METAGENOMICS OF CYANOBACTERIAL BLOOMS

1
METAGENOMICS OF CYANOBACTERIAL BLOOMS
Phillip B Pope and Bharat K.C. Patel Microbial
Gene Research and Resources Facility, School of
Biomolecular and Biomedical Sciences, Faculty of
Science, Griffith University, Brisbane,
Queensland 4111, Australia, Eskitis Institute for
cell and molecular therapies, Griffith
University, Brisbane, Queensland 4111, Australia,
and Cooperative Research Centre for Water Quality
and Treatment, Australia.
BACKGROUND AIMS Cyanobacterial blooms are
commonly associated with the production of
secondary metabolites including numerous human
health affecting toxins. Molecular analysis of
large genomic fragments recovered directly from
an environmental bloom sample represents an
approach by which genes responsible for the
formation and expression of secondary metabolites
can be studied. Consequently, we have prepared a
large-construct Bacterial Artificial Chromosome
library (BAC) from the DNA of a natural
toxin-producing cyanobacterial bloom. We have in
tandem, from the same cyanobacterial bloom DNA,
constructed and analysed PCR based single gene
libraries of 16S rRNA and polyketide synthases
(PKS). This three-fold experimental approach has
the potential to provide a phylogenetic community
snap shot of the cyanobacterial bloom community
structure and their physiological functions
within the bloom. Therefore molecular analysis of
the BAC library together with the PCR single gene
libraries provides a powerful tool for
establishing a link between secondary metabolite
production (e.g. PKS) and microbial diversity in
situ.
Collection of bloom sample Lake Samsonvale
(27?16S, 152?56E), a warm monomictic eutrophic
reservoir experiences thermal stratification from
September to May with toxic cyanobacterial blooms
a frequent occurrence during this period. Water
samples (0-3 m) were collected predefined
locations, 10001S and 10006S (SEQwater
cooperation, Brisbane, QLD), with an integrated
depth sampler in mid-September 2003. Samples were
kept at 4?C for no more than 24 hours before use.
Collection of cyanobacterial bloom sample
Fig 1. Lake Samsonvale (North Pine dam).
Extraction of high molecular weight DNA
Construction of BAC library High molecular weight
(HMW) DNA was isolated from the toxic bloom
sample and a BAC library consisting of 2850
clones constructed. The average insert size of
CBNPD1 library determined after Bgl II digestion
and pulsed field gel electrophoresis (PFGE) was
determined to be 27 kb, with size ranging from
5kb to greater than 50kb (Fig 2 and 3). Greater
than 60 of the clones contained inserts greater
than 20kb in size.
Fig 3. BAC insert size distribution.
random BAC clones
97.0 Kb
48.5 Kb
Single gene PKS library A high proportion of
cyanobacterial secondary metabolites have been
found to belong to groups of polyketides and
peptides. A single-gene library constructed from
PCR fragments amplified from cyano-bacterial
bloom HMW DNA (Fig 4) revealed the presence of 9
nucleotide sequences that encoded regions of PKS
genes (Table 1). Analysis suggests that the same
gene diversity could be represented in the BAC
library as the same HMW DNA was used to construct
both libraries.
23.1 Kb
9.0 Kb
6.0 Kb
L2 L3
vector fragment
Fig 2. PFGE of random BAC clones.
Construction of BAC library
Single gene polyketide- synthase library
Library screening sequence and expression 16S
rRNA genes from pooled clones were amplified. Two
clones identified (Fig 5) were presumed to
contain 16S rRNA genes. Sequence analysis showed
that they belonged to 2 phylogenetic groups
(Table 2). One of the 2 BAC clones containing the
16S rRNA gene clone 578 was sequenced and is
reported below (BAC clone insert sequencing and
analysis). Evidence of functional expression of
the cloned library was provided from the presence
of amylase in clone 905 (Fig 6).
Fig 4. Gel electrophoresis analysis of PCR
reactions using cyanobacterial PKS specific
primers DKF/DKR from two different DNA templates
(DNA extraction (L2), and ligation mix (L3))
used in CBNPD1 BAC library construction.
Library screening sequence and expression
Fig 6. Example of a starch hydrolysis screen
detecting CBNPD1 clone 905 producing a
metagenomic DNA encoded amylase by the clone host
E.coli.

Table 1. Protein-coding genes of highest
similarity to that of sequenced and analysed
clone insert DNA originating from PCR screens
(Fig 4).
Fig 5. PCR screening of CBNPD1 clones 577-588 to
identify clones containing 16S rDNA. Lanes 3 and
6 represent amplified 16S rDNA products from
individually examined clones (578 and 581).
Sequence originating from lake Samsonvale
Cluster possibly characteristic for bloom events
Single gene 16S rDNA library
BAC clone insert sequencing and analysis
Table 3. 16S rRNA sequences obtained from CBNPD1
BAC clones identified from 16S rDNA library
screens.
OUTCOMES FUTURE DIRECTIONS In a bid to
understand microbial community structure and
composition as well as determine the mechanisms
of blooming and toxin production in
cyanobacterial blooms, we used a
culture-independent approach and constructed a
metagenomic library, designated CBNPD1.
Preliminary molecular evidence suggests that the
library contains valuable information on genes,
gene clusters and partial or entire metabolic
pathways of yet to be cultured microbes. Our
single gene studies have demonstrated the
presence of a diverse range of PKS genes present
in the cyanobacterial bloom under investigation.
We hope to screen for their presence in our
metagenomic library in future. The marriage of
our studies will provide a much-needed
understanding of interactions within bacterial
communities associated with cyanobacterial blooms
and their distribution. It is hoped that this
information will ultimately be used to predict
bloom events and the data be included as part of
water management strategies.
Response regulator
Response regulator
RTX toxin
Fig 7. Phylogenetic tree of partial 16S rDNA
sequences obtained from Lake Samsonvale (CYN-)
affiliating to sequences obtained from GenBank of
the Domain Alpha-proteobacteria. Novel freshwater
clusters identified in this study are designated
as cyn. Sequences titled LiUU-, MCY- and CYN- are
sequences from cyanobacterial bloom associated
communities in Sweden and Australia.
Single gene 16S rDNA library The diversity of
bacterial communities associated with
cyanobacterial blooms have been poorly studied
despite their probable important role in
cyanobacterial bloom structure and function. A
phylogenetic comparison of 78 16S rDNA sequences
has identified 5 new clusters which also may be
characteristic of bloom events (one example shown
in Fig 7) showing that bacterial communities
associated with cyanobacterial blooms may have
specific groups that are distinct for bloom events
Fig 8. Linear ORF maps of the 7 completely
sequenced BAC clones from the Cyanobacterial
bloom metagenome library. ORFs are colour coded
according to their COG affiliations and to
highest ribosomal genes, where they exist
BAC clone insert sequencing and analysis BAC
inserts sequenced to completion comprised 173 kb
of a cyanobacterial bloom metagenome. 183 genes
have been identified and assigned to COG
functional categories, several of which were
found to affiliate to proteins of interest both
ecologically and in terms of potential in
industrial processes. These include a putatuve
RTX toxin and putative response regulators
involved in quorum sensing and controlling
expression of exoproteins, including toxins.