Phylogenetic and metagenomic analysis of Verrucomicrobia in former agricultural grassland soil.

Abstract

The bacterial phylum Verrucomicrobia has a widespread distribution, and is known to be one of the most common and diverse phyla in soil habitats. However, members of this phylum have typically been recalcitrant to cultivation methods, hampering the study of this presumably important bacterial group. In this study, we examine the phylogenetic diversity of the Verrucomicrobia in a former agricultural field and gain access to genomic information via a metagenomic approach. We examined Verrucomicrobia-like 16S rRNA gene sequences recovered from general bacterial and phylum-specific libraries, revealing a dominance of subdivisions 1 and 2. A PCR-based screening method was developed to identify inserts containing verrucomicrobial 16S rRNA genes within a large-insert metagenomic library, and on screening of 28,800 clones, four fosmids were identified as containing verrucomicrobial genomic DNA. Full-length sequencing of fosmid inserts and gene annotation identified a total of 98 ORFs, representing a range of functions. No conservation of gene order was observed adjacent to the ribosomal operons. Fosmid inserts were further analyzed for tetranucleotide frequencies to identify remnants of past horizontal gene transfer events. The metagenomic approach utilized proved to be suitable for the recovery of verrucomicrobial genomic DNA, thereby providing a window into the genomes of members of this important, yet poorly characterized, bacterial phylum.

Full text

Page 1

R E S E A R C H A RT IC L E
PhylogeneticandmetagenomicanalysisofVerrucomicrobia in
former agriculturalgrasslandsoil
Anna Kielak1, Jorge L.M. Rodrigues2, Eiko E. Kuramae1,3, Patrick S.G. Chain4,5, Johannes A. van Veen1,6
& George A. Kowalchuk1,7
1Department of Microbial Ecology, Netherlands Institute of Ecology (NIOO-KNAW), Heteren, The Netherlands;2Arlington Department of Biology,
University of Texas, Arlington, TX, USA;3Department of Animal Ecology, Institute of Ecological Science, Free University Amsterdam, The Netherlands;
4Microbial Program, Joint Genome Institute, Walnut Creek, CA, USA;5Center for Microbial Ecology, Michigan State University, East Lansing, MI, USA;
6Institute of Biology, Leiden University, Leiden, The Netherlands; and7Institute of Ecological Science, Free University Amsterdam, Amsterdam, The
Netherlands
Correspondence: George A. Kowalchuk,
Department of Microbial Ecology,
Netherlands Institute of Ecology (NIOO-
KNAW), PO Box 40, 6666 ZG, Heteren, The
Netherlands. Tel.: 131 026 479 1314; fax:
131 026 472 3227; e-mail:
g.kowalchuk@nioo.knaw.nl
Received 7 April 2009; revised 18 August 2009;
accepted 24 August 2009.
Final version published online 7 October 2009.
DOI:10.1111/j.1574-6941.2009.00785.x
Editor: Michael Wagner
Keywords
Verrucomicrobia; phylogeny; environmental
genomics; bacterial communities; rhizosphere;
metagenomics.
Abstract
The bacterial phylum Verrucomicrobia has awidespread distribution, and is known
to be one of the most common and diverse phyla in soil habitats. However,
members of this phylum have typically been recalcitrant to cultivation methods,
hampering the study of this presumably important bacterial group. In this study,
we examine the phylogenetic diversity of the Verrucomicrobia in a former
agricultural field and gain access to genomic information via a metagenomic
approach. We examined Verrucomicrobia-like 16S rRNA gene sequences recovered
from general bacterial and phylum-specific libraries, revealing a dominance of
subdivisions 1 and 2. A PCR-based screening method was developed to identify
inserts containing verrucomicrobial 16S rRNA genes within a large-insert metage-
nomic library, and on screening of 28800 clones, four fosmids were identified as
containing verrucomicrobial genomic DNA. Full-length sequencing of fosmid
inserts and gene annotation identified a total of 98 ORFs, representing a range of
functions. No conservation of gene order was observed adjacent to the ribosomal
operons. Fosmid inserts were further analyzed for tetranucleotide frequencies to
identify remnants of past horizontal gene transfer events. The metagenomic
approach utilized proved to be suitable for the recovery of verrucomicrobial
genomic DNA, thereby providing a window into the genomes of members of this
important, yet poorly characterized, bacterial phylum.
Introduction
Members of the phylum Verrucomicrobia are abundant in
diverse habitats, including soil, water and sediments, as well
as extreme environments such as in the Antarctic sample
(Pearce et al., 2003), hot springs (601C) and low pH 2.0–2.5
habitats (Hou et al., 2008). The lifestyles of some Verruco-
microbia have also been linked to eukaryotic gut environ-
ments, including termites (Shinzato et al., 2005), sea
cucumbers (Sakai et al., 2003) and humans (Wang et al.,
2005), or as obligate endosymbionts of nematodes from the
genus Xiphinema (Vandekerckhove et al., 2000). In addition,
methanotrophs have been identified among the Verruco-
microbia (Dunfield et al., 2007; Pol et al., 2007). Approxi-
mately 5% of all characterized 16S rRNA genes are affiliated
with the Verrucomicrobia, and this phylum represents
1.2–11% of the total soil bacteria (Sangwan et al., 2004;
Kielak et al., 2008). Although only recognized as a separate
phylum in the pastdecade (Garrity & Holt, 2001), their large
numbers and broad distribution in soils suggest that they are
important members of terrestrial ecosystems.
Although recent advances in cultivation methods have
increased the number and diversity of available Verrucomi-
crobia isolates (Janssen et al., 1997; Janssen et al., 2002;
Sangwan et al., 2004, 2005; Stevenson et al., 2004; Yoon etal.,
2007a,b), the general poor cultivation of Verrucomicro-
bia has impaired progress in understanding their ecology.
The Verrucomicrobia has been organized into seven
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
MICROBIOLOGY ECOLOGY

