A mannose-sensitive haemagglutinin (MSHA)-like
pilus promotes attachment of Pseudoalteromonas
tunicata cells to the surface of the green alga Ulva
Doralyn S. Dalisay,1Jeremy S. Webb,2,33 Andre ´ Scheffel,4
Charles Svenson,2Sally James,2,3Carola Holmstro ¨m,2,3Suhelen Egan2,3
and Staffan Kjelleberg2,3
1Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman
Drive, La Jolla, CA 92093, USA
2,3School of Biotechnology and Biomolecular Sciences2and Centre for Marine Biofouling and
Bio-innovation3, University of New South Wales, Sydney, NSW 2052, Australia
4Max-Planck-Institute for Marine Microbiology, Celsiusstraße 1 28359, Bremen, Germany
Received 25 May 2006
10 July 2006
Accepted 12 July 2006
This study demonstrates that attachment of the marine bacterium Pseudoalteromonas tunicata to
haemagglutinin (MSHA-like) pilus. We have identified an MSHA pilus biogenesis gene locus in
P. tunicata, termed mshI1I2JKLMNEGFBACDOPQ, which shows significant homology, with
respect to its genetic characteristics and organization, to the MSHA pilus biogenesis gene locus of
Vibrio cholerae. Electron microscopy studies revealed that P. tunicata wild-type cells express
insertion in the mshJ region displayed a non-piliated phenotype. Using SM5, it has been
demonstrated that the MSHA pilus promotes attachment of P. tunicata wild-type cells in
polystyrene microtitre plates, as well as to microcrystalline cellulose and to the living surface of U.
australis. P. tunicata also demonstrated increased pilus production in response to cellulose and its
monomer constituent cellobiose. The MSHA pilus thus functions as a determinant of attachment in
P. tunicata, and it is proposed that an understanding of surface sensing mechanisms displayed
by P. tunicata will provide insight into specific ecological interactions that occur between this
bacterium and higher marine organisms.
Marine surfaces are colonized by a diversity of micro-
organisms and sessile marine organisms collectively known
as biofouling communities. The biofouling process is
initiated by the attachment of bacteria to a surface followed
by the settlement and adherence of diatoms, free-swimming
algal spores and invertebrate larvae (Bryers & Characklis,
1982). Some sessile higher organisms employ chemical
defences against biofouling through the production of
secondary metabolites that inhibit the development and
formation of a biofouling community (Harrison, 1992;
Mary et al., 1993; Maximilien et al., 1995). For example,
furanones produced by the red alga Delisea pulchra have
been reported to inhibit the settlement of common fouling
organisms (de Nys et al., 1995). For marine organisms
without intrinsic defence mechanisms, it has been proposed
that protection against fouling is maintained by the
secondary metabolites produced by surface-associated
bacteria (Egan et al., 2001a, b; Holmstro ¨m et al., 1992,
1996, 1998; Holmstro ¨m & Kjelleberg, 1999; James et al.,
Many species of the genus Pseudoalteromonas have been
found to produce bioactive compounds against different
classes of fouling organisms and are frequently found in
association with the surfaces of living marine eukaryotes
(Holmstro ¨m & Kjelleberg, 1999). A well-studied surface-
associated bacterium is Pseudoalteromonas tunicata. This
3Present address: School of Biological Sciences, University of
Southampton, Southampton SO16 7PX, UK.
Abbreviations: CLSM, confocal laser scanning microscopy; GFP, green
fluorescent protein; MSHA, mannose-sensitive haemagglutinin; RFP,
red fluorescent protein; TEM, transmission electron microscopy.
The GenBank/EMBL/DDBJ accession numbers for the sequences
0002-9158 G 2006 SGMPrinted in Great Britain2875
Microbiology (2006), 152, 2875–2883
green-pigmented bacterium was first isolated from the
surface of a tunicate, Ciona intestinalis, in Sweden
(Holmstro ¨m et al., 1998) and later from a green alga,
Ulva australis (formally Ulva lactuca), in Australian waters
(Egan et al., 2001a). P. tunicata produces a number of
extracellular bioactive compounds, each of which has a
specific inhibitory activity against target organisms such as
algal spores, fungi, invertebrate larvae or bacteria (Egan
et al., 2001b; Holmstro ¨m et al., 1992, 2002; James et al.,
1996). While several studies have addressed the production
of bioactive compounds, no data have so far been reported
on the means by which P. tunicata colonizes surfaces,
such mechanisms will contribute to a greater understanding
of the ecology of P. tunicata and its interaction with marine
The successful colonization of surfaces by bacteria is often
For example, the attachment of Escherichia coli to abiotic
surfaces is promoted by the presence of both type 1 pili and
flagella (O’Toole & Kolter, 1998). In Pseudomonas aerugi-
nosa, flagella are believed to be important for initial
attachment to a surface, while type 4 pili promote the
formation of microcolonies on the surface (O’Toole &
Kolter, 1998). It has also been reported that type 4 pili
mediate attachment of pathogenic bacteria such as Neisseria
gonorrhoeae (Morand et al., 2001) and Pseudomonas
aeruginosa (Zolfaghar et al., 2003) to host epithelial cells.
