Identification and characterization of ϕH111-1

Abstract

Characterization of prophages in sequenced bacterial genomes is important for virulence assessment, evolutionary analysis, and phage application development. The objective of this study was to identify complete, inducible prophages in the cystic fibrosis (CF) clinical isolate Burkholderia cenocepacia H111. Using the prophage-finding program PHAge Search Tool (PHAST), we identified three putative intact prophages in the H111 sequence. Virions were readily isolated from H111 culture supernatants following extended incubation. Using shotgun cloning and sequencing, one of these virions (designated ϕH111-1 [vB_BceM_ϕH111-1]) was identified as the infective particle of a PHAST-detected intact prophage. ϕH111-1 has an extremely broad host range with respect to B. cenocepacia strains and is predicted to use lipopolysaccharide (LPS) as a receptor. Bioinformatics analysis indicates that the prophage is 42,972 base pairs in length, encodes 54 proteins, and shows relatedness to the virion morphogenesis modules of AcaML1 and “Vhmllikevirus” myoviruses. As ϕH111-1 is active against a broad panel of clinical strains and encodes no putative virulence factors, it may be therapeutically effective for Burkholderia infections.

Introduction

The development and continual improvement of next-generation sequencing technologies now allows for the rapid genomic analysis of diverse populations of previously uncharacterized bacteria. Although not necessarily a focus of such studies, the characterization of prophage sequences within these genomes can provide important insights into mechanisms of virulence, horizontal transfer, and the evolution of both host and phage.¹ Furthermore, these newly sequenced genomes (along with their embedded prophage sequences) have the potential to be a repository of novel inducible phages that could be exploited for biotechnological and/or medical applications.

Within the genus Burkholderia, prophages have been intensively studied to assess their contribution to host virulence and evolution and to identify appropriate inducible phage candidates for diagnostic or therapeutic use.^2-5 For the Burkholderia cepacia complex (BCC)—a group of opportunistic pathogens infecting cystic fibrosis (CF) patients—characterization studies generally focus on the potential for medical application of a specific phage. Of particular importance are phages infecting B. cenocepacia due to the clinical predominance and virulence of this species.⁴ Although the therapeutic use of temperate phages is generally discouraged, the limited availability of obligately lytic BCC phages has necessitated the use of confirmed, putative, or modified temperate phages for several in vivo efficacy studies.^6-8 The use of such phages against Burkholderia is arguably safer than it is against many other pathogens because virulence factors in this genus have not been discovered to be encoded by temperate phages.²

One of the challenges of prophage identification is the differentiation of inducible prophages from defective prophage remnants, a distinction with both evolutionary and practical implications. From an evolutionary standpoint, inducible prophages can transfer bacterial or phage genes through transduction, while prophage remnants do not actively facilitate horizontal exchange (with some exceptions, such as gene transfer agents).^1,9 From a practical standpoint, the ability to independently propagate, characterize, modify, and utilize temperate phages is extremely limited if the prophage cannot be induced. PHAge Search Tool (PHAST) is a newly developed prophage identification program developed in part to address this challenge.¹⁰ It can predict if a prophage is intact, incomplete, or questionable. Here, we use PHAST to facilitate the identification and further classical and molecular characterization of an inducible prophage in the newly sequenced genome of the CF isolate Burkholderia cenocepacia H111. This work demonstrates how the integration of improved bioinformatics tools with next-generation DNA sequencing can greatly accelerate the identification and isolation of biomedically important inducible phages.