Page 2

subdivisions based on 16S rRNAgene analyses, although not
all subdivisions contain culture representatives (Schlesner
et al., 2006). Characterized genera include Verrucomicro-
bium (subdivision 1 – Verrucomicrobiae) (Schlesner, 1987),
Prosthecobacter (subdivision 1) (Staley et al., 1976), Akker-
mansia (subdivision 1) (Derrien et al., 2004), Xiphinemato-
bacter (subdivision 2 – Spartobacteria) (Vandekerckhove
et al., 2000), Chthoniobacter (subdivision 2) (Sangwan
et al., 2004), Opitutus (subdivision 4 – Opitutaceae) (Chin
et al., 2001), Rubritalea (subdivision 4) (Scheuermayer et al.,
2006) and Victivallis (subdivision 7) (Zoetendal et al., 2003).
Complete or draft sequences have recently become available
for eight Verrucomicrobia strains (bacterium Ellin514, Ver-
rucomicrobiae bacterium DG1235, Akkermansia muciniphila
ATCC BAA-835, Opitutaceae bacterium TAV2, Opitutus
terrae PB90-1, Methylacidiphilum infernorum V4, Chthonio-
bacter flavus Ellin428 and Verrucomicrobium spinosum DSM
4136), providing a first insight into the genetic makeup of
cultivated members of the Verrucomicrobia. As with other
bacteria, it is thought that cultivated members of the
Verrucomicrobia represent a very biased sampling of the
phylum due to the inherent selection of isolates capable of
growth under (novel) laboratory conditions. To date, of the
89 verrucomicrobial 16S rRNA gene sequences deposited in
the RDPII database (Cole et al., 2005) derived from isolates,
40 are affiliated to subdivision 4, followed by 18, 14 and six
sequences from subdivisions 1, 2 and 3, respectively. Only 17
of these sequences have come from soil isolates, 11 of which
are affiliated to subdivision 4 and six to subdivision 6.
Although Verrucomicrobia have commonly been recov-
ered from diverse soil environments, relatively little is
known about their diversity and distribution with respect
to the environmental conditions present in different soil-
borne habitats. Soil management and nutritional regimes
appear to influence Verrucomicrobia community structure
(Buckley & Schmidt, 2001), and bacterial 16S rRNA gene
library and quantitative PCR analyses have shown that
Verrucomicrobia tend to be more dominant in the rhizo-
sphere (Kielak et al., 2008). However, differences between
the rhizospheres of different plants tended not to be
significant (Kielak et al., 2008). Thus, it appears that
Verrucomicrobia represent important rhizosphere coloni-
zers, and yet virtually no information is available to date on
the genotypic or the phenotypic characteristics of these
bacteria. Furthermore, their importance in terrestrial eco-
systems and potential as sourcesof natural products remains
unexplored.
In this study, we therefore sought to gain an insight into
the phylogenetic diversity of Verrucomicrobia in a former
agricultural field and to develop and apply a cultivation-
independent approach towards gaining access to genomic
information from this phylum. To address the first of these
goals, Verrucomicrobia-like 16S rRNA gene sequences were
examined from the libraries recovered using general bacter-
ial and Verrucomicrobia-specific PCR primers. In addition, a
large-insert fosmid library (28800 clones) was constructed
from high-molecular-weight DNA isolated from the rhizo-
sphere of Festuca ovina grown at the same field site. A PCR-
based screening procedure was developed and implemented
to identify clones containing inserts with Verrucomicrobia-
like 16S rRNA genes. Four inserts, positively determined to
be of verrucomicrobial origin, were subsequently subjected
to full-length DNA sequence analysis, followed by identifi-
cation and annotation of ORFs. The genomic properties of
verrucomicrobial genomic inserts were studied and patterns
of oligonucleotide usage were examined to identify sites of
potential horizontal gene transfer (HGT).
Materials and methods
Field site and sampling
All soil samples were collected from a former arable field site
near Ede (521040N, 51450E), the Netherlands, where an
ecosystem restoration experiment was initiated after the
1995 harvest season within the European project ‘Changing
land usage: enhancement of biodiversity and ecosystem
development’. Detailed characteristics of the soil and plant
diversity treatment plots are provided in Van der Putten
et al. (2000) and Bezemer & Van der Putten (2007).
Samples for the construction of general bacterial 16S
rRNA gene libraries were derived from collective rhizo-
sphere and bulk soil from each of the plant diversity
treatments in the above experiment as described by Kielak
et al. (2008). For the construction of Verrucomicrobia-
specific 16S rRNA gene libraries, rhizosphere soil samples
were collected in August 2005 from F. ovina L. plants from
within monoculture plots of this species, adjacent to the
main plant diversity treatment experiment, as well as from
Lotus corniculatus L. plants from various locations within
the borders of the plots described above (Bezemer et al.,
2006; Kielak et al., 2008).
Rhizosphere and bulk soil samples were taken as follows:
roots were first separated and shaken to remove loosely
adhering soil. Pieces of roots with the remaining adhering
soil were cut and taken as a combined sample of rhizosphere
and root surface (rhizoplane), further referred to as rhizo-
sphere samples. Soil remaining behind after the removal of
plant material was defined as bulk soil. All soil samples were
frozen at 201C for approximately 2 weeks before DNA
extraction.
DNA isolation
For the construction of all 16S rRNA gene libraries, DNA
was isolated from 250mg wet weight soil using the Ultra
CleanTMDNA Isolation Kit,according to the manufacturer’s
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
24
A. Kielak et al.

