resulted in an elongated overall plot (Å2) shape (Fig. 2B). Although the overall shape is overestimated by a the SAXS analysis, it is consistent with the tetramer structure of Throughout the table, the values in parentheses are for the outermost resolu- tion shell. the Arc RHH structure bound to DNA (PDB code 1PAR, see b Rmerge = Rh | Îh – Ih,i | / RhRi Ih,i , where Îh = (1/nh) Ri Ih,i and nh is the number of Fig. 2C) and the tetrameric configuration determined previously times a reflection is measured. using analytical ultracentrifugation [26]. The Rg, Rmax and VPorod c Rmeas = [Rh (nh/[nh-1])½ Ri | Îh – Ih,i |] / RhRi Ih,i , where Îh = (1/nh) Ri Ih,i and nh is are also slightly overestimated, which is likely related with the for- the number of times a reflection is measured. d Rcryst = Rhkl | |Fobs| - k |Fcalc| | / Rhkl |Fobs| mation of higher oligomerization states, which has also been e Rfree = Rhkl T | |Fobs| - k |Fcalc| | / Rhkl T |Fobs| where T represents a test set observed in the SEC analysis, where a second species with lower comprising  5% of all reflections excluded during refinement. elution volume appears (Suppl. Fig. 1). There is additional density in the SAXS volume compared to Arc, indicating that Aux1pLS20 forms a triangle instead of a tubular structure. The tip of this trian- individual purified proteins (Suppl. Fig. 1) were used to calculate gle is likely occupied by the additional helical residues present in the MW based on a calibration of the columns using standard pro- the Aux1pLS20 sequence but not in Arc [51]. teins with known MW. The results of the MW estimation are shown in Table 2. The MW of the full length relaxase (RelpLS20), its N- 3.2. Crystal structures of the tetramerization domain of Aux2pLS20 terminal domain (RelN-terpLS20) and its C-terminal domain (RelC- terpLS20), respectively, correspond well to the theoretical values of The function of Aux2pLS20 has so far not been determined and a monomer, indicating that the protein is globular and monomeric cannot be easily inferred from the sequence. A high-resolution in solution in the absence of other interaction partners. However, structure could provide insights into its function by structural the estimated MW of Aux1pLS20 and Aux2pLS20 SEC both correspond homology to known proteins and we therefore set out to deter- to a pentamer and therefore deviate from the expected values mine its structure. We were able to determine two crystal struc- based on the fact they are RHH proteins. tures, named Aux2-L and Aux2-S, both containing essentially the We then set out to map the interactions between respective same structure of a C-terminal fragment of Aux2pLS20 (Fig. 3), lack- pairs of the three pLS20 encoded proteins that have been shown ing the RHH domains. Since we attempted the crystallization of the to be essential for conjugation: Aux1pLS20, Aux2pLS20, and RelpLS20 full-length protein, these fragments occurred through in situ degra- [23,26], performing SEC analysis with mixtures of Aux1pLS20, dation, indicating flexibility between the RHH and C-terminal Aux2pLS20, full length RelpLS20, and the N- and C- terminal domains domains. Both structures show a tetrameric oligomerization state, of RelpLS20. We found that Aux1pLS20 interacts with the full-length where each protein chain folds into two consecutive alpha helices. RelpLS20 and RelC-terpLS20 (Fig. 1 A-C), as evidenced by the appear- The N-terminal helix consists of 12 helical turns and the C-terminal ance of new peaks with a Vel of 1.97 ml for the Aux1pLS20/RelpLS20 helix consists of 3 turns. The short and long alpha helices are con- 760 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 Table 2 Estimated molecular weights (MW) of the relaxosome proteins of pLS20. Protein Vel (ml) Mw, monomer (kDa) Estimated Mw based on Vel (kDa) Calculated oligomerization state Predicted oligomerization state