Mar Biotechnol (2010) 12:439–451
DOI 10.1007/s10126-009-9233-y
ORIGINAL ARTICLE
Antibacterial Activity of Marine Culturable Bacteria
Collected from a Global Sampling of Ocean Surface Waters
and Surface Swabs of Marine Organisms
Lone Gram & Jette Melchiorsen &
Jesper Bartholin Bruhn
Received: 11 June 2009 / Accepted: 20 September 2009 / Published online: 13 October 2009
# Springer Science + Business Media, LLC 2009
Abstract The purpose of the present study was to isolate
marine culturable bacteria with antibacterial activity and
hence a potential biotechnological use. Seawater samples
(244) and 309 swab samples from biotic or abiotic surfaces
were collected on a global Danish marine research
expedition (Galathea 3). Total cell counts at the seawater
surface were 5×105 to 106 cells/ml, of which 0.1–0.2%
were culturable on dilute marine agar (20°C). Three percent
of the colonies cultured from seawater inhibited Vibrio
anguillarum, whereas a significantly higher proportion
(13%) of colonies from inert or biotic surfaces was
inhibitory. It was not possible to relate a specific kind of
eukaryotic surface or a specific geographic location to a
general high occurrence of antagonistic bacteria. Five
hundred and nineteen strains representing all samples and
geographic locations were identified on the basis of partial
16S rRNA gene sequence homology and belonged to three
major groups: Vibrionaceae (309 strains), Pseudoalteromo-
Work was done at National Institute of Aquatic Sciences and on board
the vessel “Vædderen” during a global marine research cruise.
Electronic supplementary material The online version of this article
(doi:10.1007/s10126-009-9233-y) contains supplementary material,
which is available to authorized users.
L. Gram (*) : J. Melchiorsen : J. B. Bruhn
National Institute of Aquatic Resources,
Technical University of Denmark,
Søltofts Plads Building 221,
2800 Kgs. Lyngby, Denmark
e-mail: gram@aqua.dtu.dk
Present Address:
J. B. Bruhn
Chr. Hansen A/S,
Bøge Allé 10-12,
2970 Hørsholm, Denmark
nas spp. (128 strains), and the Roseobacter clade (29
strains). Of the latter, 25 strains were identified as Ruegeria
mobilis or pelagia. When re-testing against V. anguillarum,
only 409 (79%) retained some level of inhibitory activity.
Many strains, especially Pseudoalteromonas spp. and
Ruegeria spp., also inhibited Staphylococcus aureus. The
most pronounced antibacterial strains were pigmented
Pseudoalteromonas strains and Ruegeria spp. The inhibi-
tory, pigmented Pseudoalteromonas were predominantly
isolated in warmer waters from swabs of live or inert
surfaces. Ruegeria strains were isolated from all ocean
areas except for Arctic and Antarctic waters and inhibitory
activity caused by production of tropodithietic acid.
Keywords Bacterial antagonism . Biotechnology .
Pseudoalteromonas . Roseobacter clade . Vibrionaceae
Introduction
Marine microorganisms are essential for nutrient turnover
in the oceans, and the growth and metabolism of marine
microorganisms has been studied to understand their
ecology and role in biochemical processes (Azam and
Malfatti 2007; Falkowski et al. 1998; Field et al. 1998).
Marine microorganisms are also the focus of attention due
to their production of secondary metabolites that may have
a range of pharmaceutical and biotechnological applications
(Blunt et al. 2008; Bull and Stach 2007; Burgess et al.
1999; Debnath et al. 2007; Egan et al. 2008; Fenical and
Jensen 2006; Jensen and Fenical 1996). Some marine
bacteria are inhibitory to other bacteria (Barja et al. 1989;
Burkhold et al. 1966; Gauthier and Flatau 1976), suggest-
ing that bacterial interactions could play an important role
in marine ecology (Long and Azam 2001; Strom 2008).
440
Control of unwanted microorganisms is essential in all
aspects of life, and microbial diseases must be treated in
humans, animals, and plants. We have, over the past
decades, seen the rapid emergence of antibiotic-resistant
bacteria and there is an urgent need for discovering novel
antimicrobial compounds (Vicente et al. 2006; Wagner-
Döbler and Biebl 2006). The marine environment including
marine microorganisms may, as indicated, serve as source
of such compounds. Only a minor fraction of the marine
microbiota is culturable (Bernard et al. 2000; Kogure et al.
1980), and assessing the complete potential of bioactivity
of the entire marine microbial population would require
inclusion of both culturable and non-culturable organisms.
In principle, (meta) genomic data can be used to determine
the presence of genes encoding possible antibiotic activity
(Baker et al. 2007); however, it requires that the mecha-
nisms (and the genetic background) of antagonistic activity
are known. Also, if genes encoding antagonistic activity are
only present in specific niches, they may not be detected in
metagenomic data. For instance, luminous genes in Vibrio
fisheri were not found in data from the GOS expedition
(Nealson and Venter 2007). Therefore, we in the present
study focused on culturable bacteria to identify organisms
which could be further explored from a biotechnological
perspective. Further studies of physiology and assessment
of potential biotechnological use (Donia and Hamann 2003;
Fenical and Jensen 2006) are greatly facilitated by having
the organism in culture.
Marine microorganisms occupy different niches in the
oceanic environment. They are either planktonic, associated
with inert or biotic surfaces, or they inhabit the sediments.
Marine aggregates and surfaces of both live organisms and
inert objects are characterized by high bacterial densities
typically 1,000–10,000-fold higher than in the surrounding
water (DeLong et al. 1993; Grossart et al. 2005). Also,
culturability of surface-associated bacteria may be signifi-
cantly higher than in the surrounding water (Jensen et al.
1996). Long and Azam (2001) have suggested that bacteria
from marine particles may be an underutilized source of
novel antibiotics, and the high density and potential
interaction with a eukaryotic host may facilitate or select
for microorganisms with unique bioactivities (Egan et al.
2008). Indeed, both Nair and Simidu (1987) and Long et al.
(2005) found that a higher percentage of attached bacteria
than planktonic bacteria were antagonistic against other
bacteria.
Several species of within the Pseudoalteromonas genus
produce antibacterial compounds (Bowman 2007), and the
group is often associated with surfaces (Holmstrom and
Kjelleberg 1999), supporting the notion that attached
bacteria have a higher production of antibacterials than
their planktonic counterparts. The attachment and biofilm
formation may itself be an inducing factor for production of
Mar Biotechnol (2010) 12:439–451
antibacterials. The production of antibacterial compounds
such as tropodithietic acid (TDA) in several Roseobacter
clade strains seems associated with formation of a biofilm
at the air–liquid interface (Bruhn et al. 2005, 2007). Hence,
we in the present study sampled from both water and
surfaces of marine organisms or inert materials, addressing
the hypothesis that the bioactive potential of surface-
associated bacteria would be higher. Whilst this phenom-
enon has been studied in more local environments in a few
previous studies, our data represent the first attempt to
address the issue on a broader geographical scale.
Materials and Methods
Sample Collection and Preparation Samples for bacterial
counts were taken from seawater and as swabs from
surfaces of marine animals, weeds, or material (wood, plastic,
feathers) floating in the sea during the Danish marine
expedition, Galathea 3. The expedition left Denmark 11th
August 2006 and returned 25th April 2007, having visited six
continents and sailed 52,000 nautical miles. Seawater
samples were collected from 122 locations (Fig. 1) from
either CTD casts (114 samples), the stern tube inlet (five
samples; inlet 5 m below sea surface), or from coastal
waters (five samples). The conductivity–temperature–depth
(CTD) profiler was equipped with Sea-Bird sensors for
salinity, temperature, and pressure and a SBE32 carousel for
30 liter Niskin water sampler bottles. Seawater samples
were taken at two depths: from the surface (5 or 10 m
depending on latitude) and the deep chlorophyll maximum
(DCM; Supplementary Table 1). At the South American
West coast, CTD casts included four water samples from
the oxygen minimum zone. Surface-attached bacteria
were obtained by swabbing surfaces of marine animals
(snakes, mussels, fish, etc.), plants (seaweed), fungi,
stones from sediment samples, or material (wood, plastic)
floating in the ocean using sterile cotton swabs. When
analyzing material of small sizes, the complete animal or
plant was placed in a sterile tube, covered with sterile
seawater, and vortexed to remove surface-attached bacte-
ria. A total of 217 swab samples from 54 different
locations were analyzed (Fig. 1, Supplementary Table 2).
A number of swab samples were taken from animals or
plants caught in the sea chest filter, which was rinsed with
1–2-week intervals. A total of 92 samples from 18
locations were analyzed (Fig. 1, Supplementary Table 3).
The majority of seawater samples were divided into
three sub-samples: one being the raw sample, one compris-
ing the assumed particle-associated bacteria (>5µm), and
one comprising the non-associated, planktonic bacteria (<5
µm). Five milliliters of seawater was filtered through a 5-
µm-pore-size polycarbonate filter (GE Water & Process
Mar Biotechnol (2010) 12:439–451 441