Page 3

specifications (MoBio Laboratories, Solana Beach, CA). For
large-insert metagenomic library construction, high-mole-
cular-weight DNA was extracted from 10g of soil (wet
weight) as described previously (Kielak et al., 2009).
The construction of the metagenome library utilized the
CopyControl Fosmid Library Production Kit (Epicentre,
Madison, WI) according to the manufacturer’s protocol.
50-phosphorylated blunt-end DNAs were ligated into the
pCC1FOS vector and transformed into Escherichia coli
EPI300 (Epicentre). Recovered clones (28800) were stored
in 96-well plates at ?801C until use.
PCR-based cloning strategies
Bacterial 16S rRNA gene libraries were constructed as
described by Kielak et al. (2008). For Verrucomicrobia-
specific libraries, a phylum-specific forward primer, Ver53F,
was combined with a bacterial reverse primer, 1378r (Ste-
venson et al., 2004). The specificity of the primer was
confirmed by probe match check against the Silva 16S rRNA
gene database (Pruesse et al., 2007). Thermocycling con-
sisted one 2-min denaturation step at 961C, followed by
30 cycles of 30s at 921C; 60s at 651C and 45s at 681C. The
terminal elongation step was extended by 10min, and
reactions were cooled to 101C upon completion.
PCR product size (approximately 1300bp) and integrity
were examined by standard agarose gel electrophoresis and
ethidium bromide staining. PCR products were purified
using the Qiagen PCR Purification Kit (Qiagen, Venlo, the
Netherlands) and cloned into the pGEM-T vector, as per the
manufacturer’s protocol (Promega, Madison, WI), for
transformation into E. coli JM109 competent cells using the
pGEM-T Vector System. Fifteen positive transformants were
selected randomly for each rhizosphere soil library, screened
for inserts of proper length by PCR with the vector-encoded
primers T7 (50-TAATACGACTCACTATAGGG-30) and SP6
(50-TATTTAGGTGACACTATAG-30) and sequenced using
the same two primers (Macrogen, Seoul, South Korea).
Given the virtual identity of Verrucomicrobia-specific PCR
denaturing gradient gel electrophoresis community profiles
from F. ovina L. and L. corniculatus L. rhizosphere samples
(Kielak et al., 2008), and the low number of clones per
library, these two libraries were treated as one phylum-
specific library for all further analyses.
Selection of clones affiliated with
the Verrucomicrobia
All clones derived from PCR-based cloning strategies were
screened for potential chimera structures using CHIMERA-
CHECK (http://rdp.cme.msu.edu) and BELLEROPHON (Huber
et al., 2004) software. After the screening of libraries, 26
clones from the Verrucomicrobia-specific libraries and nine
clones previously identified from the general bacterial
libraries (Kielak et al., 2008) were selected for further
analysis. All sequences werealso examined by BLAST (Altschul
et al., 1997) to identify the most closely related database
sequences.
Screening of the metagenomic library for
verrucomicrobial inserts
A total of 28800 clones in the metagenome library were
subjected to a PCR-based screening strategy using a semi-
nested clone pooling strategy (Kielak et al., 2008, 2009).
Briefly, clones were cultured overnight in 96-well plates,
with the contents of four plates (384 clones) combined in a
single plasmid extraction (QIAprep Spin Miniprep Kit;
Qiagen). The resulting mixed templates were subjected to
PCR with the pAF/1378R primers (Edwards et al., 1989),
followed by primers Ver53F/1378R (Stevenson et al., 2004)
as described above. If a PCR product was detected, clone
pooling was reduced to a single plate. Pooling was further
reduced to a single row of a culture plate and eventually a
single clone for subsequent reactions that produced positive
results. Four clones were identified of putative verruco-
microbial origin.
Fosmid clone sequencing and assembly
Four fosmid DNA preparations for each positive clone
(designated clones 106, 118, 184 and 286) were performed
using the Qiagen Midi Kit, and DNA concentrations were
quantified using a Nanodrop (Thermo Scientific) instru-
ment. We then pooled 2.5mg from each fosmid DNA sample
for a total of 10mg DNA. The pooled fosmid DNA was
subjected to pyrosequencing using a 1/2 FLX run on a
Roche/454 Life Sciences Genome Sequencer. The resulting
data were assembled using
v1.1.02.15 (Roche/454 Life Sciences). A total of 32Mb of
sequence data were assembled into 52 (500kb) contigs. All
large contigs were ‘shredded’ into 1000-bp fragments, over-
lapping one another by 100bp. The resulting 192 ‘shreds’
were reassembled with PHRED/PHRAP software, yielding 21
contigs, totalling 137016bp. The four fosmid clones were
also end-sequenced using the vector-targeting primers T7
and T7 RV primer (50-ATGACCATGATTACGCCAAG-30)
to facilitate orientation of the fosmid assignment. Gap
closure by primer walking was completed with the aid of 32
additional custom primers.
NEWBLERASSEMBLY software
ORF characterization
ORFs were assigned using the GLIMMER (http://www.ncbi.
nlm.nih.gov/genomes/MICROBES/glimmer_3.cgi)
FGENESB (http://linux1.softberry.com) software tools. Anno-
tation of the identified ORFs was based on a similarity
search against nonredundant protein databases using BLASTX
and
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
25
Verrucomicrobia: phylogenetic and metagenomic analysis

