The Bacteriophage T4 MotB Protein, a DNA-Binding Protein, Improves Phage Fitness

The lytic bacteriophage T4 employs multiple phage-encoded early proteins to takeover the Escherichia coli host. However, the functions of many of these proteins are not known. In this study, we have characterized the T4 early gene motB, located in a dispensable region of the T4 genome. We show that heterologous production of MotB is highly toxic to E. coli, resulting in cell death or growth arrest depending on the strain and that the presence of motB increases T4 burst size 2-fold. Previous work suggested that motB affects middle gene expression, but our transcriptome analyses of T4 motBam vs. T4 wt infections reveal that only a few late genes are mildly impaired at 5 min post-infection, and expression of early and middle genes is unaffected. We find that MotB is a DNA-binding protein that binds both unmodified host and T4 modified [(glucosylated, hydroxymethylated-5 cytosine, (GHme-C)] DNA with no detectable sequence specificity. Interestingly, MotB copurifies with the host histone-like proteins, H-NS and StpA, either directly or through cobinding to DNA. We show that H-NS also binds modified T4 DNA and speculate that MotB may alter how H-NS interacts with T4 DNA, host DNA, or both, thereby improving the growth of the phage.


Introduction
An increase in the occurrence of antibiotic-resistant bacteria has sparked an interest in phage-host interactions [1,2]. As bacteriophages have evolved multiple mechanisms to take over their hosts, hosts have responded with mechanisms to thwart takeover. An understanding of these mechanisms would be beneficial to the development of new antibacterial strategies. However, the lack of information about the functions of a large portion of phage genes [3] has hampered this investigation.
The lytic bacteriophage T4 infects Escherichia coli, resulting in rapid cell lysis after~20 min at 37 • C. Despite the use of T4 as a model organism for decades, the functions of many genes, especially those expressed early during infection, remain unknown [4]. As T4 does not encode its own RNA polymerase (RNAP), it must use the host's RNAP to program its temporal pattern of early, middle, and late gene expression (reviewed in [5]). T4 directs RNAP to early, middle and late promoters by encoding factors that change the specificity of RNAP as infection proceeds. Early RNAs, which are expressed immediately after infection, arise from T4 early promoters (Pe's). Pe's, whose activation does not require phage-encoded factors, contain a very strong match to the consensus sequences of host promoters, allowing them to compete with host DNA for the same pool of RNAP. T4 middle transcripts are expressed approximately 1 to 2 min post-infection at 37 • C from middle promoters (Pm's) and
Wild-type T4D + (T4 wt), T4 amG1 (T4 motA am ) [22], and T4 motB am were used for infections. T4 motB am was obtained by recombination of pTE103-motB am into the T4 genome. BL21(DE3)/ pTE103-motB am was infected with T4 wt at a multiplicity of infection (MOI) of 1 during exponential phase and incubated for 75 min. Cells were lysed with chloroform and resulting phage were titered on NapIV S. Plaques were screened for the presence of the amber mutation by hybridization of a 32 P-labeled probe containing either the wt or mutant sequence as described [23]. Potential mutant plaques were used to generate phage stocks by infecting NapIV S. Phage stocks containing the amber mutation were subjected to a subsequent round of single plaque selection to ensure a homogenous stock of T4 motB am . To confirm the presence of the amber mutation, the motB gene was amplified by PCR then sequenced by Macrogen (Rockville, MD, USA).

MotB Purification (Method I)
MotB containing a C-terminal intein tag with a chitin binding domain (MotB-Intein/CBD) was isolated and purified from BL21(DE3)/pLysE containing pTXB1-MotB. Overnight cultures were grown in LB containing 25 µg/mL chloramphenicol and 100 µg/mL carbenicillin. Cells were diluted to an OD 600~0 .1 in LB containing 25 µg/mL chloramphenicol and 100 µg/mL carbenicillin and then grown at 25 • C with shaking at 250 rpm. At an OD 600 between 0.3 and 0.4, synthesis of MotB-Intein/CBD was induced by the addition of 4 mM IPTG (final concentration) for 2 h. Cells were harvested by centrifugation at 13,000× g for 10 min and stored at −80 • C. Unless otherwise noted, the following procedures were performed on ice or at 4 • C. Cells were resuspended in CB Buffer (20 mM HEPES-OH (pH 8.5), 50 mM NaCl, 1 mM EDTA, 0.01% (v/v) Triton X-100) containing 1 mM benzamidine, then lysed by sonication until OD 600 was reduced~3-fold. Clarified supernatant was obtained by centrifugation at 15,000× g for 30 min followed by filtration through a 0.4 µm syringe filter. Chitin resin (New England BioLabs) equilibrated with CB Buffer (5 mL slurry per 125 mL starting culture) was added to the supernatant, and the suspension was gently rocked overnight. Resin was transferred to a 30 mL disposable column (Bio-Rad, Hercules, CA, USA) and then washed with~25 column volumes of CB Buffer containing 1 M NaCl at a flow rate of <2 mL/min (gravity drip) until the A 260 reached a minimum. The resin was subsequently washed with an additional 7 column volumes of CB Buffer containing 500 mM NaCl. On-column cleavage of the Intein/CBD-tag was initiated by quickly washing the column with 3 column volumes of CB Buffer containing 500 mM NaCl and 1 mM dithiothreitol (DTT). After ≥40 h, MotB was eluted by the addition of 2 column volumes of CB Buffer containing 500 mM NaCl and 2 mL fractions were collected. Fractions containing MotB were identified by SDS-PAGE and then dialyzed into Buffer A (25 mM Tris-HCl (pH 8.0), 50 mM NaCl, 6 M deionized Urea). The denatured MotB was isolated after passage through a 1 mL HiTrap Q column (GE healthcare), in which MotB was in the flow-through fraction while residual nucleic acid bound to the column. MotB was then refolded using the dialysis procedure previously described for E. coli σ 70 renaturation [24] and stored at −20 • C in MotB storage buffer (50 mM Tris-HCl (pH 8.0), 1 mM EDTA, 50 mM NaCl, 0.01% v/v Triton X-100, 50% v/v glycerol, 0.1 mM DTT).