Fig. 1 Sampling locations for

seawater samples, surface swab
samples, and samples from the
sea chest filter. All seawater
samples were analyzed for total
cell numbers and the fraction of
culturable bacteria, whereas sur-
face swab samples were only
analyzed for culturable bacteria.
Subsequently, all samples were
analyzed for the ability of the
colonies to inhibit Vibrio
anguillarum

Technologies cat. no. K50BP02500) to collect particle-
associated material. The filter was re-suspended in 5 ml of
sterile seawater and vortexed. The filtrate was analyzed as
the non-associated material.
Total Bacterial Cell Density Total bacterial densities were
determined using epifluorescence counts with SYBR Gold
as dye (Noble and Fuhrman 1998). Five milliliters of raw
seawater was filtered onto a 5-µm polycarbonate (PC) filter
(GE Water & Process Technologies cat. no. K50BP02500)
supported by an 8-µm mixed cellulose ester (MCE) filter
(GE Water & Process Technologies cat. no. E80WP02500).
The filtrate was subsequently filtered through a 0.02-µm
Anodisc filter (Whatman cat. no. 6809-6002) supported by
a 1.0-µm MCE filter (Advantes MFS Inc., CA, USA; no.
A100A025A). The filters were stained in a 0.35% SYBR
442
Gold solution for 15 min in the dark, washed twice in 0.02
µm filtered MilliQ water, mounted on a glass slide in
mounting buffer (50% glycerol, 50% phosphate buffered
saline; Oxoid BR0014) with 0.1% p-phenylenediamine
(P6001, Sigma Aldrich) and stored at −20°C. Fluorescent
cells were counted using an Olympus BX51 fluorescence
microscope with a 460–490-nm excitation filter and a
>510-nm barrier filter. Ten randomly chosen fields of vision
were counted from each filter.
Culturable Bacteria Aliquots of the raw seawater, the re-
suspended assumed particle-associated material, and the
filtrate were serially diluted 10-fold in sterile seawater and
plated on dilute (50%) marine agar (MA; Difco 2216) plates.
Dilute marine agar (50% MA) was prepared by mixing MA
with an equal volume of local seawater before autoclaving.
The swabs from surfaces of eukaryotic organisms or wood or
sediment stones were covered with 3–5 ml of sterile
seawater and vigorously vortexed. Samples were serially
diluted 10-fold and plated on MA (50%) plates. Plates were
incubated at 20°C until visible colony growth (3–7 days) and
colonies counted to determine culturable CFU/ml or CFU/
sample. The areas swabbed or rinsed could not be accurately
measured (e.g., a copepod) but it is estimated that they varied
between 1 and 10 cm2.
Isolation of Antibacterial Culturable Marine Bacteria Mar-
ine agar (50%) plates with visible colonies obtained from
both surfaces and water samples (see above) were replica-
plated onto an agar containing 10 ml solidified seawater
agar (local seawater with 1% agar, 0.3% Bacto casamino
acids [BD, product number 223050] supplemented with
0.4% glucose). The melted agar was held at 45–46°C and
mixed with 5–10µl liquid culture of Vibrio anguillarum
strain 90-11-287 per 10 ml agar (Hjelm et al. 2004). V.
anguilllarum strain 90-11-287 is a serotype O1 strain (Skov
et al. 1995) and was grown in marine broth (Difco 2216) at
20°C for 1–3 days before being embedded in agar. The
replica plates were incubated at 25°C for 1–2 days and
inspected for clearing zones in the agar, in which turbidity
was caused by growth of V. anguillarum. Colonies causing
inhibition of V. anguillarum were isolated from the original
MA (50%) plate and pure cultured. Isolates were frozen at
−80°C in marine broth (50%; MB diluted 1:1 with sterile
seawater) containing 30% glycerol. The initial replica
plating was done on board the research vessel, Vædderen,
and conditions did not allow screening against human
pathogenic bacteria.
Verification of Antibacterial Activity A total of 876 poten-
tial antibacterial strains were isolated from the seawater and
surface samples. All strains were pure cultured and re-tested
for antagonistic activity against V. anguillarum strain 90-
Mar Biotechnol (2010) 12:439–451
11-287 and Staphylococcus aureus strain 8325 (Novick
1967). Artificial seawater agar was prepared as described
above replacing ambient seawater with 3% Instant Ocean
(IO; Aquarium Systems Inc., Sarrebourg, France). In assays
with S. aureus as target organism, 1% of peptone (Difco
cat. no. 211677) was added to the medium. All marine
strains were grown on MA at 20°C until visible growth. A
colony of each strain was then spotted onto the agar cast
with either V. anguillarum or S. aureus, and plates were
incubated at 25°C for 24–48 h. Clearing zones in the turbid
growth of either V. anguillarum or S. aureus were scored as
antibacterial activity. Zones of clearing were graded in
categories depending on size of the clearing zone (mea-
sured outside the well area). Also, 0.5-mm zones were
scored with one plus, 0.5–1.5 mm with two plusses, and
larger zones with three or four plusses.
The marine strains were grown in MB (50%) using 3%
IO instead of seawater for 3 days at 25°C. Supernatants
were sterile-filtered and tested for antibacterial compounds
against V. anguillarum or S. aureus in a well diffusion assay
(Hjelm et al. 2004). IO-based seawater agar with the two
target bacteria was prepared as described above and 60µl of
supernatant pipetted into 6-mm wells. Plates were incubat-
ed at 25°C for 24 h and observed for clearing zones around
the wells. No attempt was made to vary or optimize growth
conditions before testing antibacterial activity of the culture
supernatant.
Strains identified as Ruegeria mobilis or pelagia were
consistently inhibitory and a sub-set of six strains were
tested for production of tropodithietic acid using previously
described culturing, extraction, and liquid chromatography–
mass spectroscopy procedures (Bruhn et al. 2005).
Phenotypic Grouping of Marine Bacteria A set of simple
phenotypic tests were used to group the culturable marine
bacteria (Bagge-Ravn et al. 2003; Hjelm et al. 2004). Gram
reaction was tested using Bactident® aminopeptidase
strips (Merck 1.13301) for detecting L-alanine aminopep-
tidase in Gram-negative bacteria, oxidase reaction tested
on oxidase slides (BBLTM Dryslide™ Oxidase slides,
Becton Dickinson and Company, Sparks, USA), and
catalase reaction by exposure to 3% H2O2. Shape and
motility was observed at 1,000× magnification in a phase
contrast microscope (Olympus BH2) after growth in MB
(50%) for 2–4 days at 20°C. The ability to metabolize
glucose by oxidation and/or fermentation was tested in OF
