University of South Florida University of South Florida
Scholar Commons Scholar Commons
Marine Science Faculty Publications College of Marine Science
2017
Isolation and Characterization of a Shewanella Phage–Host Isolation and Characterization of a Shewanella Phage–Host
System from the Gut of the Tunicate, Ciona intestinalis System from the Gut of the Tunicate, Ciona intestinalis
Brittany Leigh
University of South Florida
Charlotte Karrer
University of South Florida
John P. Cannon
University of South Florida
Mya Breitbart
University of South Florida
, mya@usf.edu
Larry J. Dishaw
University of South Florida
Follow this and additional works at: https://scholarcommons.usf.edu/msc_facpub
Part of the Life Sciences Commons
Scholar Commons Citation Scholar Commons Citation
Leigh, Brittany; Karrer, Charlotte; Cannon, John P.; Breitbart, Mya; and Dishaw, Larry J., "Isolation and
Characterization of a Shewanella Phage–Host System from the Gut of the Tunicate, Ciona intestinalis"
(2017).
Marine Science Faculty Publications
. 694.
https://scholarcommons.usf.edu/msc_facpub/694
This Article is brought to you for free and open access by the College of Marine Science at Scholar Commons. It has
been accepted for inclusion in Marine Science Faculty Publications by an authorized administrator of Scholar
Commons. For more information, please contact scholarcommons@usf.edu.
viruses
Article
Isolation and Characterization of a Shewanella
Phage–Host System from the Gut of the Tunicate,
Ciona intestinalis
Brittany Leigh
1
, Charlotte Karrer
2
, John P. Cannon
2
, Mya Breitbart
1
and Larry J. Dishaw
2,
*
1
College of Marine Science, University of South Florida, St. Petersburg, FL 33701, USA;
bleigh@mail.usf.edu (B.L.); mya@usf.edu (M.B.)
2
Department of Pediatrics, University of South Florida, St. Petersburg, FL 33701, USA;
ckarrer@mail.usf.edu (C.K.); jcannon@health.usf.edu (J.P.C.)
* Correspondence: LJDishaw@health.usf.edu; Tel.: +1-727-553-3608
Academic Editors: Mathias Middelboe and Corina P. D. Brussaard
Received: 29 January 2017; Accepted: 17 March 2017; Published: 22 March 2017
Abstract:
Outnumbering all other biological entities on earth, bacteriophages (phages) play critical
roles in structuring microbial communities through bacterial infection and subsequent lysis,
as well as
through horizontal gene transfer. While numerous studies have examined the effects of phages on
free-living bacterial cells, much less is known regarding the role of phage infection in host-associated
biofilms, which help to stabilize adherent microbial communities. Here we report the cultivation
and characterization of a novel strain of Shewanella fidelis from the gut of the marine tunicate
Ciona intestinalis, inducible prophages from the S. fidelis genome, and a strain-specific lytic phage
recovered from surrounding seawater.
In vitro
biofilm assays demonstrated that lytic phage infection
affects biofilm formation in a process likely influenced by the accumulation and integration of the
extracellular DNA released during cell lysis, similar to the mechanism that has been previously
shown for prophage induction.
Keywords: Shewanella; bacteriophage; biofilm; extracellular DNA
1. Introduction
A significant proportion of microbes in the marine environment, including both bacteria and
bacteriophages (i.e., phages), are associated with eukaryotic hosts, where they form stable symbiotic
relationships. These symbiotic relationships are often specific and necessary in maintaining animal
health via carefully orchestrated exchanges (i.e., homeostasis). Although phages are the most abundant
biological entities in the natural world [
1
,
2
], little is known about their role in structuring and
maintaining host-associated microbial communities, or how they influence bacteria within a biofilm,
a lifestyle many aquatic bacteria exhibit. Even less is known about how perturbation of these microbial
communities influences the eukaryotic host. Many animals maintain a “core” assemblage of bacteria
(i.e., a core microbiome) that likely provides advantages to the host [
3
–
5
]. Some of these bacteria are
consistently found within the same environments (e.g., animal intestines) and across diverse animal
hosts, where they are presumed to serve distinct functions for either the animal host and/or the
surrounding microbes. One such bacterial genus is Shewanella [6–9].
