Plasmid
abstract 
Conjugation plays an important role in the horizontal movement of DNA between bacterial species and even genera . 
Large conjugative plasmids in Gram-negative bacteria are associated with multi-drug resistance and have been implicated in the spread of these phenotypes to pathogenic organisms . 
A/C plasmids often carry genes that confer resistance to multiple classes of antibiotics . 
Recently , transcription factors were characterized that regulate A/C conjugation . 
In this work , we expanded the regulon of the negative regulator Acr2 . 
We developed an A/C variant , pARK01 , by precise removal of resistance genes carried by the plasmid in order to make it more genetically tractable . 
Using pARK01 , we conducted RNA-Seq and ChAP-Seq experiments to characterize the regulon of Acr2 , an H-NS-like protein . 
We found that Acr2 binds several loci on the plasmid . 
We showed , in vitro , that Acr2 can bind specific promoter regions directly and identify key amino acids which are important for this binding . 
This study further characterizes Acr2 and suggests its role in modulating gene expression of multiple plasmid and chromosomal loci . 
© 2016 Elsevier Inc. . 
All rights reserved . 
Received in revised form 26 July 2016 Accepted 28 July 2016 Available online 1 August 2016
1. Introduction
The IncA/C plasmid type has spread globally and is now a major contributor to multidrug resistance in enteric pathogens ( Welch et al. , 2007 ; Fernández-Alarcón et al. , 2011 ; Fricke et al. , 2009 ; Harmer and Hall , 2014 ) . 
This plasmid type is modular in nature , containing several variable regions encoding antimicrobial resistance mechanisms and a conserved backbone of core genes for plasmid replication and maintenance ( Fernández-Alarcón et al. , 2011 ; Harmer and Hall , 2014 ; Fricke et al. , 2011 ; Carraro et al. , 2014a ) . 
Recent work characterized several transcriptional regulators encoded by the core backbone ( Carraro et al. , 2014b ) . 
These regulators were found to regulate conjugative transfer of IncA/C plasmids . 
Interestingly , one of the negative regulators , Acr2 , is similar to H-NS , a chromosomally encoded , global transcriptional regulator ( Fernández-Alarcón et al. , 2011 ) . 
There are several nucleoid-associated proteins encoded by E. coli . 
H-NS is among the most well studied ( Dillon and Dorman , 2010 ) . 
H-NS binds preferentially to DNA with a low G + C content ( Fang and Rimsky , 2008 ) . 
These sequences serve as sites of initial binding . 
After these initial interactions , other copies of H-NS can bind to each other via pro-tein-protein interactions , as well as the DNA , forming large complexes ( Fang and Rimsky , 2008 ; Bouffartigues et al. , 2007 ) . 
This is achieved through two distinct domains . 
The N-terminus of H-NS contains a oligomerization domain and the C-terminus contains a DNA binding 
Corresponding author at : University of Minnesota , Department of Veterinary and Biomedical Sciences , 1971 Commonwealth Ave , St. Paul , MN 55108 , United States . 
E-mail address : joh04207@umn.edu ( T.J. Johnson ) . 
domain ( Ali et al. , 2013 ) . 
Primarily , H-NS binding results in silencing of adjacent gene transcription ( Navarre et al. , 2006 ; Stoebel et al. , 2008 ) . 
Chromosomal copies of H-NS play an important role in regulation of horizontally acquired DNA , such as pathogenicity islands ( Dorman , 2007 ; Ali et al. , 2014 ; Navarre et al. , 2007 ) . 
In some cases , mobile elements encode their own copies of H-NS homologs ( Yun et al. , 2010 ; Dillon et al. , 2010 ; Müller et al. , 2010 ) . 
These horizontally encoded homologs of H-NS have been shown to antagonize the binding of the chromosomal copies of H-NS to horizontally acquired DNA ( Stoebel et al. , 2008 ; Bustamante et al. , 2001 ) . 
These antagonistic H-NS homologs have been found only on genomic islands of pathogenic E. coli , which is intuitive , given their specific function . 
There have been a few studies focusing on plasmid encoded H-NS homologs ( Yun et al. , 2010 ; Dillon et al. , 2010 ; Forns et al. , 2005 ; Doyle et al. , 2007 ) . 
H-NS homologs from plasmid pSfR27 and pCAR ( Sfh and Pmr , respectively ) seem to play roles in regulating a diverse set of genes , some of which are regulated by the chromosomally encoded H-NS copies ( Yun et al. , 2010 ; Dillon et al. , 2010 ; Doyle et al. , 2007 ) . 
