Microbiology (2004), 150, 28892898 DOI 10.1099/mic.0.

27250-0

Functional and phylogenetic analysis of a

plant-inducible oligoribonuclease (orn) gene
from an indigenous Pseudomonas plasmid
Xue-Xian Zhang,1,2,3 Andrew K. Lilley,1 Mark J. Bailey1
and Paul B. Rainey2,3
1
Correspondence Centre for Ecology and Hydrology NERC, Mansfield Road, Oxford OX1 3SR, UK
Xue-Xian Zhang 2
Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK
xx.zhang@auckland.ac.nz 3
School of Biological Sciences, University of Auckland, Private Bag 92019, Auckland,
New Zealand

Application of a promoter-trapping strategy to identify plant-inducible genes carried

on an indigenous Pseudomonas plasmid, pQBR103, revealed the presence of a putative
oligoribonuclease (orn) gene that encodes a highly conserved 39 to 59 exoribonuclease specific
for small oligoribonucleotides. The deduced amino acid sequence of the plasmid-derived orn
(ornpl) showed three conserved motifs characteristic of Orn from both prokaryotes and eukaryotes.
Deletion of ornpl generated no observable phenotype, but inactivation of the chromosomal
copy caused slow growth in Pseudomonas putida KT2440. This defect was fully restored by
complementation with orn from Escherichia coli (ornE.coli). Plasmid-derived ornpl was capable
of partially complementing the P. putida orn mutant, demonstrating functionality of ornpl.
Phylogenetic analysis showed that plasmid-encoded Orn was distinct from Orn encoded by
the chromosome of proteobacteria. A survey of ornpl from related Pseudomonas plasmids
Received 16 April 2004 showed a sporadic distribution but no sequence diversity. These data suggest that the ornpl
Revised 4 June 2004 was acquired by pQBR103 in a single gene-transfer event: the donor is unknown, but is
Accepted 9 June 2004 unlikely to be a member of the Proteobacteria.

INTRODUCTION genes that contribute to the ecological performance of

bacteria in natural environments is challenging.
The genus Pseudomonas includes several species of plant-
colonizing bacteria (Palleroni, 1984). Strains of some Analysis of the phytosphere microflora of sugar beet at a
species can cause disease (e.g. Pseudomonas syringae), field site in Oxford (UK) in the mid-1990s revealed an
whereas others, for example strains of Pseudomonas fluo- abundance of self-transmissible mercury-resistance plas-
rescens, can promote plant growth (reviewed by Bloemberg mids that move freely within the resident Pseudomonas
& Lugtenberg, 2001). community (Lilley et al., 1996; Lilley & Bailey, 1997a). These
plasmids have molecular sizes ranging from 60 to 425 kb
The genomes of plant-associated Pseudomonas species and have been grouped into five distinct categories (groups I
frequently contain a significant component of extra- to V) by RFLP analysis. In field-grown sugar beet these
chromosomal DNA in the form of one or more plasmids plasmids are most commonly encountered in bacteria
(Vivian et al., 2001; Thomas, 2000); the transfer of these isolated from mature plants (Lilley & Bailey, 1997a). One
plasmids within and between bacterial populations is particular group I plasmid, pQBR103, has been the subject
thought to facilitate adaptation to novel environments of extensive analysis. Interestingly, carriage of this plasmid
(Hartl et al., 1984). Despite their ecological importance, reduces fitness in bacteria colonizing sugar beet seedlings,
knowledge of the traits encoded on naturally occurring but not bacteria colonizing mature plants (Lilley & Bailey,
plasmids is limited. Such plasmids rarely encode selectable 1997b). These data suggest a significant genotype-by-
phenotypes, and the identification of plasmid-encoded environment interaction and the existence of pQBR103-
encoded traits that confer some selectable advantage to
Abbreviations: DAP, diaminopimelic acid; Orn, oligoribonuclease; bacteria colonizing mature sugar beet plants.
CFC, cetrimide, fucidin and cephalosporin; IVET, in vivo expression
technology. In a parallel study (unpublished) we used a promoter
The GenBank/EMBL/DDBJ accession number for the sequence trapping strategy (IVET) to isolate pQBR103-encoded
reported in this paper is AJ617292. genes of possible relevance to plant colonization. The
Downloaded from www.microbiologyresearch.org by
0002-7250 G 2004 SGM Printed in Great Britain 2889
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
X.-X. Zhang and others

