RESEARCH ARTICLE
Rhodobacteraceae are the key members of the microbial
community of the initial biofilm formed in Eastern
Mediterranean coastal seawater
Hila Elifantz1,2, Gilad Horn3, Meir Ayon1, Yehuda Cohen1 & Dror Minz2
1
Department of Plant and Environmental Sciences, The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem,
Jerusalem, Israel; 2Department of Soil, Water and Environmental Sciences, Agricultural Research Organization - the Volcani Center, Bet-Dagan,
Israel; and 3Global Environmental Solutions Ltd., Petach Tikva, Israel
Correspondence: Dr Dror Minz, Department
of Soil, Water and Environmental Sciences,
Agricultural Research Organization - the
Volcani Center, P.O.B. 6, Bet-Dagan 50250,
Israel. Tel.: +972 3 9683316; fax:
+972 3 9604017; e-mail: [email protected].
gov.il
Received 26 November 2012; revised
9 March 2013; accepted 24 March 2013.
Final version published online 18 April 2013.
DOI: 10.1111/1574-6941.12122
MICROBIOLOGY ECOLOGY
Editor: Gary King
Keywords
16S rRNA; FISH-CLSM; initial biofilm;
Rhodobacteraceae.
Abstract
The formation of biofilms and biofouling is a common feature in aquatic environments. The aim of this study was to identify the primary colonizers of biofilm formed in Eastern Mediterranean Coastal water at different seasons and
follow early dynamics of biofilm community development. Pre-treated coastal
seawater and biofilm samples were collected from six different sampling events
of 2 weeks’ duration each during 1 year. The microbial community composition and specific abundance were estimated by 16S rRNA gene clone libraries
and fluorescence in situ hybridization–confocal laser scanning microscopy
(FISH-CLSM), respectively. The biofilm formed over the course of the year
was fairly consistent in terms of community composition and overall abundance with the exception of spring season. Alphaproteobacteria (30–70% of
total bacteria), in particular Rhodobacteraceae, were the dominant bacteria in
the biofilm, regardless of season, followed by Bacteroidetes (5–35%) and
Gammaproteobacteria (6–35%). There was a decrease in relative abundance of
Alphaproteobacteria and an increase in the abundance of Bacteroidetes between
the initial and 2-week-old biofilm. This observation may aid man-made facilities that have to deal with biofilm formation and help the development of
appropriate strategies to control those biofilms.
Introduction
Biofouling is a common phenomenon in aquatic environments. Biofilms are formed naturally on flora and fauna
(Rao et al., 2005; Egan et al., 2008), sediments (Heijs
et al., 2008) and man-made structures such as ships
(Candries & Atlar, 2003), pipelines (L
opez et al., 2006),
nanofiltration membranes (Ivnitsky et al., 2010) and
reverse osmosis membranes (Vrouwenvelder & van der
Kooij, 2001). Some of these biofilms can cause costly
damages such as corrosion (Al-Malahy & Hodgkiess,
2003), drag force of ships (Candries & Atlar, 2003) and a
decrease in desalination process effectiveness (Vrouwenvelder
& van der Kooij, 2001). In desalination plants, biofilm can
form on various components employed in the process
(pipes, holding tanks, membranes) and tremendously
increase operational costs, and consequently water
prices (Schneider et al., 2005). Therefore, it is important
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
to understand the nature of these biofilms in order to
develop optimal operational strategies.
A basic understanding of marine biofilms has been
acquired in recent years via culture-independent techniques, including clone libraries and fluorescent in situ
hybridization (FISH). The most dominant bacteria on submerged surfaces in the coastal Atlantic and Pacific Oceans
were shown to be affiliated with the Alphaproteobacteria
class, and in particular the Rhodobacteraceae family, followed by the Gammaproteobacteria class and the Bacteroidetes phylum (Dang & Lovell, 2000, 2002; Jones et al.,
2006; Dang et al., 2008). In contrast, Gammaproteobacteria
are the dominant bacteria in biofilms formed on pipelines
transporting seawater from the Gulf of Mexico (L
opez
et al., 2006), as well as on glass slides submerged in estuarine water at East Sabine Bay, Florida (Moss et al., 2006).
The aim of this study was to identify and quantify the
primary colonizers responsible for initial biofilm
FEMS Microbiol Ecol 85 (2013) 348–357
349
Primary colonizers of biofilm in Mediterranean coastal seawater
development in Mediterranean coastal seawater over a
course of a year. A comprehensive seasonal perspective of
biofilm dynamics may help develop tactics that can
reduce biofilm formation and henceforth improve operational efficiency of man-made constructions. Recent studies have described the microbial community composition
of biofilms from various components of desalination
plants. One study suggested that the initial biofilm community was dominated by Gammaproteobacteria, and that
the mature community was dominated by Alpha- and
Betaproteobacteria (H€
orsch et al., 2005). Another study
suggested that the Sphingomonas subgroup, which belongs
to the Alphaproteobacteria, is the main component in the
biofilm (Bereschenko et al., 2010). Whereas those studies
focused on one season, the current study evaluated biofilm formation dynamics at different times during 1 year.