Page 4

and BLASTP with an expectation-value (e-value) threshold
of oe15. ORFs shorter than 150bp were discarded. tRNA
genes were identified using the
(version1.21; http://lowelab.ucsc.edu/tRNAscan-SE/).
ORFs were classified according to the clusters of ortholo-
gous groups of proteins (COGs) database using COGNITOR
(http://www.ncbi.nlm.nih.gov/COG/old/xognitor.html).
TRNASCAN-SE program
Phylogenetic analyses
All recovered Verrucomicrobium-like 16S rRNA gene se-
quences (four from metagenomic, nine from general bacter-
ialand 26 from phylum-specific libraries) werealigned using
the ARB software package (Ludwig et al., 2004) along with 68
sequences obtained from public databases, and the align-
ment was manually edited based on the secondary structure.
Phylogenetic reconstruction was performed by neighbor-
joining (NJ), maximum likelihood (ML) and Bayesian
inference (BI). MRMODELTEST 2.2 (Nylander, 2004) was used
to determine the most appropriate nucleotide substitution
model. The likelihood ratio test, as well as the Akaike
information criterion, identified the general time-reversible
model with invariable sites and the G-shaped distribution of
substitution rates as the best-fitting model. The NJ and ML
trees were constructed using the PHYLIP 3.67 package (Felsen-
stein, 2005) with the model parameters calculated by
MRMODELTEST. The NJ tree was bootstrapped 1000 times, and
the ML tree 100 times. The Bayesian tree was constructed
using the MRBAYES 3.1.2 program (Ronquist & Huelsenbeck,
2003). Four independent computations, with random start-
ing trees and Markov chains, were run for 3000000 genera-
tions with a sampling frequency of 100 generations. The first
600000 generations were discarded as burn-in.
Comparative genomics and identification of
potential sites of HGT
The gene orders within the recovered verrucomicrobial
fosmid inserts and ORF similarities were compared with
the eight publicly available verrucomicrobial draft genome
sequences by pairwise TBLASTX using BLAST v2.2.18 (ftp://ftp.
ncbi.nih.gov; Altschul et al., 1990) and ACTsoftwarefor gene/
ORF visualization (Teeling et al., 2004; Carver et al., 2005).
The e-value threshold for ORFs compared with these
genomes was 10?30, because the diversity within the Verru-
comicrobia phylum and the low number of available Verru-
comicrobia genomes limit the chance for perfect match
detection.
Searches for GC islands were performed using CPGFINDER
(SoftBerry, http://linux1.softberry.com). For identification
of potential HGT regions, the method described by
Tamames & Moya (2008) was applied to examine tetranu-
cleotide frequencies. Fosmid sequences were edited to
include all ORFs after removal of rRNA operons, and the
resulting sequences werecut into 300-bp fragments and tiled
with an overlap of 290bp for each subsequent fragment.
Pearson’s correlation coefficients of tetranucleotide frequen-
cies for all fragments were calculated using TETRA software
(Teeling et al., 2004), and correlation matrix visualization
was performed in R (R statistical programming environ-
ment) using the ‘GPLOTS’ package.
Sequence accession numbers
The 16S rRNA gene sequences described in this study have
been deposited in GenBank under accession numbers
FJ822616–FJ822655. The accession numbers of fosmid clone
sequences are as follows: clone 106, FJ872372; clone 118,
FJ872373; clone 184, FJ872374; and clone 286, FJ872375.
Results and discussion
Several studies based on sequencing of 16S rRNA genes have
suggested that Verrucomicrobia represent one of the more
dominant phyla in soil environments (Janssen, 2006; Kielak
et al., 2008). They typically represent between 2% and 8% of
the total bacterial community, with the highest densities
observed in the rhizosphere (Kielak et al., 2008). This
relatively high abundance suggests that members of this
phylum may play significant roles in the soil environment
and have important interactions with plants. This study was
therefore geared towards recovering information regarding
the phylogenetic diversity of Verrucomicrobia in soil, as well
as providing tools for gainingaccessto genomic information
from this potentially important phylum.
Phylogenetic analyses (NJ, ML and BI) of 16S rRNA gene
sequences recovered from three independent cloning ap-
proaches showed strong phylogenetic support for the estab-
lished clustering systemwithin
(Hugenholtz et al., 1998; Schlesner et al., 2006) (NJ tree in
Fig. 1; ML and BI trees not shown). Subdivision 4 was found
to be the basal group in all phylogenetic trees, followed by
subdivision 3. Subdivisions 1 and 2 appear to be sister
groups (Fig. 1).
In agreement with other studies (Janssen, 2006), we
identified subdivision 2 (Spartobacteria) to be the most
dominantly detected cluster within all libraries, representing
18 of 27, six of eight and three of four clones for the
Verrucomicrobia phylum-specific, general bacterial and
metagenomic libraries, respectively. Subdivision 1 (Verruco-
microbiae) was also represented in all libraries with one
identified metagenomic clone, two clones in the bacterial-
specific library and four in the Verrucomicrobia-specific
library. Subdivisions 3 and 4 were relatively rare and were
only identified in the Verrucomicrobia phylum-specific
libraries (three clones and one clone for subdivisions 3 and
4, respectively). Although all three cloning strategies yielded
highly similar results, it is premature to state that the three
theVerrucomicrobia
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
26
A. Kielak et al.

Page 5

Fig. 1. NJ tree of the phylum Verrucomicrobia, highlighting the sequences recovered in this study. Names in italics indicate sequences retrieved from
GenBank, red, general bacterial library; blue, metagenome library; green, phylum specific library; underlined, genomes being sequenced. Numbers near
nodes indicate bootstrap values. Sequences from the phylum Planctomycetes were used as an outgroup.
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
27
Verrucomicrobia: phylogenetic and metagenomic analysis