Shewanella species from a wide range of environments are known for their highly versatile
metabolic capabilities that utilize diverse electron acceptors including nitrate, nitrite, thiosulfate,
elemental sulfur, iron oxide and manganese oxide [
10
–
12
]. Shewanella species shuttle electrons across
their membranes during anaerobic respiration, resulting in electrical activity within their biofilms and
the transformation of insoluble compounds to bioavailable ones. Interestingly, biofilms with electrical
Viruses 2017, 9, 60; doi:10.3390/v9030060 www.mdpi.com/journal/viruses
Viruses 2017, 9, 60 2 of 16
activity have been documented to influence host cellular responses [
13
]. These bacteria make stable
biofilms and because they can respire almost any compound, they likely represent important symbionts
of animals as well. Despite extensive genomic rearrangements within Shewanella genomes [
14
],
members of the genus retain a core set of metabolic genes that facilitate their survival in diverse
environments [
15
], including the gut of a number of organisms [
6
–
9
]. Shewanella putrefaciens, which is
closely related to S. fidelis, has shown promise as a probiotic for aquaculture [
16
], further emphasizing
important roles for Shewanella in aquatic animal-microbe relationships.
To date, a number of Shewanella phages (both lytic and temperate) have been described from
marine and freshwater environments [
17
–
21
], and in Shewanella oneidensis, prophages have also been
implicated as vital for biofilm formation through excision-mediated lysis [
21
]. Stably integrated
prophage-like elements are common within the genomes of most marine bacterial species [
22
],
and prophages are also thought to be important among bacteria that colonize the gut mucosa of
animals [
23
], often forming biofilms [
24
,
25
]. These biofilms are thought to serve as physical structures
that can enhance pathogen defense by contributing to physical barriers, and through the production of
diverse antimicrobials [26,27].
To begin to understand the role of phages in shaping the microbiome of sessile, filter-feeding
marine invertebrates, we isolated and characterized a core member of the gut microbiome in the
tunicate, Ciona intestinalis [
4
]. This novel strain of Shewanella fidelis (3313) was sequenced, and its
inducible prophages and a strain-specific lytic phage (SFCi1, which was isolated separately from
seawater) were characterized. Previously, it has been shown that spontaneous prophage induction can
augment biofilms in some strains of bacteria [
28
–
31
]; we demonstrate here that infection of S. fidelis 3313
by lytic phage SFCi1 also enhances biofilm formation in vitro in a similar DNA-dependent manner.
2. Materials and Methods
2.1. Bacterial Isolation from the Gut of Ciona Intestinalis
Ciona intestinalis specimens were collected from Mission Bay in San Diego (M-REP Animal
Collection Services, San Diego, CA, USA) during the Spring of 2014. Animals were cleared for 48 h in
seawater filtered through a 0.22
µ
m pore size filter (Millipore Sterivex, Merck, Darmstadt, Germany)
(with water changes every several hours), before the entire gut (stomach, midgut, hindgut) of five
animals was dissected and homogenized using a dounce homogenizer. The gut homogenate was
filtered through a 0.45
µ
m pore size filter (Millipore Sterivex, Merck) to remove host tissue, and the
bacteria were pelleted by centrifugation at 12,500 xg for 10 min and washed three times through
resuspension and centrifugation in 1 mL of sterile (filtered through a 0.22
µ
m pore size filter and
autoclaved) artificial seawater (Instant Ocean AS9519, Marine Depot, Garden Grove, CA, USA).
Serial dilutions of the bacterial homogenate were plated on marine agar (MA) 2216 (Becton Dickinson
Company, Franklin Lakes, NJ, USA). Colonies displaying distinct phenotypes were randomly chosen,
purified by streaking, and grown separately in the corresponding liquid broth (marine broth (MB) 2216,
pH 7.6) at 20
◦
C; subsequently, a 20% glycerol stock was made for each isolate and stored at
−
80
◦
C.
DNA was isolated using the PowerSoil DNA Kit (MoBio Laboratories, Carlsbad, CA, USA) and the 16S
rRNA gene amplified using universal primers 27F and 1492R [
32
] (polymerase chain reaction (PCR)
conditions: denature at 95
◦
C for 5 min, cycle 35 times through 94
◦
C for 30 s, 56
◦
C for 30 s, 72
◦
C for
1 min 30 s, and end with a final extension at 72
◦
C for 10 min), sequenced via the Sanger platform and
identified using BLAST against the NCBI non-redundant database [33].