MotB Purification (Method II)
MotB containing a C-terminal His 6 -tag (MotB-His) was isolated and purified from TOP10F' containing pNW129-MotB-His. Overnight cultures were grown in LB containing 12 µg/mL tetracycline, 40 µg/mL kanamycin, and 0.05% (w/v) glucose. Cells were diluted to an OD 600~0 .1 in LB containing 12 µg/mL tetracycline and 40 µg/mL kanamycin and grown to an OD 600 around 0.3, when synthesis of MotB-His was induced by the addition of 0.2% (w/v) arabinose (final concentration) for 1 h. Cells were harvested by centrifugation at 13,000× g for 10 min and stored at −80 • C. Unless otherwise noted, the following procedures were performed on ice or at 4 • C. Cells were resuspended in 40 mL of Lysis Buffer (25 mM Tris-HCl (pH 8.0), 50 mM NaCl, 5 mM imidazole, 0.01% (v/v) Triton X-100, 1 mM benzamidine) per 1 L culture, lysed by sonication as described above and then centrifuged at 17,500× g for 20 min. The pellet, containing MotB-His, was resuspended in Lysis Buffer and centrifuged again at 17,500× g for 15 min. To recover MotB-His, the pellet was resuspended in 20 mL Binding Buffer (20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 5 mM imidazole, 6 M urea, 1 mM benzamidine, pH adjusted to 7.9) per 1 L culture, incubated for 1 h with gentle mixing, and then centrifuged at 17,500× g for 15 min. The supernatant, containing MotB-His, was then incubated with gentle mixing for a minimum of 1 h with 5 mL (per 1 L culture) Ni-NTA His-binding resin (Millipore, Burlington, MA, USA) that was pre-equilibrated with Binding Buffer. Resin was transferred to a 30 mL disposable column (Bio-Rad) then washed with four column volumes of Binding Buffer followed by six column volumes of Wash Buffer (25 mM Tris-HCl (pH 8.0), 500 mM NaCl, 60 mM imidazole, 6 M urea, pH adjusted to 7.9) at a flow rate of <2 mL/min (gravity drip). The major fraction of MotB-His eluted with 2 column volumes of Elution Buffer 1 (20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 200 mM imidazole, 6 M urea, pH adjusted to 7.9). Residual MotB-His was subsequently eluted with 2 column volumes of Elution Buffer 2 (20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 400 mM imidazole, 6 M urea, pH adjusted to 7.9). For both elution conditions, 2.5 mL fractions were collected. Fractions containing MotB-His, identified by SDS-PAGE, were refolded as described in Method I.

H-NS Purification
H-NS containing a C-terminal intein tag with a chitin binding domain (H-NS-Intein/CBD) was isolated and purified from BL21(DE3)/pLysE containing pTXB1-H-NS. Overnight cultures were grown in LB containing 25 µg/mL chloramphenicol and 100 µg/mL carbenicillin. Cells were diluted to an OD 600~0 .1 in LB containing 25 µg/mL chloramphenicol and 100 µg/mL carbenicillin and then grown at 25 • C with shaking at 250 rpm. At an OD 600 between 0.3 and 0.4, synthesis of H-NS-Intein/CBD was induced by the addition of 4 mM IPTG (final concentration) for 2 h. Cells were harvested by centrifugation at 13,000× g for 10 min and stored at −80 • C. Unless otherwise noted, the following procedures were performed on ice or at 4 • C. Cells were resuspended in 30 mL of H-NS Sonication Buffer (20 mM Tris-HCl (pH 8.0), 750 mM NaCl, 1 mM EDTA, 0.01% (v/v) Triton X-100, 1 mM Benzamidine) then lysed by sonication until OD 600 was reduced~3-fold. Clarified supernatant was obtained by centrifugation at 15,000× g for 30 min followed by filtration through a 0.45 µm syringe filter. Chitin resin (New England BioLabs) equilibrated with H-NS Sonication Buffer (5 mL slurry per 250 mL starting culture) was added to the supernatant and rocked overnight. Resin/cell lysate was transferred to a 30 mL disposable column (Bio-Rad) and washed with approximately 20 column volumes of CCB buffer (20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 1 mM EDTA, 0.01% (v/v) Triton X-100) at a flow rate of <2 mL/min (gravity drip). On-column cleavage of the Intein/CBD-tag was initiated by quickly washing the column with 3 column volumes of CCB buffer containing 50 mM DTT. After ≥40 h, H-NS was eluted by the addition of 3 column volumes of CCB buffer and 1 mL fractions were collected. Fractions containing H-NS were identified by SDS-PAGE and pooled. As this pool contained a small amount of intein tagged protein, it was passed over a new column containing 5 mL of fresh, equilibrated chitin slurry, and fractions from the flow-through plus 2 column volume washes were collected. The now pure H-NS was dialyzed into H-NS Storage Buffer (10 mM potassium phosphate (pH 7.5), 200 mM NaCl, 0.1 mM EDTA, 50% (v/v) glycerol) and stored at −20 • C.

Identification of Nucleic Acid Copurifying with MotB
MotB samples used to identify copurifying nucleic acid were purified using MotB purification Method I with the following modifications. Elution fractions from the chitin column that contained MotB as determined by SDS-PAGE were directly dialyzed into MotB Storage Buffer and stored at −80 • C. Protein was removed by chloroform:isoamyl alcohol (24:1) extraction, and sodium acetate and glycogen (Thermo Scientific, Waltham, MA, USA) were added to the aqueous phase at a final concentration of 496 mM and 0.17 mg/mL, respectively. The nucleic acid was then ethanol precipitated, resuspended in nuclease-free water, and quantified by NanoDrop 2000c (Thermo Scientific).
To identify nucleic acid, 1× DNase buffer (Ambion, Carlsbad, CA, USA) was added to~300-400 ng nucleic acid followed by the addition of either 2 units of RNase-free DNase I (Ambion) or 100 µg of DNase-free RNase A and incubation at 37 • C for 10 min. Treated samples and the untreated control were electrophoresed on 0.8% (w/v) agarose gels at 120 V for 45 min in 1× TAE (Quality Biological) at room temperature. Gels were stained with ethidium bromide for 1 min and visualized using UV illumination. Two independent replicates were performed.