Aux1pLS20 2.05 9.02 46.94 Pentamer (5.20) Tetramer Aux2pLS20 1.82 17.03 85.78 Pentamer (5.03) Tetramer RelpLS20N-ter 2.26 26.92 27.06 Monomer (1.01) Monomer RelpLS20C-ter 2.21 28.27 30.85 Monomer (1.07) Monomer RelpLS20 2.10 48.75 41.17 Monomer (0.84) Monomer Fig. 1. In vitro interactions between relaxosome proteins. A) The SEC elution profile of RelpLS20:Aux1pLS20 mixtures at different stoichiometries. B) the SEC elution profile of RelC-terpLS20:Aux1pLS20 mixtures at different stoichiometries. C) The SEC elution profile of RelpLS20:Aux2pLS20 mixtures at different stoichiometries. In panels A-C, the green, orange and purple lines represent the profile of mixtures at 1:1, 5:1 and 10:1 M ratio, respectively, the blue line in A-C marks the peak of the corresponding RelpLS20 domain, the red line in all plots represents the Aux1pLS20 or Aux2pLS20 and the purple line marks the peak of the complex, D) Thermogram of an Aux1pLS20:RelpLS20 titration as determined by isothermal titration calorimetry. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) nected by a loop of 16 amino acids (A98-E113). The first three N- through extensive hydrophobic packing interactions, thereby terminal turns of the long alpha helices are kinked with respect forming tetramers. This creates a tubular structure with a narrow the rest of the a-helix. The long a-helices form a coiled coil struc- channel at the center. The short alpha helices act as a lock to sta- ture, interacting with the long a-helices of the other chains bilize the structure. They are oriented in a parallel fashion to the 761 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 Fig 2. SAXS analysis of Aux1pLS20. A) Sequence of Aux1pLS20, showing the residues predicted as forming a b-strand in yellow and residues predicted to form a-helices in pink. B) Side (top panel) and top (bottom panel) view of the volume reconstruction based on the Aux1pLS20 SAXS data. C) Cartoon representation of the Arc tetramer in complex with the DNA (not shown) observed in the structure of Arc (PDB code 1PAR). The residues are rainbow colored, blue for the N-terminus changing to red at the C-terminus in a gradual fashion. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 3. Crystallographic structure of the tetramerization domain of Aux2pLS20. A) Sequence of the full-length protein, predicted b-strands are colored yellow and a-helices pink. The green box indicates the fragment crystallized. B) Cartoon representation of the Aux2pLS20 structure, showing the four monomers in different colors. The N-terminal helices are the inner and longer helices, the C-terminal helices are the external, shorter helices shown at the top. C) Top view of the Aux2pLS20 structure along the a-helices, coloring as in B). D) Structure of the C-terminal domain of the C-terminal domain of plasmid F TraM (PDB code 3D8A). The TraMpF chains are colored in the same order as the Aux2pLS20 inner helices whereas the peptide molecules of the coupling protein TraDpF are colored red. Note the relative shift of the C-terminal helices with respect to the inner helices compared to the Aux2pLS20 structure shown in panel C. E) Graphical summary of the results described in this article. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) long helices, but interact with the long alpha helices of chains pro- Interestingly, the structure is similar to the C-terminal domain tomers n + 2 and n + 3, thanks to a long loop that connects both of TraMpF protein from plasmid F (PDB codes 3D8A and 2G7O) helices (Fig. 3C). and that of pED208 (PDB code 3ON0) as shown in Fig. 3D [52– 762 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 54]. TraM also consists of two a-helices that fold into a similar tet- state of Aux1pLS20 should be a multiple of dimers. Our combined rameric stalk structure, consisting of a central a-helix followed by data and previously published