It has been proposed that uncontrolled expression of these genes , caused by plasmid acquisition , could lead to a reduction in fitness and subsequent loss of the plasmid from the population . 
The H-NS homolog encoded on the R27 plasmid of E. coli , H-NSR27 , has been shown to directly interact with the plasmid 's origin of replication , oriT and other transfer associated genes to regulate conjugation ( Forns et al. , 2005 ) . 
H-NSR27 was shown to be involved in an intricate interplay of chromosomally encoded H-NS homologs to thermally regulate the expression of the conjugative transfer apparatus of R27 . 
These recent studies exemplify the diverse roles these plasmid encoded H-NS homologs serve . 
Given how widely distributed thes homologs are among plasmid types , the true diversity of roles for H-NS homologs is unknown ( Shintani et al. , 2015 ) . 
In this study , we characterize Acr2 , an H-NS-like protein that was found to negatively regulate conjugative transfer . 
RNA-Seq and ChAP-Seq ( Chromatin Affinity Precipitation-Seq ) were used to characterize Acr2 binding sites and regulatory network . 
We show that Acr2 binds multiple loci on the plasmid , specifically in regions of transfer genes and transposons carried by the plasmid . 
Additionally , we found that Acr2 binds several loci on the host bacterial chromosome and may directly alter host gene expression . 
Our sequence analysis indicates that Acr2 shares a DNA binding motif with that of other H-NS homologs and using site-directed mutagenesis we demonstrate that these amino acids are critical for its function as a repressor of conjugation . 
2. Materials and methods
2.1. Bacterial strains, plasmids and growth conditions
The bacterial strains and plasmids are listed in Table 1 . 
All strains were routinely grown in Difco ™ Luria-Bertani ( LB ) broth or LB agar at 37 °C unless otherwise noted . 
Broth cultures were grown in a shaking incubator at 37 °C with shaking ( 200 RPM ) unless otherwise noted . 
Supplementation of ampicillin ( Amp , 100 μg / mL ) , chloramphenicol ( Cm , 20 μg / mL ) , nalidixic acid ( Nal , 30 μg / mL ) , kanamycin ( Kan , 100 μg / mL ) , rifampicin ( Rif , 100 μg / mL ) and tetracycline ( 12.5 μg / mL ) were used as needed . 
Counter-selections were done using M9 minimal agar supplemented with 0.2 % rhamnose . 
Arabinose induction of pBAD22 vector constructs was achieved with a final concentration of 0.02 % . 
X-gal added to agar plates at a concentration of 40 μg / mL . 
Diaminopimelic acid ( DAP ) was add to a final concentration of 300 μM to facilitate growth of WM3064 . 
2.2. Strain construction via recombineering
All deletions and mutations were done via λ-Red mediated recombination ( recombineering ) with some variations . 
In order to delete the resistance genes from pAR060302 , strain DY331 was used because it expresses recombineering genes from the chromosome and not from a plasmid which must be selected for . 
To ease transfer of pAR060302 , it was moved into a nalidixic acid resistant variant of strain WM3064 , which is a DAP auxotroph . 
To obtain a nalidixic acid resistant WM3064 , it was grown at 37 °C with shaking ( 200 RPM ) in LB broth supplemented with DAP overnight . 
This culture was used to inoculate a new 10 mL LB + Dap broth culture , which was incubated for 4 h at 37 °C with shaking ( 200 RPM ) . 
The cells were pelleted by centrifugation at 8000 RPM for 10 min . 
The pellet was resuspended with 100 μL of LB broth and used to inoculate an LB agar plate supplement with Nal and Dap to select a spontaneous nalidixic acid resistant mutant ( WM3064nalR ) . 
This plate was incubated at 37 °C overnight . 
Isolated colonies were streak purified on a new LB + Nal + Dap agar plate . 
The pAR060302 parental E. coli strain , AR060302 , was mated with WM3064nalR . 
Briefly , WM3064nalR was struck on an LB + Dap plate and the WT strain AR060302 was struck over the top . 
The plate was incubated overnight . 
The resulting growth was struck onto an LB + Nal + CM + Dap plate to select for WM3064nalR ( pAR0603020 ) transconjugants . 
WM3064nalR ( pAR060302 ) was used to transfer pAR060302 into strain DY331 in a similar manner except selection for transconjugants was achieved by growth on LB + Cm plates ( No DAP ) to select for DY331 ( pAR060302 ) . 
PCR was used to generate the tet knockout amplicon by amplifying the neo-ccdB cassette from pKD45 using the primers pARdeltaTet-fw and rv ( Table 2 ) ( 12.5 μL Phusion 2 × master mix ( Life Technologies ™ ) , 500 nM of each primer , 1 μL of template ( boiled cells ) , PCR conditions : 95 °C 5 min , 25 cycles of 95 °C 30 s , 55 ° C 30 s , 72 °C for 2 min ; then a final incubation at 72 °C for 10 min ) . 