DNase I Footprinting
Solutions were assembled in a total volume of 10 µL using 0.005 µM DNA (1 µL) and 0 to 3.2 µM of MotB (4 µL), as indicated, in the same buffer used for gel shift assays with oligonucleotides as described above. After 10 min of incubation at 37 • C, 0.8 units of DNase I (1.1 µL) was added to the reaction. Solutions were incubated for an additional 45 s at 37 • C and quenched with equal volumes of phenol. DNA was extracted, ethanol precipitated, and electrophoresed on 5% (w/v) polyacrylamide (19:1, acrylamide:bis), 7 M urea denaturing gels for 2500 volt-h in 0.5× TBE. After electrophoresis, gels were visualized by autoradiography and analyzed as described above.

Pull-Down Assay
BL21(DE3)/pLysE containing pTXB1-MotB or pTXB1 was grown in LB containing 25 µg/mL chloramphenicol and 100 µg/mL carbenicillin at 25 • C. MotB was isolated as described in Method I except that after binding to the chitin column, the resin was washed with 5 column volumes of CB buffer containing 500 mM NaCl followed by 5 column volumes of CB buffer containing 1 M NaCl. Proteins, isolated from the SDS-PAGE gel, were analyzed by mass spectrometry (Whitehead Institute, Cambridge, MA, USA).

Bacteriophage Fitness Assays
Overnight cultures were diluted in fresh LB to an OD 600~0 .1, grown to early log phase (OD 600 0.3), and then transferred to ice until ready to titer. For each titer,~50-100 phages in 100 µL LB were incubated with 200 µL of culture for 5 min at room temperature. Following addition of 2.5 mL 0.65% (w/v) Drake top agar at 45 • C [25], cells were plated on 1.5% (w/v) LB agar (Sigma) and incubated at 37 • C overnight. For infections using cells containing pNW129-MotB or pNW129, cultures were grown to an OD 600~0 .4 as described above and then protein synthesis was induced by the addition of 0.1% (w/v) arabinose (final concentration) for 20 min before collecting on ice.
Burst size experiments were performed as described [26] for single-step growth experiments with the following modifications. NapIV S and NapIV NS were grown to mid log phase (OD 600~0 .5). The NapIV S culture was transferred to ice for use as the plating strain. NapIV NS (900 µL) was transferred to a 1.5 microcentrifuge tube containing 100 µL of phage (MOI of~0.004) pre-warmed at 37 • C. At 10 min post-infection, the phage-culture sample was diluted 10,000-fold in LB warmed to 37 • C with gentle vortexing to mix and returned to 37 • C. For pre-burst (≤30 min) and post-burst (35 to 70 min) time points, 100 µL or 10 µL of the diluted phage culture, respectively, were added to 100 µL of NapIV S followed by the addition of 3 mL of 0.65% Drake top agar (at 50 • C). The cells were immediately plated on 1.5% (w/v) LB agar and incubated overnight at 37 • C. Burst size was calculated by dividing the average number of plaques for the post-burst time points (50 to 70 min) by the average number of plaques for pre-burst time points (15 to 30 min).

Purification of Total RNA
NapIV NS were grown to a cell density of~4 × 10 8 cells/mL (OD 600~0 .4) and then infected with either T4 wt or T4 motB am at a MOI of 10. As indicated, RNA was isolated at 1, 5 or 10 min post-infection using method II as described [27]. Briefly, cells were chemically lysed and nucleic acid was isolated by hot phenol extraction followed by ethanol precipitation. Samples were then treated with DNase, and total RNA was isolated by phenol extraction followed by ethanol precipitation. Total RNA used for RNA-seq was run on a Bioanalyzer using the Agilent RNA 6000 Nano Kit (Agilent Technologies, Santa Clara, CA, USA) to evaluate the sample quality [28]. RNA used for RT-qPCR and primer extensions was evaluated using a 0.8% (w/v) agarose gel stained with ethidium bromide.

Transcriptome Analysis
rRNA subtraction was performed with the bacterial RiboMinus Kit (Ambion) according to manufacturer instructions. cDNA was prepared using the NEBNext strand specific kit (New England BioLabs) according to manufacturer instructions for libraries with 300-450 bp insert size with the following modifications. Illumina adaptor sequences based on TruSeq HT Sample Prep Kits were purchased from Integrated DNA Technologies and used in the ligation step. TruSeq-1 and TruSeq-2 primer were used for PCR enrichment of adaptor ligated DNA. Library size was verified with a Bioanalyzer using an Agilent High Sensitivity DNA kit (Agilent Technologies). The concentration of each library was determined using the KAPA Library Quantification Kit (Roche, Basel, Switzerland) for Illumina platforms. Sequencing was performed by the NIDDK Genomics Core facility using a MiSeq system with the MiSeq 2 × 250 bp Sequencing Kit (Illumina, San Diego, CA, USA).
Quality trimming (max ambiguity = 2) and trimming of 15 nucleotides from the 5' terminal of all sequences were performed with CLC Genomics Workbench (version 8.5, Qiagen, Hilden, Germany). The quality of raw sequencing data was accessed in CLC Biomedical Workbench software (Qiagen). A bimodal distribution of GC-content was observed in the data set with peak maximums at 36% GC and 55% GC ( Figure S1). This distribution is consistent with the presence of both host-encoded (~55% GC) and T4-encoded (34.5% GC) transcripts within the data set.
Trimmed data were aligned to the E. coli reference genome (NC_012971.2) and an unmapped reads file was generated. Approximately 44-54% of total reads mapped to the host genome. A significant number of host-encoded transcripts is consistent with other transcriptome analysis of bacteriophage infections during early and middle stages of temporal expression [29,30]. The unmapped reads file, presumably containing sequences derived from the T4 genome, was subsequently aligned to the bacteriophage T4 reference genome (NC_000866.4). To mitigate the occurrence of multi-mapping reads, overlapping annotations of genes with alternative start sites or introns were resolved to a single annotation in the T4 reference genome to enable unique gene assignment by CLC Biomedical Workbench software. Differential expression experiments consisted of 3 replicates of NapIV NS infected with either T4 wt or T4 motB am . Statistical analysis was performed in CLC Genomics Workbench using the Baggerley's Test with False Discovery Rate (FDR) corrected p values. RNA-seq data is available in the NCBI database (GEO number GSE113425) and in Table S1.