The adherence of Vibrio cholerae to environmental surfaces
is directly associated with the presence of the mannose-
sensitive haemagglutinin (MSHA) pilus, which belongs to a
family of type 4 pili (Marsh & Taylor, 1999). In V. cholerae,
the MSHA pilus has been demonstrated to play a direct role
in both colonization and subsequent biofilm formation on
abiotic as well as biotic surfaces (Chiavelli et al., 2001;
Watnick et al., 1999; Watnick & Kolter, 1999).
that carries a transposon insertion in an ORF termed mshJ,
a MSHA pilus biogenesis protein. DNA sequencing
upstream and downstream of mshJ revealed the presence
of a gene locus with 17 ORFs that are homologous to the
MSHA pilus biogenesis gene locus of V. cholerae, and we
propose that this gene locus is involved in the assembly and
transport of an MSHA pilus in P. tunicata. Here we
demonstrate a role for the P. tunicata MSHA pilus in
and propose that P. tunicata demonstrates surface sensing
presence of cellulose, one of the major surface polymers of
Bacterial strains, plasmids and culture conditions. P. tunicata
wild-type was maintained on the complex marine medium VNSS
(Marden et al., 1985). For attachment assays, P. tunicata was grown
in nine salt solution (NSS), pH 7?0, or Marine Minimal Medium
(3M) (Neidhardt et al., 1974), each medium containing either
0?05% (w/v) glucose, 0?25% microcrystalline cellulose (Avicel;
Fluka) or 0?1% cellobiose (Sigma) as the sole source of carbon. P.
tunicata SM5 was isolated from a P. tunicata mini-Tn10 transposon
library (generated according to the method of Egan et al., 2002a)
and was selected on the basis of reduced attachment to polystyrene
microtitre plates according to the method of O’Toole & Kolter
(1998). The P. tunicata SM5 strain was grown on VNSS agar plates
containing 100 mg streptomycin ml21and 85 mg kanamycin ml21.
P. tunicata wild-type cells were provided with a green fluorescent
protein (GFP) colour tag by transconjugation using the constitutive
GFP expression plasmid pCJS10. This plasmid contains the gfpmut3
range vector pHRP304 (Bagdasarian et al., 1981). In addition, a red
fluorescent protein (RFP) colour tag was provided to SM5 cells using
the pCJS10-derived plasmid pCJS10R. This plasmid contains the RFP
gene dsred (Clontech) inplace of gfpmut3 on pCJS10 (Raoet al., 2005).
tunicata (Egan et al., 2002a) and labelled transconjugants were grown
on VNSS agar plates containing 15 mg chloramphenicol ml21and
100 mg streptomycin ml21. GFP- and RFP-labelled strains showed
bright fluorescence after overnight culture and we observed no
differences in the growth rate or surface attachment properties of the
labelled strains versus the unlabelled parent strain (data not shown).
Panhandle PCR, DNA sequencing and sequence analysis. To
obtain sequence information from the genes disrupted by the mini-
Tn10 transposon in the SM5 mutant, panhandle PCR was carried
out as described previously using adaptor-specific primer AP1 (59-
GGATCCTAATACGACTCACTATAGGGC-39) and transposon-spe-
cific primers Tn10C (59-GCTGACTTGACGGGACGGCG-39) and
Tn10D (59-CCTCGAGCAAGACGTTTCCCG-39) (Egan et al., 2002a,
b). Panhandle PCR products were visualized in 1% agarose gel and
purified using a PCR purification kit (Qiagen), according to the
manufacturer’s instructions. PCR products were sequenced using
transposon-specific primers Tn10C and Tn10D and a primer walk-
ing strategy. For sequencing, between 50 and 100 ng double-
stranded template DNA, 1 ml specific primer (10 pmol), 4 ml CSA
buffer and 4 ml BigDye terminator cycle sequencing reaction mix
(Applied Biosystems) were mixed in a final volume of 20 ml.
Amplification of DNA was conducted using the following para-
meters: 94uC for the initial denaturation step which was followed by
25 cycles of 94uC for 10 s, 50uC for 5 s and 60uC for 4 min. After
cycling, the sequencing mixture was cleaned and purified using a
butanol purification protocol (Tillett & Neilan, 1999). Separation of
sequencing products was performed on an ABI 377 DNA sequencing
system at the Sydney University Prince Alfred Macromolecular
Analysis Centre (SUPAMAC). The DNA sequence electrophero-
grams were analysed with ABI-PRISM software. Multiple sequence
alignments were performed using the Staden Package System
(Medical Research Council – Laboratory of Molecular Biology,
University of Cambridge). The complete consensus DNA sequence
was compared with sequences in the GenBank database using the
BLAST-search algorithm (Altschul et al., 1990) and ORFs were
defined using the ORF finder program made available through the
National Center for Biotechnology Information (NCBI) website
(www.ncbi.nlm.nih.gov). Promoter prediction was performed using
the neural network promoter prediction tool (Reese, 2001) available
though the Berkley Drosophila genome project website (www.
fruitfly.org/seq_tools/promoter.html). The GCG software package
provided by the Australian National Genomic Information Service
(ANGIS) website (www.angis.org.au/WebANGIS/) and the ExPASy
(Expert Protein Analysis System) site (http://expasy.proteome.
org.au/index.html) of the Swiss Institute of Bioinformatics (SIB)
D. S. Dalisay and others
were also used for sequence analysis. Additional sequence informa-
tion was obtained by analysis of the draft genome sequence for P.
tunicata using the BLAST-search algorithm (Altschul et al., 1990).