Materials and methods
Site description and sample collection
The Palmachim desalination plant is located on the central Mediterranean coast of Israel (about 15 km south of
Tel-Aviv). The plant desalinates shallow seawater, close to
the shore. To study biofilm formation, a medium pressure (up to 100 psi) biofilm device (Robbins Device,
Tyler Research, Edmonton, Canada) was used since it is
a convenient apparatus for biofilm formation and sampling. The device was positioned after a 15-lm particle
removing pre-filter and before the entry of the water to
the RO systems. The device contained 12 sterile glass
coupons (1 cm diameter each) later used for microbial
community composition and FISH analyses. To investigate the initial community of biofilm at different times
of the year, six 2-week experiments were performed
between May 2008 and May 2009 with a seawater flow
rate of 0.5 m3 h 1 at all experiments. In other marine
environments, longer incubation times were selected to
investigate mature biofilms (Acu~
na et al., 2006; Wietz
et al., 2009), whereas in the current study 2 weeks’ incubation was chosen to estimate the initial biofilm. The
length of experiments was determined to fit the growth
of biofilms developing in pretreated seawater even in
winter conditions. In each experiment, water was sampled once per week and biofilms were sampled after 1
and 2 weeks of flow for microbial community analysis by
16S rRNA gene clone libraries and FISH. Water was also
sampled for dissolved organic carbon (DOC), total
organic carbon (TOC), total organic nitrogen (TON),
and chlorophyll and these analyses were performed by a
certified commercial lab (Bactochem, Ness-Ziona, Israel).
Temperature was measured once a week, and the data
for each month was averaged.
FEMS Microbiol Ecol 85 (2013) 348–357
Samples of the planktonic community were collected
by filtering 750 mL of the cells collected onto the membrane 0.2-lm polycarbonate membrane (GE Water and
Process Technologies) for phylogenetic affiliation analysis
using a 16S rRNA gene clone library. For biofilm community composition, glass coupons were removed gently
from the medium pressure biofilm device on site and
placed in a 15-mL tube. All samples for DNA extraction
were stored at 80 °C until further processing. Seawater
samples of 50 mL for FISH analysis were fixed in 2%
paraformaldehyde overnight and collected on a 0.2-lm
polycarbonate membrane (GE Water and Process Technologies). Biofilm samples for FISH analysis were fixed in
4% paraformaldehyde for 4 h. Following fixation, samples
were washed with phosphate-buffered saline, pH 7.2
(36% (v/v) of 0.2 M Na2HPO4, 14% (v/v) of 0.2 M
NaH2PO4, 0.7% (w/v) of NaCl) for 30 min, and then
with double-distilled water for 10 min. The glass coupons
were then air dried and kept at 20 °C until further
analysis.
DNA extraction and clone library construction
and sequencing
DNA extraction was performed using the UltraClean soil
DNA isolation kit (MoBio Laboratories, Inc., CA) with
minor modifications to the initial step for the biofilm
samples. Sterile glass beads and 1 mL extraction buffer
were added to 15 mL Falcon tubes containing the glass
coupons or membranes. The total developed biofilm on
the coupon was used for community analysis. The tubes
were vigorously vortexed for 10 min and then centrifuged
at 4000 g for 15 min. Supernatants were transferred to
1.5-mL microcentrifuge tubes and the rest of the extraction process performed according to the manufacturer’s
instructions. The 16S rRNA gene fragments were amplified with the DreamTaq kit (Fermentas, Burlington,
Canada) using the general primers 11F (Kane et al., 1993)
and 1392R (Lane, 1991). The PCR products were visualized by agarose gel (1%) electrophoresis, and positive
samples with an apparent band at c. 1400 bp were used
for cloning. PCR products were ligated directly into the
pCR2.1 TOPO TA cloning vector (Invitrogen, Inc.),
according to the manufacturer’s instructions. Ligation
reactions were cloned and sequenced by the Genome
Sequencing Center at Washington University (St. Louis,
MO). A total of 2403 sequences were obtained using the
907R primer (Muyzer et al., 1993).
16S rRNA gene library data analysis
A base-calling analysis was performed on all sequences
using BIOEDIT (http://www.mbio.ncsu.edu/BioEdit/bioedit.
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
350
html). The sequences were aligned with the phylogenetic
software package ARB (Ludwig et al., 2004). Affiliation was
determined according to the location of the aligned
sequences in a tree constructed from 500 000 reference
sequences (2010 silva database). To determine the similarities among the samples, a neighbor-joining distance
matrix with Jackknife correction was constructed and
exported into UNIFRAC (Lozupone & Knight, 2005). Percent coverage was calculated by MOTHUR (Schloss et al.,
2009). On average, the libraries covered 66% of the diversity in the environment at the 95% similarity level (Tables
2 and 3). The sequences determined in this work were
submitted to GenBank under accession numbers
JF947375–JF949718.
FISH-CLSM
FISH was performed as described previously (Manz et al.,
1992) using the following phylogenetic probes, chosen
based on the sequence data obtained from the clone
libraries: Alf968 for Alphaproteobacteria (Gl€
ockner et al.,
1999), GRb for Rhodobacteraceae (Eilers et al., 2000),
CF319a for Bacteroidetes (Manz et al., 1996), Gam42a
combined with a competitive unlabeled probe Bet42a for
Gammaproteobacteria (Manz et al., 1992) and Pla46 for
Planctomycetes (Neef et al., 1998). Non-specific hybridization was tested using the NON338 probe (Wallner et al.,
1993). Hybridization was performed for 5 h at 46 °C
using 30% (Gam42a and Pla46) or 35% formamide
(Alf968, GRb, CF319a, and NON338) in hybridization
buffer (0.9 M NaCl, 20 mM Tris-HCl, pH 7.2, 5 mM
EDTA, 0.01% sodium dodecylsulfate, SDS). Following
hybridization, samples were washed in a wash buffer
(20 mM Tris-HCl, pH 7.2, 10 mM EDTA, 0.01% SDS,
and NaCl according to formamide percentage; 102 and
80 mM for 30% and 35% formamide, respectively) for
1 h at 48 °C, to remove excess and non-specifically
bound probes. The samples were then counterstained
with 0.5 ng lL 1 4′,6′-diamidino-2-phenylindole (DAPI)
and kept at 20 °C until data collection. Confocal laser
scanning microscopy was performed using the Leica SP5
(Leica, Germany). Twenty fields of view were collected
for each sample.