Page 6

methods are truly similar in their recovery of sequences
from across the breadth of the phylum, given the low
number of clones examined. No clones closely related to
subdivisions 5, 6 or 7 were identified in any of the libraries,
and these subdivisions were therefore excluded from the
presented phylogenetic analyses.
The number of clones from the metagenomic library
determined to contain genomic inserts of verrucomicrobial
origin was rather low (four fosmid clones), but within the
range (three to nine positive clones) that would be expected,
given the library size (28800 clones), size of the insert
(32kb), estimated average genome size (6000kb), 16S rRNA
gene copies per genome (1–3) and the relative abundance of
Verrucomicrobia within the soil community sampled
(4.9–5.7%) (Kielak et al., 2008). The average insert size per
positive fosmid clone was 32?1.5kb (Table 1), yielding a
total of approximately 130kb of Verrucomicrobia genetic
information. All four genome fragments revealed typical
bacterial rrn operon organization, with 16S rRNA gene,
followed by tRNAIle, the 23S rRNA gene and then the 5S
rRNA gene. In two clones (118 and 286), the 16S rRNA and
tRNAIlegenes were separated by tRNAAla.
The total G1C content of the fosmid inserts ranged from
60% to 62%, with a range from 61% to 64% for coding
regions. These values are comparable to those found for
verrucomicrobial genomes for which drafts are available,
with G1C contents ranging from 52% for Ellin514 (http://
www.jgi.doe.gov) to 65% for O. terrae PB90-1 (NC010571).
Methylacidiphilum infernorum V4 (NC010794) is a marked
exception, with a G1C content of only 45%. Pearson’s
correlation coefficients for the four verrucomicrobial fosmid
inserts were also calculated based on tetranucleotide usage
patterns. Using this measure, clones 118 and 106 were most
similar (r=0.78), followed by the clone pair 118 and 286
(r=0.75). The lowest correlation was between clones 106
and 286 (r=0.62), while clone 184 was similarly distant to
the other three clones (r=0.64). The divergence in 16S
rRNA genes followed a similar pattern, with the lowest
dissimilarity (12%) between clones 118 and 106 and the
highest dissimilarity (20%) between clones 184 and 286
(Table 2).
The numbers of putative protein-encoding genes assigned
for each clone varied between 22 ORFs (clone 184) and 28
ORFs (clone 118) (Table 1). All ORFs were compared with
the sequences available in public databases with special
emphasis on the sequences available for the eighth Verruco-
microbia genome project currently underway. Atotal of 10%
of the predicted genes, found predominantly within clones
118 and 286, showed low or no similarity to sequences
deposited in public databases (Supporting Information,
Tables S1 and S2). The number of ORFs with best matches
to verrucomicrobial sequences ranged between 11 out of 22
Table 1. Verrucomicrobial genome fragment features
Fosmid clone ID
286118106184
Number of bases (kb)
rRNA
tRNA
Number of ORFs
Coding percentage
GC percentage in coding regions
Overall GC percentage
32226
3 (16S, 5S, 23S)
2
22
82
63.51
60.42
32436
3 (16S, 5S, 23S)
2
28
90
63.69
61.62
33568
3(16S, 5S, 23S)
1
26
85
61.29
59.65
29919
3(16S, 5S, 23S)
1
22
95
63.84
61.54
Table 2. Correlation coefficients between Verrucomicrobia 106, 118, 184 and 286 genome fragments (including and excluding the ribosomal operon)
based on tetranucleotide usage patterns and distance matrix of 16S rRNA gene sequences
Fosmid
clone ID
Fosmid clone ID
106118184
Pearson’s correlation
16S rRNA gene
distance
Pearson’s correlation
16S rRNA gene
distance
Pearson’s correlation
16S rRNA gene
distance
With RNA
operon
Without RNA
operon
With RNA
operon
Without RNA
operon
With RNA
operon
Without RNA
operon
118
184
286
0.89
0.81
0.86
0.78
0.64
0.62
0.12
0.18
0.14
0.82
0.88
0.64
0.75
0.17
0.120.800.64 0.20
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
28
A. Kielak et al.

Page 7

ribosomal operon
ribosomal operon
ribosomal operon
ribosomal operon
(a)
(b)
(d)
(c)
Fig. 2. Tetranucleotide frequency distributions of fosmid clones of verrucomicrobial origin. (a) Fosmid 106; (b) fosmid 184; (c) fosmid 286 and (d) fosmid 118.
Color indicates the tetranucleotide distribution relative to the full fragment lengths, with red representing typical (high correlation) and green atypical (low
correlation)regions.ThepositionsofORFsineachofthefosmidsareindicatedbyarrowsabovethediagrams.NumbersinarrowscorrespondtotheCDSnumber
offosmids118(TableS1); 284(TableS2);184(TableS3)and106(Table S4).Theribosomaloperonillustratedinthefigureineachfosmidwasexcludedfromthe
analysis.
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
29
Verrucomicrobia: phylogenetic and metagenomic analysis