basal medium (Merck 1.10282) supplemented with 1.5%
Instant Ocean salts. Presumed Vibrio strains were tested for
sensitivity to vibriostaticum O/129 (2,4-diamino 6,7 di-
isopropyl pteridine) 10 and 150 μg (Oxoid 129150) on
marine agar. Also, the ability of the strains to grow on
thiosulfate citrate bile salts sucrose (TCBS) medium
(CM333, Oxoid) was tested at 20°C.
Mar Biotechnol (2010) 12:439–451
Sequencing of 16S rRNA Coding Gene and Genus or
Species Allocation Five hundred and nineteen strains were
randomly selected from all samples containing antagonistic
bacteria. A maximum of three strains were selected from
each sample with inhibitory bacteria. Partial sequencing of
the 16S rRNA gene was performed by Macrogen Co. Ltd.
(South Korea). The 16S rRNA coding gene was amplified
using the universal primers 27F (5′-AGAGTTTGATCMT
GGCTCAG-3′) and 1492R (5′-TACGGYTACCTTGTTAC
GACTT-3′). Sequencing was performed by Macrogen Co.
Ltd. using 518F (5′-CCAGCAGCCGCGGTAATACG-3′)
and 800R (5′-TACCAGGGTATCTAATCC-3′) as primers.
Identification to genus level was done using the BLAST-
based program leBIBI (Devulder et al. 2003). The
phylogeny of the Pseudoalteromonas strains is currently
being analyzed in more depth at our laboratory. The
sequences have been deposited in GenBank under acces-
sion numbers FJ457121–FJ457248 (Pseudoalteromonas
strains), FJ457249–FJ457301 (so-called other strains),
FJ457302–FJ457610 (Vibrionaceae), and FJ460022–
FJ460050 (Roseobacter clade strains).
Statistical Analysis To compare the level of culturable
bacteria between, e.g., raw and filtered water samples, the
difference in logarithm of the colony-forming units per
milliliter was calculated, and it was tested if the difference
was significantly different from 0 using Student's t test
(Hald 1952). Comparison of the prevalences of inhibitory
colonies in water samples and samples taken from surfaces
was done using Fisher exact test (Fischer 1958).
Results
Total Bacterial Cell Density in Surface Marine Waters Sea-
water samples from 122 locations were analyzed during the
Galathea 3 expedition (Fig. 1). Samples from 98 of the 114
CTD casts included both samples from surface waters (5–
10 m depth, SUR) and from the DCM. The average
bacterial density was 6.9×105 cells/ml at the surface and
7.4×105 cells/ml at the DCM (Table 1), and the bacterial
densities at these depths did not differ significantly
(p<0.01). Maximum cell densities were 5×106 cells/ml
(Table 1). In the DCM samples, 65 samples (of which all
but one were from the >5-μm sub-samples) contained less
than 104 cells/ml. In contrast, 66 samples had counts above
106 cells/ml and all 66 samples were non-particle-associated
sub-samples. A similar pattern was seen for total counts in
the ocean surface water samples where 70 samples had
counts below 104 cells/ml and all were particle-associated
sub-samples. All of the 69 samples with counts above
106 cells/ml were non-particle-associated sub-samples.
443
Samples with cell densities below average were taken close
to Greenland, the Antarctic continent, and in the Sargasso
Sea. Samples with a cell density above average were mostly
sampled in the upwelling zone along the South American
coast (Fig. 2).
Culturable Bacteria Culturable counts were in average
4.9×102 CFU/ml for ocean surface water samples and
4.3×102 CFU/ml for DCM samples, representing 0.13%
and 0.15% of the total cell density, respectively (Table 1).
The culturable counts in surface water and DCM samples
were not significantly different. The culturable density
varied 2–3 log units between sampling sites. The lowest
number of culturable bacteria was less than 10 CFU/ml and
the highest was 6.6×103 CFU/ml. Samples taken from and
around Greenland, the Antarctic, and the Sargasso Sea had
culturable counts below average, and samples from along
the South American coast, the West coast of Africa (both
having a upwelling zone), and close to the Solomon Islands
were above average. A total of 630 individual water
samples were analyzed. Of these, 93 had culturable counts
of less than 10 CFU/ml, 128 (20%) between 10 and 102
CFU/ml, 238 (38%) between 102 and 103 CFU/ml, and 90
(14%) between 103 and 104 CFU/ml. One or several of the
marine agar plates of 81 samples could not be counted since
a contaminating colony (likely contaminating when pouring
the plates) was spread during surface plating (Supplemen-
tary Table 1).
The colony-forming units per milliliter of presumed
particle-associated bacteria (>5 μm) and of non-associated
bacteria (<5 μm) in surface water and DCM samples were
not different (Table 1). There was a slight loss of bacteria
during filtration (0.1–0.2 log units) which, albeit low, was
statistically significant when compared to the culturable
counts of the raw water sample.
Some samples of live or inert surfaces were obtained by
vortexing a whole copepod or a small fish, whereas other
samples were cotton swabs of, e.g., fish gills. Also, the
amount of biomass (biofilm) may have varied between
samples. Hence, we have not been able to determine the
precise surface areas of each swab sample but estimate that
the average sampling area was between 1 and 10 cm2.
Culturable counts from surface samples varied from 102 to
106 CFU/sample (Supplementary Tables 2 and 3). The
counts do indicate that the level of culturable bacteria (per
sample unit) was between one and three orders of
magnitude higher on living and abiotic surfaces than in
open waters.
Isolation of Antibacterial Culturable Marine Bacteria Ap-
proximately 77,000 bacterial colonies, of which 28,000
were from seawater samples and 49,000 from swab samples
of live or inert surfaces, were replica-plated and tested for
444 Mar Biotechnol (2010) 12:439–451

Table 1 Total cell count and culturable bacteria in seawater samples collected during the Galathea 3 expedition

Water sample Fraction of water sample No. of individual samples Direct count Culturable counts
(log (cells/ml)) (log (CFU/ml))

Min Avg Max Min Avg Max

Surface water Raw water 122 nd nd nd <1.0 2.69 3.67

Particle associated (>5 μm) 112 <3.68b <3.68 4.53 <1.0 1.77 3.08
Non-associated (<5 μm) 112 4.63 5.84 6.83 <1.0 2.53 3.53
Deep chlorophyll maximum Raw water 94 nd nd nd <1.0 2.63 3.82
Particle associated (>5 μm) 92 <3.68 <3.68 4.53 <1.0 1.70 3.08
Non-associated (<5 μm) 92 <3.68 5.87 6.70 <1.0 2.51 3.64
Oxygen minimum zonea Raw water 4 – nd – – 3.67 –
Particle associated (>5 μm) 1 – 3.65 – – 1.85 –
Non-associated (<5 μm) 1 – 5.91 – – 3.48 –