FISH-CLSM data analysis
The CLSM files were compiled in the IMAGEJ software
(http://rsbweb.nih.gov/ij/) and the cells labeled with individual probes were counted with the DAIME software
(Daims et al., 2006). The DAPI images were used for
total counts and as verification of the accuracy of the
FISH hybridization. For each phylogenetic group, relative
abundances (percent of total bacteria) as well as absolute
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
H. Elifantz et al.
abundances were calculated for each field of view and
averaged. Less than 5% of DAPI-stained cells were
hybridized non-specifically with the NON338 probe.
Results
Environmental parameters
The Mediterranean coastal seawater exhibited typical temperate (physical and chemical) characteristics throughout
the study. Water temperature ranged from about 17 °C
during winter to 30 °C during summer, and the spring
season was characterized by intermediate temperatures of
about 20 °C (Table 1). DOC concentrations of the seawater correlated with temperature (r = 0.97, P < 0.05) and
were the highest in August 2008 (1801 lM C), when the
combination of high temperature and high DOC may
have influenced biofilm biomass. These high DOC concentrations were maintained through September 2008 and
decreased towards winter, in which the concentration was
about 108 lM C. The percentage DOC of TOC changed
seasonally. In August 2008, the DOC made up only
50–60% of the TOC, but this percentage increased to
c. 100% in September 2008 to January 2009 (Table 1).
The total organic nitrogen (TON) was in the range of
32–95 lM N throughout the year (Table 1) and there
was no visible trend with season.
Microbial community composition of seawater
and biofilm
In the seawater, a few phylogenetic groups dominated the
microbial community, as reflected by 16S rRNA gene
diversity. Alphaproteobacteria was the most dominant
phylogenetic group on most sampling occasions, comprising 26–48% of the microbial community (Table 2). Three
additional phylogenetic groups were also abundant in the
seawater community and included Bacteroidetes, Actinobacteria and Cyanobacteria (Table 2). The highest relative
abundance of Cyanobacteria (27%) was detected in
August 2008 and the lowest (1%) in March 2009. The
lowest abundance of Cyanobacteria coincided with a
phytoplankton spring bloom, as indicated by elevated
chlorophyll concentration (1.5 mg L 1 in March–April
2008 vs. 0.1–0.5 mg L 1 in other months). During the
rest of the year, this group represented 9–15% of the total
microbial community. In contrast, the highest relative
abundance of Bacteroidetes was observed in March 2009
(23%) and the lowest (6%) in September 2008 (Table 2).
Actinobacteria was the second most dominant group in
the clone libraries during September 2008 (25%) and
May 2009 (35%). Less prominent (< 5%) phylogenetic
groups detected in the clone library analyses included
FEMS Microbiol Ecol 85 (2013) 348–357
351
Primary colonizers of biofilm in Mediterranean coastal seawater
Table 1. Average temperature, organic carbon (DOC and TOC), and total organic nitrogen (TON) of processed seawater at the different
sampling times during the year
Date
Temperature*
May 2008
August 2008
September 2008
January 2009
March 2009
May 2009
20.3
29.8
28.1
17.6
18.2
23.5
DOC† (lM C)
0.6
0.7
0.7
0.2
0.7
1.2
73
1801
1696
108
96
658
49
1000
700
35
22
576
% DOC of TOC
TOC† (lM C)
60.3
51.6
98.5
100
100
107
121
3488
1721
73
96
612
TON (lM N)
110
2000
670
35
22
600
95
32
64
93
29
52
67
20
30
25
25
33
*The temperature values are averages of the sampling points for each experiment (2–3 weeks).
†
The DOC, TOC and TON are averages of two to three time points for each experiment (2–3 weeks). Large variations as reflected in the standard
deviations are a result of variations between sampling dates within each experiment.
Table 2. Microbial community composition of processed seawater as determined by 16S rRNA gene clone libraries. The average ( SD)
percentages of three sampling dates in each month are presented. (n.d., not detected). Numbers in parentheses represent number of clones
analyzed for each library. Percent coverage was calculated for the 95% similarity level according to 1–(n/N) where n is the number of singeltons
and N is the number of clones
Sampling month and year
Phylum
Class
Proteobacteria
Alpha
Gamma
Delta
Bacteroidetes
Planctomycetes
Actinobacteria
Cyanobacteria
Other bacteria*
% coverage
May 2008
(217)
48
8
1
15
3
8
15
2
61
4
4
0.5
5
3
5
7
1
August 2008
(418)
26
7
4
9
4
17
27
6
60
2
2
2
2
2
3
2
3
September 2008
(215)
34
7
3
6
3
25
11
11
62
14
4
2
2
2
12
4
4
January 2009
(145)
41
6
3
18
1
12
9
10
58
16
2
2
6
1
8
5
5
March 2009
(199)
May 2009
(198)
47 2
n.d
23 2
22 1
3
77
35 5
1
9
n.d
35 11 4
72
9
2
3
2
11
1
1
8
4
1
3
9
6
3
*Other bacteria = clones that were affiliated with various other bacterial phyla which their overall percentage in the community was below 4%.
Some of these groups appeared only once in the entire year. These bacterial groups included Betaproteobacteria, Acidobacteria, Chloroflexi, Fusobacteria, Verrucomicrobia, Firmicutes and bacteria affiliated with candidate divisions TM6, OD1 and OP3.