RT-qPCR
RNA was reverse transcribed into cDNA using the iScript cDNA Synthesis Kit (BioRad), and target genes were amplified with iTaq Universal SYBR green supermix (BioRad) in a BioRad CFX95. The relative gene expression was determined with the comparative C t method using CFX manager software (Bio-Rad), also known as 2 −∆∆Ct [31]. Genes frd.2, nrdD, and motA, whose expression was unaffected in RNA-seq analysis, were used as endogenous controls. Primers were designed using Primer3web [32,33]; sequences are available upon request.

MotB is Toxic when Expressed in E. coli
Initial efforts to heterologously express motB, which had been cloned downstream of the arabinose-inducible promoter P BAD , in the E. coli B strain derivative BL21(DE3) were prevented by plasmid mutation(s) that eliminated the expression of motB. To mitigate this issue, all overnight cultures included low levels of glucose to reduce leaky expression from the P BAD promoter. To assess MotB toxicity, induction of MotB synthesis was done in both BL21(DE3) and in the E. coli K12 strain derivative TOP10F' by the addition of arabinose. In BL21(DE3), a rapid reduction in cell density, as monitored by the OD 600 , was observed approximately 90 min after induction (Figure 1a), indicating cell lysis. At 60 min, a small amount of induced protein that comigrated as MotB was detectible by SDS-PAGE (Figure 1a inset). MotB production in TOP10F' led to inhibition of cell growth around 60 min after induction without evidence of cell lysis (Figure 1b). For TOP10F' cells, a significant amount of MotB was produced at 60 min post-induction, indicating that the difference in toxicity was not due to a decrease in expression (Figure 1b inset). Genome-wide comparisons of the transcriptomes and proteomes of E. coli B and K-12 strains have shown extensive differences, which could account for the difference in MotB toxicity between the cell lines [37].

MotB is Toxic when Expressed in E. coli
Initial efforts to heterologously express motB, which had been cloned downstream of the arabinose-inducible promoter PBAD, in the E. coli B strain derivative BL21(DE3) were prevented by plasmid mutation(s) that eliminated the expression of motB. To mitigate this issue, all overnight cultures included low levels of glucose to reduce leaky expression from the PBAD promoter. To assess MotB toxicity, induction of MotB synthesis was done in both BL21(DE3) and in the E. coli K12 strain derivative TOP10F' by the addition of arabinose. In BL21(DE3), a rapid reduction in cell density, as monitored by the OD600, was observed approximately 90 min after induction (Figure 1a), indicating cell lysis. At 60 min, a small amount of induced protein that comigrated as MotB was detectible by SDS-PAGE (Figure 1a inset). MotB production in TOP10F' led to inhibition of cell growth around 60 min after induction without evidence of cell lysis (Figure 1b). For TOP10F' cells, a significant amount of MotB was produced at 60 min post-induction, indicating that the difference in toxicity was not due to a decrease in expression (Figure 1b inset). Genome-wide comparisons of the transcriptomes and proteomes of E. coli B and K-12 strains have shown extensive differences, which could account for the difference in MotB toxicity between the cell lines [37].

MotB Copurifies with DNA
MotB is highly basic with a predicted pKa of 9.23, suggesting that it might interact with nucleic acid. Furthermore, the protein (identified as p17.6) was previously found as part of a prereplicative DNA complex after T4 infection [18], again suggesting a possible interaction with DNA. Our first attempts to purify MotB were thwarted by high levels of nucleic acid that remained with the protein throughout purification. Consequently, we constructed a plasmid that produced MotB with an auto-cleavable C-terminal Intein/CBD tag in the hope that extensive salt washes while the protein was bound to the column would remove DNA/RNA. To monitor the level of nucleic acid during purifications, we measured the 260/280 nm absorbance ratio [38]. Highly purified MotB was obtained after absorption/elution from a chitin column (Figure 2, lane 1). However, even though the resin was washed extensively with buffer containing 1 M NaCl before protein elution, the purified fraction had a 260/280 ratio of between 1.21 and 1.35. To isolate the copurifying nucleic acid, an aliquot of the fraction was extracted with phenol and precipitated with ethanol ( Figure 2

MotB Copurifies with DNA
MotB is highly basic with a predicted pKa of 9.23, suggesting that it might interact with nucleic acid. Furthermore, the protein (identified as p17.6) was previously found as part of a prereplicative DNA complex after T4 infection [18], again suggesting a possible interaction with DNA. Our first attempts to purify MotB were thwarted by high levels of nucleic acid that remained with the protein throughout purification. Consequently, we constructed a plasmid that produced MotB with an autocleavable C-terminal Intein/CBD tag in the hope that extensive salt washes while the protein was bound to the column would remove DNA/RNA. To monitor the level of nucleic acid during purifications, we measured the 260/280 nm absorbance ratio [38]. Highly purified MotB was obtained after absorption/elution from a chitin column (Figure 2, lane 1). However, even though the resin was washed extensively with buffer containing 1 M NaCl before protein elution, the purified fraction had a 260/280 ratio of between 1.21 and 1.35. To isolate the copurifying nucleic acid, an aliquot of the fraction was extracted with phenol and precipitated with ethanol ( Figure 2 Figure S2b). However, it is important to note that the reduced toxicity of the MotB-Intein/CBD could also be due to growth and induction at 25 C instead of 37 C. The culture was grown at 25 C to improve folding and solubility of the fusion protein as recommended by the IMPACT kit instruction manual (New England BioLabs). Both proteins were highly purified ( Figure S3a and S3b). Subsequent assays detailed below indicate that both the refolded His6-tagged MotB and the refolded wt MotB generated after intein cleavage have similar and strong DNA-binding activity.