Transmission electron microscopy (TEM). Cell surface morphol-
ogy of P. tunicata wild-type and SM5 strains was examined using
TEM. Bacterial cells were grown on VNSS plates for 24 h (equivalent
to the stationary phase of growth) and colonies were gently resus-
pended in PBS, pH 7?4 (l21: 8?00 g NaCl, 0?20 g KCl, 1?44 g
Na2HPO4, 0?24 g KH2PO4) to a concentration of 106cells ml21.
Alternatively, bacterial cells were grown in static conditions in NSS
liquid medium containing either glucose, cellobiose or cellulose as
the sole carbon source. Carbon and Formvar-coated copper grids
were placed on a drop of cell suspension for 5 min and then nega-
tively stained with 2% phosphotungstic acid for 30 s. The cells were
examined using a Hitachi H7000 transmission electron microscope.
Haemagglutination assay. Haemagglutination assays were per-
formed as described by Gardel & Mekalanos (1996). Bacterial strains
were grown in VNSS medium at 28uC under static conditions and
assayed during both exponential and stationary growth phases.
Briefly, the cells were washed twice and resuspended in Krebs–
Ringer solution (KRT). An initial concentration of 1010cells ml21
was serially diluted in 96-well microtitre plate with each well con-
taining 100 ml 3% (v/v) washed horse erythrocyte suspension. The
mixture was incubated at room temperature for 30 min and scored
for haemagglutination. V. cholerae strain M1615 was used as a posi-
tive control. To determine if pili are sensitive to the presence of
mannose during the haemagglutination assay, 50 ml 1% a-methyl-D-
mannoside (a non-metabolizable derivative of mannose) were added
to the haemagglutination mixture. The haemagglutination assay for
each strain was carried out in three replicates.
Attachment assays. The attachment of the P. tunicata wild-type
and the SM5 mutant strains to abiotic surfaces was tested by modi-
fying an attachment assay described by O’Toole & Kolter (1998).
Bacterial isolates were grown under static conditions at 28uC on
VNSS medium and harvested after 24 h. The cell suspension was
centrifuged (6000 g, 5 min), washed twice and resuspended in 10 ml
PBS, pH 7?4, to an OD600of 0?6–0?7. Aliquots (1 ml) of cell suspen-
sion were added to wells of a 24-well polystyrene microtitre plate.
The plate was shaken slowly for 1 h at room temperature. Wells
were then washed twice using sterile distilled water and air-dried for
45 min. The attached cells were fixed at 80uC for 30 min and
stained with 0?1% crystal violet for 45 min. The cells were de-
stained with 95% ethanol and quantified by measuring OD590.
We examined attachment of GFP-labelled P. tunicata wild-type and
RFP-labelled SM5 cells to microcrystalline cellulose. Attachment was
monitored essentially as described by Bayer et al. (1983) and by direct
confocal laser scanning microscopic (CLSM) imaging of fluorescent
cells attached to cellulose particles. Cells grown using glucose or
cellobiose were harvested after 24 h and cells grown using cellulose
were harvested after 48 h. The cells were then gently washed twice with
PBS (centrifugation at 5000 g for 10 min) and resuspended in 10 ml
(~107cells ml21) (pre-grown in glucose, cellobiose or microcrystal-
line cellulose), 1 ml 20% microcrystalline cellulose in PBS and 1 ml
1 h at room temperature. The OD400of the suspension was measured
and compared to a control (identical cell suspension in PBS). To test
the effect of mannose in the attachment of cells to cellulose, the
attachment assay mixture was added with 100 mM a-methyl-D-
mannoside. The attachment assay for each strain was carried out in
threereplicates. Thesamples weremountedon aglass microscope slide
and transmission fluorescent images were captured using an Olympus
LSMGB200 confocal laser scanning microscope.
consisted of1 mlcells
Preparation of axenic thallus of green alga U. australis and
culture conditions. Axenic thallus of U. australis was obtained fol-
lowing the protocol of Rao et al. (2006). Briefly, plants of U. austra-
lis were collected from rocks at Clovelly Bay, Sydney, Australia. The
collected plants were rinsed with 50 ml autoclaved sea water and thal-
lus discs of around 0?6 cm diameter were excised from the lower part
of U. australis by using a steel punch. To remove other bacteria from
the surface of U. australis, surfaces of U. australis discs were swabbed
with sterile cotton tips and exposed to 0?012% (v/v) NaOCl for
5 min. Plant pieces were incubated in an antibiotic mixture contain-
ing 300 mg ampicillin l21, 30 mg polymyxin l21and 60 mg gentami-
cin l21for 24 h followed by 1 h recovery in 20 ml sterile sea water.
Attachment assay with axenic U. australis. GFP-tagged P. tuni-
cata wild-type and SM5 mutant strains were grown on VNSS
medium containing chloramphenicol for 24 h. After 24 h the cell
suspensions were centrifuged at 6000 g for 7 min. The bacterial cells
were rinsed twice with 10 ml PBS and resuspended in PBS to an
OD610of 0?35–0?45. The assay mixture consisted of 1 ml washed
cells, one axenic thallus disc of U. australis and 1 ml PBS. The mix-
ture was incubated for a period of 2 h with slow shaking and for 1 h
without shaking at room temperature. The U. australis pieces were
rinsed twice with 10 ml PBS to remove unattached cells. For each
sample, 10 images were manually analysed by counting the bacteria
attached to the surface using CLSM. The attachment assay was
repeated four times with three replicates for each bacterial strain.