The resulting amplicon was purified with a Qiagen ™ PCR cleanup kit . 
The neo-ccdB cassette from pKD45 allows for selection with kanamycin and then removal of the cassette ( scar-less or otherwise ) using counterselection as expression of ccdB ( via growth on minimal rhamnose plates ) is lethal . 
Strain DY331 ( pAR060302 ) was grown in LB + Cm broth at 32 °C with shaking ( 200 RPM ) overnight . 
The overnight culture was used to inoculate a new 50 mL LB + Cm broth culture ( 1:100 dilution ) . 
This was grown until an OD600 ~ 0.4 at which time it was moved to a 43 °C shaking water bath and incubated for 20 min to prime cells for recombineering . 
After the incubation the flask was cooled on ice for 
10 min . 
The cells were pelleted by centrifugation at 8000 RPM for 5 min at 4 °C . 
The cell pellet was resuspended with 25 mL ice cold H2O . 
The cells were collected again by centrifugation and washed an additional time . 
The final cell pellet was resuspended in 200 μL of ice cold H2O . 
A 40 μL aliquot of washed cells was electroporated with 0.5 -- 1 μg of puri-fied amplicon . 
Cells were recovered with LB broth and incubated at 32 ° C for 4 h . 
After recovery cells were plated on LB + Kan plates . 
Isolated colonies were checked for growth on LB + Tet plates to confirm disruption of the tet ( A ) gene . 
Successful mutants were mated on LB plates with E. coli strain K-12 MG1655 carrying pSIM5-Tet at 32 ° overnight . 
Successful transconjugants were selected on LB + Cm + Tet plates to select for pAR060302 ( Δtet : : neo-ccdB ) and pSIM5-tet . 
This strain was used to remove the neo-ccdB cassette by recombineering using a ssDNA substrate ( Sawitzke et al. , 2011 ) . 
Cells were primed for recombineering as above and the 70 nt oligonucleotide DeltaTetRepairOligo ( Table 2 ) was electroporated . 
Cells were recovered and plated on M9 + Rhamnose + Cm plates . 
This resulted in pAR060302 ( Δtet ) , where the tet locus was deleted and the neo-ccdB cassette was removed . 
Isolated colonies were checked for mutations via PCR . 
The entire insertionremoval process was then done a second time to target and remove the blaCMY-2 gene using the appropriate primers . 
This resulted in pARK01 , a tetracycline and ampicillin susceptible variant of pAR060302 . 
Deletions of predicted transcriptional regulators on pARK01 were done via recombineering in strain K12 ( pARK01 , pSIM5-tet ) . 
The plasmid pKD4 ( Datsenko and Wanner , 2000 ) carrying the FRT-neo-FRT cassette was used as a template for PCR ( 12.5 μL Phusion 2 × master mix ( Life Technologies ™ ) , 500 nM of each primer , 1 μL of template ( boiled cells ) , PCR conditions : 95 °C 5 min , 25 cycles of 95 °C 30 s , 55 ° C 30 s , 72 °C for 2 min ; then a final incubation at 72 °C for 10 min ) . 
PCR amplicons were subjected to a DpnI digestion to eliminate template plasmid and then electroporated into cells primed for recombineering . 
Resulting colonies were checked by PCR for mutations . 
Successful mutants were then grown and made electrocompetent and the plasmid pCP20 was electroporated . 
These transformants were grown at 32 °C for 48 h to `` flip '' out the neo cassette , leaving a copy of FRT . 
The plasmid pCP20 was lost via incubation at 37 °C for 48 h. Colonies were screened via PCR and amplicons were verified by sequencing . 
Alterations in the AT-Hook motif of Acr2 were achieved by first inserting the neo-ccdB cassette ( amplified using delATHOOKAcr2-kanccdB-fw and - rv primers ) into acr2 , and then through another round of recombineering using the appropriate ssDNA oligo and counterselection on minimal rhamnose plates to remove the cassette and alter the coding sequence as desired . 
2.3. Molecular cloning methods
For constructs using pLUXtet , containing a promoterless lux operon , cloning was done via double digestion with SpeI and BamHI and ligation with DNA ligase ( NEB ) . 
Oligonucleotides used to generate PCR amplicons of promoter regions with flanking ends containing restriction sites are listed in Table 2 . 
To clone genes from pAR060302 into pBAD22 ( Guzman et al. , 1995 ) , in vivo cloning via recombineering was used ( Lee et al. , 2001 ) . 