MotB Binds DNA with No Detectable Sequence Specificity
To examine the DNA-binding properties of MotB in detail, we tested the ability of MotB (or MotB-His) to bind different dsDNA oligonucleotides: (1) a 37 bp fragment containing positions −29  Figure  S2a) than wt MotB when expressed in E. coli, whereas MotB-His retained comparable toxicity ( Figure  S2b). However, it is important to note that the reduced toxicity of the MotB-Intein/CBD could also be due to growth and induction at 25 • C instead of 37 • C. The culture was grown at 25 • C to improve folding and solubility of the fusion protein as recommended by the IMPACT kit instruction manual (New England BioLabs). Both proteins were highly purified ( Figure S3a and S3b). Subsequent assays detailed below indicate that both the refolded His 6 -tagged MotB and the refolded wt MotB generated after intein cleavage have similar and strong DNA-binding activity.

MotB Binds DNA with No Detectable Sequence Specificity
To examine the DNA-binding properties of MotB in detail, we tested the ability of MotB (or MotB-His) to bind different dsDNA oligonucleotides: (1) a 37 bp fragment containing positions −29 to +8 of the promoter (Pl 8 ) for the T4 late gene 8, whose expression decreased 2.1-fold in the motB am infection 5 min post-infection as detailed below (dsP8, Figure 3a); (2) a 40 bp fragment containing positions +8 to +47 of the E. coli proU operon (ProU-40, Figure 3c); and (3) a random 32 bp oligonucleotide that is not present in either the host or viral genome (Oligo-32, Figure 3d). All three ds oligonucleotides were retarded in EMSAs, with similar K d(app) 's of 0.04 to 0.1 µM. It should be noted a comparison of MotB vs. MotB-His binding (Figure 3a,c) indicated that both proteins bind with similar affinity.  To assess whether fragment length affected MotB binding, we compared the affinity of MotB-His for ds oligonucleotides of various sizes containing the Pl8 promoter motif ( Figure S4a): 37 bp (Figure 3a), 23 bp ( Figure S4b), and 20 bp ( Figure S4c). While the 37 bp and 23 bp fragments yielded similar Kd(app)'s (0.1 ± 0.03 µM and 0.09 ± 0.03 µM, respectively), MotB affinity for the 20 bp fragment decreased ~10-fold, suggesting that a fragment size greater than ~20 bp is needed for optimal binding. Finally, we also tested the ability of MotB-His to bind ssDNA, using the 37 nucleotides top strand of the ds P8 fragment (Figure 3b). In this case, affinity decreased ~6-fold, indicating a MotB preference for ds DNA.  Figure S4c). While the 37 bp and 23 bp fragments yielded similar K d(app) 's (0.1 ± 0.03 µM and 0.09 ± 0.03 µM, respectively), MotB affinity for the 20 bp fragment decreased 10-fold, suggesting that a fragment size greater than~20 bp is needed for optimal binding. Finally, we also tested the ability of MotB-His to bind ssDNA, using the 37 nucleotides top strand of the ds P8 fragment (Figure 3b). In this case, affinity decreased~6-fold, indicating a MotB preference for ds DNA.

DNase I Footprints Indicate that MotB Can Bind Nonspecifically Along the DNA
To further investigate DNA specificity, we performed DNase I footprinting, using an unmodified 218 bp DNA fragment containing the sequence surrounding Pl 8 from positions −143 to +75 (T4 map units 85,671 to 85,888; Figure 4a,b). Thus, this region included the sequences of the P8 oligonucleotides used for EMSAs (Figure 4c). No significant protection was seen until a MotB:bp DNA ratio of 1.5:1 (MotB at 16 pmol), when there was significant protection throughout the fragment. At a 2-fold higher concentration of MotB (32 pmol), the DNA was essentially fully protected with no detectable cleavage except for a faint band corresponding to cleavage at position +28. This result suggests that there is no preferred sequence binding motif for MotB within this fragment and that once the level of MotB relative to the DNA is high enough, MotB binding "spreads" along the entire region of the fragment.

DNase I Footprints Indicate that MotB Can Bind Nonspecifically Along the DNA
To further investigate DNA specificity, we performed DNase I footprinting, using an unmodified 218 bp DNA fragment containing the sequence surrounding Pl8 from positions −143 to +75 (T4 map units 85,671 to 85,888; Figure 4a,b). Thus, this region included the sequences of the P8 oligonucleotides used for EMSAs (Figure 4c). No significant protection was seen until a MotB:bp DNA ratio of ~1.5:1 (MotB at 16 pmol), when there was significant protection throughout the fragment. At a 2-fold higher concentration of MotB (32 pmol), the DNA was essentially fully protected with no detectable cleavage except for a faint band corresponding to cleavage at position +28. This result suggests that there is no preferred sequence binding motif for MotB within this fragment and that once the level of MotB relative to the DNA is high enough, MotB binding "spreads" along the entire region of the fragment.

H-NS and StpA Copurify with MotB
To determine if MotB interacts with host proteins, pull-down assays were performed using a lysate from cells producing the MotB-Intein/CBD. To control for host proteins that might interact with the Intein/CBD tag, we used cells producing only Intein/CBD as a negative control. Cell lysates