Results and Discussion

B. cenocepacia H111 prophage screening and isolation

Preliminary sequence analysis of B. cenocepacia H111 contigs identified several regions containing prophage-like genes. To determine which regions might contain complete prophages, PHAST was used to analyze the 71 available H111 contigs (NZ_CAFQ01000001.1–NZ_CAFQ01000071.1). This program identified three potential intact prophages in contigs NZ_CAFQ01000015.1 (C15), NZ_CAFQ01000032.1 (C32), and NZ_CAFQ01000043.1 (C43). In C15, the region identified is 47.0 kilobase pairs (kbp) in length and shows similarity to proteins of the Shigella myovirus SfV and other phages (including Burkholderia phages AH2, Bcep176, BcepNazgul, Bcep22, ϕ644-2, ϕ1026b, ϕE125, BcepF1, and KS5). In C32, the prophage region is shorter (33.2 kbp) and shows extensive similarity at the protein level to the P2-like myovirus ϕE202, a prophage of Burkholderia thailandensis.³ In C43, the region identified is 26.0 kbp in length and shows similarity to proteins of the Vibrio myovirus vB_VpaM_MAR and other phages infecting species such as Burkholderia, Ralstonia, Erwinia, Salmonella, Hemophilus, Streptomyces, and Escherichia.

Based on the PHAST prediction that intact prophages were present in the H111 genome, we assayed H111 culture supernatants for spontaneous phage release following extended incubation. When filter-sterilized supernatant was plated with B. cenocepacia C6433 (a common BCC phage host) in soft-agar overlays, many small (1–2 mm) identical plaques with turbid centers and very turbid halos were observed. A single phage plaque designated as ϕH111-1 (or vB_BceM_ϕH111-1) was subsequently picked and propagated to high titer on C6433. The following analyses were performed on the single phage type isolated from a single plaque. In order to potentially isolate any other putative prophages from this strain, alternative screening procedures that vary the mode of induction and/or the propagation host may be required.

ϕH111-1 morphology, receptor, and host range

ϕH111-1 is a myovirus with a capsid diameter of approximately 65 nm (Fig. 1). To identify the putative phage receptor, we tested the ability of ϕH111-1 to infect B. cenocepacia K56-2 strains with lipopolysaccharide (LPS) mutations.^11,12 The majority of strains remained susceptible to phage infection excluding the truncated inner core wabO and waaC mutants (Table S1), indicating that ϕH111-1 likely interacts with moieties in the LPS inner core.

Figure 1. Transmission electron micrograph of a ϕH111-1 virion stained with phosphotungstic acid. The micrograph was taken at 140,000-fold magnification; scale bar represents 50 nm.

When tested against a panel of nine B. cenocepacia strains (all of which were originally isolated from CF patients), ϕH111-1 was able to infect seven of these strains: C6433, 715J, J2315, K56-2, C1257, C5424, and PC184 (Table S1). The broad host range with respect to characterized B. cenocepacia CF strains suggested that ϕH111-1 could be active against a wide variety of clinical isolates. To confirm this prediction, we tested ϕH111-1 against a B. cenocepacia panel acquired from the University of Alberta Hospital Cystic Fibrosis Clinic.¹³ ϕH111-1 was able to infect all 13 strains tested (Table S1), providing further evidence that this phage may be an appropriate candidate for therapeutic use (particularly if the lysogeny module were deleted).⁸ Excluding B. cenocepacia, the ϕH111-1 host range was found to be relatively narrow as only B. multivorans ATCC 17616 and C5274 were susceptible to phage infection from a panel of 18 other Burkholderia strains tested (representing 8 additional BCC species) (Table S1).

ϕH111-1 genome sequence

To determine if ϕH111-1 represented one of the intact prophages predicted by PHAST, we isolated phage DNA and performed shotgun cloning. Two randomly chosen EcoRI clones were partially sequenced and compared with the H111 reference sequence using BLASTN. One clone matched with C43 bp 172,986–176,784 while the other matched to C43 bp 179,518–181,496. Both of these sequences fall within the prophage region in C43 predicted by PHAST (159,103–185,124), indicating that this program correctly identified both the locus and the intact nature of the prophage.

To identify the exact prophage boundaries attL and attR, we screened for direct repeats 20 kbp upstream and downstream of the cloned sequences (from 152,986–201,496 in C43) using two-sequence BLASTN alignment. An imperfect 24 bp direct repeat (Fig. 2) was identified that flanked sequences consistent with prophage genes: one copy was found upstream of a series of hypothetical protein genes (starting with I35_4470) and one copy was found at the 5′ end of an arginine tRNA gene (I35_4520). To confirm that these repeats represented the attL and attR sequences, we designed primers (downstream of I35_4470 and upstream of I35_4520) and verified by PCR and sequencing that these regions became adjacent within packaged virion DNA, forming the attP overlap region (Fig. 2). Based on restriction analysis, DNA in the ϕH111-1 virion is predicted to be linear without cohesive ends.