In order to generate a template for PCR , pBAD22 was digested overnight using BamHI . 
The resulting digestion was used as template for PCR ( 12.5 μL Phusion 2 × master mix ( Life Technologies ™ ) , 500 nM of each primer , 1 μL of template , PCR conditions : 95 °C 5 min , 25 cycles of 95 °C 30 s , 55 °C 30 s , 72 °C for 6 min ; then a final incubation at 72 °C for 10 min ) . 
Primers used for in vivo cloning were designed to amplify the vector pBAD22 , and add 40 bp regions of homology starting with 2nd codon of the gene of interest to the reverse primer and 40 bp of homology ending with the codon next to the stop codon for the forward primer . 
PCR amplicons were purified with a Qiagen ™ PCR Cleanup Kit and electroporated into K-12 ( pARK01 , pSIM5-tet ) cells that were primed for recombineering . 
After 4 h of recovery in LB broth at 32 °C , cells were plated on LB + Amp + Cm plates to select for strains still carrying pARK01 as well as the newly recombined pBAD22 with appropriate insertion . 
Subsequent colonies were checked via PCR for insertions in pBAD22 . 
Successfully cloned plasmids were confirmed via DNA sequencing . 
The pSIM5-tet plasmid was lost by incubation at 37 °C for 24 h. 
2.4. Conjugation experiments
E. coli strain DH10 was used as a recipient in all conjugation assays . 
Donor and recipient cells were grown overnight at 37 °C in LB broth with the appropriate selection . 
The overnight cultures were used to inoculate new 5 mL cultures in LB broth with no selection . 
The new cultures were incubated for 4 h ; if arabinose induction was needed it was done at 2 h of incubation . 
After 4 h of incubation , 0.5 mL of donor and recipient cells were added to 1.5 mL centrifuge tube , mixed by pipetting and incubated at 37 °C without shaking for 1 h . 
The mating reactions were then vortexed and placed immediately on ice . 
They were subsequently diluted in 1 × phosphate buffered saline ( PBS ) and plated to select for transconjugants and donors . 
2.5. Lux reporter assays
Overnight cultures were grown with the appropriate selections in LB broth at 37 °C with shaking ( 200 RPM ) . 
These cultures were diluted 1:100 in new broth with selection . 
These cultures were grown under the same conditions for 2 h . 
At that time , the cultures were split in half . 
One half was treated with arabinose ( 0.02 % ) and the other half received no treatment . 
The cultures were allowed to grow for 2 h. All cultures were then aliquoted into a 96-well plate , 200 μL per well . 
Plates were then read on a Bio-Tek plate reader . 
Cell density and arbitrary light units were measured . 
Bioluminescence was standardized for cell density by dividing light units by the OD600 absorbance . 
Each value represents the mean of 3 experiments . 
2.6. RNA isolation and sequencing for RNA-Seq
Strains DH10B ( pAR ) and DH10B ( pARΔacr2 ) were grown until an OD600 of 0.5 was achieved . 
Cells were pelleted and RNA was purified using a commercially available RNA extraction kit ( Qiagen ) . 
Treatments were included to remove DNA contamination ( Qiagen ) and ribosomal RNA ( MicrobExpress , Ambion ) . 
Two biological replicates for each strain were pooled for paired-end library sequencing ( either 50 or 100 bp reads ) via Illumina Genome Analyzer II at the University of Minnesota Genomics Center . 
All of the sequencing data are publically available under the NCBI BioProject ID PRJNA273283 . 
2.7. RNA-Seq analysis
All Perl scripts and other computational biology resources used in this study can be found at https://github.com/kevinslang . 
cDNA reads were first trimmed so that the quality at each base position was above 30 and then mapped to the appropriate genome or plasmid sequence ( for pAR060302 , GenBank accession no . 
NC_012692 , for DH10 , the E. coli K-12 MG1655 published sequence was used GenBank accession no . 
NC_000913 ) Read mapping was done using BOWTIE ( Langmead et al. , 2009 ) . 
For each host , transcriptome maps of pAR060302 were constructed using Circos ( Krzywinski et al. , 2009 ) . 
To achieve this , a table was generated containing the average number of reads mapped per 250 bp of plasmid sequence . 
Each average was then normalized per 1 million total reads in the cognate sequence library . 
These averages were then log transformed and plotted as a line plot . 
For statistical testing of differentially expressed genes , the total number of reads mapped to each coding sequence ( CDS ) was calculated using Perl . 
These values were then analyzed using the R package EdgeR ( Robinson et al. , 2010 ; Robinson et al. , 2011 ) . 