H-NS and StpA Copurify with MotB
To determine if MotB interacts with host proteins, pull-down assays were performed using a lysate from cells producing the MotB-Intein/CBD. To control for host proteins that might interact with the Intein/CBD tag, we used cells producing only Intein/CBD as a negative control. Cell lysates were applied to chitin resin and extensively washed with high salt buffer. Resin aliquots were then removed, and proteins were detected by SDS-PAGE. For the control, the major species detected was Intein/CBD protein ( Figure 5, lane 10). As expected for the MotB-Intein/CBD construct, a large amount of MotB-Intein/CBD and a small amount of Intein/CBD were detected with the lysate (Figure 5, lane 18). However, in addition, two other novel species, indicated by the black arrows, were observed. Mass spectrometry indicated that the faint, higher molecular weight band contained MotB and Intein, consistent with partial degradation of our target protein and tag. The much more intense, lower molecular weight band was determined to contain the bacterial proteins H-NS and StpA with 91% and 70% sequence coverage, respectively. H-NS was approximately 6 times more abundant in the sample, which is consistent with significantly lower level of StpA compared to H-NS in E. coli [39]. Because MotB-Intein/CBD was not denatured before application to the chitin column, DNA was also present as shown in Figure 2. We conclude that MotB interacts with H-NS (and StpA) either directly or indirectly through the presence of DNA. were applied to chitin resin and extensively washed with high salt buffer. Resin aliquots were then removed, and proteins were detected by SDS-PAGE. For the control, the major species detected was Intein/CBD protein ( Figure 5, lane 10). As expected for the MotB-Intein/CBD construct, a large amount of MotB-Intein/CBD and a small amount of Intein/CBD were detected with the lysate ( Figure  5, lane 18). However, in addition, two other novel species, indicated by the black arrows, were observed. Mass spectrometry indicated that the faint, higher molecular weight band contained MotB and Intein, consistent with partial degradation of our target protein and tag. The much more intense, lower molecular weight band was determined to contain the bacterial proteins H-NS and StpA with 91% and 70% sequence coverage, respectively. H-NS was approximately 6 times more abundant in the sample, which is consistent with significantly lower level of StpA compared to H-NS in E. coli [39]. Because MotB-Intein/CBD was not denatured before application to the chitin column, DNA was also present as shown in Figure 2. We conclude that MotB interacts with H-NS (and StpA) either directly or indirectly through the presence of DNA.

MotB Interacts with Both Unmodified and Modified GHme-C DNA
Because MotB was purified from E. coli in the absence of a T4 infection, we deduced that the copurifying DNA present before denaturation did not contain the modification found on T4 DNA, GHme-C. To investigate if MotB could bind to the modified DNA, we performed EMSAs with T4 wt DNA, pretreated with SspI restriction enzyme, as well as unmodified lambda DNA pretreated with HindIII restriction enzyme as a control. Both DNAs were retarded by the presence of MotB, indicating that the protein binds both phage modified and unmodified DNA ( Figure 6). In both cases, some DNA was severely retarded, with a fraction that barely entered the gel.  10 and 18). Lane 1 corresponds to SeeBlue Plus2 protein standard; corresponding molecular weights are shown on the left. Bands corresponding to MotB-Intein/CBD and Intein/CBD are indicated. Species indicated with arrows were identified by mass spectrometry. The~36 kDa protein labeled as "degraded construct" was identified as partially degraded MotB-Intein/CBD. The~16 kDa protein was identified as H-NS and StpA.

MotB Interacts with Both Unmodified and Modified GHme-C DNA
Because MotB was purified from E. coli in the absence of a T4 infection, we deduced that the copurifying DNA present before denaturation did not contain the modification found on T4 DNA, GHme-C. To investigate if MotB could bind to the modified DNA, we performed EMSAs with T4 wt DNA, pretreated with SspI restriction enzyme, as well as unmodified lambda DNA pretreated with HindIII restriction enzyme as a control. Both DNAs were retarded by the presence of MotB, indicating that the protein binds both phage modified and unmodified DNA ( Figure 6). In both cases, some DNA was severely retarded, with a fraction that barely entered the gel. As expected, addition of H-NS to unmodified λ/HindIII DNA resulted in retardation of all the DNA fragments consistent with H-NS binding nonspecifically to DNA ( Figure 6) [40]. Strikingly, the resolution of specific DNA fragments of T4/SspI is completely lost. Previous work has shown that H-NS can bind to T4 GT7 DNA, which replaces GHme-C with unmodified cytosine [41]. To our knowledge, this is the first study that has shown that H-NS can bind to GHme-C modified T4 DNA ( Figure 6). Furthermore, addition of both MotB and H-NS together resulted in a more complete shift of either the λ/HindIII or T4/SspI DNA. At this point, we cannot distinguish whether H-NS and MotB bind to separate populations of DNA or bind concurrently to the same DNA. However, taking into account the length of DNA, it is conceivable that both MotB and H-NS could bind to the same fragment.

T4 motB am Reduces Burst Size and Overexpression of motB Increases Plaque Size
A T4 motB am mutant was constructed to investigate the functional phenotype of motB during T4 infection of non-suppressing E. coli. We observed ~2-fold reduction in burst size for the T4 motB am infection compared to T4 wt (Figure 7a), indicating that motB contributes to phage fitness although no difference in plaque size was observed between T4 wt and T4 motB am (Figure 7b). To ask whether an excess of MotB would improve phage growth, we infected cells expressing motB from a plasmid. Because of toxicity, synthesis of MotB was induced by the addition of 0.1% (w/v) arabinose for only 20 min before plating on LB agar containing no arabinose, thereby producing a low level of MotB prior to infection. In fact, the level of produced MotB was too low to be detected by SDS-PAGE. Nevertheless, prior production of MotB resulted in larger plaques when infected with T4 wt, T4 motB am , or T4 motA am (Figure S5), which normally yields very tiny plaques since MotA is required for Pm activation ( [22,34,35], reviewed in [5]). As expected, addition of H-NS to unmodified λ/HindIII DNA resulted in retardation of all the DNA fragments consistent with H-NS binding nonspecifically to DNA ( Figure 6) [40]. Strikingly, the resolution of specific DNA fragments of T4/SspI is completely lost. Previous work has shown that H-NS can bind to T4 GT7 DNA, which replaces GHme-C with unmodified cytosine [41]. To our knowledge, this is the first study that has shown that H-NS can bind to GHme-C modified T4 DNA ( Figure 6). Furthermore, addition of both MotB and H-NS together resulted in a more complete shift of either the λ/HindIII or T4/SspI DNA. At this point, we cannot distinguish whether H-NS and MotB bind to separate populations of DNA or bind concurrently to the same DNA. However, taking into account the length of DNA, it is conceivable that both MotB and H-NS could bind to the same fragment.