The P. tunicata genome encodes a putative
MSHA biogenesis locus
P. tunicata SM5 mutant yielded a 4291 bp nucleotide
sequence with six potential ORFs (MshI1–MshM). Analysis
of this region in the draft genome sequence of P. tunicata
revealed an additional 11 ORFs predicted to be in two
transcriptional units (Fig. 1). Promoter prediction analysis
revealed potential transcriptional start sites approximately
19 bp upstream of the mshI1 start codon (210 box
‘ATTTAAGAT’ and 235 box ‘TTGCTT’) and 34 bp
‘CAGTTAAAT’ and 235 box ‘TTGTAT’). These regions
are in agreement with the 210 and 235 regions of the E. coli
s70promoter and match closely with other predicted
promoter sequences for P. tunicata (Egan et al., 2002a).
Following the translational stop of ORF17 (mshQ) is a region
acid sequence of the 17 ORFs showed homology to the
secretory and structural genes for the MSHA pilus of various
bacteria (data not shown). Moreover, the operon structure
only major difference being in the genes encoding the MshI
protein. In P. tunicata there are two genes with similarity to
different regions of the V. cholerae mshI gene and these have
been termed mshI1 and mshI2, respectively (Fig. 1).
Electron microscopic detection of cell surface pili
Following the identification of the putative MSHA pilus
operon, we examined whether P. tunicata expresses
An MSHA pilus gene cluster in Pseudoalteromonas tunicata
functional pili. Transmission electron micrographs of P.
tunicata wild-type and the SM5 mutant cells are shown in
Fig. 2. It was observed that P. tunicata wild-type cells are
surrounded by flexible pili on their surface, each with a
diameter of 7–8 nm and a length of 90–150 nm. The mean
number of surface pili per cell±SEM was calculated on 50
randomly selected cells. P. tunicata wild-type cells showed a
mean number of surface pili per cell of 11?32±0?7. P.
tunicata SM5 mutant cells were non-piliated, confirming
that the putative MSHA biogenesis operon is required for
pilus expression. In addition, we observed that wild-type
cells grown using cellobiose or cellulose as the sole carbon
source exhibited a significant increase in the mean number
those grown in VNSS (P<0?0001).
Assay for agglutination of red blood cells
To further characterize the P. tunicata pilus, haemaggluti-
nation assays were performed. It was found that P. tunicata
wild-type cells cause agglutination of horse red blood cells.
Haemagglutination was observed when either exponential-
or stationary-phase wild-type P. tunicata cells were used
(Table 1). In contrast there was no haemagglutination
observed in the presence of the SM5 mutant when cells were
assayed during exponential phase; however, in stationary-
phase cells, a low level of haemagglutination was observed.
Addition of mannose to the assay mixture blocked
haemagglutination by P. tunicata wild-type cells.
Pili promote the attachment of P. tunicata to
Polystyrene microtitre plates, widely used in bacterial
attachment assays (O’Toole & Kolter, 1998; Taylor et al.,
2002), were used as an abiotic test surface. After 1 h, P.
tunicata wild-type cells demonstrated considerably higher
attachment compared to SM5 mutant cells as detected by
the crystal violet staining of attached cells to wells of a
polystyrene microtitre plate. The crystal-violet-stained P.
tunicata wild-type cells displayed an OD590of 0?97±0?09,
which was 2?8-fold higher than the crystal-violet-stained
SM5 mutant cells (0?34±0?01).
The attachment of P. tunicata wild-type and SM5 mutant
cells to cellulose was also tested. Cellulose is a major surface
polymer of C. intestinalis (De Leo et al., 1977) and U.
australis (Chapman, 1979), and P. tunicata is frequently
isolated from these surfaces. P. tunicata wild-type cells
(grown with glucose as the carbon source) demonstrated
Fig. 1. Schematic representation of the predicted MSHA gene locus in P. tunicata (top) and V. cholerae El Tor (bottom). The
entire locus is 17525 bp in length and consists of 17 continuous ORFs (mshI1I2JKLMNEGFBACDOPQ). The scale bar
represents approximately 2 kb. The inverted triangle indicates the location of the transposon in the P. tunicata SM5 MSHA
mutant strain. Black shading, >45% identity; dark grey, 35–45% identity; pale grey, 25–35% identity; white, <25%
Fig. 2. Transmission electron micrographs showing (a) P. tuni-
cata wild-type cells expressing pili on the surface, and (b) P.
tunicata SM5 mutant cells displaying the non-piliated pheno-
type. Both strains were grown on VNSS agar plates for 24 h.
All samples were negatively stained with 2% phosphotungstic
acid. Bars, 1 mm.
D. S. Dalisay and others
attachment to microcrystalline cellulose which resulted in a
22?2% reduction in the final OD400of the assay mixture
(Fig. 3). SM5 mutant cells, however, attached less effectively
to cellulose showing only 4% reduction. Moreover,
attachment of wild-type P. tunicata was enhanced when
cells were pre-grown on cellulose and cellobiose (29?4 and
40?7% reduction, respectively). SM5 mutant cells pregrown
in cellobiose and cellulose showed 26?7 and 14?5%
reduction of the final OD400 of the assay mixture,
respectively. It was also demonstrated that mannose
significantly inhibits attachment of P. tunicata wild-type
cellstocellulose,displaying only 3?7%reduction ofthefinal
OD400of the attachment assay mixture.