promoter-trap, based on a promoterless copy of 9dapB, Table 1. Distribution of ornpl among environmental plasmids
a gene encoding 2,3-dihydrodipicolinate reductase and isolated from the sugar beet phytosphere
required for the biosynthesis of diaminopimelic acid (DAP)
and lysine (Gal et al., 2003), led to the isolation of 37 plant- Plasmid Group* ornpl geneD Molecular size (kb)
induced fusions. These fusions are transcriptionally silent pQBR2 I Y 347
in minimal medium, but active in the phytosphere of sugar pQBR3 I Y 353
beet. The plant-induced genes are of particular interest pQBR10 I Y 287
because they are predicted to contribute to the fitness of pQBR11 I Y 294
the host bacteria in planta a prediction recently confirmed pQBR19 I N 285
for a locus encoding an acetylated cellulose polymer (Gal pQBR44 I N 130
et al., 2003; Spiers et al., 2003). Sequence analysis of the pQBR103 I Y 425
37 fusions revealed few with any similarity to sequences pQBR110 I N >275
deposited in public DNA or protein sequence databases. pQBR23 II N 305
However, one fusion, pIVETD5, is to an ORF (orf6) adjacent pQBR26 II N 383
to a gene (orn) predicted to encode oligoribonuclease, which pQBR55 III N 149
is an essential gene in Escherichia coli (Ghosh & Deutscher, pQBR57 IV N 261
1999). Orn hydrolyses small oligoribonucleotides and is
involved in the final step of mRNA degradation (Zhang *Plasmid group as indicated in Lilley & Bailey (1997a) on the basis
et al., 1998). Although putative Orn proteins have been of restriction fragment pattern.
identified in the sequenced Pseudomonas genomes, none DY, ornpl present; N, ornpl absent. The ornpl genes carried on different
of their functions have been characterized. Here we report plasmids have identical nucleotide sequences.
the comparative and functional analysis of orn genes carried
on an indigenous Pseudomonas plasmid and the chromo-
some of P. putida KT2440 and show that the plasmid- with an annealing temperature of 58 uC according to the manufac-
derived orn gene is plant inducible and widespread among turers protocol. The template DNA was either purified genomic
group I plasmids. DNA or whole-cell suspensions. All oligonucleotide primers were
obtained from MWG Biotech. Plasmid conjugation was carried out
using the filter plate mating method as previously described (Lilley
METHODS et al., 1996) except that a growing temperature of 28 uC was used.
Bacterial strains, plasmids and growing conditions. P. fluo-
rescens SBW25 was isolated from field-grown sugar beet at the Sequencing and deletion analysis of ornpl. Plasmid pIVETD5
University of Oxford farm, Wytham, Oxford, UK (Bailey et al., is one of the original IVET clones isolated from pQBR103 contain-
1995). Plasmid pQBR103 is representative of a group of large ing genes that show elevated levels of expression on plant surfaces
plasmids conferring resistance to mercuric chloride and acquired by (unpublished). The insert fragment of pIVETD5 was initially
genetically marked P. fluorescens (SBW25EeZY6KX) during the sequenced from both ends using primers to the 59 end of the dapB
course of field release experiments at Wytham (Lilley & Bailey, gene (Pdap) and the 59 end of the bla gene (Pbla) in the pIVETD
1997a). All the indigenous plasmids (prefixed with pQBR) used in vector (Gal et al., 2003). The whole 2?3 kb insert was then
this study were exogenously isolated from plant-colonizing Pseudo- sequenced on both strands by primer walking, whereby subsequent
monas (Lilley et al., 1996) and are listed in Table 1. Bacterial strains primers were designed based on the newly derived sequence data. A
and other recombinant plasmids used in this study are listed in random shotgun genomic library of pQBR103, which was composed
Table 2. of 2000 single colonies containing 13 kb DNA fragments at the
SmaI site of pUC18, was sequenced (unpublished data). Sequence
Pseudomonas and E. coli strains were routinely grown in LuriaBertani alignment of pIVETD5 with the shotgun sequence database of
medium (LB) at 28 uC and 37 uC, respectively. Minimal M9 medium pQBR103 identified five clones that contained overlapping DNA
(Sambrook et al., 1989) was also used for the propagation of Pseudo- fragments from the ornpl region (Fig. 1a). Sequence alignment was
monas. Pseudomonas transconjugants containing pQBR plasmids performed using the Sequencher software (Gene Codes Corpora-
were grown in LB supplemented with HgCl2 (27 mg ml21). When tion). The sequence gaps were then filled in by primer walking using
necessary, antibiotics were supplemented at the following concentra- the five pUC18 clones as sequencing templates. Details of the oligo-
tions (mg ml21): rifampicin, 50; tetracycline, 10; kanamycin, 100; nucleotides used in DNA sequencing are available on request. A
nalidixic acid, 50; ampicillin, 50; spectinomycin, 100. The IVET fusion 6773 bp sequence was finally obtained and it has been deposited in
strains based on P. fluorescens SBW25DdapB were cultivated in GenBank (accession number AJ617292).
minimal M9 medium with addition of lysine (60 mg ml21) and DAP
(800 mg ml21). Half-strength CFC (cetrimide, fucidin and cephalo- In order to construct an ornpl deletion mutant, shotgun plasmid
sporin; Oxoid) was used to select P. fluorescens recovered from the p9d5 (Fig. 1a) was digested with XhoI to remove the 378 bp ornpl
plant. fragment and transformed into E. coli DH5a after ligation. The
resulting plasmid p9d5X contains a 1?97 kb insert composed of two
DNA manipulations. General recombinant DNA techniques were DNA fragments of similar lengths flanking the ornpl deletion. This
performed according to standard protocols (Sambrook et al., 1989). 1?97 kb insert was then PCR-amplified using general M13 forward
Restriction enzymes and T4 DNA ligase were obtained from New and reverse primers incorporating BglII sites and cloned into the
England Biolabs (NEB). DNA was sequenced using Big-Dye termi- integration vector pIVETD. The insertion direction was determined
nators (Applied Biosystems) on an automated DNA sequencer, by restriction analysis. The recombinant plasmid containing the
model 310 (Perkin Elmer) following the manufacturers instructions. putative ORFs including the undeleted 200 bp of ornpl in the same
PCR reactions were performed using Taq DNA polymerase (Qiagen) transcriptional direction as the promoterless 9dapB was designated
Downloaded from www.microbiologyresearch.org by
2890 Microbiology 150
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
 Oligoribonucleases from a Pseudomonas plasmid

Table 2. Bacterial strains and plasmids

Strain or plasmid Relevant characteristics Source/reference

Strains
P. fluorescens
SBW25 Wild-type Bailey et al. (1995)
SBW25DdapB DAP/lysine auxotroph of SBW25 Gal et al. (2003)
PBR424 orf6 : : 9dapB9lacZ IVET fusion strain This study
PBR442 ornpl : : 9dapB9lacZ IVET fusion strain This study
P. putida
KT2440 plasmid-cured derivative of mt-2 Nakazawa (2002)
UWC1 Rifr derivative of KT2440 McClure et al. (1990)
PBR447 KT2440 ornP.putida : : VTc, Rif r Tcr This study
E. coli DH5alpir supE44 DlacU169 hsdR17 recA1 endA1 gyrA96 thi-1 relA1 lpir M. Mahan*
Plasmids
pUC18 Cloning vector, Apr Sambrook et al. (1989)
p9d5 2?35 kb ornpl fragment cloned in pUC18 This study
p9d5X Deletion derivative of p9d5 containing 1?97 kb insert This study
pIVETD DAP-based IVET vector containing promoterless 9dapB9lacZ, Tcr Apr Gal et al. (2003)
pIVETD5 orf6 : : 9dapB9lacZ fusion plasmid of pIVETD, Tcr Apr This study
pIVETD9 ornpl : : 9dapB9lacZ fusion plasmid of pIVETD, Tcr Apr This study
pIVETD10 pIVETD containing 1?97 kb insert from p9d5X, Tcr Apr This study
pIVET10 Delivery plasmid for ornpl deletion; pIVETD10 without the 9dapB This study
fragment, Tcr Apr
pIVET11 pIVETD without 9dapB and the gene for Tc resistance, 9lacZ Apr This study
pIVET11-1 Delivery plasmid for ornP.putida inactivation; pIVET11 containing This study
ornP.putida : : VTc; bla, 9lacZ
pCR2.1 PCR cloning vector Invitrogen
pRK2013 Helper plasmid, Tra+ Kmr Ditta et al. (1980)
pHP45V Source of the omega fragment, Tcr Fellay et al. (1987)
pME6010 Broad-host-range vector, Tcr Heeb et al. (2000)
pME6010V pME6010 : : VSpe, Sper This study
pME6010 : : ornP.putidaV ornP.putida construct for complementation This study
pME6010 : : ornplV ornpl construct for complementation This study
pME6010 : : ornE.coliV ornE.coli construct for complementation This study
pQBR103Dorn ornpl deletion mutant of pQBR103 This study