T4 motB am Reduces Burst Size and Overexpression of motB Increases Plaque Size
A T4 motB am mutant was constructed to investigate the functional phenotype of motB during T4 infection of non-suppressing E. coli. We observed~2-fold reduction in burst size for the T4 motB am infection compared to T4 wt (Figure 7a), indicating that motB contributes to phage fitness although no difference in plaque size was observed between T4 wt and T4 motB am (Figure 7b). To ask whether an excess of MotB would improve phage growth, we infected cells expressing motB from a plasmid. Because of toxicity, synthesis of MotB was induced by the addition of 0.1% (w/v) arabinose for only 20 min before plating on LB agar containing no arabinose, thereby producing a low level of MotB prior to infection. In fact, the level of produced MotB was too low to be detected by SDS-PAGE. Nevertheless, prior production of MotB resulted in larger plaques when infected with T4 wt, T4 motB am , or T4 motA am (Figure S5), which normally yields very tiny plaques since MotA is required for Pm activation ( [22,34,35], reviewed in [5]).

motB Knockdown has Only Mild Effects on T4 RNA Levels
Previous work had suggested that motB might be needed for optimal expression of certain middle genes [16] and that the MotB protein could be isolated from T4 infected cells in a complex containing DNA, RNAP, and T4-encoded factors that associate with RNAP [18]. Thus, it seemed plausible that MotB might function as a factor needed for optimal T4 gene expression. To investigate this, we performed RNA-seq and RT-qPCR analyses on RNA isolated from T4 wt and T4 motB am infections of non-suppressing E. coli 5 min post-infection. In a T4 wt infection, the 5 min time point corresponds to the peak of middle transcription, the end of early transcription, and the beginning of late transcription.
The RNA-seq data revealed that the expression of only six late genes, which decreased from 2 to 4.8-fold, were significantly affected in the T4 motB am infection relative to T4 wt (Table S1, Table 1) at 5 min post-infection. The expression of early and middle genes did not change. Five of the late genes were subsequently validated by 8,15,20 and 27 (Table 1). These genes encode proteins that contribute to the phage base plate, head, or tail.

motB Knockdown has Only Mild Effects on T4 RNA Levels
Previous work had suggested that motB might be needed for optimal expression of certain middle genes [16] and that the MotB protein could be isolated from T4 infected cells in a complex containing DNA, RNAP, and T4-encoded factors that associate with RNAP [18]. Thus, it seemed plausible that MotB might function as a factor needed for optimal T4 gene expression. To investigate this, we performed RNA-seq and RT-qPCR analyses on RNA isolated from T4 wt and T4 motB am infections of non-suppressing E. coli 5 min post-infection. In a T4 wt infection, the 5 min time point corresponds to the peak of middle transcription, the end of early transcription, and the beginning of late transcription.
The RNA-seq data revealed that the expression of only six late genes, which decreased from 2 to 4.8-fold, were significantly affected in the T4 motB am infection relative to T4 wt (Table S1, Table 1) (Table 1). These genes encode proteins that contribute to the phage base plate, head, or tail.
To investigate this further, we observed transcripts from selected Pe's and Pm's as well as the specific promoters upstream of differentially expressed late genes by primer extension of RNA isolated at 1 min, 5 min, and 10 min post-infection. At 1 min post-infection, only Pe's are significantly active; at 5 min, early transcription is waning, Pm's are very active, and late transcription is beginning; at 10 min, most transcription is from Pl's [42]. Primer extension analyses of Pe 73 , Pe 72.6 , and Pe 144.6 for RNA isolated at 1 or 5 min agreed with the conclusion that Pe activity was not significantly affected (i.e., changed by more than twofold) by the motB knock-down ( Figures S6a and S6b). Although the level of Pm RNA at 1 min was too low for analysis, primer extension products from 5 min middle RNAs were easily detected, and as expected from our RNA-seq analysis, transcript levels from the tested promoters were not significantly changed in the T4 motB am infection ( Figure S6B). Using the 5 min RNA, we were unable to quantify the levels of all the differentially expressed late RNAs because the level of late transcription at this time is so low. However, with Pl 22 and Pl 27 , we observed that the motB mutation did not significantly affect transcription. At 10 min post-infection, when late transcription is robust, products derived from late promoters were easily detected, and there was no difference between levels of RNA from T4 wt or the T4 motB am infections for all the Pl's tested ( Figure 8). In particular, the levels of primer extension products from the promoters upstream of the late genes whose RNA was lower in the RNA-seq and RT-qPCR analyses (Pl 8 , Pl 15 , Pl 20 , Pl 22 , and Pl 27 ) were unchanged. We conclude that transcription of late genes 7, 8, 15, 20 and 27, which is lower at 5 min in the T4 motB am infection, is fully restored by 10 min.
RT-qPCR, genes frd.2, nrdD, and motA, which were unaffected in the RNA-seq data, were used as endogenous controls. To investigate this further, we observed transcripts from selected Pe's and Pm's as well as the specific promoters upstream of differentially expressed late genes by primer extension of RNA isolated at 1 min, 5 min, and 10 min post-infection. At 1 min post-infection, only Pe's are significantly active; at 5 min, early transcription is waning, Pm's are very active, and late transcription is beginning; at 10 min, most transcription is from Pl's [42]. Primer extension analyses of Pe73, Pe72.6, and Pe144.6 for RNA isolated at 1 or 5 min agreed with the conclusion that Pe activity was not significantly affected (i.e., changed by more than twofold) by the motB knock-down ( Figures S6a and S6b). Although the level of Pm RNA at 1 min was too low for analysis, primer extension products from 5 min middle RNAs were easily detected, and as expected from our RNA-seq analysis, transcript levels from the tested promoters were not significantly changed in the T4 motB am infection ( Figure S6B).
Using the 5 min RNA, we were unable to quantify the levels of all the differentially expressed late RNAs because the level of late transcription at this time is so low. However, with Pl22 and Pl27, we observed that the motB mutation did not significantly affect transcription. At 10 min postinfection, when late transcription is robust, products derived from late promoters were easily detected, and there was no difference between levels of RNA from T4 wt or the T4 motB am infections for all the Pl's tested (Figure 8). In particular, the levels of primer extension products from the promoters upstream of the late genes whose RNA was lower in the RNA-seq and RT-qPCR analyses (Pl8, Pl15, Pl20, Pl22, and Pl27) were unchanged. We conclude that transcription of late genes 7, 8, 15, 20, and 27, which is lower at 5 min in the T4 motB am infection, is fully restored by 10 min.  Finally, we used primer extensions within a specific early, middle, or late region at 1, 5 or 10 min post-infection, respectively, to investigate whether the absence of motB altered the overall transcription pattern. As seen in Figure S7, the detected 5' ends within these regions are identical, again indicating that the presence of MotB does not affect the steady-state levels of these early, middle, and late transcripts at the times when they are highly expressed. Taken together, we conclude that MotB is not needed for normal levels of T4 transcription.