Figure 2. Alignment of the ϕH111-1 attL (above), attP overlap region (center), and attR (below). The 24 base pair region common to all three sequences is underlined.

The ϕH111-1 prophage is 42,972 bp in length (including both attL and attR), has a 62% GC content (lower than the H111 GC content of 67%), and integrates at an arginine tRNA gene (as noted above). Based on GeneMark predictions, this prophage sequence contains 54 open reading frames (Fig. 3, Table 1). Putative functional annotations were assigned to these proteins based on BLASTP (Table 1) and HHpred (Table S2) analysis. No putative toxin genes were identified using BTXpred. As shown in Figure 3, ϕH111-1 genes are arranged in function-specific modules involved in DNA binding, lysis, tail morphogenesis, and capsid morphogenesis/DNA packaging (discussed below).

Figure 3. Map of the ϕH111-1 prophage. Arrows indicate gene transcription in the forward or reverse direction. Small numbers indicate base pairs within the prophage (above) or H111 contig NZ_CAFQ01000043.1 (below). Black, attL, or attR; gray, unknown function; red, DNA binding; blue, lysis; pink, tail morphogenesis; purple, capsid morphogenesis and DNA packaging.

Table 1. ϕH111-1 genome annotation.

Gene	Prophage start	Prophage end	H111 contig start	H111 contig end	Strand	Length (amino acids)	Putative function	Closest relative (excluding H111 proteins)	BLASTP alignment region (amino acids)	Percent identity	Organism	GenBank accession number
attL	1	24	156838	156861
1	243	809	157080	157646	+	188	hypothetical	hypothetical protein PMI16_01842	19–200/207	37	Herbaspirillum sp CF444	ZP_10720925.1
2	939	1628	157776	158465	-	229	hypothetical	hypothetical protein BuboB_19482	1–218/220	64	Burkholderia ubonensis Bu	ZP_02379924.1
3	1668	2009	158505	158846	-	113	hypothetical	hypothetical protein bgla_2p0890	1–113/135	42	Burkholderia gladioli BSR3	YP_004351005.1
4	2266	2766	159103	159603	+	166	hypothetical	gp31	2–144/153	48	Burkholderia pseudomallei 1710a	ZP_04953065.1
5	2797	3513	159634	160350	-	238	hypothetical	hypothetical protein bgla_1g11110	1–238/238	61	Burkholderia gladioli BSR3	YP_004359751.1
6	3526	4137	160363	160974	-	203	hypothetical	hypothetical protein BP1026B_I2070	1–200/200	58	Burkholderia pseudomallei 1026b	YP_006275083.1
7	4187	4510	161024	161347	+	107	hypothetical	peptidase M23	74–132/432	37	Ferrimonas balearica DSM 9799	YP_003914362.1
8	4485	4982	161322	161819	-	165	hypothetical	hypothetical protein bgla_1g11080	1–165/165	67	Burkholderia gladioli BSR3	YP_004359748.1
9	4989	5237	161826	162074	-	82	hypothetical	hypothetical protein bgla_1g11070	62–142/143	62	Burkholderia gladioli BSR3	YP_004359747.1
10	5592	6377	162429	163214	-	261	DNA adenine methylase	D12 class N6 adenine-specific DNA methyltransferase	1–262/262	84	Burkholderia ambifaria AMMD	YP_773740.1
11	6655	7296	163492	164133	-	213	hypothetical	hypothetical protein bgla_1g27190	3–198/209	89	Burkholderia gladioli BSR3	YP_004361292.1
12	7299	7865	164136	164702	-	188	endolysin	hypothetical protein bgla_1g27200	1–188/188	89	Burkholderia gladioli BSR3	YP_004361293.1