*University of California, Santa Barbara.

pIVETD10. To remove the 9dapB gene pIVETD10 was digested with
SpeI and transformed into E. coli DH5alpir. The resulting plasmid,
pIVET10, was used to deliver the ornpl mutation. It was mobilized into
P. fluorescens SBW25(pQBR103) by conjugation with the help of
pRK2013 (Tra+). Integration into pQBR103 by a single homologous
recombination event was selected for on LB agar supplemented
with HgCl2, tetracycline and X-Gal. Allelic-exchange mutants were
selected as previously described (Gal et al., 2003) by growing the
blue-coloured transconjugant for two successive 24 hour periods in
LB broth without antibiotic selection. The bacterial cells were plated
on LB agar containing X-Gal. White tetracycline-sensitive colonies
were checked for loss of ornpl by PCR using primers 103orn-F/R
(Table 3).
Construction of the ornpl : : dapB fusions and plant coloni-
zation assay. To construct an ornpl : : 9dapB fusion, plasmid
pIVETD10 was digested by XhoI, which removed a 900 bp fragment
downstream of the ornpl gene. The cleaved plasmid was religated
and transformed into E. coli DH5alpir to generate pIVETD9. It was
then conjugated into SBW25DdapB(pQBR103) by triparental mating
Downloaded from www.microbiologyresearch.org by
http://mic.sgmjournals.org
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
with the help of pRK2013. Integration into the ornpl locus was
confirmed by PCR using primers Pdap (Gal et al., 2003) and p10b5-
F, which is located upstream of ornpl but outside the cloned region
of pIVETD9.
Competitive colonization assays of the IVET fusions against the
competitor strain of SBW25(pQBR103) were performed as previously
described by Rainey (1999). Bacteria were inoculated onto coated sugar
beet (Beta vulgaris var. Amethyst) seeds at a ratio of 1 : 1 (~103 cells
per seed). The seeds were germinated and cultivated in 5 ml
scintillation vials using non-sterile vermiculite as a growth substrate.
After 14 days, bacteria were counted on selective plates (M9 with
lysine, DAP, CFC, tetracycline and X-Gal) and non-selective plates
(M9 with CFC) following recovery from the shoot (the photosynthetic
parts) and rhizosphere (roots with attached vermiculite).
Inactivation of ornP.putida and heterologous complementation.
Using the genome sequence of P. putida KT2440, a pair of primers
ornF3/ornR3 (Table 3) were designed to amplify a 2?0 kb DNA frag-
ment containing ornP.putida (Fig. 1b). The PCR product was first
2891
X.-X. Zhang and others
Fig. 1. Genetic map of orn and flanking DNA. (a) Map of the
6?773 kb sequenced ornpl region of pQBR103. The ornpl gene
is shown as a filled arrow while other ORFs (orf) that have no
homologues in the databases are shown as empty arrows.
Lines indicate the corresponding fragments cloned into fusion
plasmids. Lines with arrows show the fusion positions and the
direction of transcription of the promoterless 9dapB9lacZ genes
in IVET clones pIVETD5 and pIVETD9. (b) Solid arrows indicate
orn and its flanking yjeQ and yjeS genes identified in E. coli
and the corresponding homologues in P. fluorescens (SBW25
and PfO-1), P. syringae DC3000 (PS), P. aeruginosa PAO1
(PA) and P. putida KT2440 (PP). Open squares refer to tRNA
genes downstream of ornE.coli. Discontinuous arrows indicate
a conserved hypothetical protein, which locates elsewhere in P.
aeruginosa (PA1037). Open triangles indicate the locations of
the PCR primers ornF3/ornR3, which were used to amplify the
2?0 kb orn region of P. putida, and the insertion position of the
omega cassette into ornP.putida.
cloned into the PCR cloning vector pCR2.1 (Invitrogen) and
sequenced to check for errors. The 2?0 kb ornP.putida fragment was
liberated by EcoRI digestion and cloned into pUC18, where a 2?0 kb
omega cassette with tetracycline resistance (VTc) was inserted in the
middle of ornP.putida at the XhoI site. The omega cassette was
obtained by PCR and the entire 4?0 kb insert was subcloned into an
integration plasmid vector pIVET11, which was prepared from
pIVETD by removing the 9dapB gene and the gene encoding tetra-
cycline resistance. The resultant plasmid, designated pIVET11-1, was
mobilized into P. putida UWC1, a spontaneous rifampicin-resistant
derivative of KT2440, by the standard procedure of conjugation
with the help of pRK2013. Transconjugants were selected on M9
agar supplemented with rifampicin, tetracycline and X-Gal. Allelic-
exchange mutants were identified on the basis of white colour on
X-Gal-containing media and sensitivity to pipercillin (150 mg ml21;
carried on pIVET11-1). Mutants were confirmed by PCR analysis
using primers orn-F/R (Table 3).
Complementation of the P. putida ornP.putida mutant was performed
Downloaded from www.microbiologyresearch.org by
2892
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
by cloning the orn genes from P. putida, pQBR103 and E. coli into
pME6010 (Heeb et al., 2000), a plasmid vector that can replicate in
Pseudomonas. Primers used to amplify the coding regions of the
ornP.putida, ornpl, and ornE.coli are listed in Table 3. To ensure efficient
expression of the cloned orn genes, a Pseudomonas-specific ribosome-
binding site (RBS) GAGGA was introduced into the primers at the
position of 210 from the ORF start codon. Restriction sites of KpnI
and HindIII were incorporated into the PCR primers to allow for
cloning of the products into the corresponding sites of pME6010. An
omega cassette with spectinomycin resistance was then inserted
downstream of the cloned orn gene in vector pME6010 at the HindIII
site. This vector is designed such that genes inserted in the multi-
cloning site are transcribed under the control of a constitutive
promoter, Pk (Heeb et al., 2000). Plasmids were mobilized into the
ornP.putida mutant PBR447 by conjugation. Growth was assessed by
culturing each strain in both LB broth and minimal M9 broth.
Overnight cultures were diluted in the same medium to an initial
OD600 of ~0?02. The OD600 values were measured using a GENESYS
20 spectrophotometer (Thermo Electron).
Phylogenetic analysis. Amino acid sequences were aligned using
CLUSTAL_X (Thompson et al., 1997) and a phylogenetic tree con-
structed using the neighbour-joining method (Saitou & Nei, 1987).
The tree was displayed in TreeView (Page, 1996). The orn genes and
the flanking sequences of the following species were obtained from
genome databases (in parentheses): E. coli (http://www.genome.wisc.
edu), P. fluorescens SBW25 (http://www.sanger.ac.uk), P. fluorescens
PfO-1 (http://www.jgi.doe.gov), P. syringae DC3000 and P. putida
KT2440 (http://www.tigr.org/) and P. aeruginosa PAO1 (http://
www.pseudomonas.com/). Other Orn sequences were collected from
NCBI database (Mycobacterium tuberculosis, G7227904; Streptomyces
griseus, AB036424; Caenorhabditis elegans, NP_505606; Xylella fasti-
diosa, NP_778746; Xanthomonas axonopodis, NP_642365; Ralstonia
solanacearum, NP_519063; Neisseria meningitidis, NP_283406;
Drosophila melanogaster, NP_651184; Arabidopsis thaliana, Q9ZVE0;
Homo sapiens, NP_056338; Schizosaccharomyces pombe, CAB37438;
Ectocarpus siliculosus virus, NP_077624).
RESULTS
The orn gene carried on plasmid pQBR103 is
plant inducible
DNA sequence analysis of pQBR103 IVET clone pIVETD5
revealed the presence of a putative oligoribonuclease gene
(ornpl) (Fig. 1a). Additional sequencing of pQBR103 DNA
flanking the 2?3 kb insert cloned in pIVETD5 revealed nine
predicted ORFs, none of which, with the exception of ornpl,
showed any significant similarity to genes or proteins within
EMBL or GenBank databases. Seven of the putative ORFs
are orientated in the same direction as ornpl, suggesting the
possibility that they are transcribed from a single plant-
inducible promoter (Fig. 1a). Given that the original 9dapB
transcriptional fusion in pIVETD5 is to the sixth of these
seven genes and that ornpl is the third, it was possible that
ornpl was not plant-inducible and that the plant responsive
element was downstream of ornpl.
To determine whether ornpl is specifically induced in the
plant environment, a transcriptional fusion was generated
between ornpl and a promoterless copy of 9dapB. We then
determined, once the fusion was integrated into the genome
of P. fluorescens SBW25DdapB(pQBR103), whether this
Microbiology 150
 Oligoribonucleases from a Pseudomonas plasmid