We conservatively estimated the dispersion a 
0.001 . 
A fold-change cutoff of N2 or b − 2 and an adjusted p-value of b0 .05 were used to define significantly differentially expressed genes . 
2.8. ChAP-Seq experiments
K12 ( pARK01 , pBacr26xHis ) was grown overnight in LB + Cm + Amp at 37 °C with shaking ( 200 RPM ) . 
The overnight culture was used to inoculate a new 10 mL LB + Cm + Amp culture ( 1:100 dilution ) . 
This was grown for 2 h at 37 °C with shaking ( 200 RPM ) . 
Arabinose was added to a final concentration of 0.02 % and the culture was incubated for 2 additional hours . 
Cells were fixed with the addition of formalin to a final concentration of 1 % . 
Fixed cells were incubated at RT for 20 min . 
The formalin was quenched by addition of glycine to a final concentration of 0.5 M and incubation for 5 min at RT. . 
Cells were collected by centrifugation at 8000 RPM for 5 min . 
The cell pellet was washed 1 × with 10 mL of cold PBS and collected again . 
The resulting pellet was resuspended in 1 × lysis solution ( MagneHIS Kit ™ , Promega ) . 
The cells were sonicated on ice at 5 watts for 30 s intervals and 1 min rest times . 
This was done 10 times . 
The cell lysates were spun at 12,000 g for 5 min and the supernatant was moved to a 1.5 mL centrifuge tube containing 50 μL of MagneHIS particles . 
This suspension was taken through the MagneHIS kit protocol . 
After elution , cross-linked DNA was released via incubation at 65 °C for 18 h . 
The resulting DNA was purified with a Qiagen ™ PCR Clean up Kit and sent for paired-end sequencing using the Illumina MiSeq platform . 
The library generation and sequencing was done at the University of Minnesota Genomics Center . 
In total , 2 biological rep-licates were sequenced separately . 
As a negative control , two genomic DNA preparations of the same cells were done using a Qiagen ™ DNeasy ™ kit and ~ 400 ng of each was sent for sequencing . 
The reads that resulted from the MiSeq run were all trimmed to 50 bp from the 3 ′ end using the tool Trimmomatic ( Bolger et al. , 2014 ) . 
The trimmed reads were then mapped to either the pAR060302 sequence or the K-12 genome sequence using BWA using the default parameters ( Li and Durbin , 2009 ) . 
Reads that were mapped correctly were filtered using SamTools ( Li et al. , 2009 ) . 
The replicate libraries were then merged using SamTools . 
The resulting read alignments files were then analyzed using MACS ( Zhang et al. , 2008 ) . 
Peak summit coordinates were used to extract 200 bp of sequence surrounding them using BedTools ( Quinlan and Hall , 2010 ) . 
These sequences were combined into a multifasta file and submitted to MEME for motif analysis ( Bailey et al. , 2009 ) . 
ChAP-Seq read alignments were also subjected to read counting on 150 bp windows of the pAR060302 sequence for visualization in Circos ( Krzywinski et al. , 2009 ) . 
2.9. Protein purification and EMSA
The pBacr26xHis construct was transformed into the BL21 ( DE3 ) strain . 
The resulting strains were cultured in LB + AMP at 37 °C for 1 h with shaking ( 250 RPM ) . 
Arabinose was added to a final concentration of 0.2 % prior to growing the cultures for 5 h at 28 °C . 
Cells were spun at 4500 × g for 30 min , resuspended in 5 mL cell lysis buffer ( 20 mM Tris pH 8 , 500 mM NaCl , 5 mM imidazole , 5 mM β-mercaptoethanol ) and sonicated . 
The cellular debris was removed by centrifugation at 12,000 g for 15 min . 
200 μL of MagnePURE beads were added to the supernatant and these were incubated for 1 h on a rocking platform , washed twice with high salt washing buffer ( 20 mM Tris pH 8 , 1 M NaCl , 5 mM β-mercaptoethanol ) and twice with low salt washing buffer ( 20 mM Tris pH 8 , 0.5 M NaCl , 30 mM imidazole , 5 mM β-mercaptoethanol ) , and then eluted with 100 μL elution buffer ( 20 mM Tris pH 8 , 500 mM NaCl , 500 mM imidazole ) . 
Pefabloc ™ ( Roche ) and glycerol were added to final concentrations of 1 mg/mL and 5 % , respectively . 
Purified proteins were analyzed by SDS-PAGE gel electrophoresis and stored at − 80 °C . 
Two DNA fragments were used for EMSA analysis . 
The region upstream of acr1 ( ~ 480 bp ) and the floR promotor ( ~ 150 bp ) . 