Page 8

for clone 184 (Table S3) and 23 out of 26 for clone 106
(Table S4). For clone 184, the genome of C. flavus Ellin428
provided the best match for four of the 11 ORFs with
verrucomicrobial best matches, whereas all 23 of the ORFs
from clone 106 matched this genome. However, the 16S
rRNA gene from clone 106 showed lower similarity to the
Ellin428 strain compared with clone 118 (96% identity),
which also showed several best BLAST hit matches with
Proteobacteria, other verrucomicrobial strains and even
Planctomycetes (Table S1).
Assignment of ORFS to COG functional categories re-
vealed that all four clones encoded proteins involved in
information storage/processing, metabolism and cellular
processes/signalling, with clone 106 having the greatest
proportion of COG hits belonging to metabolism or cellular
processes and signalling. A large number of ORFs (between
nine and 15 per clone) could not be classified into any COG
category (Table S5).
The three fosmid clones affiliated withsubdivision 2 (106,
118 and 286) were compared with the draft genome of
Ellin428 by TBLASTX to examine patterns of gene synteny and
homology. Homologous protein-encoding regions identi-
fied in these comparisons were no longer than 5kb and
contained a maximum of four ORFs (Fig. S1). Clone 184,
affiliated with subgroup 1, contained no regions of con-
served gene order compared with the other clones or
available genomes. Thus, even though some of the recovered
clones were very close to sequenced verrucomicrobial
strains, based on 16S rRNA gene comparisons, large levels
of divergence appear at the level of gene order and origin.
Together, these observations highlight the fact that available
sequence information represents but a small fraction of the
depth of genomic information across the phylum as awhole.
Oligonucleotide frequencies within DNA sequences fol-
low species-specific patterns (Nei & Gojobori, 1986; Sueoka,
1988; Karlin et al., 1998; Bellgard et al., 2001; Bentley &
Parkhill, 2004). Such patterns carry an innate phylogenetic
signal, and aberrant patterns of oligonucleotide usage may
thus be indicative of regions recently acquired from foreign
genetic sources (Pride et al., 2003; Abe et al., 2005). Upon
analysis of tetranucleotide usage patterns (Tamames &
Moya, 2008), clones 106 (Fig. 2a) and 184 (Fig. 2b) each
appeared to contain one region of potential horizontal
transfer. Several regions of aberrant tetranucleotide usage
were observed in fosmid 286 (Fig. 2c), while no such regions
were found in fosmid 118 (Fig. 2d). The ORFs identified
within the regions of aberrant tetranucleotide usage typi-
cally exhibited only low levels of similarity to available
database entries (29% identity, e-value=7?10?45for clone
106; 43% identity, e-value=0.002 for clone 184; and 32%
identity, e-value=7.9 for clone 286). Interestingly, with the
exception of fosmid 106, these regions corresponded with
ORFsthat did not yield verrucomicrobial BLASTmatches. The
largest potential HGT region detected (in clone 286) con-
tained several genes (UvrB/UvrC protein, putative peptido-
glycan peptidase and ankyrin) flanked by insertion sequence
elements, an organization typical of mobile DNA regions. It
is noteworthy that the putative ankyrin repeat belongs to a
functionally diverse group of eukaryotic proteins, for which
HGT has been implicated in its transfer to prokaryotic
genomes (Bork, 1993).
The cultivation-independent approach used in this study
allowed for genomic information to be recovered specifically
from the Verrucomicrobia, but a large number of clones was
required for the identification of arelativelysmallnumberof
verrucomicrobial clones. Despite the recent advances in
shotgun metagenomic analysis, it remains clear that meta-
genomes derived from highly diverse soil environments will
not yield large genomic scaffolds (Tringe et al., 2005;
Kowalchuk et al., 2007). Thus, directed approaches, such as
the one used here, will continue to be necessary for the
recovery of extended regions of verrucomicrobial genomic
information from soil. Although this study has provided
some insight into the genome content and genomic proper-
ties of some members of the phylum Verrucomicrobia,
defining the ecological and evolutionary properties from
relatively small genomic regions (32kb) remains proble-
matic. Much larger sample sizes will be necessary in order to
retrieve sufficient information to make ecological inferences,
and novel high-throughput screening methods will need to
be used to achieve this goal (Park et al., 2008). Also, the
approach utilized here is currently restricted to regions
adjacent to rRNA operons. Now that verrucomicrobial
genomic sequences are becoming available, detecting verru-
comicrobial genomic fragments could be achieved by end-
sequencing of large insert libraries followed by sequence co-
mparison and binning. Particularly exciting are methods of
single-cell sequencing and microcultivation (Abulencia
et al., 2006; Ingham et al., 2007; Podar et al., 2007), which
could provide novel windows into the genomic diversity of
this important bacterial phylum.
Acknowledgements
We thank Stephanie Malfatti and Mari Christensen from
Joint Genome Institute (JGI) for help with DNA sequencing
and assembling. Ellin514 sequence data were produced by
the US Department of EnergyJGI (http://www.jgi.doe.gov/).
This work was funded by the BSIK programme ‘Ecoge-
nomics’ (http://www.ecogenomics.nl/). This is Netherlands
Institute of Ecology (NIOO-KNAW) publication 4606.
References
Abe T, Sugawara H, Kinouchi M, Kanaya S & Ikemura T (2005)
Novel phylogenetic studies of genomic sequence fragments
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
30
A. Kielak et al.