13	7862	8311	164699	165148	-	149	holin	putative kinetochore protein spc25 protein	80–200/259	26	Neofusicoccum parvum UCRNP2	EOD48207.1
14	8388	9437	165225	166274	-	349	tail protein D	phage late control D family protein	1–349/349	91	Burkholderia glumae BGR1	YP_002912485.1
15	9447	9653	166284	166490	-	68	tail protein X	phage tail X family protein	1–68/68	96	Burkholderia ambifaria AMMD	YP_773745.1
16	9628	10506	166465	167343	-	292	tail protein U	phage P2 GpU family protein	1–292/292	90	Burkholderia ambifaria AMMD	YP_773746.1
17	10517	12958	167354	169795	-	813	tail tape measure protein T	pyocin R2_PP, tail length determination protein	1–814/814	89	Burkholderia ambifaria AMMD	YP_773747.1
18	13039	13341	169876	170178	-	100	hypothetical	hypothetical protein Bamb_1858	1–100/100	82	Burkholderia ambifaria AMMD	YP_773748.1
19	13439	13942	170276	170779	-	167	tail tube protein FII	phage major tail tube protein	1–167/167	92	Burkholderia ambifaria AMMD	YP_773749.1
20	13953	15122	170790	171959	-	389	tail sheath protein FI	phage tail sheath protein	1–389/389	93	Burkholderia ambifaria AMMD	YP_773750.1
21	15207	15986	172044	172823	-	259	tail fiber assembly protein G	bacteriophage-acquired protein	1–231/233	60	Burkholderia thailandensis TXDOH	ZP_02376040.1
22	16002	18020	172839	174857	-	672	tail fiber protein	phage tail fiber protein	1–670/670	59	Burkholderia glumae BGR1	YP_002912494.1
23	18008	18586	174845	175423	-	192	baseplate assembly protein I	phage tail protein I	1–192/192	92	Burkholderia ambifaria AMMD	YP_773753.1
24	18576	19472	175413	176309	-	298	baseplate assembly protein J	baseplate J family protein	1–298/298	88	Burkholderia ambifaria AMMD	YP_773754.1
25	19469	19804	176306	176641	-	111	baseplate assembly protein W	GPW/gp25 family protein	1–111/111	93	Burkholderia ambifaria AMMD	YP_773755.1
26	19804	20004	176641	176841	-	66	hypothetical	hypothetical protein Bamb_1866	1–66/66	85	Burkholderia ambifaria AMMD	YP_773756.1
27	20064	20744	176901	177581	-	226	baseplate assembly protein V	phage baseplate assembly protein V	1–226/226	90	Burkholderia ambifaria AMMD	YP_773758.1
28	20748	21272	177585	178109	-	174	tail protein	hypothetical protein Bamb_1869	1–174/174	81	Burkholderia ambifaria AMMD	YP_773759.1
29	21262	21792	178099	178629	-	176	tail protein	hypothetical protein Bamb_1870	1–176/176	88	Burkholderia ambifaria AMMD	YP_773760.1
30	21795	22082	178632	178919	-	95	head-tail joining protein	hypothetical protein	1–95/96	77	Burkholderia glumae BGR1	YP_002912503.1
31	22084	23079	178921	179916	-	331	major capsid protein	hypothetical protein Bamb_1872	1–331/331	92	Burkholderia ambifaria AMMD	YP_773762.1
32	23153	23497	179990	180334	-	114	head decoration protein	hypothetical protein BURMUCF1_2022	1–114/114	91	Burkholderia multivorans ATCC BAA-247	ZP_15920090.1
33	23528	24595	180365	181432	-	355	Clp protease	ATP-dependent Clp endopeptidase, proteolytic subunit ClpP	1–355/357	84	Burkholderia multivorans ATCC BAA-247	ZP_15921695.1