Table 3. PCR primers

Primer Sequence* Application

ornF3/ornR3 59-cagatctgatatggccgtcgagttgtt-39/ Amplification of ornP.putida DNA region (2 kb)

59-cagatctctgctggtcgttgtagtgga-39
Omega-X 59-cctcgaggatccggtgattgattgag-39 Amplification of the omega cassette (2 kb)
orn-F/R 59-ggtaccgaggataggctaatatgaccgcc-39/ Amplification of ornP.putida (600 bp)
59-aagcttcgcagatacaatatctgc-39
103orn-F/R 59-ggtaccgaggacatcaacttatggtggat-39/ Amplification of ornpl (620 bp)
59-aagcttcctaggtcatgagcagctgt-39
coli-orn-F/R 59-ggtaccgaggagaaaatagcatgagt gccaatg-39/ Amplification of ornE.coli (530 bp)
59-aagctttggattccgg cgcttacgtg-39
CVR3/gsp41 59-agccatgcgcgtcgtgccca-39/ Amplification of the putative oriV (710 bp)
59-gaagatactggcctggtag-39
p10b5-F 59-ggttacactgacccttgc-39 ornpl : : lacZ fusion confirmation

*Restriction sites incorporated into the primers are underlined and the artificial RBSs are highlighted in bold.