Discussion
Although many aspects of bacteriophage T4 biology have been extensively characterized, dozens of T4's~300 genes encode proteins whose specific functions remain unknown. Most of these are nonessential genes that are expressed early, leading to the presumption that they are helpful for host takeover, particularly under certain conditions. In this study, we have characterized one such early gene, motB, located in a large dispensable region of the T4 genome, called ∆12 (from 69 to 39; map units 16,280 to 3778) [17]. Other genes located in ∆12 have been shown to be important for phage fitness under certain environmental conditions or in certain host strains and/or to be toxic when expressed in E. coli [17,43]. The toxicity of motB when it is heterologously expressed in E. coli and the phage burst increase in its presence lead to the conclusion that MotB is needed for optimal phage growth, perhaps through a host takeover function.
Previous work indicated that MotB can be isolated in a pre-replicative complex containing DNA, RNAP, and other polymerase-associated and DNA-binding proteins [18]. Other work reported that the expression of particular middle genes was enhanced by the presence of motB in a motA knockdown infection [16]. Taken together, these results suggested that MotB might be needed for optimal middle gene expression, hence the name, modifier of transcription B. However, our transcriptome analyses indicate that at 5 min after infection, when middle gene expression is robust, neither early nor middle gene RNA levels are affected in a T4 motB am infection. Rather only the expression of a few late genes is lowered at 5 min post-infection. By 10 min when late transcripts are highly expressed, transcription of these genes has recovered. Therefore, it seems unlikely then that MotB directly regulates T4 gene expression unless this function is redundant within the genome.
Our biochemical analyses indicate that MotB is a protein that binds to both unmodified and T4 wt GHme-modified DNA. It binds very tightly to host DNA when produced heterologously in vivo, it binds with~100 nanomolar affinity to various ds oligonucleotides, and in EMSAs it retards each of the λ/HindIII fragments. The DNase I footprint of MotB on an unmodified 218 bp fragment, which includes the T4 late promoter P8, does not reveal a high affinity binding site despite the down-regulation of late gene 8 in the T4 motB am infection. Instead, we observe MotB protection of the entire region of the DNA once a sufficient protein concentration is reached. This type of binding profile is indicative of a protein that can "spread" along the DNA with little, if any sequence specificity, perhaps through protein-protein cooperativity. However, we cannot eliminate the possibility that MotB has a high affinity site that was not present in the tested DNAs.
Curiously, MotB copurifies with H-NS and StpA, E. coli histone-like proteins that also can spread along the DNA with limited DNA specificity. H-NS, and its less abundant homolog StpA, target AT-rich DNA sequences, condensing genomic DNA through the formation of ordered and looped structures, and typically repress transcription at the affected region (reviewed in [8]). Our gel retardation assays indicate that H-NS binds to T4 wt GHMe-modified DNA and that the retardation of either unmodified or GHMe-modified DNA increases in the presence of both MotB and H-NS over that observed with either alone (Figure 6). At this point it is unclear whether both H-NS and MotB bind to the same region of DNA simultaneously. Furthermore, despite multiple efforts to show a direct interaction between H-NS and MotB, we have not observed an interaction in the absence of DNA.
Given the high predicted pKa of MotB (9.23) and its presence in the T4 dispensable region ∆12, we can determine its relative abundance during infection from previous 2-D gels of radiolabeled T4 proteins at various times after infection [44]. We identify a highly basic protein, which migrates as predicted for MotB's size and predicted pKa and is missing in the T4 ∆12 infection. The level of this protein is high when T4 proteins are labeled from 1 to 3 min post-infection and remains high even with 7-9 min post-infection labeling. This identification is supported by another finding that MotB (designated as p17.6) production initiates early after infection (0-4 min) and continues as infection progresses (7-12 min) [18]. It is known that T4 factors begin to degrade the host DNA shortly after infection, but degradation of host DNA to acid-soluble products is still less than 50% at 15 min post-infection [45]. It seems reasonable then to conclude that the high levels of MotB would be able to bind host and/or T4 DNA throughout the time it is present. Given the copurification of MotB with H-NS and StpA, two proteins that would be present on the host DNA at the time of infection and could interact with T4 wt DNA, it is tempting to speculate that MotB binding to the DNA might work to counteract H-NS/StpA binding. However, additional work is needed to determine the localization of MotB and H-NS relative to host and phage DNA. Nonetheless, it is clear that heterologous expression of motB is highly toxic, resulting in rapid cell lysis of BL21(DE3) cells and growth arrest of TOP10F' cells. The bactericidal properties of phage proteins when expressed in their host is one of the driving forces of phage research. The mechanisms by which proteins such as MotB inhibit bacterial growth could provide insight for the development of future antibiotics.
Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4915/10/7/343/s1, Figure S1: GC-contents of total RNA have a bimodal distribution for T4 wt infections; Figure S2: Toxicity of C-terminally tagged MotB constructs in E. coli; Figure S3: Purification of MotB; Figure S4: MotB binding to P8 oligonucleotides of various lengths; Figure S5: Effect of heterologous production of MotB on T4 plaque size; Figure S6: Primer extension products from selected T4 promoters in a T4 wt or a T4 motB am infection at 1 min and 5 min post-infection; Figure S7: Transcripts in early, middle, and late regions are similar in T4 wt (+) or T4 motB am (am) infections of NapIV NS; and Table S1: Summary of all RNA-seq and RT-qPCR data.