These fragments were amplified by PCR using the primers Orf184-EMSA-F and Orf184-EMSA-R ( acr1 ) , and Flo-F and Flo-R ( floR ) . 
The amplicons were purified using a PCR cleanup kit ( Qiagen ) . 
Various concentrations of purified Acr2 were incubated with 10 nM DNA in binding buffer ( 15 mM HEPES pH 7.9 , 40 mM KCl , 1 mM EDTA , 1 mM DTT , 5 % glycerol ) for 30 min . 
The reactions were separated by gel electrophoresis for 2.5 h at 70 V on a 7.5 % native polyacrylamide gel at 4 °C ( buffered with Tris glycine pH 8.0 ) . 
Gels were stained with SYBR Green for 20 min at room temperature , washed twice with ddH2O , and DNA complexes were visualized with ultraviolet light . 
2.10 . 
Development of acr1-lacZ fusion construct and detection of LacZ activity 
Recombineering was used to replace the E. coli chromosomal genes lacY and lacA with an FRT-neo-FRT cassette . 
The lacZ gene , starting at the 4th codon was then amplified , along with the FRT-neo-FRT cassette with primers that contained 40 bp of homology with acr1 and the region upstream of acrC . 
This amplicon was used to replace acr1-acrC starting with the 4th codon of acr1 in strain MC4100 ( pARK01 , pSIM5-Tet ) via recombineering and the neo-FRT cassette was removed by introduction of pCP20 expressing the FLP recombinase . 
In the same strain , acr2 was disrupted via recombineering . 
These reporter cells were then transformed with constructs over expressing Acr2 and the relevant Acr2 variants . 
They were struck onto LB + Amp + Xgal plates with and without arabinose . 
3. Results
3.1 . 
Acr2 is an H-NS-like protein that represses conjugative transfer of IncA/C plasmids 
To further investigate IncA/C plasmid conjugation , we developed a plasmid variant by systematically deleting the tetracycline and betalactam resistance genes ( see Materials and Methods ) to generate pARK01 . 
We demonstrate that this variant is no different in terms of conjugation frequency than the wild-type plasmid ( Fig . 
S1A ) . 
Recently , several transcriptional regulators on the IncA/C plasmid pVCR94ΔX were characterized in terms of their ability to regulate conjugative transfer ( Carraro et al. , 2014b ) . 
However , pVCR94ΔX was isolated from a Vibrio isolate and a deletion of unknown content was made in its development for genetic studies ( Carraro et al. , 2014a ) . 
To be certain that the functions were conserved in our IncA/C plasmid , pAR060302 ( isolated from an E. coli ) , we repeated several experiments using mutants of pAR060302 . 
Our results were mostly consistent with what was previously found ( Fig . 
S1C and D ) , which indicated that pAR060302 was regulated in a similar matter to pVCR94ΔX . 
The putative protein encoded by orf183 shares homology only with hypothetical proteins . 
Because of its proximity to the other predicted transcriptional regulators , we wondered if it may be involved in regulating conjugative transfer ( Fig . 
S1B ) . 
Deletion of this gene does result in a slight increase in transconjugants , however we could not rule out that this was due to a polar effect of the deletion ( Fig . 
S1C ) . 
Given the small magnitude of the effect , the putative protein produced by orf183 is unlikely to be a major contributor to repression of conjugation . 
In our experiments , deletion of acr1 did not alter the frequency of conjugative transfer of pARK01 . 
However , given the strong evidence previously reported on acr1 ( Carraro et al. , 2014b ) , we acknowledge that this could be due to differences in our experimental set up . 
To further investigate Acr2 , we generated an Acr26xHis construct and confirmed its ability to complement an acr2 deletion mutant by conducting transfer experiments . 
Plasmids lacking acr2 exhibit a 10-fold increase in conjugation frequency compared to wild-type ( Fig. 1 ) . 
The observation , from our data and that of other 's ( Carraro et al. , 2014b ) , that Acr2 negatively regulates conjugation corroborated our RNA-Seq results comparing WT and Δacr2 plasmids . 
Nearly all of the genes that are predicted to be involved in conjugative transfer wer significantly up-regulated ( at least 2-fold and p-value b0 .05 ) in the Δacr2 strain ( Fig. 2 , 4th ring ) . 
In addition to genes involved with transfer being up-regulated , the region encoding many putative hypothetical proteins and putative phage-like proteins ( bp positions ~ 82 k -- 100 k ) were also up-regulated in our experiment . 
Again these results are congruent with what has previously been reported , demonstrating that pARK01 is regulated in the same manner as other IncA/C plasmids and that our Acr26xHis construct is functionally analogous to the wild-type protein . 