2.2. Phage Isolation, Propagation, and Purification for Transmission Electron Microscopy
S. fidelis 3313 recovered from the Ciona gut homogenate was screened for lytic phages via standard
plaque assays using seawater from which the animals were shipped (i.e., bag water) filtered through a
0.22
µ
m pore size filter. Approximately 500 mL of the filtered seawater was concentrated using Amicon
Ultra-15 concentration units (molecular weight cut-off (MWCO) 100 kDa; EMD (Merck Millipore,
Viruses 2017, 9, 60 3 of 16
Darmstadt, Germany) by centrifugation to a final volume of ~15 mL. Lytic phages were isolated with
the double agar method (0.5% low-melt top agar) [
34
] using the prepared seawater concentrate and the
bacterial host grown to log phase (OD
600
= 0.25) in MB. Each plaque was then cored, plaque-purified
three times and resuspended in 500
µ
L of sterile modified sodium magnesium (MSM) buffer (450 mM
NaCl, 102 mM MgSO
4
, 50 mM Tris Base, pH 8). The purified phage was propagated on S. fidelis 3313
lawns on MA at room temperature. The resulting lysate was filtered through a 0.22
µ
m pore size filter
and stored in MSM buffer at 4
◦
C.
To estimate phage–host growth dynamics including the latent period and burst size, a one-step
infection curve was performed according to Hyman and Abedon [
35
], with slight modifications.
Latent period is defined as the period between the adsorption time and the initial phage lysis of the
bacterial culture, prior to any significant rise in phage particles [
35
]. For this procedure, a 10 min
adsorption step at a multiplicity of infection (MOI) of 1 was followed by centrifugation at 13,000
×
g for
30 s to pellet the bacteria with adsorbed phages. The pellet was then resuspended in 1 mL of sterile MB.
Triplicate samples were taken at 10 min intervals for up to 2 h and directly plated using the double
agar method to determine phage titer. Additionally, burst size was measured as the ratio of final phage
particles to the number of bacterial cells at the onset of phage exposure.
A portion of the lysate was further purified via cesium chloride (CsCl) gradient
ultracentrifugation [
36
] for morphological analysis by transmission electron microscopy (TEM) using
an Hitachi 7100 (Hitachi Ltd., Tokyo, Japan). The purified virus particles were prepared for imaging
on a formvar grid (Electron Microscopy Sciences, Hatfield, PA, USA) using a negative stain with 2%
uranyl acetate, as described previously [
37
]. Images were captured using an Orius SC600 bottom
mount camera (Gatan Inc., Pleasanton, CA, USA) at 100 kV. A separate aliquot of this purified viral
suspension was reserved for DNA extraction using the QIAmp MinElute Virus Spin Kit (Qiagen Inc.,
Valencia, CA, USA) and for sequencing, as described below.
2.3. DNA Extraction, Sequencing and Analysis
Shewanella fidelis 3313 was cultured in MB overnight at 20
◦
C with shaking at 90 RPM, and its
lytic phage was propagated and purified as described above. Bacterial DNA was extracted using the
PowerSoil DNA Kit (MoBio Laboratories, Carlsbad, CA, USA) as described above. All viral DNA
was amplified using a GenomiPhi V2 DNA amplification kit (GE Healthcare Life Sciences, Pittsburgh,
PA, USA) to generate adequate template for sequencing (~1
µ
g). To minimize bias introduced by the
amplification process, three identical reactions were prepared and pooled.
Bacterial, phage and prophage DNA were sequenced with the Illumina MiSeq platform generating
mate-pair (2
×
250) libraries (Operon, Eurofins MWG Operon LLC, Huntsville, AL, USA). The NuGen
UltraLow DNA kit was used (Eurofins, Louisville, KY, USA) to prepare libraries, and DNA quality
was assessed on a BioAnalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA), with size selection
(400–600 bp) conducted to remove outlier DNA fragments after sonication. The resulting mate-pair
reads were assembled using approaches described in Deng et al. [
38
], first by a de Bruijn graph
assembler (Velvet de novo assembler with a k-mer of 35 (phage) and 27 (bacteria) [
39
]), followed by
the default consensus algorithm in Geneious 8.1.7 (Biomatters Ltd, Auckland, New Zealand) [
40
].
Viral genome open reading frames (ORFs) were identified using Glimmer3 [
41
] through Geneious 8.1.7;
annotations were improved with the BLASTX algorithm against non-redundant protein databases
in GenBank, Protein Data Bank (PDB), SwissProt, Protein Information Resource (PIR) and Protein
Research Foundation (PRF) via Geneious 8.1.7. All resulting bacterial contigs were uploaded to the
Rapid Annotation using Subsystem Technology (RAST) server [
42
,
43
] under sample ID mgs422948,
with full annotations based on the SEED Database. All contigs passed Metagenomics (MG)-RAST
quality control and all predicted proteins were annotated. The complete 16S rRNA gene of S. fidelis
3313 was compared via PHYML maximum likelihood trees to other described Shewanella species
in the GenBank database (Figure S1). The closest species was determined to be Shewanella fidelis
(ATCC BAA-318; GenBank ID: 17801). All assembled contigs greater than 1500 bp (a total of 24) were
Viruses 2017, 9, 60 4 of 16
compared to this nearest neighbor to determine average nucleotide identity (ANI) of the entire genome
using the ANI calculator [
44
]. Genomes with ANI values above 95% are considered to belong to the
same species [45].