We also compared the attachment of the two strains to
cellulose using fluorescently tagged cells and CLSM.
Transmission fluorescence images revealed that substantial
numbers of P. tunicata wild-type cells attached to micro-
crystalline cellulose (Fig. 4b). In contrast, the SM5 mutant
cells attached less to cellulose both during attachment in
pure culture and mixed with wild-type P. tunicata cells
(Fig. 4c and a, respectively).
Pili promote attachment of P. tunicata to the
surface of the green alga U. australis
Axenic discs of U. australis were prepared and resulted in
over 90% reduction of the natural microbial community as
previously demonstrated (Rao et al., 2006). Axenic U.
australis thallus discs were incubated with P. tunicata wild-
type:GFP and P. tunicata SM5:GFP cells, and attachment
assays were carried out in PBS, pH 7?4. The total cell counts
per mm2of algal surface for P. tunicata wild-type and the
SM5 mutant were 2377±607 and 996±362 cells, respec-
tively. P. tunicata wild-type cells demonstrated 2?4-fold
enhanced attachment compared to the SM5 mutant on the
surface of U. australis (Fig. 5).
Surface-associatedbacteria onmarine sessileorganismsmay
provide benefits to their hosts. For example, P. tunicata
produces a range of inhibitory compounds and may offer
protection against colonization of biofouling and host-
pathogenic organisms (Egan et al., 2000, 2001b; Holmstro ¨m
et al., 1992; James et al., 1996). Such interactions are
initiated by the attachment and initial colonization of
bacteria to the host surface, and may be facilitated by the
expression of surface adhesins. In V. cholerae El Tor, the
attachment to phytoplankton and zooplankton surfaces is
mediated by the MSHA pilus (Chiavelli et al., 2001).
Recently, it was reported that an MSHA pilus promotes
interaction between V. cholerae El Tor and haemolymph of
this study, we investigated the role of an MSHA-like pilus in
the attachment of P. tunicata to the surface of U. australis, a
livingsurfacefromwhichP.tunicata hasbeenisolated inthe
This study describes a gene locus proposed to be involved in
have sequenced the DNA flanking a Tn10-disrupted ORF
with high homology to mshJ of V. cholerae (Marsh & Taylor,
recently obtained draft genome of P. tunicata, revealed 17
contiguous ORFs (mshI1I2JKLMNEGFBACDOPQ) with
Table 1. Haemagglutination of horse erythrocytes by bacteria at different growth stages
Values represent the reciprocal of the dilution in which haemagglutination can be observed. The results
of the assay performed in the presence of 1% a-methyl-D-mannoside were zero for all strains tested in
both growth phases.
Bacterial strainStationary phase Exponential phase
Vibrio cholerae M1615 (+ control)
P. tunicata wild-type
P. tunicata mshJ (SM5) mutant
Fig. 3. Attachment of P. tunicata wild-type (black bars) and
SM5 mutant (grey bars) to microcrystalline cellulose (Avicel).
The microcrystalline cellulose was mixed with bacterial cells
(~107c.f.u. ml”1) for 1 h and the final OD400of the cellulose
and bacterial cell suspension was measured after 1 h incuba-
tion. The attachment of bacteria to cellulose was scored by the
reduction in the final OD400of the suspension.a, Bacteria were
grown in this medium prior to the attachment assay which was
performed in the absence of mannose;
in this medium prior to the attachment assay which was per-
formed in the presence of mannose. Error bars represent the
SEM of triplicate cultures.
b, bacteria were grown
An MSHA pilus gene cluster in Pseudoalteromonas tunicata
similar genetic organization and high homology to MSHA
pili biogenesis and secretory proteins of various Gram-
negative bacteria, including Pseudoalteromonas haloplanktis,
Shewanella baltica, V. cholerae, Vibrio parahaemolyticus and
Vibrio vulnificus. The best characterized MSHA pilus gene
locus is that of V. cholerae El Tor (Marsh & Taylor, 1999).
The 17 P. tunicata ORFs identified in this study share major
similarities with the MSHA pili secretory and structural
genesofV.cholerae,including (1)homology ofthepredicted
of the gene products, (3) similarity in their organization,
orientation and arrangement, and (4) the presence of
polycistronic genes, where overlapping stop and start
codons of ORFs have been identified.
TEM studies revealed the presence of flexible pili on the cell
surface of P. tunicata, with ultrastructural characteristics
similar to pili of other Gram-negative bacteria. In contrast,
the P. tunicata SM5 mutant, disrupted in the MSHA gene
locus, showed no expression of pili on the cell surface.
Biogenesis of pili requires numerous gene products,
including a structural prepilin subunit, ancillary proteins
with prepilin-like leader sequences, inner- and outer-
membrane proteins and nucleotide-binding proteins (Alm
& Mattick, 1997). It has been reported that a mutation in
any of these genes is sufficient to prevent the assembly of
of any one of the secretory genes pilO, pilP or pilQ in
Pseudomonas aeruginosa resulted in loss of pili, confirming
the importance of these mutated genes in pilus biogenesis
(Martin et al., 1995). These secretory genes belong to an
operon for secretion and export which includes two other
genes (pilM and pilN) required in the biogenesis of fimbriae
in Pseudomonas aeruginosa (Martin et al., 1995). In V.
cholerae, it has been reported that the expression of MSHA
to form functional pili on the bacterial surface is completely
dependent on the transcription and expression of two
operons: secretory and structural (Martin et al., 1995).