data show that the tetrameric a second antiparallel helix that laterally packs against the central arrangement is most likely. helices. The latter structure reveals the complex of the full-length The data obtained for the relaxase suggests that the protein is protein bound to DNA. Comparison of the Aux2pLS20 four-helix monomeric in solution, which is consistent with previous dynamic bundle with that of the 3D8A and 3ON0 structures shows that light scattering and AUC data [23]. Many relaxases are found in a the tetramerization domain in Aux2pLS20 is considerably longer monomeric state, both in solution and in the crystal and single par- than in the other analyzed structures, suggesting that the N- ticle electron microscopy structures, which is particularly true for terminal turns of the long alpha helices in Aux2pLS20 may form part G+ relaxases [21,57]. of the DNA binding domain. This region is not included in the crys- The structures of the C-terminal domain of Aux2pLS20 show that tal structures of the TraMpF in absence of DNA, indicating that the it consists of two alpha helices connected by a short loop (Fig. 3). density is not well defined in that region and that these residues The N-terminal helix is kinked and forms a four-helix bundle, com- are disordered. The structure of the DNA complex of TraMpF shows bining four monomers through extensive and tightly packed that this region becomes ordered upon binding of DNA. Therefore, hydrophobic interactions. As a result, the structure of Aux2pLS20 the DNA binding may induce a conformational change in this shows a tetrameric oligomerization state. However, the SEC data region that would displace the N-terminal part of the long alpha indicate a pentamer. As stated above, RHH domains oligomerize helix, allowing the RHH domain to change position to adjust to as multiples of dimers, and a pentameric structure for the the configuration of the DNA binding sites in the oriT. Aux2pLS20 is highly unlikely given that the protein contains dimeric When comparing the structure of the tetramerization domain of RHH domain. Interestingly, the oligomerization of Aux2pLS20 was Aux2pLS20 with that of TraMpF (PDB codes 3D8A and 3ON0), the previously determined as a hexamer using AUC data [26]. From most pronounced difference is in the position of the C-terminal the structures, it is hard to envisage how the C-terminal stalk helices with respect to the long helix. The loop connecting the domain could reorganize into a hexameric structure, since the small and long helices is shorter in TraMpF. As a result, the interac- four-helix bundle is held together by a tightly packed hydrophobic tions between the C-terminal and N-terminal helices in the four- core. This packing would likely be lost when adding an additional helix bundle occur between adjacent monomers, in contrast to two helices to form a hexamer. However, we cannot exclude the Aux2pLS20, where the short helix interacts with the n + 2 monomer. possibility of the formation of a hexamer, and additional work will It should be noted that the positions of the shorter helices are be required to determine the oligomerization states of this protein equivalent with those observed in the TraMpF structure despite at different stages of the conjugation process. the swap in position. The domain architecture and the C-terminal structure of TraMpF interacts with the plasmid F coupling protein, TraD, and Aux2pLS20 are reminiscent of the TraM protein of plasmid F and the structure of the complex has been determined (PDB code pED208 [52–54,58]. Thus, Aux2pLS20 is a structural analog of the 3D8A) [53]. The TraDpF peptide binds to the loop region that con- TraMIncF proteins. This type of protein is a common factor in gram nects the N- and C-terminal helices (Fig. 3D). The helices in the negative relaxosomes of plasmids from different incompatibility 3ON0 structure