strain could grow in the plant environment. Growth of P. putida KT2440 (hosts of pQBR103), and Orn from E. coli,
P. fluorescens SBW25DdapB can only occur if transcription Streptomyces and human, whose functions have been
of the promoterless 9dapB gene occurs in the plant investigated biochemically, is shown in Fig. 3. The plasmid-
environment (Gal et al., 2003). In this instance this would derived Orn possesses four characteristic invariant acidic
require the promoter driving expression of ornpl to be active amino acids distributed in three separate motifs; the fifth
in the plant environment. highly conserved acidic residue between motifs II and III,
and the catalytically important histidine (Fig. 3), suggest
The ornpl : : 9dapB fusion strain PBR442 [P. fluorescens that the plasmid-borne orn is a functional gene.
SBW25DdapB(pQBR103 ornpl : : pIVETD9)], is unable to
grow on minimal M9 medium in the absence of DAP and The phylogenetic relationships of plasmid-derived Orn and
lysine showing that the promoter driving ornpl expression its representative homologues in prokaryotes, eukaryotes
is inactive on this medium. In order to test whether the and viruses are shown in Fig. 4. Orn from proteobacteria,
promoter is active in the plant environment, PBR442 was
inoculated onto sugar beet seeds. In addition, further seeds
were inoculated with: (1) the original orf6 : : 9dapB IVET
fusion strain PBR424; (2) an IVET strain with a fusion
between promoterless 9dapB and a constitutive promoter
(PBR391; positive control) and (3) two IVET strains with
fusions between fragments of DNA with no detectable
promoter activity (PBR393, PBR394; negative controls).
Results are shown in Fig. 2. Both PBR442 and PBR424
increased in frequency over the course of 2 weeks when
inoculated onto seeds at ~26103 cells per seed, as did the
positive control strain PBR391. In contrast, the negative
controls failed to increase in frequency. The ability of
PBR442 to grow in both the rhizosphere and phyllosphere,
but not on minimal M9 medium, demonstrates that the Fig. 2. Induction of the plasmid-derived orn gene in the rhizo-
promoter driving expression of ornpl is induced in the sphere (filled columns) and shoot (open columns) of sugar beet
plant environment. seedlings. PBR393 and PBR394 are two preselected Lac
strains (negative controls) with 9dapB9lacZ genes fused to
Comparative and phylogenetic analysis of pQBR103 DNA without detectable promoter activity. PBR391
plasmid-derived orn gene is an IVET strain (positive control) with a fusion between
9dapB9lacZ and a constitutive promoter. PBR424 and PBR442
The deduced amino acid sequence of the plasmid-derived are orf6 : : 9dapB9lacZ and ornpl : : 9dapB9lacZ fusion strains,
orn (ornpl) gene showed 34 % identity and 51 % similarity respectively. All fusion strains were grown in direct competition
with the well-defined E. coli oligoribonuclease, which with the wild-type SBW25 (pQBR103). After 14 days, the
belongs to the DEDDh subfamily of 39 to 59 exoribo- competitor strain was present at similar levels to the positive
nucleases (Zuo & Deutscher, 2001). Sequence alignment control across all treatments (data not shown). Data are means
of chromosome-derived Orn from P. fluorescens SBW25, and standard errors from five plants.
Downloaded from www.microbiologyresearch.org by
http://mic.sgmjournals.org 2893
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
X.-X. Zhang and others
which in our analysis included four representatives from
the genus Pseudomonas, clustered within a single group,
whereas the plasmid-derived Orn formed a separate
cluster. The G+C content of the plasmid-derived Orn
gene is 53?06 mol%, which is similar to the estimated
mean G+C content (53?16 mol%) of plasmid pQBR103
but much lower than that of Pseudomonas chromosomes
(5867 mol%). As shown in Fig. 1(b), the E. coli orn gene
Downloaded from www.microbiologyresearch.org by
2894
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
Fig. 3. Alignment of the deduced amino
acid sequence of pQBR103-derived orn
(pQBR) with oligoribonuclease of E. coli
(Ec) and homologues from Streptomyces
griseus (Str), Homo sapiens (Hs), P. fluores-
cens SBW25 (PF) and P. putida KT2440
(PP). Identical sequences are indicated by
an asterisk (*) and amino acid residues of
pQBR103-derived Orn matching one and
more than one amino acids of other four
proteins are shown as a dot (.) and a colon
(:), respectively. The three conserved motifs
of the 39 to 59 exoribonuclease DEDD family
are indicated by boxes (Zuo & Deutscher,
2001). The four highly conserved negatively
charged residues (D-E-D-D) within the three
motifs and the fifth (D) between motifs II
and III, as well as the catalytically important
histidine (H), are highlighted in bold.
(ornE.coli) is flanked by two genes encoding hypothetical
proteins (YjeQ and YjeS). This organization is highly
conserved among Pseudomonas species (Fig. 1b). However,
ornpl is flanked by sequences that have no homologue in
the public gene databases. These data suggest that the
plasmid-derived Orn gene was acquired by horizontal gene
transfer from an unknown organism: the donor appears
not to be a member of the Proteobacteria.
Fig. 4. Phylogenetic tree showing the
evolutionary relationships of plasmid-derived
Orn and selected homologues representing
different origins including P. fluorescens
SBW25 (PF), P. putida KT2440 (PP),
P. aeruginosa PAO1 (PA) and P. syringae
DC3000 (PS). Accession numbers of the
Orn homologues are listed in Methods.
Percentage bootstrap values larger than 50
obtained after 1000 iterations are shown
on branches. The scale bar refers to the
number of substitutions per site.
Microbiology 150
 Oligoribonucleases from a Pseudomonas plasmid

The plasmid-derived orn gene is functional (a) 1.6

Functionality of ornpl could in principle be demonstrated 1.4
by showing that ornpl was capable of complementing a
knockout mutant of the chromosomal orn. We began 1.2
by constructing an ornpl mutant (pQBR103Dorn, see 1.0

OD600
Methods). Growth of the mutant in the genetic background 0.8
of SBW25 was assessed on minimal (M9) and complex (LB)
0.6
media, but no obvious defects were observed. To determine
whether a phenotype might manifest in the rhizosphere 0.4
and phyllosphere, the fitness of SBW25(pQBR103Dorn) 0.2
relative to SBW25(pQBR103) was determined by com-
petitive colonization assays on sugar beet seedlings. No 1 2 3 4 5 6 7 8
significant difference was observed between mutant and Time (h)
wild-type in this environment (data not shown). (b)
Orn+
_
Orn+ Orn
_
Orn
We next sought to inactivate the chromosomal copy of orn.
We chose to do this work in P. putida KT2440 rather than
P. fluorescens SBW25 because genome sequence from
SBW25 was not available at the time. In addition, KT2440
is a permissive host of pQBR103 and similar plasmids LB M9
(Lilley et al., 1996). On the basis of the analysis of E. coli (c)
orn mutants we expected the ornP.putida mutant to be 1.0
lethal. To our surprise, insertion of an omega (tetracycline
resistance) cassette into the middle of the ornP.putida gene 0.8
of KT2440 (KT2440 ornP.putida : : VTc) was not lethal, but
did result in a mutant (designated PBR447) that had an 0.6
OD600