Deletion in any of the putative promoter regions upstream
of mshI, a secretory gene, or mshB, a structural gene,
abolished MSHA pilus assembly, secretion and expression
(Martin et al., 1995). Additionally, mutation in mshE, a
secretory gene in the MSHA biogenesis locus, showed
abolished haemagglutination (Hase et al., 1994). The fact
that the P. tunicata SM5 mutant displayed a non-piliated
phenotype and was unable to mediate haemagglutination
suggests that the MSHA gene locus is required for pilus
biogenesis in P. tunicata.
Gram-negative bacteria bind to surfaces via the tip adhesins
of the pili (Strom & Lory, 1993). These adhesins may have
different specific receptors. For example, V. cholerae El Tor
strains express a haemagglutinin pilus with a preference for
mannose receptors (Jonson et al., 1991). We have demon-
strated that the P. tunicata pilus causes haemagglutination
of horse red blood cells. As for V. cholerae El Tor, the
Fig. 4. Transmission fluorescent micrographs
using CLSM showing attachment of bacterial
strains to microcrystalline cellulose (Avicel)
after the adherence assay. (a) P. tunicata wild-
type: GFP and SM5 mutant: RFP added to
the plant as a 1:1 mixture; (b) P. tunicata
wild-type:GFP only and (c) SM5 mutant:RFP
only. Bar, 20 mm.
D. S. Dalisay and others
clumping of red blood cells was abolished in the presence of
The MSHA pili expressed by V. cholerae cells are known to
be involved in the attachment to and colonization of
in aquatic environments and has been shown to mediate
attachment to solid substrates (Chiavelli et al., 2001;
Watnick et al., 1999). It was reported that the V. cholerae
El Tor MSHA mutant strain is unable to form biofilms and
shows decreased adherence to abiotic and biotic surfaces
(Watnick et al., 1999). In this study, we first tested
attachment of P. tunicata wild-type and SM5 mutant cells
to polystyrene and to cellulose, the surface polymer of U.
australis and C. intestinalis from which P. tunicata has been
isolated. The SM5 mutant cells showed less attachment to
both polystyrene and cellulose surfaces, in comparison with
the piliated wild-type strain. This suggests a key role for the
P. tunicata MSHA pilus in the attachment to both surfaces.
To determine if the MSHA pilus is involved in mediating
the attachment of P. tunicata cells to living surfaces we
compared the ability of the piliated and non-piliated strains
to attach directly to the surface of the green alga U. australis.
of attached P. tunicata wild-type cells compared to SM5
mutant cells at the surface of U. australis. However, a small
number of SM5 cells did attach to the algal surface,
suggesting that other mechanisms may be involved in the
adhesion of P. tunicata to the surface of U. australis (for
example non-specific physicochemical interactions between
the bacterial cell and plant surface). Nevertheless, our
findings indicate that the MSHA pilus is a major
determinant for attachment of P. tunicata to U. australis.
This study also showed that specific growth conditions affect
grown in cellulose or cellobiose were found to be
hyperpiliated when examined under TEM. These specific
We observed that wild-type cells pregrown in cellulose or
cellobiose displayed enhanced attachment to cellulose
compared to cells grown on other carbon sources (a smaller
increase in adhesion to cellulose of the SM5 mutant grown in
cellulose or cellobiose was also observed, although the reason
for this is unclear). These studies indicate that cellulose or
cellobiose may serve as environmental signals which induce
expression of the MSHA pilus and thus promote attachment
of P. tunicata to cellulose-containing surfaces. Our findings
can control pilus expression. In Pseudomonas aeruginosa,
expression of pilA is controlled by a two-component sensor–
regulator gene pair, pilS and pilR (Hobbs et al., 1993). The
PilS protein is a sensor protein located upstream of the
regulator protein PilR, thought to be responsible for sensing
unknown environmental signals (Boyd, 2000).In V. cholerae,
the response of tcp (toxin-coregulated pili) genes to
environmental stimuli is mediated by the ToxR regulon
(Strom & Lory, 1993). The ToxR protein is a major sensor
and regulator protein that transmits signals from the
periplasmic side of the inner membrane and regulates
transcription of virulence factors, including the toxin-
coregulated pili. Interestingly, a putative transcriptional
regulation of the expression of antifouling compounds in P.
tunicata(Eganetal., 2002b).The P. tunicatawmpRmutant is
devoid of pili on its surface when examined using TEM (data
may explain the surface-sensing mechanisms demonstrated
by P. tunicata in response to environmental stimuli. In the
marine ecosystem, the P. tunicata wmpR gene product may
sense an environmental signal (e.g. cellulose or surface
polymers of C. intestinalis and U. australis) and respond by
increasing the expression of MSHA pili on the cell surface,
of these marine organisms.