34	24592	26085	181429	182922	-	497	portal protein	phage portal protein, lambda family	1–485/496	90	Burkholderia multivorans ATCC BAA-247	ZP_15921707.1
35	26082	26288	182919	183125	-	68	head-tail joining protein	hypothetical protein Bcenmc03_1109	1–68/68	100	Burkholderia cenocepacia MC0–3	YP_001764407.1
36	26302	28221	183139	185058	-	639	terminase large subunit	terminase GpA	1–639/639	97	Burkholderia cenocepacia MC0–3	YP_001764406.1
37	28250	28819	185087	185656	-	189	terminase small subunit	hypothetical protein Bcenmc03_1107	1–189/189	94	Burkholderia cenocepacia MC0–3	YP_001764405.1
38	28910	29104	185747	185941	-	64	hypothetical	hypothetical protein Bcenmc03_1105	1–64/64	86	Burkholderia cenocepacia MC0–3	YP_001764403.1
39	29352	30125	186189	186962	-	257	DnaJ chaperone	hypothetical protein BURMUCF1_2377	1–257/257	99	Burkholderia multivorans ATCC BAA-247	ZP_15916016.1
40	30346	32853	187183	189690	-	835	DNA primase	virulence-associated E family protein	1–835/835	95	Burkholderia ambifaria AMMD	YP_773770.1
41	33114	33476	189951	190313	-	120	hypothetical	hypothetical protein Bamb_1881	8–127/127	90	Burkholderia ambifaria AMMD	YP_773771.1
42	33484	34026	190321	190863	-	180	transcriptional regulator	hypothetical protein Bamb_1882	15–194/194	93	Burkholderia ambifaria AMMD	YP_773772.1
43	34108	34344	190945	191181	-	78	transcriptional regulator	hypothetical protein PLA107_31961	3–62/74	57	Pseudomonas syringae pv lachrymans str. M301315	ZP_16673553.1
44	34448	34843	191285	191680	+	131	transcriptional regulator	XRE family transcriptional regulator	12–141/143	92	Burkholderia ambifaria AMMD	YP_773773.1
45	35300	35518	192137	192355	+	72	hypothetical	hypothetical protein BURMUCF1_2052	1–72/72	79	Burkholderia multivorans ATCC BAA-247	ZP_15921714.1
46	35568	35930	192405	192767	+	120	hypothetical	hypothetical protein BURMUCF1_2384	1–120/120	82	Burkholderia multivorans ATCC BAA-247	ZP_15916012.1
47	35930	37228	192767	194065	+	432	ParB-like protein	hypothetical protein BURMUCF1_2385	1–434/434	64	Burkholderia multivorans ATCC BAA-247	ZP_15916017.1
48	37225	37689	194062	194526	+	154	hypothetical	hypothetical protein BURMUCF1_2386	1–152/152	89	Burkholderia multivorans ATCC BAA-247	ZP_15916020.1
49	37682	38059	194519	194896	+	125	hypothetical	hypothetical protein	408–466/532	39	Burkholderia glumae BGR1	YP_002911887.1
50	38056	38289	194893	195126	+	77	excisionase	hypothetical protein Bpse14_41058	1–76/76	88	Burkholderia pseudomallei 14	ZP_02417306.1
51	38342	39295	195179	196132	+	317	DNA cytosine methylase	DNA-cytosine methyltransferase	1–317/317	91	Burkholderia phytofirmans PsJN	YP_001894795.1
52	39341	40351	196178	197188	-	336	restriction endonuclease	hypothetical protein Bphyt_1154	1–336/336	75	Burkholderia phytofirmans PsJN	YP_001894794.1
53	40341	41480	197178	198317	-	379	ParB-like protein	hypothetical protein Bphyt_1153	1–379/379	84	Burkholderia phytofirmans PsJN	YP_001894793.1
54	41575	42732	198412	199569	-	385	integrase	site-specific recombinase, phage integrase family	15–372/379	96	Burkholderia multivorans ATCC BAA-247	ZP_15916025.1
attR	42949	42972	199786	199809