TraMpF, not bound to the peptide, show a similar groups [5,59]. In plasmid F and other systems, one of the functions layout to the structure of the TraDpF/TraMpF complex, with only of this protein is to interact with the coupling protein through the minor adjustments of the residues in the loop between the helices. C-terminal domain and the relaxase, functioning as a connector In contrast, in Aux2pLS20, the loop between the helices of Aux2pLS20 protein. Thus, the structural homology of the Aux2pLS20 C- traverses the region corresponding to the binding site of TraDpF, terminal domain with that of TraMpF suggests that one of the func- which is due to the swap of the position of the short C-terminal tions of Aux2pLS20 is to interact with the coupling protein of pLS20. helix described above. Unlike TraMpF, which was proposed to also interact with the relaxase TraIpF [60], no interactions could be detected between Aux2pLS20 and RelpLS20. This is consistent with previous AUC data 4. Discussion which did not detect any interaction between these proteins [26]. Thus, it seems that the interaction between the T4SS and In this paper, we report the low-resolution structure of the pLS20 relaxosome is not accomplished by a direct interaction Aux1pLS20 and two high-resolution structures of the tetrameriza- between Aux2pLS20 and the relaxase and it is likely that Aux2pLS20 tion domain of Aux2pLS20. In addition, we analyze the possible bridges the relaxosome and T4SS channel through interactions interactions between the auxiliary proteins and the relaxase using with DNA regions of the oriT [26]. ITC and SEC, which showed that Aux1pLS20 and RelpLS20 interact, but The comparison of the structure of the tetramerization domain the other proteins do not. of Aux2pLS20 with that of TraMpF also provides clues on how differ- Our results indicate that the RHH protein Aux1pLS20 tetramer- ent connector proteins select their substrates. The swapped posi- izes and forms an elongated structure, reminiscent of the RHH pro- tion of the two helices with respect to TraMpF and the teins, of which the transcriptional repressor Arc from the concomitant differences in the position of the loop between the Salmonella phage P22 is a founding member [27]. The sequential two helices, result in a substantially altered binding pocket of the length, function and structural features of Aux1pLS20 are similar peptide fragment of the coupling protein. As a consequence, the to protein TraY of IncF plasmids: both TraY and Aux1pLS20 are interaction between the coupling protein of pLS20 and Aux2pLS20, RHH proteins and are able to bind to inverted sequences on their if it exists, is likely different from that observed for plasmid F- respective oriTs [55,56]. Our SAXS data suggest that Aux1pLS20 is like connector proteins. Future work should elucidate if Aux2pLS20 a tetramer like TraYpF, which is in agreement with the AUC data interacts directly with the pLS20 coupling protein, and if the previously published [26]. From the SEC experiments, however, a TraMpF binding pocket is preserved or is located in a different pentameric configuration was derived. It should be noted that region of the protein. the elution volume of the proteins in SEC can vary according to We found that Aux1pLS20 is able to interact with the C-terminal their shape. The discrepancies of the MW derived from SEC data domain of RelpLS20 in SEC and in ITC. This is to the best of our compared to SAXS and AUC can be explained by the elongated, knowledge unique among the relaxosomes studied so far. The non-globular shape of Aux1pLS20. Given that RHH domains are interaction is consistent in that it occurs with the full length intrinsically dimeric in nature, it follows that the oligomerization RelpLS20 as well as its C-terminal domain, but not the N-terminal 763 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 domain (Fig. 1 and Suppl Fig. 2). It should be noted, however, that Acknowledgement this interaction was not observed in AUC data previously published [26]. An explanation of this apparent discrepancy may lie in the We thank Rafel Prohens of the polymorphism and calorimetry different experimental conditions used in both techniques and that unit of the CCiT at the University of Barcelona. We acknowledge the interaction is weak as determined by ITC measurements. It is the assistance of the staff of the XALOC and NCD-SWEET beamli- likely that it can only be detected under favorable conditions, nes, as well as the Floor Coordinators at the ALBA synchrotron. which may explain why AUC experiments failed to detect this interaction. Finally, it should be noted that the experiments described above were performed in absence of DNA, and it is Appendix A. Supplementary data expected that the presence of DNA sequences with affinity for these proteins may induce stronger binding and stabilization of Supplementary data to this article can be found online at the complexes. Future experiments are needed to investigate these https://doi.org/10.1016/j.csbj.2021.12.041. questions further. The findings in this paper are summarized graphically in Fig. 3E. The combined data presented here show that Aux1pLS20 and References Aux2pLS20 likely bridge the various components of the relaxosome and mediate the interaction with the T4SS channel. We identify [1] Bellanger X, Payot S, Leblond-Bourget N, Guédon G. Conjugative and Aux2pLS20 as a structural analog of TraMpF, which suggests that mobilizable genomic islands in bacteria: evolution and diversity. FEMS Microbiol Rev 2014;38(4):720–60. https://doi.org/10.1111/1574-6976.12058. Aux2pLS20 links the relaxosome to the T4SS channel through inter- [2] Grohmann E, Muth G, Espinosa M. Conjugative plasmid transfer in gram- actions with the oriT. Furthermore, we show that Aux1pLS20 inter- positive bacteria. Microbiol Mol Biol Rev 2003;67(2):277–301. acts with RelpLS20, which suggests that it stabilizes the binding of [3] Goessweiner-Mohr N, Arends K, Keller W, Grohmann E. Conjugative type IV the relaxase to the DNA sequence, as both that RelpLS20 and secretion systems in Gram-positive bacteria. Plasmid 2013;70(3):289–302. https://doi.org/10.1016/j.plasmid.2013.09.005. Aux1pLS20 contain DNA binding domains. In addition, the interac- [4] Grohmann E, Christie PJ, Waksman G, Backert S. Type IV secretion in Gram- tion provides a possible function to the C-terminal domain of the negative and Gram-positive bacteria. Mol Microbiol 2018;107(4):455–71. relaxase. https://doi.org/10.1111/mmi.13896. [5] De La Cruz F, Frost LS, Meyer RJ, Zechner EL. Conjugative DNA metabolism in Gram-negative bacteria. FEMS Microbiol Rev 2010;34(1):18–40. https://doi. org/10.1111/j.1574-6976.2009.00195.x. 5. Conclusions [6] Smillie C, Garcillán-Barcia MP, Francia MV, Rocha EPC, de la Cruz F. Mobility of Plasmids. Microbiol Mol Biol Rev 2010;74(3):434–52. [7] Christie PJ, Whitaker N, González-Rivera C. Mechanism and structure of the We present the envelope of Aux1pLS20 and two X-ray structures bacterial type IV secretion systems. Biochim Biophys Acta - Mol. Cell Res of Aux2pLS20. The combined data presented here show that 2014;1843(8):1578–91. https://doi.org/10.1016/j.bbamcr.2013.12.019. Aux1pLS20 and Aux2pLS20 are likely tetramers in solution. We iden- [8] Low HH, Gubellini F, Rivera-Calzada A, Braun N, Connery S, Dujeancourt A, et al. Structure of a type IV secretion system. Nature 2014;508(7497):550–3. tify Aux2pLS20 as a TraMpF analog and show that Aux1pLS20 interacts https://doi.org/10.1038/nature13081. with the C-terminal domain of RelpLS20. [9] Li YG, Hu Bo, Christie PJ, Sandkvist M, Cascales E, Christie PJ. Biological and structural diversity of type IV secretion systems. Microbiol Spectr 2019;7(2). [10] Zechner EL, Lang S, Schildbach JF. Assembly