Fig. 5. CLSM images of (a) P. tunicata wild-type GFP and (b)
P. tunicata (SM5) mshJ mutant GFP attached to the surface of
U. australis. Bars, 50 mm.
An MSHA pilus gene cluster in Pseudoalteromonas tunicata
organization and high homology to MSHA pilus biogenesis
operons of marine vibrios, and have shown that the msh
gene locus is required for theexpressionofa MSHA piluson
the P. tunicata cell surface. Our findings demonstrate that
production of the MSHA pilus mediates the attachment of
P. tunicata to surfaces and that this pilus is a key
determinant of attachment of P. tunicata to both abiotic
and living surfaces. We also provide evidence that
expression of MSHA pilus is enhanced in the presence of
cellulose or cellobiose and propose that substrate-sensing
mechanisms and pilus expression in P. tunicata have a
profound effect on the ecological distribution of this
bacterium in marine surface ecosystems.
This work was supported by the Australian Research Council and by
the Centre for Marine Biofouling and Bio-Innovation, University of
New South Wales, Australia.
Alm, R. A. & Mattick, J. S. (1997). Genes involved in the biogenesis
and function of type-4 fimbriae in Pseudomonas aeruginosa. Gene
Altschul, S. F., Gish, W., Miller, W., Meyers, E. W. & Lipman, D. J.
(1990). Basic Local Alignment Search Tool. J Mol Biol 215, 403–410.
Bagdasarian, M., Lurz, R., Ruckert, B., Franklin, F. C., Bagdasarian,
M. M., Frey, J. & Timmis, K. N. (1981). Specific-purpose plasmid
cloning vectors. II. Broad host range, high copy number, RSF1010-
derived vectors, and a host–vector system for gene cloning in
Pseudomonas. Gene 16, 237–247.
Bayer, E., Kenig, R. & Lamed, R. (1983). Adherence of Clostridium
thermocellum to cellulose. J Bacteriol 156, 818–827.
Boyd, J. M. (2000). Localization of the histidine kinase PilS to the
poles of Pseudomonas aeruginosa and identification of a localization
domain. Mol Microbiol 36, 153–162.
Bryers, J. & Characklis, W. (1982). Processes governing primary
biofilm formation. Biotechnol Bioeng 24, 2451–2476.
Organization. Baltimore, MD: University Park Press.
A.R.O.(1979). Biology of Seaweeds: Levels of
Chiavelli, D. A., Marsh, J. W. & Taylor, R. K. (2001). The mannose-
sensitive hemagglutinin of Vibrio cholerae promotes adherence to
zooplankton. Appl Environ Microbiol 67, 3220–3225.
Cormack, B. P., Valdivia, R. H. & Falkow, S. (1996). FACS-optimized
mutants of the green fluorescent protein (GFP). Gene 173, 33–38.
De Leo, G., Patricolo, E. & D’Ancona-Lunetta, G. (1977). Studies on
the fibrous components of the test of Ciona intestinalis Linnaes.
Cellulose-like polysaccharide. Acta Zoo 58, 135–141.
de Nys, R., Steinberg, P., Willemsen, P., Dworjanyn, S., Gabelish, C.
& King, R. (1995). Broad spectrum effects of secondary metabolites
from the red alga Delisea pulchra in antifouling assays. Biofouling 8,
Egan, S., Thomas, T., Holmstro ¨m, C. & Kjelleberg, S. (2000).
Phylogenetic relationship and antifouling activities of bacterial
epiphytes from the marine alga Ulva lactuca. Environ Microbiol 2,
Egan, S., Holmstro ¨m, C. & Kjelleberg, S. (2001a). Pseudoalteromonas
ulvae sp. nov., a bacterium with antifouling activities isolated from the
surface of a marine alga. Int J Syst Evol Microbiol 51, 1499–1504.
Egan, S., James, S., Holmstro ¨m, C. & Kjelleberg, S. (2001b).
Inhibition of algal spore germination by the marine bacterium
Pseudoalteromonas tunicata. FEMS Microbiol Ecol 35, 67–73.
Egan, S., James, S., Holmstro ¨m, C. & Kjelleberg, S. (2002a).
Correlation between pigmentation and antifouling compounds
produced byPseudoalteromonastunicata. Environ Microbiol4, 433–442.
Egan, S., James, S. & Kjelleberg, S. (2002b). Identification and
characterization of a putative transcriptional regulator controlling
the expression of fouling inhibitors in Pseudoalteromonas tunicata.
Appl Environ Microbiol 68, 372–378.
Gardel, C. & Mekalanos, J. (1996). Alterations in Vibrio cholerae
motility phenotypes correlate with changes in virulence factor
expression. Infect Immun 64, 2246–2255.
Harrison, P. (1992). Control of microbial growth and of amphipod
gazing by water-soluble compounds from leaves of Zostera marina.
Mar Biol 67, 25–30.
Hase, C. C., Bauer, M. E. & Finkelstein, R. A. (1994). Genetic
characterization of mannose-sensitive hemagglutinin (MSHA)-negative
Hobbs, M., Collie, E., Free, P., Livingston, S. & Mattick, J. (1993).
PilS and PilR, a two component transcriptional regulatory system
controlling expression of type 4 fimbriae in Pseudomonas aeruginosa.
Mol Microbiol 7, 669–682.
Holmstro ¨m, C. & Kjelleberg, S. (1999). Marine Pseudoalteromonas
species are associated with higher organisms and produce biologi-
cally active extracellular agents. FEMS Microbiol Ecol 30, 285–293.