Samples were taken from 114 CTD casts, of which 98 included samples from both surface waters and the deep chlorophyll maximum. At 16 CTD
casts, only one depth was sampled and these have been counted only as surface samples. At four casts, water from the oxygen minimum zone was
sampled. Five samples originated from the stern tube inlet (surface) and five samples from coastal water (surface). A total of 630 sub-samples
were analyzed
nd not done
a
Only data from one sample
b
Estimated limit of detection from fluorescence microscopy based on field of vision, magnification, and dilution

inhibitory activity against V. anguillarum. Of those, 900
(3%) from seawater and 6,200 (12.6%) from swabs of live
or inert surfaces were inhibitory against V. anguillarum in
the primary replica plating. This difference between
prevalence of inhibitory organisms on water and swab
surface samples was highly significant in the Fischer’s
exact test (p<0.001). Colonies were isolated to represent as
many different inhibition zones (differing in size and
clarity) as possible from each sample and a total of
approximately 900 strains were isolated and pure cultured.
The number of antagonistic isolates from each sample
8
Greenland W Australia
SW Africa Solomon Is
Log (cells or CFU/ml)
varied. In some samples, only one or two inhibitory
colonies were present; in others, as much as half of the
colonies on the replica plates were inhibitory.
Inhibitory bacteria were present in 40–50% of the
water samples. In the seawater samples (CTD casts and
stern tube), 40 of the surface water samples (38%) and
45 (46%) of the DCM samples harbored culturable
bacteria that caused inhibition of V. anguillarum after
replica plating. The level of inhibitory colonies was, on
average, 5% of the culturable count (Supplementary
Table 1). The DCM samples contained a higher percentage
of inhibitory culturable bacteria than did the surface water
samples (p<0.05) since 13% and 4% of the colonies from
Antarctic Sargasso
particle and non-particle samples were antagonistic,
SW America N Atlantic whereas the numbers for surface water samples were 8%
and 2%, respectively.

6
A total of 217 swab samples from live or inert surfaces
from 54 different locations were tested for presence of
4
inhibitory colonies (Supplementary Tables 2 and 3).
Colonies capable of inhibiting V. anguillarum were found
in 132 (61%) of these samples with an average of 13% of
2 the culturable bacterial count. The percentage of inhibitory
colonies varied in positive samples from 92% to 0.1%. The
material sampled varied along the route depending on the
0
Sampling along the Galathea 3 route
collaborating research projects on board (e.g., Antarctic fish
or sediment organisms at the Solomon Islands), and it was
Fig. 2 Total cell counts (filled squares—surface water, open squares— not possible systematically to sample on the complete
deep chlorophyll maximum) and culturable counts of marine bacteria
cruise from particular fish or algal genera. Therefore, we
(filled triangles—surface water, open triangles—deep chlorophyll
maximum) along the 52,000 nautical sea mile route of the Galathea 3 have not sub-divided the samples into more detail and
expedition (see Fig. 1 for map) cannot, based on this material, determine if particular
Mar Biotechnol (2010) 12:439–451
surfaces or organisms, in particular geographic locations,
harbor a higher level or prevalence of antagonistic bacteria.
A total of 92 swab samples from live or inert surfaces
samples from 18 sea chest filter rinsings were tested and
inhibitory colonies were found in 69 (75%) of the samples
(Supplementary Table 3). The average percentage of
inhibitory colonies was 31% (of culturable bacteria) in
positive samples, which, however, varied from 92% to 0.3%.
This variation did not appear to depend systematically on
sample type or geographic location (Supplementary Table 3)
Repeatedly, several different samples were taken from the
same sea chest rinsing containing different levels of
culturable bacteria and different colony morphologies,
indicating that the sampled microbiota remained alive and
associated with their particular host despite being caught in
the filter.
Phenotypic Grouping of Antagonistic Bacteria Of the 900
presumed antagonistic strains isolated, 876 were re-
culturable after frozen storage and ocean cruising. The vast
majority were Gram-negative motile rods with positive
oxidase and catalase reaction. Only two strains were clearly
Gram-positive and four to five strains were Gram-variable.
Five hundred and twenty seven strains fermented glucose,
whereas the remaining strains were either oxidative or
negative in OF medium. All glucose-fermenting strains
grew on TCBS and were sensitive to vibriostaticum,
indicating that they could be identified as Vibrio spp. The
non-fermentative strains were heterogeneous. Several
strains were intensely pigmented and appeared with yellow,
purple, red, or orange pigmentation. When a sub-set of
these were identified using 16S rRNA gene sequence
analysis, they clustered in the Pseudoalteromonas genus.
Several strains produced a brownish water-soluble pigment
and grew in a rosette-like manner. Subsequent 16S analyses
of these indicated that they were part of the Roseobacter
clade (R. mobilis or pelagia).
The 876 strains were re-tested for antibacterial activity
against V. anguillarum, and 680 strains (78%) remained
inhibitory against V. anguillarum. The strains were also
tested for inhibition of S. aureus, and 554 strains (63%)
were inhibitory against the Gram-positive organism. Eight
strains were only inhibitory against S. aureus and 132
strains only inhibited V. anguillarum. The antibacterial
activity varied in intensity with some bacteria causing only
a faint hazy zone of inhibition in the layer of target bacteria,
whereas others (primarily Pseudoalteromonas and Rose-
obacter clade strains) caused large zones of inhibition.
Identification of Marine Bacteria by 16S rRNA Gene
Sequencing Between one and five were isolated from each
sample, except for a few cases where a larger number (up to
30) were isolated (Supplementary Tables 1, 2, and 3). The
445
presumed inhibitory strains within a sample were often very
similar by phenotypic grouping (data not shown). There-
fore, to provide a more representative assessment of the
collection, 519 of the 876 strains were selected from the
complete sample material. The strains were selected to
ensure that each sample was represented (one to three
strains) and to ensure that each sample containing strains
being inhibitory in the re-test were included.
More than 50% of these 519 strains (58%) were identified
as Vibrionaceae (Vibrio or Photobacterium), and the
grouping based on partial 16S rRNA sequence homology
matched the phenotypic grouping. Twenty five percent
were identified as Pseudoalteromonas and the remaining
strains were members of the Roseobacter clade (6%) and
Halomonas (4%) (Table 2). Four hundred and nine of the
strains selected for sequencing were inhibitory upon re-
testing. For 110 strains, however, inhibitory activity could
not be detected upon sub-culturing and re-testing. Strains
that did not retain antagonistic activity were mostly found
within specific groups, e.g., Halomonas (Table 2).
BLAST analyses using leBiBi gave, for several strains,
an indication of the species identity (Supplementary
Table 4); however, a thorough phylogenetic analysis has
not been conducted as part of this study. Therefore, Table 2
presents genus identity. Twenty six of the 29 Roseobacter
clade strains were identified as Ruegeria spp. and 25 strains
were most likely R. mobilis or R. pelagia, which cannot be
distinguished on the basis of 16S rRNA gene homology.
Within the Pseudoalteromonas group, several strains were
suggested to be P. luteoviolaceum, P. rubra, and P.
piscicida, and this coincided with detection of colorful
pigments in the groups (purple, red, or yellow pigments).
The Vibrionaceae group was heterogeneous and several
different Vibrio species were likely to be present in our
collection, including potential fish and human pathogens.
Origin and Activity of Strains Antagonistic members of
Vibrio, Pseudoalteromonas, and the Roseobacter clade
were widely distributed around the globe, and no single
hot spots were found for these three major groups. The
degree of inhibitory activity as measured in a direct plating
against V. anguillarum varied. Large and clear zones of
inhibition were repeatedly caused by Ruegeria species and
several (pigmented) Pseudoalteromonas species. The in-
hibitory activity of the Vibrio group was much less
pronounced and only approximately 40 Vibrio strains gave
distinct zones of inhibition. Inhibitory strains were detected
at all legs of the route.
Presumed inhibitory Vibrio strains were isolated on all
legs of the voyage—in cold waters (south or north of 65°)
and also close to the equator—and the majority of
inhibitory Vibrio strains were isolated from swab samples
of surfaces (Table 2).
446 Mar Biotechnol (2010) 12:439–451