and mechanisms of bacterial type IV secretion machines. Philos Trans R Soc B Biol Sci 2012;367(1592):1073–87. Funding https://doi.org/10.1098/rstb.2011.0207. [11] Christie PJ, Valero LG, Buchrieser C. Biological diversity and evolution of type This work was supported by Ministry of Economy and Compet- IV secretion systems. Curr Top Microbiol Immunol 2017;413. https://doi.org/ 10.1007/978-3-319-75241-9_1. itiveness of the Spanish Government grants BIO2016-77883-C2-1- [12] Wong JJW, Lu J, Glover JNM. Relaxosome function and conjugation regulation P and PID2019-108778GB-C21 to W.J.J.M, PID2020-117028 GB-I00 in F-like plasmids – a structural biology perspective. Mol Microbiol (AEI/FEDER, EU), BIO2016-77883-C2-2-P (AEI/FEDER, EU) and 2012;85:602–17. https://doi.org/10.1111/j.1365-2958.2012.08131.x. [13] Christie PJ, Vogel JP. Bacterial type IV secretion: Conjugation systems adapted FIS2015-72574-EXP (AEI/FEDER, EU), which also supported N.B., to deliver effector molecules to host cells. Trends Microbiol 2000;8(8):354–60. to R.B. Funding for open access charge: Ministry of Economy and https://doi.org/10.1016/S0966-842X(00)01792-3. Competitiveness of the Spanish Government PID2020- [14] Nagai H, Kubori T. Type IVB secretion systems of Legionella and other gram- negative bacteria. Front Microbiol 2011;2. https://doi.org/10.3389/ 117028 GB-I00. The funders had no role in study design, data col- fmicb.2011.00136. lection and analysis, decision to publish, or preparation of the [15] Costa TRD, Harb L, Khara P, Zeng L, Hu Bo, Christie PJ. Type IV secretion manuscript. systems: advances in structure, function, and activation. Mol Microbiol 2021;115(3):436–52. https://doi.org/10.1111/mmi.14670. [16] Kohler V, Keller W, Grohmann E. Regulation of gram-positive conjugation. Front Microbiol 2019;10:1134. https://doi.org/10.3389/fmicb.2019.01134. Author contributions [17] Wawrzyniak P, Plucienniczak G, Bartosik D. The different faces of rolling-circle replication and its multifunctional initiator proteins. Front Microbiol 2017;8. https://doi.org/10.3389/fmicb.2017.02353. IC, NB and AC designed and performed the experiments, BMC, [18] Guzmán-Herrador DL, Llosa M. The secret life of conjugative relaxases. Plasmid XC and FGO assisted in the technical preparation of the experi- 2019;104:102415. https://doi.org/10.1016/j.plasmid.2019.102415. ments. JVC, AMA, and WJJM provided pET28-based clones and cell [19] Garcillán-Barcia MP, Redondo-Salvo S, Vielva L, de la Cruz F. MOBscan: automated annotation of MOB relaxases. Methods Mol Biol 2020;2075. culture material of some of the protein samples (or subdomains) https://doi.org/10.1007/978-1-4939-9877-7_21. used in the initial and/or final studies. MM analyzed the SAXS data, [20] Ilangovan A, Kay CWM, Roier S, El Mkami H, Salvadori E, Zechner EL, et al. RB designed the research and wrote the paper. Cryo-EM structure of a relaxase reveals the molecular basis of DNA unwinding during bacterial conjugation. Cell 2017;169(4):708–721.e12. https://doi.org/ 10.1016/j.cell.2017.04.010. [21] Edwards JS, Betts L, Frazier ML, Pollet RM, Kwong SM, Walton WG, et al. Declaration of Competing Interest Molecular basis of antibiotic multiresistance transfer in Staphylococcus aureus. Proc Natl Acad Sci U S A 2013;110(8):2804–9. https://doi.org/10.1073/ The authors declare that they have no known competing finan- pnas.1219701110. [22] Gomis-Rüth FX, Coll M. Cut and move: protein machinery for DNA processing cial interests or personal relationships that could have appeared in bacterial conjugation. Curr Opin Struct Biol 2006;16(6):744–52. https://doi. to influence the work reported in this paper. org/10.1016/j.sbi.2006.10.004. 