3.2. Characterization of Acr2 binding sites
We performed a ChAP-Seq experiment in vivo to better understand where Acr2 binds on both the plasmid and host bacterial chromosome . 
Using nickel affinity chromatography to pull down Acr26xHis , cross-linked DNA was extracted and subsequently sequenced . 
Analysis of the ChAP-Seq data revealed that Acr26xHis binds to several loci on the plasmid ( Fig. 2 , rings 1 and 2 ) . 
Some of the ChAP peaks overlapped with genes of unknown function , such as near base pair coordinates 19 k , 45 k , 140 k , and 145 k . 
The regions that showed greatest binding were the entire ISEcp1 region , the regions within the traFHG and the region upstream of acr1 . 
Binding upstream of acr1 would suggest a direct repression of the operon encoding the positive regulators acrDC . 
Our ChAP data suggests that Acr26xHis binds to chromosomal DNA sequences . 
However , like other H-NS-like proteins , the interactions are not easily interpretable . 
For example , we only found 3 instances of binding of genes that were subsequently found to be significantly differentially expressed in our RNA-Seq data ( Fig. 3 ) . 
Interestingly , all three of these genes have functions associated with metabolism , most notably the glcB gene which was found previously as being up regulated in E. coli carrying pAR060302 , compared to cells lacking the plasmid ( Lang and Johnson , 2015 ) . 
We used MACS ( Model-based Analysis of ChAP-Seq ) ( Zhang et al. , 2008 ) to determine significantly enriched peaks in DNA pulled down with Acr26xHis . 
The sequence surrounding the summits of each of the significant ChAP peaks found by MACS was submitted to MEME ( Multiple Em for Motif Elicitation ) to determine if there were common motifs within the significant peaks . 
The motif discovered in a majority of the peaks is A + T rich , similar to that of other H-NS-like proteins ( Fig. 4B ) . 
Electromobility shift assays ( EMSA ) were used to confirm that Acr26xHis binds to the DNA fragment upstream of acr1 containing the motif , but not a different A + T rich promoter sequence located upstream of the floR gene on plasmid pAR060302 ( Fig. 4A ) . 
Analysis using the tool Bprom ( Solovyev and Salamov , 2011 ) to computationally predict bacterial promoters shows a predicted transcriptional start site 287 bp upstream of the start codon of acr1 ( Fig. 4C ) . 
These results demonstrate that Acr2 directly represses transcription of the operon containing acrDC by binding to the sequence upstream of acr1 and that there is sequence specificity to Acr2 binding regardless of A + T content . 
3.3. The C-terminal domain of Acr2 is crucial for its activity
The C-terminal domain of Acr2 is similar to that of other H-NS-like proteins ( Fig. 5A ) . 
It contains an AT-hook motif , with conserved amino acids Q/RGR , that has been shown in structure studies to be critical to contact with DNA ( Gordon et al. , 2011 ) . 
We examined the possibility that this motif was important for Acr2 activity by making mutations in this region using recombination with a ssDNA substrate ( see Materials and methods ) . 
An in-frame deletion of the codons for QGRRPD ( ΔAT-hook ) resulted in abolishment of the ability of Acr2 to repress conjugative transfer ( Fig. 5B ) . 
Although substitution of alanine for glutamine at position 116 resulted in increased transfer frequency , it was not as a dramatic as the increase observed for alanine substitutions for the arginine residues at positions 118 and 119 ( Fig. 5B ) . 
Given that these substitutions could have led to proteins targeted for degradation , which could also explain the acr2 − phenotypes observed , we cloned each of these Acr2 variants in an arabinose inducible vector . 
We used these constructs to test the ability to repress the LacZ activity of an acr1-lacZ fusion construct ( Fig. 5C ) . 
Only the WT Acr2 and the Q116A variant were able to repress LacZ activity . 
The Q116A result contrasts with that of the conjugation experiment . 
In the conjugation experiment the Acr2 variants were under native promotion . 
This is probably much lower than that achieved by arabinose induction of the pBAD vector , however , we can not rule out the possibility of degradation . 
Taken together , these results suggest that the QGRR motif is critical for the ability of Acr2 to repress the promoter upstream of acr1 . 
4. Discussion
4.2. Acr2 is an H-NS-like protein with a conserved DNA binding motif
Our results demonstrate that , like other H-NS homologs , Acr2 binds A + T rich DNA . 
Our in vitro results suggest that Acr2 binds in a sequence specific manner . 
It has been proposed that the amino acid composition in the AT-hook motif is the mechanism by which different H-NS homologs distinguish between chromosomal and horizontally acquired DNA ( Gordon et al. , 2011 ) . 