Betaproteobacteria, Acidobacteria, Chloroflexi, Fusobacteria,
Verrucomicrobia, Firmicutes and sequences affiliated with
candidate divisions TM6, OD1 and OP3 (Table 2).
In the biofilm, several reoccurring trends could be
identified during the year. Similar to the seawater, the
biofilm microbial community was dominated by Alphaproteobacteria (Table 3). In the one week old biofilm
64–78% of the clones at all sampling dates (with the
exception of January 2009, 32%) were affiliated with this
class. In January 2009 the other dominant phylogenetic
group was Gammaproteobacteria, making up 35% of the
community. As biofilm matured, the overall percentage of
Alphaproteobacteria decreased in May–September 2008
and March 2009 (Table 3). This decrease was accompanied by an increase in the relative abundance of Gammaproteobacteria, Bacteroidetes or both. In January, the 1week-old biofilm was composed of equal amounts of
Alpha- and Gammaproteobacteria; as the biofilm matured,
the relative abundance of Alphaproteobacteria increased,
whereas that of Gammaproteobacteria decreased (Table 3).
FEMS Microbiol Ecol 85 (2013) 348–357
Overall, we could not detect a clear relationship
between the biofilm communities by UNIFRAC analysis
(Fig. 1). The communities sampled in August 2008 and
March 2009 were the only two to cluster within themselves, and together made up cluster A (Fig. 1). Two
other major clusters (B and C) contained communities
from various sampling seasons and biofilm ages, and no
particular association could be observed. Whereas biofilm
samples from either September 2008 or May 2009 could
be detected in each of the clusters (clusters C and B,
respectively), the communities from May 2008 and January 2009 were spread between clusters B and C, in no
particular order.
As mentioned above, Alphaproteobacteria was the most
dominant component of the water and biofilm microbial
communities. In the water, Rhodobacteraceae and SAR11
were the two major subgroups, comprising 50–90% of all
Alphaproteobacteria (Fig. 2). In May and August 2008 the
percentage of these two groups was similar, in September
2008 and January 2009 SAR11 was the dominant subgroup
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
352
H. Elifantz et al.
Table 3. Microbial community composition (percentages of total amount of clones for each sample) as determined by 16S rRNA gene clone
libraries of biofilm formed in Mediterranean seawater at the different sampling times during the year. W1 and W2 represent biofilm age of 1
and 2 weeks, respectively, for each sampling experiment. (n.d., not detected). Numbers in parentheses represents number of clones analyzed for
each library. Percent coverage was calculated for the 95% similarity level according to 1–(n/N) where n is the number of singeltons and N is the
number of clones
Month, year and biofilm age
Phylum
Class
Proteobacteria
Alpha
Gamma
Delta
Bacteroidetes
Planctomycetes
Other bacteria*
% coverage
May 2008
August 2008
September
2008
January 2009
March 2009
May 2009
W1
(67)
W2
(63)
W1
(83)
W2
(90)
W1
(62)
W2
(71)
W1
(63)
W2
(30)
W1
(82)
W2
(76)
W1
(83)
W2
(57)
64
12
1
5
7
9
64
48
11
3
35
3
n.d.
56
78
4
1
1
8
6
63
54
6
6
10
9
16
75
75
3
2
4
15
4
68
44
20
n.d.
17
14
6
73
32
35
6
13
8
7
55
73
10
7
n.d.
n.d.
10
60
72
4
5
13
2
3
76
58
11
4
24
3
1
72
75
7
4
5
2
7
70
76
2
4
5
n.d.
14
69
*Other bacteria = clones that were affiliated with various other bacterial phyla in which their overall percentage in the community was below
4%. Some of these groups appeared only once in the entire year. These bacterial groups included Betaproteobacteria, Acidobacteria, Chloroflexi,
Fusobacteria, Verrucomicrobia, Firmicutes and bacteria affiliated with candidate divisions TM6, OD1 and OP3.
Cluster A
Cluster B
Cluster C
Fig. 1. Community similartity dendrogram of biofilm clone libraries
by UNIFRAC analysis. W1 and W2 represent 1- and 2-week-old biofilms,
respectively. Months and year of each library are indicated.
in the water (50–70%), and in March and May 2009
Rhodobacteraceae became the dominant Alphaproteobacteria
group (c. 40%). The Surface 1 cluster (Brown & Fuhrman,
2005) was the most dominant SAR11 clade in the water
communities. This group was detected only once as a
minor component of 1-week-old biofilm community in
January 2009, and was completely absent from all other
biofilm communities (Fig. 2).
In the biofilm, Rhodobacteraceae was the most dominant alphaproteobacterial subgroup (Fig. 2a), making up
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
90% of all Alphaproteobacteria in August and September
2008 in the 2-week-old biofilms (Fig. 2a). The lowest
(30%) relative abundance of Rhodobacteraceae was
detected in a 1-week-old biofilm in January 2009. In the
remaining biofilm samples this family represented 40–
80% of all Alphaproteobacteria. The Roseobacter genus
represented a substantial fraction of the Rhodobacteraceae
family, constituting on some occasions the majority of
this group in the biofilms (September 2008, March and
May 2009) (Fig. 2a). The Roseobacter clade CHAB-I-5
made up 10–70% of the Alphaproteobacteria community
in biofilm samples collected throughout the year
(Fig. 2b). The highest percentage of this subgroup was
detected in 2-week-old biofilm formed in September
2008, followed by a 1-week-old biofilm in March and
May of 2009.
Microbial community composition by FISH
FISH analyses were performed concomitant with clone
library analyses on the water and biofilms to quantify the
abundance and spatial distribution (biofilm only) of
selected microbial groups. For the water samples, hybridization was below detection limits, likely due to a low
content of ribosomes in the cells (data is not shown).