obvious growth defect. This defect was noticeable in LB

broth, but more pronounced in minimal M9 medium
(Fig. 5), in which PBR447 failed to reach the same cell
density as the wild-type strain even when the incubation
time was prolonged to 48 h. In LB agar plate culture,
PBR447 formed small colonies with a dimpled centre
distinguishable from the convex colony formed by the
wild-type strain. Interestingly, no such morphological
differences were evident on minimal M9 medium agar
plates, although colonies of the mutant were significantly
smaller (Fig. 5).
Lack of lethality due to inactivation of ornP.putida allowed
investigation of the functionality of the plasmid-derived
orn gene through genetic complementation. We first asked
whether presence of the entire pQBR103 was capable of
complementing PBR447. The plasmid was introduced into
PBR447 and its ability to restore growth and colony
morphology was assessed. Fig. 5 shows that pQBR103 failed
to complement the lesion. This result was not surprising
and is consistent with our knowledge that ornpl is not active
on minimal medium.
Since the factor(s) that induces ornpl expression in the
plant environment is not known, the putative coding
region of ornpl was cloned (by PCR) into the broad-host-
range plasmid pME6010, placing ornpl under the control
of a constitutive kanamycin-resistance promoter (Pk). An
artificial RBS was incorporated into the forward primer to
ensure efficient translation of ornpl; a spectinomycin-
resistant omega cassette was inserted downstream of ornpl
to ensure transcriptional termination and to provide a
Downloaded from www.microbiologyresearch.org by
http://mic.sgmjournals.org
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
0.4
0.2
1 2 3 4 5 6 7 8
Time (h)
Fig. 5. Growth of P. putida KT2440 (#) and KT2440
(pQBR103) ($), PBR447 (KT2440ornP.putida : : VTc; n) and
PBR447 (pQBR103) (m) in liquid medium of LB (a) and M9
(c). Data are means and standard errors of five replicates. The
colonies shown in (b) are typical of P. putida KT2440 (Orn+)
and PBR447 (Orn) growing in LB or M9 agar (1?8 %) after
2 days incubation at 28 6C. The two strains were plated in the
same Petri dish.
selectable marker for plasmid conjugation. In addition
(and in parallel) the coding region of ornp.putida was
cloned into pME6010. The two plasmids one carrying
ornpl (pME6010 : : ornplV) and the other ornP.putida
(pME6010 : : ornP.putidaV) were independently introduced
into the ornP.putida mutant PBR447 and growth of the trans-
conjugants was compared in M9 broth. The growth defect of
the ornP.putida mutant was fully restored upon introduction
of the chromosomal copy of orn (pME6010 : : ornP.putidaV)
(Fig. 6a). This demonstrates that the slow growth of PBR447
was caused by the ornP.putida mutation. The plasmid-derived
2895
X.-X. Zhang and others

Outside the context of E. coli, orn has received relatively

little attention. There are no functional studies of orn from
Pseudomonas, but evidence that orn from the chromosome
of KT2440 is an oligoribonuclease could be obtained by
showing that the well-characterized oligoribonuclease gene
from E. coli (ornE.coli) is able to complement the defect
in PBR447. The ornE.coli gene from E. coli DH5a was
amplified by PCR and cloned into pME6010 exactly as
described above. Plasmid pME6010 : : ornE.coliV was then
introduced into the ornP.putida mutant PBR447 and its
ability to restore growth of PBR447 was examined. PBR447
(pME6010 : : ornE.coliV) grew as quickly as the wild-type
strain KT2440, showing that the E. coli-derived orn gene
can fully complement the ornP.putida mutant. Together
these data provide strong support for the claim that both
ornP.putida and ornpl are functional oligoribonucleases.

The orn gene is not required for conjugative

transfer of pQBR103
The yeast mitochondrial genome harbours an orn gene
that has been implicated in escape of mitochondrial DNA
to the nucleus (Hanekamp & Thorsness, 1999). This led
us to consider the possibility that orn might be asso-
ciated with conjugative transfer. To determine whether
pQBR103Dornpl is deficient in conjugation, its ability to
transfer to wild-type P. putida KT2440 and the ornP.putida
mutant of KT2440(PBR447) was assessed. Both pQBR103
and pQBR103Dornpl were able to transfer at a similar
frequency (1?06610261?661027, transconjugants per
recipient) from P. fluorescens SBW25 (donor) to KT2440
(recipient). We were similarly able to obtain transconju-
gants from matings between SBW25(pQBR103Dornpl) and
PBR447.

Distribution of ornpl among indigenous

Pseudomonas plasmids
Fig. 6. Growth of P. putida PBR447 (KT2440ornP.putida : : VTc)
carrying orn genes from pQBR103 (a) and E. coli (b) and a
To determine whether the ornpl gene is widely distributed
schematic map (c) showing the construct used to express orn. among indigenous plasmids occurring at the same field
(a, b) All strains were grown in minimal M9 medium: KT2440 site, 12 plasmids representing four different plasmid groups
(#); PBR447 containing pME6010 : : ornP.putidaV (%); PBR447 (Lilley & Bailey, 1997a) were selected for a PCR assay of
containing pME6010 : : ornplV ($); PBR447 containing the ornpl gene. Primers were designed to amplify the
pME6010 : : ornE.coliV (m); PBR447 containing the vector 620 bp coding region of ornpl and the results are shown
pME6010V (n). Data are means and standard errors of five in Table 1. The orn gene was found in five plasmids,
repeats (error bars lie within the symbols). (c) The filled bar is including pQBR103, which have molecular sizes ranging
the multi-cloning site of pME6010 although only the restriction from 287 kb to 425 kb. The five PCR-amplified ornpl
sites used in this study are shown (X, XhoI; K, KpnI; H, genes were subsequently sequenced and shown to be
HindIII). The orn gene was expressed under the control of a identical at the DNA level.
constitutive kanamycin resistance promoter (Pk).
The five plasmids carrying orn are all members of the
dominant group I plasmids (Table 1), which were defined
orn gene (pME6010 : : ornplV) was also capable of comple- originally on the basis of restriction fragment patterns
menting the defect in PBR447, but not to the same extent (Lilley et al., 1996). Sequence identity of the orn genes
as ornP.putida. Partial complementation of PBR447 by the suggested the possibility that the variation among these
plasmid-derived copy of ornpl is not unexpected given the plasmids is primarily the result of spontaneous recombina-
divergent phylogenetic origins of this gene, but nonethe- tion events. To this end we sequenced 300 bp from the
less, the ability of this gene to increase growth of PBR447 plasmid replicative origin (oriV) of the five plasmids with
provides evidence of its functionality. identical orn sequences as accomplished previously for
Downloaded from www.microbiologyresearch.org by
2896 Microbiology 150
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23