Table 2 Identification of 519 culturable marine bacteria isolated from seawater samples, surface swab samples, and sea chest rinsing samples
during the Galathea 3 expedition

Identification based No. of Percent No. of No. of strains Percent No. of No. of strains Percent
on 16S rRNA gene strains of total strains retaining inhibitory strains retaining inhibitory
sequence for 16S from inhibitiona surface from inhibitiona seawater
sequencing surfaces strains seawater strains

Vibriob 301 58 207 199 96 94 81 86

Pseudoalteromonasc 128 25 92 69 75 36 12 33
Ruegeriad,e 26 5 6 5 96 20 20 100
Cobetia 17 3 13 1 8 4 0 0
Photobacterium 8 2 5 3 60 3 2 67
Alteromonas 7 1 4 0 0 3 1 33
Pseudomonas 3 1 0 0 – 3 3 100
Flexibacter 3 1 3 0 0 0 0
Acinetobacter 2 0 0 0 – 2 2 100
Micrococcus 3 1 2 0 0 1 0 0
Spongiobacter 3 1 3 3 100 0 0 –
Halomonas 2 0 2 0 0 0 0 –
Marinomonas 2 0 2 1 50 0 0 –
Paracoccus 2 0 2 2 100 0 0 –
Psychrobacter 2 0 1 0 0 1 0 0
Psychromonas 2 0 1 1 100 1 1 100
Flavobacterium 1 0 1 1 100 0 1 –
Idiomarina 1 0 1 0 0 0 0 –
Joostella 1 0 1 0 0 0 0 –
Leeuwenhoekiella 1 0 0 0 – 1 0 0
Loknatellad 1 0 1 0 – 0 0 –
Mesonia 1 0 0 0 – 1 0 0
Roseobacterd 1 0 1 0 – 0 0 –
Sulfitobacterd 1 0 0 0 – 1 0 –
Total 519 100 348 284 82 171 123 72

The strains were isolated from 249 samples. Strains were isolated as presumed inhibitory following replica plating on Vibrio anguillarum. Also,
409 of the 519 strains retained inhibitory activity against either V. anguillarum or S. aureus after freeze storage, re-culturing, and re-testing in an
agar-spot assay. Most Vibrio strains had only a faint inhibition zone in re-testing
a
Inhibitory in spot assay against either V. anguillarum, S. aureus, or both
b
Many Vibrio strains with very faint inhibition zone in re-testing
c
Includes one strain of Glaciecola mesophila (Ps. atlantica)
d
Belonging to the Roseobacter clade
e
Non-inhibitory strain R. atlantica

Also, the majority of presumed inhibitory Pseudoalter-
omonas strains were isolated from swab samples from
surfaces and a higher percentage of the surface-associated
strains retained antibacterial activity when sub-cultured
(Table 2). Thus, 92 of 128 Pseudoalteromonas originated
from swabs of surfaces, and 69 of these were repeatedly
inhibitory. In contrast, only 36 Pseudoalteromonas strains
were from water samples and only 12 of these retained
some degree of antibacterial activity. The inhibitory
Pseudoalteromonas appeared to be specifically surface-
associated as they were not isolated from water sampled
from the same areas as the surfaces (e.g., copepods)
harboring antagonistic Pseudoalteromonas. When several
strains of inhibitory Pseudoalteromonas were isolated from
the same sample, they were mostly identified as the same
species. However, a few samples (e.g., a copepod and a sea
weed) harbored different species (rubra and piscicida or
phenolica and rubra and luteoviolaceae). Most (54) of the
70 strains of Pseudoalteromonas that retained inhibitory
activity upon sub-culture were isolated in sub-tropical and
tropical waters (Supplementary Tables 1, 2, and 3). Ps.
ruthenica constituted an exception, as 11 of the 15 strains
tentatively identified as Ps. ruthenica were isolated from
Antarctic areas.
Mar Biotechnol (2010) 12:439–451 447

In the smaller collection of Roseobacter clade strains, the
surface-antagonism pattern was reversed, and eight of 29
strains originated from swabs of surfaces and 21 from
ocean water samples. In the swab sample group, five strains
retained inhibition, whereas 20 of 21 ocean water strains
repeatedly gave large clear zones of inhibition. The four
strains where inhibition was not reproduced were tentative-
ly identified as Roseobacter prionitis, Loktanella spp.
(swab of surfaces), Sulfitobacter pontiacus (water sample),
and Ruegeria atlantica (algae). Twenty five of the 26
Ruegeria strains were repeatedly inhibitory, and 16S
homology identified these as R. mobilis or R. pelagia.
They all produced a brownish pigment which was most
pronounced under static growth conditions. The Ruegeria
strains were isolated along the entire route; however, no
isolates were found north of latitude 53.49° or south of
latitude −42.57° (Table 3, Fig. 3), indicating that their
prevalence in Arctic and Antarctic waters was limited. Six
strains were randomly selected and all produced tropodi-
thietic acid which is believed to be the antibacterial
compounds of several roseobacters (Brinkhoff et al. 2004;
Bruhn et al. 2007; Porsby et al. 2008).
When cultured in liquid MB, only 79 of the 418 strains
produced inhibitory compound(s) that could be detected in
the sterile-filtered culture supernatant. These 79 strains
belonged to the Roseobacter clade (28 strains), to Pseu-
doalteromonas (32 strains), to Vibrio (ten strains), and the
two remaining strains were identified as Photobacterium
spp. Spent culture medium from the Ruegeria strains was
primarily inhibitory to V. anguillarum, whereas super-
natants from the other groups mostly were inhibitory to
S. aureus.
Discussion
Several marine microorganisms are capable of producing
antibacterial compounds, and we demonstrate here that it is
a trait found in marine bacteria on a global scale. In the
present study, we found that, in the fraction of the
microbiota culturable under the conditions chosen (dilute
MA, 20°C), this trait was most predominant in three groups
of marine bacteria: Vibrio, Pseudoalteromonas, and Rue-
geria. In agreement with this, Long and Azam (2001) found
that members of the Vibrionales and Alteromonadales were
the most prominent producers of antibacterial substances,
and antibacterial secondary metabolites have been detected
in several Roseobacter clade strains (Buchan et al. 2005;
Wagner-Döbler and Biebl 2006), including very recently
also in Ruegeria species (Porsby et al. 2008). All three
groups are readily culturable—or contain species that are
readily culturable—and therefore likely members of our
strain collection. Pseudomonads which are easily culturable

Table 3 Origin and geographical location of 25 inhibitory Ruegeria mobilis/pelagia isolated on the Galathea 3 expedition

Arbitrary sample no. No. of strains Latitude (+=N) Longitude (+=E) Source TDA detected at
stagnant growtha