Page 9

derived from uncultured microbe mixtures in environmental
and clinical samples. DNA Res 12: 281–290.
Abulencia CB, Wyborski DL, Garcia JA et al. (2006)
Environmental whole-genome amplification to access
microbial populations in contaminated sediments. Appl
Environ Microb 72: 3291–3301.
Altschul SF, Gish W, Miller W, Myers EW & Lipman DJ (1990)
Basic local alignment search tool. J Mol Biol 215: 403–410.
Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller
W & Lipman DJ (1997) Gapped BLASTand PSI-BLAST: a new
generation of protein database search programs. Nucleic Acids
Res 25: 3389–3402.
Bellgard M, Schibeci D, Trifonov E & Gojobori T (2001) Early
detection of G1C differences in bacterial species inferred from
the comparative analysis of the two completely sequenced
Helicobacter pylori strains. J Mol Evol 53: 465–468.
Bentley SD & Parkhill J (2004) Comparative genomic structure of
prokaryotes. Annu Rev Genet 38: 771–792.
Bezemer TM & van der Putten WH (2007) Ecology – diversity
and stability in plant communities. Nature 446: E6–E7.
Bezemer TM, Lawson CS, Hedlund K, Edwards AR, Brook AJ,
Igual JM, Mortimer SR & Van der Putten WH (2006) Plant
species and functional group effects on abiotic and microbial
soil properties and plant–soil feedback responses in two
grasslands. J Ecol 94: 893–904.
Bork P (1993) Hundreds of Ankyrin-like repeats in functionally
diverse proteins – mobile modules that cross phyla
horizontally. Proteins 17: 363–374.
Buckley DH & Schmidt TM (2001) Environmental factors
influencing the distribution of rRNA from Verrucomicrobia in
soil. FEMS Microbiol Ecol 35: 105–112.
Carver TJ, Rutherford KM, Berriman M, Rajandream MA, Barrell
BG & Parkhill J (2005) ACT: the Artemis comparison tool.
Bioinformatics 21: 3422–3423.
Chin KJ, Liesack W & Janssen PH (2001) Opitutus terrae gen.
nov., sp. nov., to accommodate novel strains of the division
‘Verrucomicrobia’ isolated from rice paddy soil. Int J Syst Evol
Micr 51: 1965–1968.
Cole JR, Chai B, Farris RJ, Wang Q, Kulam SA, McGarrell DM,
Garrity GM & Tiedje JM (2005) The Ribosomal Database
Project (RDP-II): sequences and tools for high-throughput
rRNA analysis. Nucleic Acids Res 33: D294–D296.
Derrien M, Vaughan EE, Plugge CM & de Vos WM (2004)
Akkermansia muciniphila gen. nov., sp. nov., a human
intestinal mucin-degrading bacterium. Int J Syst Evol Micr 54:
1469–1476.
Dunfield PF, YuryevA, Senin Petal. (2007) Methane oxidation by
an extremely acidophilic bacterium of the phylum
Verrucomicrobia. Nature 450: 879–870.
Edwards U, Rogall T, Blocker H, Emde M & Bottger EC (1989)
Isolation and direct complete nucleotide determination of
entire genes – characterization of a gene coding for 16S-
ribosomal RNA. Nucleic Acids Res 17: 7843–7853.
Felsenstein J (2005) PHYLIP (Phylogeny Inference Package)
Version 3.6. Distributed by the Author, Departmentof Genome
Sciences, University of Washington, Seattle, WA.
Garrity GM & Holt JG (2001) The Road Map to the Manual,
Vol. 1. pp. 119–166. Springer-Verlag, New York, NY.
Hou SB, Makarova KS, Saw JHWet al. (2008) Complete genome
sequence of the extremely acidophilic methanotroph isolate
V4, Methylacidiphilum infernorum, a representative of the
bacterial phylum Verrucomicrobia. Biol Direct 3: 26.
Huber T, Faulkner G & Hugenholtz P (2004) Bellerophon: a
program to detect chimeric sequences in multiple sequence
alignments. Bioinformatics 20: 2317–2319.
Hugenholtz P, Goebel BM & Pace NR (1998) Impact of culture-
independent studies on the emerging phylogenetic view of
bacterial diversity. J Bacteriol 180: 4765–4774.
Ingham CJ, Sprenkels A et al. (2007) The micro-Petri dish, a
million-well growth chip for the culture and high-throughput
screening of microorganisms. P Natl Acad Sci USA 104:
18217–18222.
Janssen PH (2006) Identifying the dominant soil bacterial taxa in
libraries of 16S rRNA and 16S rRNA genes. Appl Environ
Microb 72: 1719–1728.
Janssen PH, Schuhmann A, Morschel E & Rainey FA (1997)
Novel anaerobic ultramicrobacteria belonging to the
Verrucomicrobiales lineage of bacterial descent isolated by
dilution culture from anoxic rice paddy soil. Appl Environ
Microb 63: 1382–1388.
Janssen PH, Yates PS, Grinton BE, Taylor PM & Sait M (2002)
Improved culturability of soil bacteria and isolation in pure
culture of novel members of the divisions Acidobacteria,
Actinobacteria, Proteobacteria, and Verrucomicrobia. Appl
Environ Microb 68: 2391–2396.
Karlin S, Campbell AM & Mrazek J (1998) Comparative DNA
analysis across diverse genomes. Annu Rev Genet 32: 185–225.
Kielak A, Pijl AS, van Veen JA & Kowalchuk GA (2008)
Differences in vegetation composition and plant species
identity lead to only minor changes in soil-borne microbial
communities in a former arable field. FEMS Microbiol Ecol 63:
372–382.
Kielak A, Pijl AS, van Veen JA & Kowalchuk GA (2009)
Phylogenetic diversity of Acidobacteria in a former agricultural
soil. ISME J 3: 378–382.
Kowalchuk GA, Speksnijder AGCL, Zhang K, Goodman RM &
van Veen JA (2007) Finding the needles in the meta-genome
haystack. Microb Ecol 53: 475–485.
Ludwig W, Strunk O, Westram R et al. (2004) ARB: a software
environment for sequence data. Nucleic Acids Res 32:
1363–1371.
Nei M & Gojobori T (1986) Simple methods for estimating the
numbers of synonymous and nonsynonymous nucleotide
substitutions. Mol Biol Evol 3: 418–426.
Nylander JAA (2004) MrModeltest v2. Program. Distributed by
the Author, Evolutionary Biology Centre, Uppsala University,
Uppsala, Sweden.
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
31
Verrucomicrobia: phylogenetic and metagenomic analysis