764 I. Crespo, N. Bernardo, A. Cuppari et al. Computational and Structural Biotechnology Journal 20 (2022) 757–765 [23] Ramachandran G, Miguel-Arribas A, Abia D, Singh PK, Crespo I, Gago-Córdoba [42] Evans PR, Murshudov GN. How good are my data and what is the resolution? C, et al. Discovery of a new family of relaxases in Firmicutes bacteria. PLOS Acta Crystallogr D Biol Crystallogr 2013;69(7):1204–14. Genet 2017;13(2):e1006586. https://doi.org/10.1371/journal.pgen.1006586. [43] Sammito M, Millán C, Rodríguez DD, de Ilarduya IM, Meindl K, De Marino I, [24] Zheng B, Tomita H, Inoue T, Ike Y. Isolation of VanB-type Enterococcus faecalis et al. Exploiting tertiary structure through local folds for crystallographic strains from nosocomial infections: First report of the isolation and phasing. Nat Methods 2013;10(11):1099–101. https://doi.org/10.1038/ identification of the pheromone-responsive plasmids pMG2200, encoding nmeth.2644. VanB-type vancomycin resistance and a Bac41-type bacteriocin, a. Antimicrob [44] Buchan DWA, Jones DT. The PSIPRED Protein Analysis Workbench: 20 years on. Agents Chemother 2009;53. https://doi.org/10.1128/AAC.00754-08. Nucleic Acids Res 2019;47. doi: 10.1093/nar/gkz297. [25] Tomita H, Ike Y. Genetic analysis of transfer-related regions of the vancomycin [45] Cowtan K. The Buccaneer software for automated model building. 1. Tracing resistance Enterococcus conjugative plasmid pHTb: Identification of oriT and a protein chains. Acta Crystallogr D Biol Crystallogr 2006;62(9):1002–11. putative relaxase gene. J Bacteriol 2005;187(22):7727–37. https://doi.org/ [46] Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. 10.1128/JB.187.22.7727-7737.2005. Acta Crystallogr D Biol Crystallogr 2010;66(4):486–501. https://doi.org/ [26] Miguel-Arribas A, Hao J-A, Luque-Ortega JR, Ramachandran G, Val-Calvo J, 10.1107/S0907444910007493. Gago-Córdoba C, et al. The Bacillus subtilis conjugative plasmid pLS20 encodes [47] Liebschner D, Afonine PV, Baker ML, Bunkóczi G, Chen VB, Croll TI, et al. two ribbon-helix-helix type auxiliary relaxosome proteins that are essential Macromolecular structure determination using X-rays, neutrons and for conjugation. Front Microbiol 2017;8. https://doi.org/10.3389/ electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol fmicb.2017.02138. 2019;75(10):861–77. [27] Schreiter ER, Drennan CL. Ribbon-helix-helix transcription factors: variations [48] Williams PA, Hughes CE, Lim GK, Kariuki BM, Harris KDM. Discovery of a new on a theme. Nat Rev Microbiol 2007;5(9):710–20. https://doi.org/10.1038/ system exhibiting abundant polymorphism: M -aminobenzoic acid. Cryst nrmicro1717. Growth Des 2012;12(6):3104–13. https://doi.org/10.1021/cg3003178. [28] Srivastava VK, Yadav R. Isothermal titration calorimetry. Elsevier Inc.; 2019. [49] Cramer J, Jiang X, Schönemann W, Silbermann M, Zihlmann P, Siegrist S, et al. doi: 10.1016/B978-0-12-816548-5.00009-5. Enhancing the enthalpic contribution of hydrogen bonds by solvent shielding. [29] Huang TC, Toraya H, Blanton TN, Wu Y. X-ray powder diffraction analysis of RSC Chem Biol 2020;1(4):281–7. silver behenate, a possible low-angle diffraction standard. J Appl Crystallogr [50] Jelesarov I, Bosshard HR. Isothermal titration calorimetry and differential 1993;26(2):180–4. https://doi.org/10.1107/S0021889892009762. scanning calorimetry as complementary tools to investigate the energetics of [30] Kieffer J, Karkoulis D. PyFAI, a versatile library for azimuthal regrouping. J. biomolecular recognition. J Mol Recognit 1999;12:3–18. https://doi.org/ Phys. Conf. Ser. 2013;425(20):202012. https://doi.org/10.1088/1742-6596/ 10.1002/(SICI)1099-1352(199901/02)12:1<3::AID-JMR441>3.0.CO;2-6. 