Holmstro ¨m, C., Rittschof, D. & Kjelleberg, S. (1992). Inhibition of
settlement by larvae of Balanus amphitrite and Ciona intestinalis by a
surface-colonizing marine bacterium. Appl Environ Microbiol 58,
Holmstro ¨m, 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
Holmstro ¨m, C., James, S., Neilan, B., White, D. & Kjelleberg, S.
(1998). Pseudoalteromonas tunicata sp. nov., a bacterium that
produces antifouling agents. Int J Syst Bacteriol 48, 1205–1212.
Holmstro ¨m, C., Egan, S., Franks, A., McCloy, S. & Kjelleberg, S.
(2002). Antifouling activities expressed by marine surface associated
Pseudoalteromonas species. FEMS Microbiol Ecol 41, 47–58.
James, S., Holmstro ¨m, C. & Kjelleberg, S. (1996). Purification and
characterization of a novel antibacterial protein from the marine
bacterium D2. Appl Environ Microbiol 62, 2783–2788.
Jonson, G., Holmgren, J. & Svennerholm, A. (1991). Identification of
a mannose-binding pilus on Vibrio cholerae El Tor. Microb Pathog
Marden, P., Tunlid, A., Malmcrona-Friberg, K., Odham, G. &
Kjelleberg, S. (1985). Physiological and morphological changes
during short term starvation of marine bacterial isolates. Arch
Microbiol 142, 326–332.
Marsh, J. W. & Taylor, R. K. (1999). Genetic and transcriptional
analyses of the Vibrio cholerae mannose-sensitive hemagglutinin type
4 pilus gene locus. J Bacteriol 181, 1110–1117.
Martin, P., Watson, A., McCaul, T. & Mattick, J. (1995). Characterization
of a five-gene cluster required for the biogenesis of type 4 fimbriae in
Pseudomonas aeruginosa. Mol Microbiol 16, 497–508.
Mary, A., Mary, V., Rittschof, D. & Nagabhushanam, R. (1993).
Bacterial–barnacle interaction: potential for using juncellins and
D. S. Dalisay and others
antibiotics to alter structure of bacterial communities. J Chem Ecol
Mathews, C. K. & van Holde, K. E. (1990). Biochemistry. Redwood
City, CA: The Benjamin/Cummings Publishing Company.
Maximilien, R., de Nys, R., Holmstro ¨m, C., Gram, L., Givskov, M.,
Crass, K., Kjelleberg, S. & Steinberg, P. (1995). Chemical mediation
of bacterial surface colonization by secondary metabolites from red
algal Delisea pulchra. Aquat Microb Ecol 15, 233–246.
Morand,P.C.,Tattevin,P., Eugene,E., Beretti,J.-L. & Nassif, X.(2001).
The adhesive property of the type IV pilus-associated component PilC1
of pathogenic Neisseria is supported by the conformational structure of
the N-terminal part of the molecule. Mol Microbiol 40, 846–856.
Neidhardt, F., Bloch, P. & Smith, D. (1974). Culture medium for
enterobacteria. J Bacteriol 119, 736–747.
O’Toole, G. A. & Kolter, R. (1998). Flagellar and twitching motility
are necessary for Pseudomonas aeruginosa biofilm development. Mol
Microbiol 30, 295–304.
Rao, D., Webb, J. S. & Kjelleberg, S. (2005). Competitive interactions
in mixed-species biofilmscontaining
Pseudoalteromonas tunicata. Appl Environ Microbiol 71, 1729–1736.
Rao, D., Webb, J. S. & Kjelleberg, S. (2006). Microbial colonization
and competition on the marine alga Ulva australis. Appl Environ
Microbiol 72, 5547–5555.
Reese, M. (2001). Application of a time-delay neural network to
promoter annotation in the Drosophila melanogaster genome. Comp
Chem 26, 51–56.
Strom, M. S. & Lory, S. (1993). Structure–function and biogenesis of
the type IV pili. Annu Rev Microbiol 47, 565–596.
Taylor, C. M., Beresford, M., Epton, H. A. S., Sigee, D. C., Shama, G.,
Andrew, P. W. & Roberts, I. S. (2002). Listeria monocytogenes relA
and hpt mutants are impaired in surface-attached growth and
virulence. J Bacteriol 184, 621–628.
Tillett, D. & Neilan, B. A. (1999). n-Butanol purification of dye
terminator sequencing reactions. Biotechniques 26, 606–608, 610.
Watnick, P. I. & Kolter, R. (1999). Steps in the development of a
Vibrio cholerae El Tor biofilm. Mol Microbiol 34, 586–595.
Watnick, P. I., Fullner, K. J. & Kolter, R. (1999). A role for the
mannose-sensitive hemagglutinin in biofilm formation by Vibrio
cholerae El Tor. J Bacteriol 181, 3606–3609.
Zampini, M., Canesi, L., Betti, M., Ciacci, C., Tarsi, R., Gallo, G. &
Pruzzo, C. (2003). Role for mannose-sensitive hemagglutinin in
promoting interactions between Vibrio cholerae El Tor and mussel
hemolymph. Appl Environ Microbiol 69, 5711–5715.
Zolfaghar, I., Evans, D. J. & Fleiszig, S. M. J. (2003). Twitching
motility contributes to the role of pili in corneal infection caused by
Pseudomonas aeruginosa. Infect Immun 71, 5389–5393.
An MSHA pilus gene cluster in Pseudoalteromonas tunicata