pQBR11 (Viegas et al., 1997). Sequence results showed that
the oriV regions of the five plasmids are identical. Together,
these data suggest that the plasmid-derived orn genes were
acquired by a single horizontal gene transfer event.
DISCUSSION
Functional analyses of plasmids from natural environments
pose special challenges. Plasmid pQBR103 is representative
of a large group of conjugative plasmids present in plant-
colonizing Pseudomonas species. pQBR103 was isolated by
its ability to confer resistance to mercuric chloride on
recipient bacteria, but other than this marker, the functional
traits encoded by the plasmid were unknown (Bailey et al.,
2001).
Here we have described the identification and functional
characterization of a novel oligoribonuclease gene carried
on pQBR103. Plasmid-encoded orn (ornpl) was initially
isolated using a promoter-trapping strategy that allowed
recovery of pQBR103 genes that were transcriptionally
active on sugar beet plants, but inactive on minimal M9
medium (unpublished). While ornpl was not directly fused
to the gene trapped by the promoterless 9dapB reporter,
fusions between ornpl and a promoterless copy of 9dapB
demonstrate that the gene is plant inducible.
Evidence that ornpl encodes a functional oligoribonuclease
was initially inferred from in silico analyses of the predicted
protein. Subsequent experiments showed that ornpl could
partially complement an orn mutant (ornP.putida) of P. putida
KT2440. At the same time we showed that the orn mutant
(ornP.putida) of P. putida KT2440 could be complemented
by the orn gene from E. coli. Together these data provide
compelling evidence that ornpl is functional.
Oligoribonucleases have received little attention. Never-
theless, studies to date show that Orn is a 39 to 59
exoribonuclease and an important component of the
mRNA degradation pathway in bacteria (Ghosh &
Deutscher, 1999; Ohnishi et al., 2000); a similar role has
been suggested for the conserved human homologue
(Nguyen et al., 2000). In E. coli, Orn is one of eight
known exoribonucleases and the only one known to be
essential for cell viability (Zuo & Deutscher, 2001).
Inactivation of the orn gene results in accumulation of
small oligoribonucleotides (25 residues in length), which
are thought to be toxic to E. coli cells (Ghosh & Deutscher,
1999).
Little is known about mRNA degradation in the genus
Pseudomonas. However, given the phylogenetic closeness
of E. coli and Pseudomonas (both members of the gamma
subdivision of the Proteobacteria) and similarity of Orn
from E. coli and Pseudomonas (67 % identical), we were
surprised to find that inactivation of ornP.putida was not
lethal. The growth defect we observed was also noted in
an orn mutant of Streptomyces griseus (Ohnishi et al., 2000).
Downloaded from www.microbiologyresearch.org by
http://mic.sgmjournals.org
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
Oligoribonucleases from a Pseudomonas plasmid
The reason for the difference between phenotypes of orn
mutants in E. coli relative to P. putida is unclear. One
possibility is that in P. putida there is functional redundancy
among oligoribonucleases, but analysis of the KT2440
genome sequence did not reveal obvious Orn homologues.
A further possibility is that mRNA degradation pathways
are different in the two bacteria. For instance, E. coli
contains a gene encoding RNase II, which is a principal
39 to 59 hydrolytic exoribonuclease, but this is absent in
Pseudomonas. Further evidence of differences between
Pseudomonas and E. coli comes from a recent study of
another key exoribonuclease PNPase (polynucleotide
phosphorylase) in P. putida (Favaro & Deho, 2003),
which when mutated did not exhibit cold sensitivity as
expected from parallel studies in E. coli.
The precise role of ornpl and the reason for its presence
in the group I Pseudomonas plasmids from Wytham is
unclear. The selective forces operating on plasmids, com-
bined with the evidence presented here of functionality and
plant-inducibility, suggest that ornpl is likely to have
ecological relevance. Precisely what this might be remains
unclear, but it is possible that Ornpl functions in a role
other than mRNA degradation, a suggestion that has
some credibility given the demonstration that the human
Orn homologue, Sfn (small fragment nuclease), is capable
of hydrolysing both RNA and DNA oligomers (Nguyen
et al., 2000).
Finally, our discovery that identical copies of ornpl are
present in a subset of group I Pseudomonas plasmids
from Wytham and yet share little in common with chromo-
somal orn from c-Proteobacteria indicates that ornpl is of
foreign origin. The complete lack of allelic diversity in
ornpl suggests that ornpl was acquired just once before
dissemination by lateral gene transfer. The lack of diver-
sity suggests that the transfer event was recent, but in the
absence of complete sequence analysis of these plasmids
it is not possible to exclude the possibility that intragenetic
recombination combined with plasmid instability generates
the restriction fragment differences previously reported
among the five ornpl-containing plasmids (Lilley et al.,
1996). Indeed, the corresponding lack of allelic diversity
in putative oriV from the same five plasmids suggests a
single plasmid type capable of converting between different
semi-stable states.
Our data do not permit us to identify the ancestral organism
from which ornpl originated, although it seems certain that
ornpl does not originate from the Proteobacteria. However,
even with the usual caveats associated with long-branch
attraction (Baxevanis & Ouellette, 2001), our data do
suggest that ornpl originates from a eukaryotic source. This
is of some interest because it is consistent with data that
suggest that orn first evolved in eukaryotes and was then
transferred to eubacteria (but not the archaea) by lateral
gene transfer (Zuo & Deutscher, 2001). Perhaps ornpl
reflects an additional dissemination event from a eukaryote.
2897
X.-X. Zhang and others
ACKNOWLEDGEMENTS
We thank Andrew Spiers, Sarah Turner and Robert Jackson for helpful