Consistent with the clone library results, the dominant
phylogenetic groups in the biofilm were affiliated with
Alphaproteobacteria and Bacteroidetes, which represented
26–70% and 10–50% of the bacterial community (as
determined by DAPI), respectively (Fig. 3). Gammaproteobacteria made up only up to 10% of the community
according to the FISH analysis and on some occasions
FEMS Microbiol Ecol 85 (2013) 348–357
353
Primary colonizers of biofilm in Mediterranean coastal seawater
Relative abundance of Roseobacter and SAR11
(% of Alphaproteobacteria)
100
non-Roseobacter Rhodobacteraceae
Roseobacter
Surface 1 SAR11
non-Surface 1 SAR11
(a)
80
60
40
20
0
W B1 B2
100
W B1 B2
W B1 B2
W B1 B2
W B1 B2
W B1 B2
(b)
Fig. 2. Relative abundance of
Rhodobacteraceae and SAR11 [Surface 1
SAR11 cluster; (Brown & Fuhrman, 2005)] (a)
and the Rhodobacteraceae genus CHAB-I-5 (b)
in processed seawater and biofilms at the
different sampling times during the year. The
data from the clone libraries of the water
samples were averaged (three consecutive
weeks) for each experiment. W, B1 and B2,
water, 1- and 2-week-old biofilm, respectively.
Relative abundnace of CHAB-I-5
(% of Alphaproteobacteria)
CHAB-I-5
80
60
40
20
0
W B1 B2
W B1 B2
W B1 B2
W B1 B2
W B1 B2
W B1 B2
May
2008
August
2008
September
2008
January
2009
March
2009
May
2009
were below detection limit (Fig. 3). The FISH analysis
managed to detect over 70% of the DAPI-stained cells
with the exception of one, a 2-week-old biofilm formed
in May 2009, in which only 50% of the DAPI-stained
cells were detected by FISH.
While relative abundance is a common form of presenting FISH data, it is necessary to calculate absolute cell
quantities in order to compare the changes in biofilm
biomass on an annual basis. In general, the total amount
of bacteria per cm2 was in the range of 1–7 9 106
throughout the year (Fig. 4a). One exception was documented in the 1-week-old biofilm of March 2009, when
FEMS Microbiol Ecol 85 (2013) 348–357
Sample type and date
the total amount was 3 9 107 bacteria cm 2 (ANOVA
P < 0.0001, Tukey HSD). A similar trend was observed
with the specific bacterial groups, with a peak in cell
amounts in March 2009. The absolute amount of Alphaproteobacteria (majority of which always belonged to the
Rhodobacteraceae) was usually higher than Bacteroidetes at
all sampling times (Fig. 4b).
Discussion
The aim of the current study was to identify and quantify
the early bacterial colonizers of biofilms developed in
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
354
H. Elifantz et al.
Relative abundance of
phylogenetic group in the community (%)
Mediterranean coastal seawater. The majority of the
planktonic bacteria in seawater were affiliated with the
Alphaproteobacteria, Bacteroidetes, Cyanobacteria and
Actinobacteria taxa. Whereas the relative abundance of
the first three was typical for marine environments
(Giovannoni & Stingl, 2005), the relative abundance of
100
Actinobacteria was unexpectedly high. This discrepancy
could be explained by the fact that the water used in the
study was pumped close to the shore, which could be
affected by terrestrial inputs, as described previously in other
environments (Crump et al., 1999; Garneau et al., 2006).
The most important phylogenetic group in the water
and biofilms was the Alphaproteobacteria. In the planktonic community, the relative abundance of the SAR11
clade was equal to or higher than that of the Rhodobacteraceae. SAR11 related sequences represented c. 15% of the
whole community obtained from the water samples, similar to what was documented in other coastal environments (Rappe et al., 1997, 2000; Crump et al., 1999) but
was lower than observed in the Western Mediterranean
Sea (Alonso-Saez et al., 2007) and the world’s oceans
(Morris et al., 2002). However, bacteria affiliated with the
SAR11 clade were almost completely absent from the biofilm communities. This was not surprising, as SAR11 is
usually absent from particles and other marine biofilms
(Crump et al., 1999; Jones et al., 2006) and generally is
considered to be a planktonic group (Giovannoni &
Stingl, 2005). Similarly, Cyanobacteria and Actinobacteria,
although of significant levels in the water, were also
absent from the biofilms.
In contrast, Rhodobacteraceae, and in particular the
Roseobacter genus, was the dominant bacterial group in
the biofilm, constituting about 25% of the community.
The highest fraction of Rhodobacteraceae in the water was
Alphaproteobacteria
Bacteroidetes
Gammaproteobacteria
Planctomycetes
80
60
40
20
N.A
0
B1 B2
May
2008
B1 B2
August
2008
B1 B2
B1 B2
September January
2008
2009
B1 B2
March
2009
B1 B2
May
2009
Sample type and date
Fig. 3. Percentage of the major phylogenetic groups as determined
by FISH-CLSM in biofilms sampled from processed seawater at the
different sampling times during the year. B1 and B2 refer to 1- and
2-week-old biofilms, respectively. NA, sample not available.