H111 contig start and end values correspond with B. cenocepacia H111 accession number NZ_CAFQ01000043.1.

Sequence analysis

At the nucleotide level, ϕH111-1 is most similar to putative prophage elements in chromosome 1 of Burkholderia gladioli BSR3, Burkholderia glumae BGR1, Burkholderia ambifaria AMMD, and Burkholderia pseudomallei BPC006, 1026b, and MSHR346. Based on a BLASTN comparison, these sequences share 62–78% coverage with the ϕH111-1 prophage (Table S3). As PHAST analysis predicts that each of these regions represents an intact prophage (Table S3), ϕH111-1 may be the first isolated representative of a group of closely related but broadly distributed temperate phages in the genus Burkholderia.

To assess protein relatedness, we used CoreGenesUniqueGenes (CGUG) to compare ϕH111-1 to previously sequenced phages. This program assesses the percentage of proteins that are shared between a genome of interest and a reference genome (determined based on a defined BLASTP threshold).¹⁴ Based on BLASTP analysis, the ϕH111-1 tail proteins show similarity to those of enterobacteria phage P2 (NC_001895.1) and other phages in the genus P2likevirus. Comparing ϕH111-1 and P2 with CGUG (Table 2), the proteomes are 25.58% similar with respect to P2, placing ϕH111-1 in the same subfamily (Peduovirinae) but a different genus.¹⁵ We could not identify any previously characterized phages with ≥ 40% similarity that would belong to the same genus as ϕH111-1. Currently, phages with the most similar proteomes are AcaML1 of Acidithiobacillus caldus (JX507079.1; 28.17% similar) and the “Vhmllikevirus” phages VHML of Vibrio harveyi (NC_004456.1; 28.07% similar) and both VP58.5 (FN297812.1; 31.03% similar) and vB_VpaM_MAR (NC_019722.1; 29.03% similar) of Vibrio parahemolyticus.^16-19 Although these phages share subfamily-level similarity, they are very distinct with respect to aspects such as host, gene content, and lifestyle. AcaML1 lysogenizes a thermophilic and acidophilic γ-proteobacterium and has a significantly larger 59 kbp genome with two insertion sequences.¹⁶ The “Vhmllikevirus” phages VHML, VP58.5, and vB_VpaM_MAR have similar genome sizes to ϕH111-1 (41–43 kbp) but are thought to lysogenize as linear plasmids with telomeres in Vibrio species.^17-19

Table 2. CoreGenesUniqueGenes comparison of ϕH111-1, P2, AcaML1, and VHML.

ϕH111-1 protein	Putative function	Similar protein in P2	Similar protein in AcaML1	Similar protein in VHML
gp1	Hypothetical	None	None	None
gp2	Hypothetical	None	None	None
gp3	Hypothetical	None	None	None
gp4	Hypothetical	None	None	None
gp5	Hypothetical	None	None	None
gp6	Hypothetical	None	None	None
gp7	hypothetical	None	None	None
gp8	hypothetical	None	None	None
gp9	hypothetical	None	None	None
gp10	DNA adenine methylase	None	None	None
gp11	hypothetical	None	None	None
gp12	endolysin	None	None	ORF19
gp13	holin	None	None	None
gp14	tail protein D	tail protein	phage tail protein X	tail protein
gp15	tail protein X	gpX	phage late control protein D	ORF45
gp16	tail protein U	gpU	phage tail formation protein U	ORF44
gp17	tail tape measure protein T	gpT	phage tail length tape measure protein	ORF43
gp18	hypothetical	None	hypothetical protein	None
gp19	tail tube protein FII	major tail tube protein	phage tail tube protein FII	major tail tube protein
gp20	tail sheath protein FI	major tail sheath protein	phage tail sheath protein FI	major tail sheath protein
gp21	tail fiber assembly protein G	gpG	None	None
gp22	tail fiber protein	None	None	None
gp23	baseplate assembly protein I	gpI	phage baseplate assembly protein gpI	ORF33
gp24	baseplate assembly protein J	baseplate assembly protein	phage baseplate assembly protein gpJ	baseplate assembly protein
gp25	baseplate assembly protein W	baseplate wedge subunit	phage baseplate assembly protein gpW	baseplate wedge subunit
gp26	hypothetical	None	None	None
gp27	baseplate assembly protein V	gpV	phage baseplate assembly protein gpV	ORF30
gp28	tail protein	None	None	ORF29
gp29	tail protein	None	None	ORF28
gp30	head-tail joining protein	None	hypothetical protein	None
gp31	major capsid protein	None	major capsid protein	ORF26
gp32	head decoration protein	None	hypothetical protein	None
gp33	Clp protease	None	periplasmic serine proteases (ClpP class)	None
gp34	portal protein	None	portal protein	ORF23
gp35	head-tail joining protein	None	hypothetical protein	None
gp36	terminase large subunit	None	packaging terminase large subunit gpA	ORF22
gp37	terminase small subunit	None	None	None
gp38	hypothetical	None	None	None
gp39	DnaJ chaperone	None	None	None
gp40	DNA primase	None	None	None
gp41	hypothetical	None	None	None
gp42	transcriptional regulator	None	None	None
gp43	transcriptional regulator	None	None	None
gp44	transcriptional regulator	None	None	None
gp45	hypothetical	None	None	None
gp46	hypothetical	None	None	None
gp47	ParB-like protein	None	site specific recombinase large subunit	None
gp48	hypothetical	None	None	None
gp49	hypothetical	None	None	None
gp50	excisionase	None	None	None
gp51	DNA cytosine methylase	None	site specific DNA modification methylase Dcm	None
gp52	restriction endonuclease	None	None	None
gp53	ParB-like protein	None	None	None
gp54	integrase	None	None	None