discussions and Lena Ciric for help with DNA sequencing. This work
was supported by NERC (UK).
REFERENCES
Bailey, M. J., Lilley, A. K., Thompson, I. P., Rainey, P. B. & Ellis, R. J.
(1995). Site directed chromosomal marking of a fluorescent
pseudomonad isolated from the phytosphere of sugar beet; stability
and potential for marker gene transfer. Mol Ecol 4, 755763.
Bailey, M. J., Rainey, P. B., Zhang, X. X. & Lilley, A. K. (2001).
Population dynamics, gene transfer and gene expression in plasmids,
the role of the horizontal gene pool in local adaptation at the plant
surface. In Microbiology of Aerial Plant Surfaces, pp. 171189. Edited
by S. E. Lindow, E. I. Hecht-Poinar & V. J. Elliott. St Paul, MN:
American Phytopathological Society Press.
Baxevanis, A. D. & Ouellette, B. F. F. (2001). Bioinformatics: a
Practical Guide to the Analysis of Genes and Proteins, 2nd edn. New
York: Wiley Interscience.
Bloemberg, G. V. & Lugtenberg, B. J. J. (2001). Molecular basis of
plant growth promotion and biocontrol by rhizobacteria. Curr Opin
Plant Biol 4, 343350.
Ditta, G., Stanfield, S., Corbin, D. & Helinski, D. R. (1980). Broad
host range DNA cloning system for Gram negative bacteria:
construction of a gene bank of Rhizobium meliloti. Proc Natl Acad
Sci U S A 77, 73477351.
Favaro, R. & Deho, G. (2003). Polynucleotide phosphorylase-
deficient mutants of Pseudomonas putida. J Bacteriol 185, 52795286.
Fellay, R., Frey, J. & Krisch, H. (1987). Interposon mutagenesis of
soil and water bacteria: a family of DNA fragments designed for in
vitro insertional mutagenesis of Gram-negative bacteria. Gene 52,
147154.
Gal, M., Preston, G. M., Massey, R. C., Spiers, A. J. & Rainey, P. B.
(2003). Genes encoding a cellulose polymer contribute toward the
ecological success of Pseudomonas fluorescens SBW25 on plant
surfaces. Mol Ecol 12, 31093121.
Ghosh, S. & Deutscher, M. P. (1999). Oligoribonuclease is an
essential component of the mRNA decay pathway. Proc Natl Acad Sci
U S A 96, 43724377.
Hanekamp, T. & Thorsness, P. E. (1999). YNT20, a bypass
suppressor of yme1 yme2, encodes a putative 39-59 exonuclease
localized in mitochondria of Saccharomyces cerevisiae. Curr Genet 34,
438448.
Hartl, D. L., Dykhuizen, D. E. & Berg, D. E. (1984). Accessory DNAs
in the bacterial gene pool: playground for coevolution. In Origins
and Development of Adaptations (Ciba Foundation Symposium 102),
pp. 233245. London: Pitman Books.
Heeb, S., Itoh, Y., Nishijyo, T., Schnider, U., Keel, C., Wade, J.,
Walsh, U., OGara, F. & Haas, D. (2000). Small, stable shuttle vectors
based on the minimal pVS1 replicon for use in gram-negative, plant-
associated bacteria. Mol PlantMicrobe Interact 13, 232237.
Lilley, A. K. & Bailey, M. J. (1997a). The acquisition of indigenous
plasmids by a genetically marked pseudomonad population coloniz-
ing the sugar beet phytosphere is related to local environmental
conditions. Appl Environ Microbiol 63, 15771583.
Downloaded from www.microbiologyresearch.org by
2898
IP: 125.164.3.181
On: Tue, 07 Nov 2017 10:32:23
Lilley, A. K. & Bailey, M. J. (1997b). Impact of plasmid pQBR103
acquisition and carriage on the phytosphere fitness of Pseudomonas
fluorescens SBW25: burden and benefit. Appl Environ Microbiol 63,
15841587.
Lilley, A. K., Bailey, M. J., Day, M. J. & Fry, J. C. (1996). Diversity of
mercury resistance plasmid obtained by exogenous isolation from
the bacteria of sugar beet in three successive years. FEMS Microbiol
Ecol 20, 211227.
McClure, N. C., Fry, J. C. & Weightman, A. J. (1990). Gene transfer in
activated sludge. In Bacterial Genetics in Natural Environments,
pp. 111129. Edited by J. C. Fry & M. J. Day. London: Chapman &
Hall.
Nakazawa, T. (2002). Travels of a Pseudomonas, from Japan around
the world. Environ Microbiol 4, 782786.
Nguyen, L. H., Erzberger, J. P., Root, J. & Wilson, D. M., III (2000).
The human homolog of Escherichia coli Orn degrades small single-
stranded RNA and DNA oligomers. J Biol Chem 275, 2590025906.
Ohnishi, Y., Nishiyama, Y., Sato, R., Kameyama, S. & Horinouchi, S.
(2000). An oligoribonuclease gene in Streptomyces griseus. J Bacteriol
182, 46474653.
Page, R. D. M. (1996). TreeView: an application to display
phylogenetic trees on personal computers. Comput Appl Biosci 12,
357358.
Palleroni, N. J. (1984). Pseudomonadaceae. In Bergeys Manual of
Systematic Bacteriology, vol. 1, pp. 141199. Edited by N. R. Kreig &
J. G. Holt. Baltimore: Williams & Wilkins.
Rainey, P. B. (1999). Adaptation of Pseudomonas fluorescens to the
plant rhizosphere. Environ Microbiol 1, 243257.
Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Mol Biol Evol 4,
374377.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning:
a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory.
Spiers, A. J., Bohannon, J., Gerhrig, S. M. & Rainey, P. B. (2003).
Biofilm formation at the air-liquid interface by the Pseudomonas
fluorescens SBW25 wrinkly spreader requires an acetylated form of
cellulose. Mol Microbiol 50, 1527.
Thomas, C. M. (2000). The Horizontal Gene Pool: Bacterial Plasmids
and Gene Spread. Amsterdam: Harwood Academic Publishers.
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. &
Higgins, D. G. (1997). The CLUSTAL_X windows interface: flexible
strategies for multiple sequence alignment aided by quality analysis
tools. Nucleic Acids Res 25, 48764882.
Viegas, C. A., Lilley, A. K., Bruce, K. B. & Bailey, M. J. (1997).
Description of a novel plasmid replicative origin from a genetically
distinct family of conjugative plasmids associated with phytosphere
microflora. FEMS Microbiol Lett 149, 121127.
Vivian, A., Murillo, J. & Jackson, R. W. (2001). The roles of plasmids
in phytopathogenic bacteria: mobile arsenals? Microbiology 147,
763780.
Zhang, X., Zhu, L. & Deutscher, M. P. (1998). Oligoribonuclease is
encoded by a highly conserved gene in the 39-59 exonuclease
superfamily. J Bacteriol 180, 27792781.
Zuo, Y. & Deutscher, M. P. (2001). Survey and summary:
exoribonuclease superfamilies: structural analysis and phylogenetic
distribution. Nucleic Acids Res 29, 10171026.
Microbiology 150