Total cells abundance in biofilm
(cells cm–2)
4e+7
Total
3e+7
2e+7
1e+7
a
a
a
a
a
a
N.A
Abundance of specific bacterial groups
(cells cm–2)
(a)
c
a
a
b
b
2e+7
Alphaproteobacteria
Rhodobacteraceae
Bacteroidetes
(b)
*
*
1e+7
*
W1
W2
May
2008
W1
W2
August
2008
N.A
** *
W1 W2
W1
September
2008
** *
W2
January
2009
Sample type and date
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
W1
W2
March
2009
W1
2W
May
2009
Fig. 4. Amount of (a) total bacteria and
(b) bacteria in specific groups in the biofilm
formed at the different sampling times during
the year in seawater as determined by FISHCLSM. Error bars indicate SE of 20 fields of
view. Significantly different total amounts of
cells (ANOVA, P < 0.005) are indicated by the
letters a, b and c in panel A. The significantly
different amounts of bacteria (ANOVA,
P < 0.005) within each phylogenetic group are
indicated with * in panel B. W1 and W2
represent 1 and 2 weeks of biofilm age,
respectively. N.A., result not available.
FEMS Microbiol Ecol 85 (2013) 348–357
Primary colonizers of biofilm in Mediterranean coastal seawater
detected in March, which could be attributed to a spring
bloom that occurred at the time, as suggested by the high
chlorophyll level obtained for that sampling time. Roseobacter are known for their close association with algal
blooms that produce dimethylsulfoniopropionate (DMSP),
which in turn is recycled by various Roseobacter strains
(Wagner-D€
obler & Biebl, 2006). However, as Rhodobacteraceae were not always abundant in the water, their high
relative abundance in the biofilm suggested a selection for
certain types of bacteria to settle and form a biofilm. The
dominance of Rhodobacteraceae is not unique to the
Mediterranean region, and this group has also been
identified as a major biofilm component in various
locations of the Atlantic and Pacific Oceans (Dang &
Lovell, 2002; Jones et al., 2006).
A recent genomic study of an isolated marine Roseobacter, Phaeobacter gallaeciensis, which is a primary colonizer
of surfaces, revealed capabilities to produce antibiotics as
a mechanism to eliminate potential competitors, as well
as the production of siderophores which may aid in
scavenging iron from the nutritionally poor environment
(Thole et al., 2012). In a recent review, the genomes of
various Roseobacter isolates were screened for surface colonization traits (Slightom & Buchan, 2009). A comparison
of the 16S rRNA gene sequences retrieved in the current
study with some of these isolates indicated phylogenetic
similarities and suggested that the Roseobacter community
that occupied the biofilm in the current study may have a
similar capability.
The Roseobacter strain related to Roseobacter CHAB-I-5
was dominant in the biofilm communities throughout the
year. This subgroup composed 10–40% of the entire biofilm community and 10–70% of all Alphaproteobacteria in
the biofilms, but comprised only a very low percentage of
total bacteria in the water samples. A quantitative analysis
of surface water bacterioplankton in the Chesapeake Bay
suggested that the overall abundance of this subgroup
was low than that of other Roseobacters (Buchan et al.,
2009). Although CHAB-I-5-like organisms seem important in biofilm formation in the current study, little is
known about them as they have no cultured representatives (Wagner-D€
obler & Biebl, 2006) and, as such, cannot
be related as yet to any potential function. Further environmental genomic studies, however, may reveal some of
its functional traits.
Members of the Bacteroidetes phylum also represented
a significant fraction of the bacterial community in the
water and biofilm samples. Whereas Alphaproteobacteria
made up the majority of the microbial community in the
1-week-old biofilm, the relative abundance of Bacteroidetes increased between week 1 and 2 on most sampling
occasions. This suggests that Bacteroidetes may be secondary colonizers. Alphaproteobacteria are known to survive
FEMS Microbiol Ecol 85 (2013) 348–357
355
better in poor environments (Alonso & Pernthaler, 2006),
as in the case of initial phase of biofilm formation, when
the amount of nutrients and organic carbon close to the
surface is limiting (Flemming, 2002). In contrast, Bacteroidetes requires higher levels of nutrients and organic
matter (Kirchman, 2002; Elifantz et al., 2007), which
could be available with biofilm maturation and matrix
development. The mature biofilm may include nutritional
niches for this group and therefore one can expect a
higher abundance of this phylogenetic group in the biofilm (Flemming, 2002). One exception was observed for
January 2009, during which Alphaproteobacteria abundance increased with time, whereas that of Bacteroidetes
decreased. A possible explanation could be attributed to
the low temperature of the seawater. However, the
UNIFRAC analysis for community similarities suggested that
on most occasions the biofilm community of 2-week-old
biofilm was more similar to the 1-week-old biofilm from
that period than to other 2-week-old biofilm communities obtained from other periods, indicating that the initial colonization community may determine subsequent
community composition.
In addition to the clone libraries, the relative and absolute abundances of the major phylogenetic groups and
total bacteria were quantified using FISH-CLSM. As mentioned above, diversity studies only estimate relative
abundance, whereas microscopy allowed us to count
accurately the actual amount of bacteria in the community. The concentration of bacteria per cm2 in the biofilm
formed in seawater was fairly constant at the different
sampling times during the year. One exception was noted
in March 2009, when the amount of bacteria in the
1-week-old biofilm was 7.5 times higher than in other
sampling dates. During this month, there was a phytoplankton bloom typical for this region after winter mixing
(Krom et al., 1992). This was also reflected in the higher
chlorophyll levels measured during that season, although
surprisingly not in the DOC measurements.
In conclusion, this study suggests that overall the
dynamics of the initial biofilm formation in Eastern Mediterranean coastal seawater was consistent over a year and
may be easy to predict in this type of water in the future.
Whereas Rhodobacteraceae are the primary colonizers of
surfaces, Bacteroidetes seem to be the secondary colonizers. However, careful attention should be paid to the
spring season, in which phytoplankton blooms occur in
the region and may affect the microbial community composition dynamics and overall biomass of these biofilms,
as suggested by the cell counts in the biofilm developed
in the current study. This may be of a particular interest
for man-made facilities that use coastal seawater for desalination, industrial water cooling, aquaculture, etc. The
information collected in the current and similar studies
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
356
may aid in developing strategies to eliminate or remove
biofilm from these facilities.