CoreGenesUniqueGenes analysis was performed using a cutoff score of 75.

The commonalities among ϕH111-1, AcaML1, and the “Vhmllikevirus” phages are largely restricted to the morphogenesis genes. These phages all have P2-like tail proteins, but encode capsid morphogenesis/DNA packaging and accessory proteins that are unrelated to those of P2 (Table 2). The ϕH111-1 tail morphogenesis module extends from gene 14-29, encoding only three proteins that lack homologs in either AcaML1 or VHML: tail fiber assembly protein gp21, tail fiber protein gp22, and hypothetical protein gp26 (Table 2). The dissimilarity of the tail fiber protein (predicted to be the phage anti-receptor) is expected based on the differences in host specificity.²⁰ Based on CGUG analysis, ϕH111-1 encodes a protein similar to each P2 tail protein excluding E/E+E′, H, R, and S (Table 2). The ϕH111-1 capsid morphogenesis and DNA packaging proteins are more closely related to those of AcaML1 than the “Vhmllikevirus” phages. This module includes genes 30–37, encoding the head-tail joining proteins, major capsid protein, head decoration protein, Clp protease, portal protein, and terminase subunits (Table 1). AcaML1 encodes proteins similar to each of these (excluding the terminase small subunit), while VHML only encodes similar major capsid, portal, and terminase large subunit proteins (Table 2).

The predicted ϕH111-1 lysis and DNA binding proteins are largely unrelated to those of AcaML1 and the “Vhmllikevirus” phages (Table 2). A putative holin and N-acetylmuramidase endolysin are encoded proximal to the tail morphogenesis module, with the latter being similar to a VHML protein (Table 2). The predicted DNA binding proteins of ϕH111-1 have a range of functions based on HHpred analysis (Table S2): adenine and cytosine methylases, DnaJ chaperone, primase, transcriptional regulators, ParB-like proteins, excisionase, restriction endonuclease, and integrase (Table 1). Excluding the DNA adenine methylase gp10, each of these proteins is encoded near the right end of the prophage (Fig. 3). Only the gp47 ParB-like protein and gp51 DNA cytosine methylase are similar to AcaML1 proteins (Table 2). There is no evidence that ϕH111-1 carries AcaML1-type transposase or “Vhmllikevirus”-type protelomerase genes.

Conclusions

In order for phage therapy to be a viable alternative to antibiotic treatment for Burkholderia infections, phages must be identified that have activity against an array of clinical isolates without encoding potential virulence factors. By using the PHAST program as a rapid screening tool prior to classical and molecular characterization, we were able to identify such a phage in the chromosome of B. cenocepacia H111. ϕH111-1 has a broad B. cenocepacia host range and lacks genes associated with pathogenicity, making it one of the most clinically promising BCC temperate phages isolated to date. While the results of this study were highly informative with respect to ϕH111-1 morphology, host range, receptor binding, and genetic content, further analyses are required to characterize related prophages and to assess the safety and activity of this phage in vivo.

