Available online at www.sciencedirect.
com
ScienceDirect
Taking the pseudo out of pseudogenes
Ian Goodhead and Alistair C Darby
Pseudogenes are defined as fragments of once-functional
genes that have been silenced by one or more nonsense,
frameshift or missense mutations. Despite continuing
increases in the speed of sequencing and annotating bacterial
genomes, the identification and categorisation of pseudogenes
remains problematic. Even when identified, pseudogenes are
considered to be rare and tend to be ignored. On the contrary,
pseudogenes are surprisingly prevalent and can persist for long
evolutionary time periods, representing a record of once-
functional genetic characteristics. Most importantly,
pseudogenes provide an insight into prokaryotic evolutionary
history as a record of phenotypic traits that have been lost.
Focusing on the intracellular and symbiotic bacteria in which
pseudogenes predominate, this review discusses the
importance of identifying pseudogenes to fully understand the
abilities of bacteria, and to understand prokaryotes within their
evolutionary context.
Addresses
Functional and Comparative Genomics, School of Biological Sciences,
University of Liverpool, Crown St., Liverpool L69 7ZB, UK
Current Opinion in Microbiology 2015, 23:102–109
This review comes from a themed issue on Genomics
Edited by Neil Hall and Jay Hinton
http://dx.doi.org/10.1016/j.mib.2014.11.012
1369-5274/# 2014 Published by Elsevier Ltd.
Introduction
The paradigm for bacterial genomes is that they are small
and efficient. Coding sequences are short — approxi-
mately 1 kb in length — lack introns and are arranged
in such a manner that little coding capacity (i.e. the
amount of ‘active’ coding sequence as a percentage of
the total genome size) is wasted. Operons are organised
‘end-to-end’ with few gaps in between. Generally, pro-
karyotic genome size correlates well with gene number,
unlike in some eukaryotes, where expansions in genome
size tend to show a reduced overall percentage of the
genome coding for genes [1]. Generally, intergenic and
non-functional sequences are scarce in bacteria, probably
due to a propensity for redundant sequences to be
deleted, although decreased coding capacities can be
observed in intracellular bacteria [2]. It is assumed that
Current Opinion in Microbiology 2015, 23:102–109
the increased amount of non-coding, intergenic
sequences in intracellular bacteria are the remnants of
genes that have been systematically scrambled by
mutation [3].
In bacteria, pseudogenes, defined here as ‘genes that have
been silenced by one or more deleterious mutations,’ are
commonly detected as fragments of previously described
homologs, and therefore can become difficult to annotate
once significant degradation to the original sequence has
occurred. A comparison between the genomes of any two
species can identify polymorphisms between related
genes that may have altered or ablated their function.
Pseudogenes, therefore, can function as a record of the
proteins, enzymes or pathways that are no longer necess-
ary as the bacterium has adapted to a novel environment.
High throughput DNA sequencing is now an indispen-
sible tool in life sciences. The technology allows the rapid
description of genomes, transcriptomes and epigenetic
features in ways that were not possible a decade ago. In
the area of bacterial genomics and informatics we are
feeling very pleased with ourselves. We can now rapidly
sequence large panels of bacteria to high standards of
finished qualities, and describe important differences
associated with phenotypes [4,5]. Nevertheless, we are
struggling to define or categorise pseudogenes, and to
explain why they persist (i.e. have not been entirely
removed from the genome). We have not examined
whether pseudogenes have any residual function, and
it is not clear why we should invest resources to do so.
The dynamics of gene evolution
In order to emphasise the importance of pseudogenes, it
is important to understand them in the context of the
evolution of a microbial genome. Microbial pathogens
and symbionts exist at different stages along a pathway of
pseudogenisation, which ultimately ends with gene
deletion and a greatly reduced genome size, depending
on the length of time that the cell has co-evolved along-
side its host (Figure 1). From an evolutionary standpoint,
bacterial genomes both expand and contract, depending
upon the rate of acquisition of new genes (and pheno-
types) and the deletion of redundant sequences. Such
dynamics are observable within genera and species (Box
1). The ‘accessory’ genome contains genes that encode
adaptive traits and can expand through horizontal gene
transfer [6], to increase host-range or virulence [7], or
introduce tolerance to antimicrobial compounds [8],
environmental toxins [9], or genes associated with the
establishment or maintenance of symbiosis [10]. Corre-
spondingly, the ‘core’ genome, which encodes essential
www.sciencedirect.com
 The open problem of pseudogenes Goodhead and Darby 103

Figure 1

Deletion

Transposable elements
Recombination
Pseudogenisation
# Pseudogenes

Host cell

Host cell

Bacterial cell

Free-living bacterium Recent symbiont Obligate symbiont

Host-restriction Genome reduction
Large genome size < Genome size << Genome size
High % coding capacity < Coding capacity Host-specific
Few transposable elements <Transposable elements > % coding capacity
Free-living / multiple hoste <Selection AT-bias