425/20/202012. [51] Raumann BE, Rould MA, Pabo CO, Sauer RT. DNA recognition by b-sheets in the [31] Konarev PV, Volkov VV, Sokolova AV, Koch MHJ, Svergun DI. PRIMUS: a Arc represser-operator crystal structure. Nature 1994;367(6465):754–7. Windows PC-based system for small-angle scattering data analysis. J Appl https://doi.org/10.1038/367754a0. Crystallogr 2003;36(5):1277–82. https://doi.org/10.1107/ [52] Wong JJW, Lu J, Edwards RA, Frost LS, Glover JNM. Structural basis of S0021889803012779. cooperative DNA recognition by the plasmid conjugation factor, TraM. Nucl [32] Svergun DI. Determination of the regularization parameter in indirect- Acids Res 2011;39. doi: 10.1093/nar/gkr296. transform methods using perceptual criteria. J Appl Crystallogr 1992;25 [53] Lu J, Wong JJW, Edwards RA, Manchak J, Frost LS, Glover JNM. Structural basis (4):495–503. https://doi.org/10.1107/S0021889892001663. of specific TraD-TraM recognition during F plasmid-mediated bacterial [33] Franke D, Svergun DI. DAMMIF, a program for rapid ab-initio shape conjugation. Mol Microbiol 2008;70(1):89–99. https://doi.org/10.1111/ determination in small-angle scattering. J Appl Crystallogr 2009;42 j.1365-2958.2008.06391.x. (2):342–6. https://doi.org/10.1107/S0021889809000338. [54] Lu J, Edwards RA, Wong JJW, Manchak J, Scott PG, Frost LS, et al. Protonation- [34] Juanhuix J, Gil-Ortiz F, Cuní G, Colldelram C, Nicolás J, Lidón J, et al. mediated structural flexibility in the F conjugation regulatory protein, TraM. Developments in optics and performance at BL13-XALOC, the EMBO J 2006;25(12):2930–9. https://doi.org/10.1038/sj.emboj.7601151. macromolecular crystallography beamline at the Alba Synchrotron. J [55] Frost LS, Ippen-Ihler K, Skurray RA. Analysis of the sequence and gene products Synchrotron Radiat 2014;21(4):679–89. https://doi.org/10.1107/ of the transfer region of the F sex factor. Microbiol Rev 1994;58(2):162–210. S160057751400825X10.1107/S160057751400825X/rv5017sup1.avi10.1107/ [56] Fernandez-Lopez R, de Toro M, Moncalian G, Garcillan-Barcia MP, de la Cruz F. S160057751400825X/rv5017sup2.pdf. Comparative genomics of the conjugation region of F-like plasmids: Five [35] Fauth F, Boer R, Gil-Ortiz F, Popescu C, Vallcorba O, Peral I, et al. The shades of F. Front Mol Biosci 2016;3. doi: 10.3389/fmolb.2016.00071. crystallography stations at the Alba synchrotron. Eur Phys J Plus 2015;130(8). [57] Pluta R, Boer DR, Lorenzo-Díaz F, Russi S, Gómez H, Fernández-López C, et al. https://doi.org/10.1140/epjp/i2015-15160-y. Structural basis of a histidine-DNA nicking/joining mechanism for gene [36] Oscarsson M, Beteva A, Flot D, Gordon E, Guijarro M, Leonard G, et al. transfer and promiscuous spread of antibiotic resistance. Proc Natl Acad Sci MXCuBE2: the dawn of MXCuBE Collaboration. J Synchrotron Radiat USA 2017;114(32):E6526–35. 2019;26:393–405. https://doi.org/10.1107/s1600577519001267. [58] Peng Y, Lu J, Wong JJW, Edwards RA, Frost LS, Mark Glover JN. Mechanistic [37] Tickle IJ, Flensburg C, Keller P, Paciorek W, Sharff A, Vonrhein C, et al. basis of plasmid-specific DNA binding of the F plasmid regulatory protein, STARANISO 2018. TraM. J Mol Biol 2014;426(22):3783–95. https://doi.org/10.1016/j. [38] Vonrhein C, Flensburg C, Keller P, Sharff A, Smart O, Paciorek W, et al. Data jmb.2014.09.018. processing and analysis with the {\it autoPROC} toolbox. Acta Crystallogr Sect [59] Guglielmini J, Néron B, Abby SS, Garcillán-Barcia MP, La Cruz DF, Rocha EPC. D 2011;67:293–302. https://doi.org/10.1107/S0907444911007773. Key components of the eight classes of type IV secretion systems involved in [39] Kabsch W. XDS. Acta Crystallogr D Biol Crystallogr 2010;66(2):125–32. bacterial conjugation or protein secretion. Nucl Acids Res 2014;42. doi: [40] Evans P. Scaling and assessment of data quality. Acta Crystallogr D Biol 10.1093/nar/gku194. Crystallogr 2006;62(1):72–82. [60] Ragonese H, Haisch D, Villareal E, Choi J-H, Matson SW. The F plasmid-encoded [41] Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, et al. TraM protein stimulates relaxosome-mediated cleavage at oriT through an Overview of the CCP 4 suite and current developments. Acta Crystallogr D Biol interaction with TraI. Mol Microbiol 2007;63(4):1173–84. Crystallogr 2011;67(4):235–42. 765