Page 10

Park SJ, Kang CH et al. (2008) Metagenome microarray for
screening of fosmid clones containing specific genes. FEMS
Microbiol Lett 284: 28–34.
Pearce DA, van der Gast CJ, Lawley B & Ellis-Evans JC (2003)
Bacterioplankton community diversity in a maritime
Antarctic lake, determined by culture-dependent and
culture-independent techniques. FEMS Microbiol Ecol 45:
59–70.
Podar M, Abulencia CB, Walcher M, Hutchison D, Zengler K,
Garcia JA, Holland T, Cotton D, Hauser L & Keller M (2007)
Targeted access to the genomes of low-abundance organisms
in complex microbial communities. Appl Environ Microb 73:
3205–3214.
Pol A, Heijmans K, Harhangi HR, Tedesco D, Jetten MSM & den
Camp H (2007) Methanotrophy below pH1 by a new
Verrucomicrobia species. Nature 450: 874–870.
Pride DT, Meinersmann RJ, Wassenaar TM & Blaser MJ (2003)
Evolutionary implications of microbial genome
tetranucleotide frequency biases. Genome Res 13: 145–158.
Pruesse E, Quast C, Knittel K, Fuchs B, Ludwig W, Peplies J &
Gl¨ ockner FO (2007) SILVA: a comprehensive online resource
for quality checked and aligned ribosomal RNA sequence data
compatible with ARB. Nucleic Acids Res 35: 7188–7196.
Ronquist F & Huelsenbeck JP (2003) MrBayes 3: Bayesian
phylogenetic inference under mixed models. Bioinformatics 19:
1572–1574.
Sakai T, Ishizuka K & Kato I (2003) Isolation and characterization
of a fucoidan-degrading marine bacterium. Mar Biotechnol 5:
409–416.
Sangwan P, Chen XL, Hugenholtz P & Janssen PH (2004)
Chthoniobacter flavus gen. nov., sp nov., the first pure-culture
representative of subdivision two, Spartobacteria classis nov.,
of the phylum Verrucomicrobia. Appl Environ Microb 70:
5875–5881.
Sangwan P, Kovac S, Davis KER, Sait M & Janssen PH (2005)
Detection and cultivation of soil Verrucomicrobia. Appl
Environ Microb 71: 8402–8410.
Scheuermayer M, Gulder TAM, Bringmann G & Hentschel U
(2006) Rubritalea marina gen. nov., sp. nov., a marine
representative of the phylum ‘Verrucomicrobia’, isolated from a
sponge (Porifera). Int J Syst Evol Micr 56: 2119–2124.
Schlesner H (1987) Verrucomicrobium spinosum gen. nov., sp.
nov. – a fimbriated prosthecate bacterium. Syst Appl Microbiol
10: 54–56.
Schlesner H, Jenkins C & Staley JT (2006) The phylum
Verrucomicrobia: a phylogenetically heterogeneous bacterial
group. Prokaryotes 7: 881–896.
Shinzato N, Muramatsu M, Matsui T & Watanabe Y (2005)
Molecular phylogenetic diversity of the bacterial community
in the gut of the termite Coptotermes formosanus. Biosci Biotech
Bioch 69: 1145–1155.
Staley JT, Bont J & Jonge KD (1976) Prosthecobacter fusiformis
nov. gen. et sp., the fusiform caulobacter. Antonie van
Leeuwenhoek 42: 333–342.
Stevenson BS, Eichorst SA, Wertz JT, Schmidt TM & Breznak JA
(2004) New strategies for cultivation and detection of
previously uncultured microbes. Appl Environ Microb 70:
4748–4755.
Sueoka N (1988) Directional mutation pressure and neutral
molecular evolution. P Natl Acad Sci USA 85: 2653–2657.
Tamames J & Moya A (2008) Estimating the extent of horizontal
gene transfer in metagenomic sequences. BMC Genomics 9:
136.
Teeling H, Waldmann J, Lombardot T, Bauer M & Gl¨ ockner FO
(2004) TETRA: a web-service and a stand-alone program for
the analysis and comparison of tetranucleotide usage patterns
in DNA sequences. BMC Bioinformatics 5: 163.
Tringe SG, von Mering C, Kobayashi A et al. (2005) Comparative
metagenomics of microbial communities. Science 22: 554–557.
Vandekerckhove TTM, Willems A, Gillis M & Coomans A (2000)
Occurrence of novel verrucomicrobial species, endosymbiotic
and associated with parthenogenesis in Xiphinema
americanum-group species (Nematoda, Longidoridae). Int J
Syst Evol Micr 50: 2197–2205.
Van der Putten WH, Mortimer SR, Hedlund Ket al. (2000) Plant
species diversity as a driver of early succession in abandoned
fields: a multi-site approach. Oecologia 124: 91–99.
Wang M, Ahrne S, Jeppsson B & Molin G (2005) Comparison of
bacterial diversity along the human intestinal tract by direct
cloning and sequencing of 16S rRNA genes. FEMS Microbiol
Ecol 54: 219–231.
Yoon J, Oku N, Matsuda S, Kasai H & Yokota A (2007a)
Pelagicoccus croceus sp. nov., a novel marine member of the
family Puniceicoccaceae within the phylum ‘Verrucomicrobia’
isolated from seagrass. Int J Syst Evol Micr 57: 2874–2880.
Yoon J, Yasumoto-Hirose M, Matsuo Y, Nozawa M, Matsuda S,
Kasai H & Yokota A (2007b) Pelagicoccus mobilis gen. nov., sp.
nov., Pelagicoccus albus sp. nov. and Pelagicoccus litoralis sp.
nov., three novel members of subdivision 4 within the phylum
Verrucomicrobia, isolated from seawater by in situ cultivation.
Int J Syst Evol Micr 57: 1377–1385.
Zoetendal EG, Plugge CM, AkkermansADL & de Vos WM (2003)
Victivallis vadensis gen. nov., sp. nov., a sugar-fermenting
anaerobe from human faeces. Int J Syst Evol Micr 53: 211–215.
SupportingInformation
Additional Supporting Information may be found in the
online version of this article:
Fig. S1. Comparative genomic organization of fosmids in
relation to Ellin428.
Table S1. Identified ribosomal genes and predicted ORF of
the verrucomicrobial genomic fragment 118.
Table S2. Identified ribosomal genes and predicted ORF of
the verrucomicrobial genomic fragment 286.
Table S3. Identified ribosomal genes and predicted ORF of
the verrucomicrobial genomic fragment 184.
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
32
A. Kielak et al.

Page 11

Table S4. Identified ribosomal genes and predicted ORF of
the verrucomicrobial genomic fragment 106.
Table S5. Number of ORFs of Verrucomicrobia genomic
fragments (106, 118, 184, and 286) assigned to different
COG functional categories.
Please note: Wiley-Blackwell is not responsible for the
content or functionality of any supporting materials sup-
plied by the authors. Any queries (other than missing
material) should be directed to the corresponding author
for the article.
FEMS Microbiol Ecol 71 (2010) 23–33
c ?2009 NIDO-KNAW
Journal compilation c ?2009 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd.
33
Verrucomicrobia: phylogenetic and metagenomic analysis