It is thought that the number or arrangement of positively charged amino acids , such as arginine , determine the DNA recognition by H-NS-like proteins . 
Acr2 shares a DNA binding motif with that of other H-NS homologs and our results show that this set of amino acids is critical for its function as a repressor . 
When we removed or altered this DNA binding motif from Acr2 , we found that , in some cases , the conjugation frequency increased higher than that of an Δacr2 mutation . 
This result could be explained in two ways : 1 ) Acr2 variants could be imparting a dominant negative effect on other negative regulators , such as Acr1 . 
Or 2 ) Acr2 variants could bind the promoter upstream of acr1 in such a way as to promote transcription . 
The AT-hook motif in Acr2 contains an additional arginine residue ( R119 ) that is not shared with other H-NS homologs . 
We have shown that this arginine is important for the function of Acr2 . 
This difference might result in Acr2 having a distinct propensity to bind specific regions of IncA/C plasmids to accomplish specific tasks , such as repressing conjugation . 
4.3. Other roles for Acr2
We have generated evidence that Acr2 might have other roles in the biology of IncA/C plasmids . 
Our ChAP-Seq data suggest that Acr2 binds nearly the entire ISEcp1 element that carries the blaCMY-2 gene . 
Comparative genomics studies have demonstrated that the acquisition of this transposon is a recent event in the evolution of IncA/C plasmids ( Fernández-Alarcón et al. , 2011 ; Fricke et al. , 2009 ; Harmer and Hall , 2014 ) . 
Could Acr2 bind newly acquired mobile elements within IncA/C plasmid backbones ? 
Given the propensity for chromosomally encoded H-NS to bind genomic islands , it seems plausible ( Navarre et al. , 2006 ) . 
Furthermore , Acr2 binds to the 3 ′ end of the rhs gene located downstream of the class 1 integron ( bp coordinate ~ 146,000 ) . 
The rhs genes have repeat sequence elements that implicates them in genome shuffling ( Lin et al. , 1984 ) . 
Recent work from Harmer et al. suggests that the rhs homolog carried on IncA/C plasmids plays an important role in the diversity of newly integrated mobile elements ( Harmer and Hall , 2014 ) . 
Binding by Acr2 may inhibit the ability for recombination to happen at the rhs locus . 
H-NS has recently been shown to play an important role in the evolution of Salmonella , providing stability for its genomic islands ( Ali et al. , 2014 ) . 
Given broad spatial distribution of highly similar variants of IncA/C plasmids , Acr2 binding of rhs could be a mechanism driving IncA/C plasmid evolution . 
Our ChAP-Seq experiment also yielded evidence that Acr2 binds chromosomal DNA . 
It is unclear what the true role of this binding is as nearly all genes bound showed no differential expression in RNA-Seq experiments . 
We can not rule out that these peaks might be an artifact of overexpressing Acr2 . 
The lack of differentially expressed genes could also be explained by the presence of chromosomally encoded nucleoid associated proteins , such as H-NS , IHF and others . 
It will take more sophisticated experiments to tease out the meaning behind these results . 
There were three bound genes , however , that were differentially expressed in our RNA-Seq data . 
All three have to do with different metabolic pathways . 
Interestingly , one of the genes was glcB , which is involved in the glyoxylate bypass pathway . 
We previously found this pathway to be modulated in several different host bacteria upon acquisition of IncA/C plasmids ( Lang and Johnson , 2015 ) . 
This lends some credence to the possibility that IncA/C plasmids encode the ability to specifically alter host metabolic pathways to improve fitness of plasmid carrying cells . 
This presents an almost phage-like scenario where the plasmid co-ops the host bacterium to become adept at carrying and disseminating the plasmid . 
It is an interesting area of study that future work must explore . 
Supplementary data to this article can be found online at http://dx . 
doi.org/10.1016/j.plasmid.2016.07.004 . 
Acknowledgments
The authors would like to thank Dr. Jeff Gralnick ( University of Minnesota ) , Dr. Fitnat Yildiz ( UC Santa Cruz ) , and Dr. Don Court ( NIH ) for sharing of strains . 
The authors thank Dr. William Navarre ( University of Toronto ) for sharing purified H-NS and EMSA technical assistance . 
Data analysis was carried out using tools available through the Minnesota Supercomputing Institute at the University of Minnesota . 
The primary author , KSL , was supported by a fellowship from the United States Department of Agriculture National Institute of Food and Agriculture grant no. 2013-67011-21276 . 
TJJ was supported through funding from the University of Minnesota College of Veterinary Medicine Signature Programs .