27 1 53.49 −23.3306 CTD +

28 3 1.61 1.64158 CTD +
29 1 −7.421 155.603 CTD nd
30 1 −9.4973 −79.6658 CTD nd
31 1 −10.179 157.594 CTD nd
32 2 −13.872 −76.8039 CTD nd
33 3 −16.023 119.323 CTD +
34 2 −17.004 120.779 Copepod +
14 1 −17.775 121.866 CTD nd
35 1 −20.052 115.812 Sediment nd
36 2 −26.304 −71.2624 CTD nd
37 2 −27.616 15.1324 CTD nd
38 1 −29.286 −71.883 CTD nd
39 1 −31.406 91.1776 CTD nd
40 1 −37.217 72.7091 Copepod +
41 2 −42.572 149.667 CTD +

Four strains isolated from surface swabs and 21 from water. Six strains were tested for production of tropodithietic acid. Samples ranked
according to latitude
nd not done
a
One strain from each sample
448
Fig. 3 Isolation sites (16) of 25
strains of Ruegeria mobilis/
pelagia on the Galathea 3 cruise.
All 25 strains were antagonistic
and produced brown pigment
when grown in marine broth
under stagnant conditions
and contain many antagonistic species were only repre-
sented with two to three strains; however, this could be due
to salinity of the waters always being 30 ppm or above and
the preference of pseudomonads for a lower salinity.
The cell density in surface waters was remarkably
constant along the Earth’s circumnavigation, and the level
of 105–106 cells/ml was in agreement with reports from
other studies (Bernard et al. 2000; DeLong et al. 1993;
Selje and Simon 2003). It is well known (Giovannoni and
Rappé 2000), as also demonstrated in the present study, that
only a fraction of the marine bacterioplankton is culturable.
Some studies report culturability rates of approximately 1%
(Bernard et al. 2000), whereas we in the present study
found only 0.15%. Clearly, the choice of medium and
incubation temperature influence the culturability rate, and
using a nutrient-rich agar (50% marine agar) may have
prevented growth of several sub-groups that require low
nutrient conditions to proliferate (Rappe et al. 2002).
We found that bacteria associated with live or inert
surfaces were more likely to display antibacterial activity.
Other studies (Long and Azam 2001; Nair and Simidu
1987) have on more limited collections suggested a similar
pattern. Surfaces are nutrient rich and hence are attractive
niches, and a microorganism with a competitive advantage
is likely to be a successful colonizer. We also attempted to
separate aggregates from water by filtering through 5-µm
filters. However, counts in these samples were not two to
three orders of magnitude higher as would be expected
(DeLong et al. 1993; Grossart et al. 2005) if we had indeed
retained the aggregates. The sampling and draining from
the CTD rosette may have disintegrated the aggregates or
the pore size of 5µm may have been slightly too large
causing disaggregation rather than retention.
Approximately 20% of the strains did not at all display any
bioactivity upon sub-culturing and re-testing. Also, the
antibacterial activity of many strains (particularly Vibrio
species) was reduced causing only hazy, small zones of
clearing in the V. anguillarum layer. The presumed “loss” of
Mar Biotechnol (2010) 12:439–451
activity was not random but associated with sub-groups or
species (Table 2, Supplementary Table 4). For instance, in
the Roseobacter clade, all three non-Ruegeria strains no
longer inhibited Vibrio whilst all the R. mobilis/pelagia
strains retained inhibitory activity. In the Pseudoalteromonas
group, it was predominantly the non-pigmented strains, e.g.,
P. marina, that did not retain inhibition. This indicated that
the inhibition by particular species may be lost upon sub-
culturing, that they in the environmental sample are tightly
associated with other, potentially inhibitory organisms that
escaped isolation, or that this phenotype is influenced by
environmental factors, e.g., from the local seawater used to
dilute the primary MA plates. Also, one can speculate that
some of the inhibitory zones in the initial screening were
caused by Vibrio bacteriophages undergoing lysis during
bacterial growth (Wommack and Colwell 2000).
We do not yet know by which mechanism(s) these
marine bacteria inhibit other bacteria; however, several
(Pseudoalteromonas and Ruegeria) produce extracellular
secondary metabolites (Bowman 2007; Wagner-Döbler and
Biebl 2006), whereas others may interact by competing for
nutrients (Fredrickson and Stephanopoulos 1981). For
instance, the Vibrio genus produces iron-sequestering side-
rophores (Crosa and Walsh 2002), and differences in
chelation efficiency may mediate interactions also with
other vibrios. Our assessment of antibacterial activity was
done with very few growth conditions (surface growth on
MA or planktonic growth in MB at 25°C), and as
environmental conditions can have a dramatic impact on
secondary metabolite profile and antagonism, other culture
conditions may reveal other antagonistic strains and,
likewise, altering the growth conditions may enhance or
suppress antagonism. For instance, the addition of a single
component (anthranilic acid) altered the secondary metab-
olite spectra of a marine Halomonas (Bitzer et al. 2006),
and surface association appears to facilitate production of
antibacterial compounds in several marine organisms
(Bruhn et al. 2005; Yan et al. 2002).
Mar Biotechnol (2010) 12:439–451
16S rRNA gene sequence homology was used to provide
a tentative identification, and on an overall scale, specific
groups appeared to be associated with antagonistic activity.
For example, all strains identified as Ps. luteoviolaceae or
Ps. piscicida were consistently inhibitory. However, within
each of these groups, different secondary metabolites may
cause the inhibition and an identification scheme including
homology of other genes (e.g., gyrase B) may reveal sub-
clusters that could have different inhibitory spectra (Grossart
et al. 2004). Also, combining a more detailed phylogeny
with the secondary metabolite profile of the strains may
reveal clusters (species or sub-species) with different
antibacterial compounds (Jensen et al. 2007).
Whilst Vibrio spp. have mostly attracted interest as
bacteria causing disease in aquatic organisms or humans,
the Pseudoalteromonas genus and Roseobacter clade are
known to harbor species that produce antibacterial com-
pounds (Bowman 2007; Brinkhoff et al. 2004; Bruhn et al.
2007; Wagner-Döbler and Biebl 2006). A few reports have
demonstrated that antagonistic marine organisms also reside
in the Vibrionaceae family (Castro et al. 2002; Long and
Azam 2001; Towse 2005) including species known to
harbor pathogens such as V. parahaemolyticus (Radjasa et
al. 2007), V. anguillarum (Hjelm et al. 2004), or V.
alginolyticus (Austin et al. 1995).
We found that several Pseudoalteromonas strains were
consistently producing antibacterial compounds. Despite
sampling living surfaces extensively, we did not isolate P.
tunicata which is a well-described bioactive species
associated with surfaces (Holmstrom et al. 1996). The
majority of pseudoalteromonads retaining antagonistic
activity were isolated from swabs of live or inert surfaces
and in several areas (e.g., Western Australia) the inhibitory
pseudoalteromonads were never isolated from parallel
water samples, indicating a preference of these organisms
for colonization of surfaces. Some strains produced
antibacterial compounds in liquid culture and some com-