Acknowledgements
We would like to thank Lisa A. Waidner, Eddie Cytryn
and Linda Popels for their valuable comments on the
manuscript. This work was funded by the Ministry of
Industry, Commerce and Employment as part as Magnet
‘Maim’ project.
References
Acu~
na N, Ortega-Morales BO & Valadez-Gonzales A (2006)
Biofilm colonization dynamics and its influence on the
corrosion resistance of austenitic UNS S31603 stainless steel
exposed to Gulf of Mexico seawater. Mar Biotechnol 8: 62–70.
Al-Malahy KSE & Hodgkiess T (2003) Comparative studies of
the seawater corrosion behaviour of a range of materials.
Desalination 158: 35–42.
Alonso C & Pernthaler J (2006) Roseobacter and SAR11
dominate microbial glucose uptake in coastal North Sea
waters. Environ Microbiol 8: 2022–2030.
Alonso-Saez L, Balague V, Sa EL, Sanchez OS, Gonzalez JM,
Pinhassi J, Massana R, Pernthaler J, Pedr
os-Ali
o C & Gasol
JM (2007) Seasonality in bacterial diversity in north-west
Mediterranean coastal waters: assessment through clone
libraries, fingerprinting and FISH. FEMS Microbiol Ecol 60:
98–112.
Bereschenko LA, Stams AJM, Wuverink GJW & van
Loosdrecht MCM (2010) Biofilm formation on reverse
osmosis membranes is initiated and dominated by
Sphingomonas spp. Appl Environ Microbiol 76: 2623–2632.
Brown MV & Fuhrman JA (2005) Marine bacterial
microdiveristy as revealed by internal transcribed spacer
analysis. Aquat Microb Ecol 41: 15–23.
Buchan A, Hadden M & Suzuki MT (2009) Development and
application of quantitative-PCR tools for subgroups of the
Roseobacter clade. Appl Environ Microbiol 75: 7542–7547.
Candries M & Atlar M (2003) Estimating the impact of newgeneration antifoulings on ship performance: the presence
of slime. Proc IMarEST - Part A - J Mar Eng Technol 2003:
13–22.
Crump BC, Armbrust EV & Baross JA (1999) Phylogenetic
analysis of particle-attached and free-living bacterial
communities in the Columbia River, its estuary and the
adjacent coastal ocean. Appl Environ Microbiol 65: 3192–3204.
Daims H, L€
ucker S & Wagner M (2006) DAIME, a novel image
analysis program for microbial ecology and biofilm research.
Environ Microbiol 8: 200–213.
Dang H & Lovell CR (2000) Bacterial primary colonization
and early succession on surfaces in marine waters as
determined by amplified rRNA gene restriction analysis and
sequence analysis of 16S rRNA genes. Appl Environ
Microbiol 66: 467–475.
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved
H. Elifantz et al.
Dang H & Lovell CR (2002) Numerical dominance and
phylotype diversity of marine Rhodobacter species during
early colonization of submerged surfaces in coastal marine
waters as determined by 16S ribosomal DNA sequence
analysis and fluorescence in situ hybridization. Appl Environ
Microbiol 68: 496–504.
Dang H, Li T, Chen M & Huang G (2008) Cross-ocean
distribution of Rhodobacterales bacteria as primary surface
colonizers in temperate coastal marine waters. Appl Environ
Microbiol 74: 52–60.
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.
Eilers H, Pernthaler J, Gl€
ockner F & Amann R (2000)
Culturability and in situ abundance of pelagic bacteria from
the North Sea. Appl Environ Microbiol 66: 3044–3051.
Elifantz H, Dittel AI, Cottrell MT & Kirchman DL (2007)
Dissolved organic matter assimilation by heterotrophic
bacterial groups in the western Arctic Ocean. Aquat Microb
Ecol 50: 39–49.
Flemming H-C (2002) Biofouling in water systems – cases,
causes and countermeasures. Appl Microbiol Biotechnol 59:
629–640.
Vincent WF, Alonso-Saez L, Gratton Y &
Garneau ME,
Lovejoy C (2006) Prokaryotic community structure and
heterotrophic production in a river-influenced coastal arctic
ecosystem. Aquat Microb Ecol 42: 27–40.
Giovannoni SJ & Stingl U (2005) Molecular diversity and
ecology of microbial plankton. Nature 437: 343–348.
Gl€
ockner FO, Fuchs BM & Amann R (1999) Bacterioplankton
compositions of lakes and oceans: a first comparison based
on fluorescence in situ hybridization. Appl Environ Microbiol
65: 3721–3726.
Heijs SK, Laverman AM, Forney LJ, Hardoim PR & Van Elsas
JD (2008) Comparison of deep-sea sediment microbial
communities in the Eastern Mediterranean. FEMS Microbiol
Ecol 64: 362–377.
H€
orsch P, Gorenflo A, Fuder C, Deleage A & Frimmel FH
(2005) Biofouling of ultra- and nanofiltration membranes
fordrinking water treatment characterized by fluorescence in
situ hybridization (FISH). Desalination 172: 41–52.
Ivnitsky H, Minz D, Kautsky L, Preis A, Osfeld A, Semiat R &
Dosoretz CG (2010) Biofouling formation and modeling in
nanofiltration membranes applied to wastewater treatment.
J Membr Sci 360: 165–173.
Jones PR, Cottrell MT, Kirchman DL & Dexter SC (2006)
Bacterial community structure of biofilms on artificial
surfaces in an estuary. Microb Ecol 53: 153–162.