Materials and Methods

Bacterial strains

B. cenocepacia H111 was originally isolated from a CF patient in Germany.²¹ Strains from the original and updated BCC experimental strain panels,^22,23B. cenocepacia K56-2 LPS mutants,^11,12 and clinical isolates from the University of Alberta Hospital Cystic Fibrosis Clinic¹³ were used for phage isolation, propagation, and host range testing. Strains were grown aerobically overnight at 30 °C in half-strength Luria-Bertani (½ LB) broth or solid medium (containing agar or, for DNA isolation, agarose). LPS mutants were grown similarly but supplemented with 100 μg/ml trimethoprim.

Phage isolation and analysis

For ϕH111-1 isolation, a 10 ml broth culture of H111 was grown aerobically with shaking for 48 h at 30 °C. One milliliter of the culture was pelleted 2 min at 10,000 rcf and the supernatant was filter-sterilized using a Millex-HA 0.45 μm syringe-driven filter unit (Millipore). The supernatant was diluted in modified suspension medium (modified SM; 50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 10 mM MgSO₄), plated in soft agar overlays with B. cenocepacia C6433, and incubated overnight at 30 °C. A single plaque was picked using a sterile Pasteur pipette and suspended in modified SM. To collect a high titer lysate, the single plaque stock was replated with C6433, overlaid with modified SM following overnight incubation, pelleted, and filter-sterilized as above. Phage stocks were stored at 4 °C.

For host range analysis (BCC panel strains, clinical isolates, and K56-2 LPS mutants), strains were screened with both overlays and spot testing (10 μl spots of diluted lysate on overlays of the host strain). Electron microscopy grids were prepared by incubating filter-sterilized (0.22 μm) lysate on a carbon-coated copper grid for 5 min followed by phosphotungstic acid staining for 30 s. A Philips/FEI (Morgagni) transmission electron microscope with charge-coupled device camera was used to capture images with the assistance of the University of Alberta Department of Biological Sciences Advanced Microscopy Facility. Phage DNA was isolated, digested, shotgun cloned, and partially sequenced as described previously.²⁴ Screening for cohesive sites was also performed as described previously.²⁵

Bioinformatics

Prophage regions were identified in H111 using PHAge Search Tool (PHAST) analysis of contigs NZ_CAFQ01000001.1–NZ_CAFQ01000071.1.¹⁰ Putative prophage boundaries (i.e., flanking direct repeats attL and attR) were identified using two-sequence BLASTN.²⁶ ϕH111-1 lysate was PCR amplified (I35_4519F: TTGCTATACTC TGTCCCCGCCG; I35_4471R: CAACCATTTCGT CAGCCGGATAG) and sequenced to verify that the attL and attR sequences were found in a single copy in the phage DNA (as the attP overlap region). The prophage sequence was reannotated from the original record using GeneMark.hmm for prokaryotes.²⁷ We were unable to definitively identify either a translationally frameshifted tail protein or an Rz/Rz1 pair following manual annotation.^28,29 BLASTP, HHpred, CD-Search, and BTXpred were used to predict protein function.^26,30-32 Genome and proteome relatedness were assessed using BLASTN and CoreGenesUniqueGenes (CGUG) with a cutoff score of 75, respectively.^14,15,26 The genome map was constructed using Geneious.³³ The ϕH111-1 prophage sequence can be found in the GenBank database under B. cenocepacia H111 accession number NZ_CAFQ01000043.1 (bp 156,838–199,809).

Glossary

Abbreviations:

PHAST

PHAge Search Tool

BCC

Burkholderia cepacia complex

CF

cystic fibrosis

bp

base pairs

LPS

lipopolysaccharide

10.4161/bact.26649

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Supplemental Materials

Supplemental materials may be found here: www.landesbioscience.com/journals/bacteriophage/article/26649