pounds were only detectable after extraction in ethylacetate.
This indicates that a variety of chemical bioactive com-
pounds are present in our strain collection.
The Roseobacter clade is one of the most common
marine bacterioplankton groups constituting as much as
10–30% of prokaryotic DNA in some niches (Buchan et al.
2005), and our study is the first to demonstrate that readily
culturable species are found globally. The clade is geo-
graphically widely distributed; however, some species or
sub-groups appear to be restricted to, e.g., temperate or
polar regions (Bruhn et al. 2005; Pommier et al. 2005; Selje
et al. 2004; Yan et al. 2002). We almost exclusively isolated
R. mobilis/pelagia within the Roseobacter clade, and except
for the Arctic and Antarctic waters, they were found at all
parts of the route indicating an almost cosmopolitan
distribution. We wondered if our Ruegeria strains were
449
similar to the Roseobacter clade organisms (SCB32) which
is a true cosmopolitan (Pommier et al. 2005); however,
SCB32 is likely a Silicibacter species (BLAST search), and
in the Roseobase (www.roseobase.org) it does not cluster
with strains (such as Phaeobacter BS107) which produce
TDA such as do our Ruegeria strains. Our finding of
predominantly Ruegeria spp. is likely caused by our
combination of medium and temperature; however, it is
slightly surprising that other known antagonistic Rose-
obacter clade strains (such as Phaeobacter inhibens or
gallaciencis) that are easily cultured were not detected. We
had assumed that a larger number of surface-associated
inhibitory Roseobacter clade strains would be isolated in
our study; however, in agreement with our lack of finding
such strains, Long and Azam (2001) demonstrated that the
surface-associated alpha-Proteobacteria were not inhibitory
to Vibrio. Twenty five Ruegeria strains were, based on 16S
rRNA partial sequences, similar to R. pelagia (Lee et al.
2007) or R. mobilis (Muramatsu et al. 2007) which appear
indistinguishable on the basis of 16S rRNA or gyraseB
gene sequence homology (Porsby et al. 2008). They all
produced a brown pigment described in some Ruegeria and
Phaeobacter species (Bruhn et al. 2007; Jensen et al. 2007;
Porsby et al. 2008), and all of a sub-set of six strains tested
produced the antibacterial compound tropodithietic acid. A
similar species-specific metabolite production has been
found in marine actinomycetes Salinospora (Jensen et al.
2007), and our finding of TDA being produced by R.
mobilis/pelagia independent of or isolation site confirms
that “a consistent phenotype can be associated with a
globally distributed population” (Jensen et al. 2007) and
likely essential for species survival.
The rapid increase and spread of antibiotic-resistant
bacteria has caused a revival of research aimed at
identifying and developing novel antibacterial compounds.
This has led to a renewed interest in natural product
discovery, and the marine environment represents an
underexplored and exciting global niche of novel bioactive
compounds. While several studies have demonstrated that
marine bacteria are a promising source of bioactive
compounds (Burgess et al. 1999; Debnath et al. 2007; Lee
et al. 2007), our study represents the first study to elucidate
this potential on a global scale. We identified several groups
of highly bioactive bacteria and demonstrated that, while
antibacterial marine bacteria occur in all marine niches
investigated in this study, there appears to be a global
sectioning or niche specificity for particular species. Our
strain collection represents a major resource for biopro-
specting and development of marine biotechnology. Further
studies of the bioactivities of these organisms may give rise
to structurally novel antibacterial compounds or known
compounds with yet to be discovered biological effects.
Also, the collection will allow us to address issues of
450
evolutionary and biological importance such as the second-
ary metabolite profile within a species or genus as a
function of geographic origin.
Acknowledgments The Carlsberg Foundation and the Danish
Center for Food Research (LMC) are thanked for donations covering
international travel. Funding from the Programme Committee for
Food, Health and Welfare under the Danish Strategic Research
Council is acknowledged. We express our thankfulness to Drs. Jørn
Smedsgaard and Kristian Fog Nielsen and technician Ellen Kirstine
Lyhne for participation in and sampling done during the cruise. We
thank Kerstin Geitner and Andreas Espersen for help with Fig. 1 and
Matthias Wietz for help with Fig. 3. The present work was carried out
as part of the Galathea 3 expedition under the auspices of the Danish
Expedition Foundation. This is Galathea 3 contribution no. P50.
References
Austin B, Stuckey LF, Robertson P, Effendi I, Griffith D (1995) A
probiotic strain of Vibrio alginolyticus effective in reducing
diseases caused by Aeromonas salmonidica, Vibrio anguillarum,
and Vibrio ordalli. J Fish Dis 18:93–96
Azam F, Malfatti F (2007) Microbial structuring of marine ecosys-
tems. Nat Rev Microbiol 5:782–791
Bagge-Ravn D, Ng Y, Hjelm M, Christiansen JN, Johansen C, Gram L
(2003) The microbial ecology of processing equipment in
different fish industries—analysis of the microflora during
processing and following cleaning and disinfection. Int J Food
Microbiol 87:239–250
Baker DD, Chu M, Oza U, Rajgarhia V (2007) The value of natural
products to future pharmaceutical discovery. Nat Prod Rep
24:1225–1244
Barja JL, Lemos ML, Toranzo AE (1989) Purification and character-
ization of an antibacterial substance produced by a marine
Alteromonas species. Antimicrob Ag Chem 33:1674–1679
Bernard L, Schäfer H, Joux F, Courties C, Muyzer C, Lebaron P
(2000) Genetic diversity of total, active and culturable marine
bacteria in coastal seawater. Aquat Microb Ecol 23:1–11
Bitzer J, Grosse T, Wang LZ, Lang S, Beil W, Zeeck A (2006) New
aminophenoxazinones from a marine Halomonas sp.: fermenta-
tion, structure elucidation, and biological activity. J Antibiot
59:86–92
Blunt JW, Copp BR, Hu WP, Munro MHG, Northcote PT, Prinsep
MR (2008) Marine natural products. Nat Prod Rep 25:35–94
Bowman JP (2007) Bioactive compound synthetic capacity and
ecological significance of marine bacterial genus Pseudoalter-
omonas. Mar Drugs 5:220–241
Brinkhoff T, Bach G, Heidorn T, Liang LF, Schlingloff A, Simon M
(2004) Antibiotic production by a Roseobacter clade-affiliated
species from the German Wadden Sea and its antagonistic
effects on indigenous isolates. Appl Environ Microbiol 70:
2560–2565
Bruhn JB, Nielsen KF, Hjelm M, Hansen M, Bresciani J, Schulz S,
Gram L (2005) Ecology, inhibitory activity, and morphogenesis
of a marine antagonistic bacterium belonging to the Roseobacter
clade. Appl Environ Microbiol 71:7263–7270
Bruhn JB, Gram L, Belas R (2007) Production of antibacterial
compounds and biofilm formation by Roseobacter species are
influenced by culture conditions. Appl Environ Microbiol
73:442–450
Buchan A, Gonzalez JM, Moran MA (2005) Overview of the marine
Roseobacter lineage. Appl Environ Microbiol 71:5665–5677
Mar Biotechnol (2010) 12:439–451
Bull AT, Stach JEM (2007) Marine actinobacteria: new opportunities
for natural product search and discovery. Trends Microbiol
15:491–499
Burgess JG, Jordan EM, Bregu M, Mearns-Spragg A, Boyd KG