Kane MD, Poulsen LK & Stahl DA (1993) Monitoring the
enrichment and isolation of sulfate-reducing bacteria by
using oligonucleotide hybridization probes designed from
environmentally derived 16S rRNA sequences. Appl Environ
Microbiol 59: 682–686.
Kirchman DL (2002) The ecology of Cytophaga-Flavobacteria
in aquatic environments. FEMS Microbiol Ecol 39: 91–100.
FEMS Microbiol Ecol 85 (2013) 348–357
Primary colonizers of biofilm in Mediterranean coastal seawater
Krom MD, Brenner S, Kress N, Neori A & Gordon LI (1992)
Nutrient dynamics and new production in a warm-core
eddy from the Eastern Mediterranean Sea. Deep-Sea Res 39:
467–480.
Lane DJ (1991) 16S/23S rRNA sequencing. Nucleic Acid
Techniques in Bacterial Systematics (Stackebrandt E &
Goodfellow M, eds), pp. 115–175. John Wiley & Sons,
Chichester.
L
opez MA, Dıaz de la Serana FJZ, Jan-Roblero J, Romero JM
& Hernandez-Rodrıguez C (2006) Phylogenetic analysis of a
biofilm bacterial population in a water pipeline in the Gulf
of Mexico. FEMS Microbiol Ecol 58: 145–154.
Lozupone C & Knight R (2005) UniFrac: a new phylogenetic
method for comparing microbial communities. Appl Environ
Microbiol 71: 8228–8235.
Ludwig W, Strunk O, Westram R et al. (2004) ARB: a
software environment for sequence data. Nucleic Acids Res
32: 1363–1371.
Manz W, Amann R, Ludwig W, Wagner M & Schleifer KH
(1992) Phylogenetic oligodeoxynucleotide probes for the
major subclasses of Proteobacteria - problems and solutions.
Syst Appl Microbiol 15: 593–600.
Manz W, Amann R, Ludwig W, Vancanneyt M & Schleifer KH
(1996) Application of a suite of 16S rRNA-specific
oligonucleotide probes designed to investigate bacteria of
the phylum Cytophaga-Flavobacter-Bacteroides in the
natural environment. Microbiology 142: 1097–1106.
Morris RM, Rappe MS, Connon SA, Vergin KL, Siebold WA,
Carlson CA & Giovannoni SJ (2002) SAR11 clade
dominates ocean surface bacterioplankton communities.
Nature 420: 806–810.
Moss JA, Nocker A, Lepo JE & Snyder RA (2006) Stability and
change in estuarine biofilm bacterial community diversity.
Appl Environ Microbiol 72: 5679–5688.
Muyzer G, De Waal EC & Uitterlinden AG (1993) Profiling of
complex microbial populations by denaturing gradient gel
electrophoresis analysis of polymerase chain reactionamplified genes coding for 16S rRNA. Appl Environ
Microbiol 59: 695–700.
Neef A, Amann R, Schlesner H & Schleifer K–H (1998)
Monitoring a widespread bacterial group: in situ detection
of planctomycetes with 16S rRNA-targeted probes.
Microbiology 144: 3257–3266.
FEMS Microbiol Ecol 85 (2013) 348–357
357
Rao D, Webb JS & Kjelleberg S (2005) Competitive
interactions in mixed-species biofilms containing the marine
bacterium Pseudoalteromonas tunicate. Appl Environ
Microbiol 71: 1729–1736.
Rappe MS, Kemp PF & Giovannoni SJ (1997) Phylogenetic
diversity of marine picoplankton 16S rRNA genes cloned
from the continental shelf off Cape Hatteras, North
Carolina. Limnol Oceanogr 42: 811–826.
Rappe MS, Vergin K & Giovannoni SJ (2000) Phylogenetic
comparisons of a coastal bacterioplankton community with
its counterparts in Open Ocean and freshwater systems.
FEMS Microbiol Ecol 33: 219–232.
Schloss PD, Westcott SL, Ryabin T et al. (2009) Introducing
mothur: open-source, platform-independent, communitysupported software for describing and comparing
microbial communities. Appl Environ Microbiol 75: 7537–
7541.
Schneider RP, Ferreira LM, Binder P, Bejarano EM, G
oes KP,
Slongo E, Machad CR & Rosa GMZ (2005) Dynamics of
organic carbon and of bacterial populations in a
conventional pretreatment train of a reverse osmosis unit
experiencing severe biofouling. J Memb Sci 266: 18–29.
Slightom RN & Buchan A (2009) Surface colonization by
marine Roseobacters: Integrating genotype and phenotype.
Appl Environ Microbiol 75: 6027–6037.
Thole S, Kalhoefer D, Voget S et al. (2012) Phaeobacter
gallaeciensis genomes from globally opposite locations reveal
high similarity of adaptation to surface life. ISME J 6: 2229–
2244.
Vrouwenvelder JS & van der Kooij D (2001) Diagnosis,
prediction and prevention of biofouling of NF and RO
membranes. Desalination 139: 65–71.
Wagner-D€
obler I & Biebl H (2006) Environmental biology of
the marine Roseobacter lineage. Annu Rev Microbiol 60: 255–
280.
Wallner G, Amann R & Beisker W (1993) Optimizing
fluorescent in situ hybridization with rRNA-targeted
oligonucleotide probes for flow cytometric identification of
microorganisms. Cytometry 14: 136–143.
Wietz M, Hall MR & Hǿj L (2009) Effects of seawater
ozonation on biofilm development in aquaculture tanks.
Syst Appl Microbiol 32: 266–277.
ª 2013 Federation of European Microbiological Societies
Published by John Wiley & Sons Ltd. All rights reserved