2.4. Prophage Induction and Identification
Prophage regions were identified and annotated by screening all bacterial contigs using the
VirSorter pipeline [
46
]. To determine if S. fidelis 3313 possessed any inducible prophages, mitomycin C
was introduced to an early log-phase culture (OD
600
of 0.025) at a final concentration of 1
µ
g/mL and
incubated for 24 h [
47
]. The resulting culture, along with an untreated control culture, was then stained
with SYBR Gold nucleic acid stain and induced phage particles were enumerated using epifluorescence
microscopy, as described previously in Patel et al. [
48
]. Induced phage particles were then CsCl-purified
and treated with 2U of DNase I Turbo (Invitrogen, Carlsbad, CA, USA), before DNA was extracted
using the QIAmp MinElute Virus Spin Kit (Qiagen Inc.). Mate-pair (2
×
250) libraries were produced
using the NuGen UltraLow DNA kit, quality controlled with the BioAnalyzer as previously stated
(Operon, Eurofins MWG Operon LLC), and sequenced on the Illumina MiSeq platform. Reads were
then mapped back to the assembled S. fidelis 3313 genome utilizing the Geneious 8.1.7 software,
with default parameters to determine which of the predicted prophages were induced. Additionally,
TEM analysis of the CsCl-purified induced prophage fraction was performed as described above.
2.5. Biofilm Assays
Single colonies of S. fidelis 3313 were grown in MB at 20
◦
C overnight with shaking at 90 RPM.
Concentration was estimated by optical density at OD
600
based on previously-calibrated growth curves
and colony forming units (data not shown). For biofilm assays, stationary cultures were diluted to a
final concentration of 10
6
cells mL
−1
(early log phase OD
600
of 0.025), and phages were added at 10
6
plaque forming units (PFUs) mL
−1
immediately before plating. All bacterial treatments and controls
were plated on 12-well plates (Thermo Scientific, Waltham, MA, USA) in triplicate; bacteria were also
plated in duplicate on 35 mm glass bottom dishes (No. 1.5, uncoated; MatTek Corporation, Ashland,
MA, USA) for extracellular DNA detection using the TOTO-1 Iodide 514/533 stain (Molecular Probes,
Invitrogen) and counterstained for live cells using SYTO60 red (Molecular Probes, Invitrogen) [
49
].
Additionally, purified salmon sperm DNA (Invitrogen) was added to separate cultures at a final
volume of 300 ng mL
−1
. All stationary culture dishes were incubated at 20
◦
C for up to two days to
allow biofilm formation, before excess liquid and planktonic bacteria were removed by gentle pipetting.
To quantify biofilm formation, culture dishes were allowed to dry, and then subsequently stained
with a 0.1% crystal violet solution for 10 min, as per Merritt et al. [
50
]. Crystal violet was removed
by decanting, and the dishes were washed twice with distilled water to remove excess stain and then
allowed to dry completely. The dried crystal violet was resuspended in 30% acetic acid for ~10 min,
and the OD
590
was determined for each culture dish [
50
]. At each time point, biofilms in the 12-well
dishes and one of the MatTek dishes at 24 h (when the difference was the most drastic) were treated
with 2U of DNase I Turbo (Invitrogen), for 10 min at 37
◦
C. Both dishes for fluorescent microscopy
staining were then washed once with 1
×
PBS before TOTO-1 staining for 10 min. Excess dye was then
removed and the biofilm washed twice in 1
×
PBS before counterstaining with SYTO60 for 10 min.
Excess dye was removed, washed once in 1
×
PBS and held in 1
×
PBS for imaging of 10 random
fields using the Leica Application Suite (Leica Microsystems, Wetzlar, Germany) and the Metamorph
version 7.5 software (Molecular Devices, LLC, Sunnyvale, CA, USA), using consistent exposure,
aperture, and magnification settings. Images were exported as TIF files from Metamorph (Molecular
Devices) and imported into ImageJ 1.48v [
51
]. Without any additional enhancements, the channels
were separated, thresholds permanently set, signal intensity was averaged and standard deviations
determined over the area selected.