(1999) Microbial antagonism: a neglected avenue of natural
products research. J Biotechnol 70:27–32
Burkhold PR, Pfister RM, Leitz FH (1966) Production of a pyrrole
antibiotic by a marine bacterium. Appl Microbiol 14:649–653
Castro D, Pujalte MJ, Lopez-Cortes L, Garay E, Borrego JJ (2002)
Vibrios isolated from the cultured manila clam (Ruditapes
philippinarum): numerical taxonomy and antibacterial activities.
J Appl Microbiol 93:438–447
Crosa JH, Walsh CT (2002) Genetics and assembly line enzymology
of siderophore biosynthesis in bacteria. Microbiol Mol Biol Rev
66:223–249
Debnath M, Paul AK, Bisen PS (2007) Natural bioactive compounds
and biotechnological potential of marine bacteria. Curr Pharm
Biotechnol 8:253–260
DeLong EF, Franks DG, Alldredge AL (1993) Phylogenetic diversity
of aggregate-attached vs. free-living marine bacterial assemb-
lages. Limnol Ocean Meth 38:924–934
Devulder G, Perriere G, Baty F, Flandrois JP (2003) BIBI, a
bioinformatics bacterial identification tool. J Clin Microbiol
41:1785–1787
Donia M, Hamann MT (2003) Marine natural products and their
potential applications as anti-infective agents. Lancet Infect Dis
3:338–348
Egan S, Thomas T, Kjelleberg S (2008) Unlocking the diversity and
biotechnological potential of marine surface associated microbial
communities. Curr Opin Microbiol 11:219–225
Falkowski PG, Barber RT, Smetacek V (1998) Biogeochemical
controls and feedbacks on ocean primary production. Science
281:200–206
Fenical W, Jensen PR (2006) Developing a new resource for drug
discovery: marine actinomycete bacteria. Nat Chem Biol 2:666–
673
Field CB, Behrenfeld MJ, Randerson JT, Falkowski P (1998) Primary
production of the biosphere: integrating terrestrial and oceanic
components. Science 281:237–240
Fischer RA (1958) Statistical methods for research workers. Hafner,
New York
Fredrickson AG, Stephanopoulos G (1981) Microbial competition.
Science 213:972–979
Gauthier MJ, Flatau GN (1976) Antibacterial activity of marine violet-
pigmented Alteromonas with special reference to production of
brominated compounds. Can J Microbiol 22:1612–1619
Giovannoni SJ, Rappé M (2000) Evolution, diversity and molecular
ecology of marine procaryotes. In: Kirchman DL (ed) Microbial
ecology of the ocean. Wiley-Liss, New York, pp 47–84
Grossart HP, Schlingloff A, Bernhard M, Simon M, Brinkhoff T
(2004) Antagonistic activity of bacteria isolated from organic
aggregates of the German Wadden Sea. FEMS Microbiol Ecol
47:387–396
Grossart HP, Levold F, Allgaier M, Simon M, Brinkhoff T (2005)
Marine diatom species harbour distinct bacterial communities.
Env Microbiol 7:860–873
Hald A (1952) Statistical theory with engineering applications. Wiley,
New York
Hjelm M, Bergh O, Riaza A, Nielsen J, Melchiorsen J, Jensen S,
Duncan H, Ahrens P, Birkbeck H, Gram L (2004) Selection and
identification of autochthonous potential probiotic bacteria from
turbot larvae (Scophthalmus maximus) rearing units. Syst Appl
Microbiol 27:360–371
Holmstrom C, Kjelleberg S (1999) Marine Pseudoalteromonas species
are associated with higher organisms and produce biologically
active extracellular agents. FEMS Microbiol Ecol 30:285–293
Mar Biotechnol (2010) 12:439–451
Holmstrom C, James S, Egan S, Kjelleberg S (1996) Inhibition of
common fouling organisms by marine bacterial isolates with
special reference to the role of pigmented bacteria. Biofouling
10:251–259
Jensen PR, Fenical W (1996) Marine bacterial diversity as a resource
for novel microbial products. J Ind Microbiol Biotechnol 17:346–
351
Jensen PR, Kauffman CA, Fenical W (1996) High recovery of
culturable bacteria from the surfaces of marine algae. Mar Biol
126:1–7
Jensen PR, Williams PG, Oh DC, Zeigler L, Fenical W (2007)
Species-specific secondary metabolite production in marine
actinomycetes of the genus Salinispora. Appl Environ Microbiol
73:1146–1152
Kogure K, Simidu U, Taga N (1980) Distribution of viable marine
bacteria in neritic seawater around Japan. Can J Microbiol
26:318–323
Lee K, Choo YJ, Giovannoni SJ, Cho JC (2007) Ruegeria pelagia sp
nov., isolated from the Sargasso Sea, Atlantic Ocean. Int J
System Evol Microbiol 57:1815–1818
Long RA, Azam F (2001) Antagonistic interactions among marine
pelagic bacteria. Appl Environ Microbiol 67:4975–4983
Long RA, Rowley DC, Zamora E, Liu JY, Bartlett DH, Azam F
(2005) Antagonistic interactions among marine bacteria impede
the proliferation of Vibrio cholerae. Appl Environ Microbiol
71:8531–8536
Muramatsu Y, Uchino Y, Kasai H, Suzuki KI, Nakagawa Y (2007)
Ruegeria mobilis sp nov., a member of the Alphaproteobacteria
isolated in Japan and Palau. Int J System Evol Microbiol
57:1304–1309
Nair S, Simidu U (1987) Distribution and significance of heterotro-
phic marine bacteria with antibacterial activity. Appl Environ
Microbiol 53:2957–2962
Nealson KH, Venter JC (2007) Metagenomics and the global ocean
survey: what's in it for us, and why should we care? ISME J
1:185–187
Noble RT, Fuhrman JA (1998) Use of SYBR Green I for rapid
epifluorescence counts of marine viruses and bacteria. Aquat
Microb Ecol 14:113–118
Novick R (1967) Properties of a cryptic high-frequency transducing
phage in Staphylococcus aureus. Virology 33:155–166
451
Pommier T, Pinhassi J, Hagstrom A (2005) Biogeographic analysis of
ribosomal RNA clusters from marine bacterioplankton. Aquat
Microb Ecol 41:79–89
Porsby CH, Nielsen KF, Gram L (2008) Phaeobacter and Ruegeria
species of the Roseobacter clade colonize separate niches in a
Danish turbot (Schophthalmus maximus)-rearing farm and antag-
onize Vibrio anguillarum under different growth conditions. Appl
Environ Microbiol 74:7356–7364
Radjasa OK, Sabdono A, Zocchi J, Zocchi E (2007) Richness of
secondary metabolite-producing marine bacteria associated with
the sponge Haliclona sp. Int J Pharmacol 3:275–279
Rappe MS, Connon SA, Vergin KL, Giovannoni SJ (2002) Cultiva-
tion of the ubiquitous SAR11 marine bacterioplankton clade.
Nature 418:630–633
Selje N, Simon M (2003) Composition and dynamics of particle-
associated and free-living bacterial communities in the Weser
estuary, Germany. Aquat Microb Ecol 30:221–237
Selje N, Simon M, Brinkhoff T (2004) A newly discovered Rose-
obacter cluster in temperate and polar oceans. Nature 427:445–
448
Skov MN, Pedersen K, Larsen JL (1995) Comparison of pulsed field
gel electrophoresis, ribotyping, and plasmid profiling for typing
of Vibrio anguillarum serovar O1. Appl Environ Microbiol
61:1540–1545
Strom SL (2008) Microbial ecology of ocean biogeochemistry: a
community perspective. Science 320:1043–1045
Towse SJ (2005) Antimicrobial activity of marine Vibrio sp., isolate
NI-22. Thesis/dissertation, University of Winnipeg.
Vicente M, Hodgson J, Massidda O, Tonjum T, Henriques-Normark
B, Ron EZ (2006) The fallacies of hope: will we discover new
antibiotics to combat pathogenic bacteria in time? FEMS Micro-
biol Rev 30:841–852
Wagner-Döbler I, Biebl H (2006) Environmental biology of the
marine Roseobacter lineage. Ann Rev Microbiol 60:255–
280
Wommack KE, Colwell RR (2000) Virioplankton: viruses in aquatic
ecosystems. Microbiol Mol Biol Rev 64:69–114
Yan LM, Boyd KG, Burgess JG (2002) Surface attachment induced
production of antimicrobial compounds by marine epiphytic
bacteria using modified roller bottle cultivation. Mar Biotechnol
4:356–366