Current Opinion in Microbiology

Genome reduction with respect to endosymbionts. Schematic of genome reduction as bacteria move from a free-living environment to a more
specific (intracellular) niche. Initially having large genomes, some reduction in both size and coding capacity is seen as the organism adapts to an
intracellular environment as the selective constraints on some genes are relaxed due to protection or nutrients offered by the host cell. Such
genes are prone to ‘pseudogenisation’ by mutation. Further specialisation results in deletion of redundant sequences and a much-reduced
genome size. The long-term maintenance of reduced coding capacities in recent symbionts may be as a result of some residual function.

genes associated with functions such as metabolism, can
remain stable unless large environmental changes occur
— such as adapting to a new environment (e.g. adaptation
to an intracellular lifestyle within a host). Horizontal gene
transfers are common in bacteria, and transfer large
amounts of content, reaching rates exceeding 16 kbp
per million years in Escherichia coli, for instance [11].
Once removed from evolutionary selective constraints,
genes can accumulate deleterious single nucleotide poly-
morphisms, frameshifts or truncations relative to an
ancestral orthologue (Figure 2). A frequent feature of
intracellular prokaryotes is the inability of the organism
Box 1 Intracellular bacteria genome evolution
to correct such non-synonymous substitutions, often due
to the lack of proofreading during DNA replication. For
example, the loss of DnaQ-mediated proofreading activi-
ties of DNA polymerase III in M. leprae, the etiological
agent of leprosy [12,13], or recA/F in Buchnera, the primary
symbiont in pea aphids [14], results in increased rates of
mutation. A combination of a lack of error-correction and
the frequent bottlenecks that are associated with an
Figure 2
polymorphism
premature stop
loss / change of functional domain
loss of start codon

Compared to the free living life style the intracellular environment can promoter / RBS
loss of promoter / RBS
be viewed as a protected, homeostatic, nutrient-rich compartment.
In such an environment a bacterium will have less need for some the
pathways whose function may now be provided by the host, for premature stop
instance those involved with metabolism or defence — genes that
insertion
would be under strong positive selection to be retained — in a free-
1
living bacterium. Because of this, many obligate intracellular bacteria 2 frameshift
3
have much smaller genomes than their free-living counterparts, as
such pathways tend to have been deleted. There are a number of Current Opinion in Microbiology
good examples where bacterial genomes have been streamlined.
Wigglesworthia glossinidia, the mutualist symbiont of tsetse flies,
Methods by which pseudogenes can be formed. A schematic showing
has a genome size of just 700 kb [26]; Mycoplasma genitalium,
actions of polymorphisms on gene sequences leading to ‘pseudogene’
the obligate parasite of primate genital and respiratory tracts, 580 kb
formation. The top four panels showing polymorphism within coding
[66] and some Buchnera species, mutualist symbionts in aphids,
sequence or promoter regions. The fifth and sixth panels show
have genome sizes of between 450 kb and 650 kb [27,67].
insertions within coding sequences leading to a premature stop or a
frameshift, respectively.

www.sciencedirect.com Current Opinion in Microbiology 2015, 23:102–109

104 Genomics
intracellular lifestyle results in frequent fixation of non-
sense mutations within the population: Non-synon-
ymous/synonymous (dN/dS) substitutions in
intracellular bacteria regularly reach rates of 1, outstrip-
ping that of free-living bacteria [15]. Non-sense
mutations and frameshifts frequently create pseudogenes
by introducing premature stop-codons, or by destroying
promoters [16,17], start-codons [18], or functional
domains [16,19]. Horizontal gene transfers are probably
the major source of pseudogenes in prokaryotes, as the
majority of the genes transferred likely provide few
benefits and are rapidly silenced due to a lack of positive
selective pressure. Some genes may even be toxic increas-
ing negative selective pressure [20]. Deletion, therefore
represents a mechanism to defend against deleterious
products of gene acquisition and maintains small gen-
omes sizes [21]. Prokaryotic pseudogenisation differs
from the mechanism in eukaryotes, in which retrotran-
sposition (‘processed’), and gene duplication and sub-
sequent degradation of additional gene copies (‘non-
processed’), predominate [22,23]. There are also
examples in intracellular bacteria where transposons or
mobile elements disrupt gene function [24,25].
In obligate mutualist symbionts, genome reduction may
be coupled with dependence upon the symbiont for the
production of certain metabolites that are not provided by
standard nutrition, such as Wigglesworthia providing essen-
tial B-vitamins that are not present in a tsetse fly’s regular
blood meal [26], or Buchnera providing aphids with essen-
tial amino acids that are absent from phloem sap [27].
Arguably, the best example of genome reduction is the
mitochondrion, which is an ancient symbiosis that has
become a eukaryotic organelle in its own right [28];
phylogenetic analysis of cytochrome B and cytochrome
C oxidase I genes suggest mitochondria may have
diverged from a free-living Rickettsiaceae ancestor
approximately 1.5–2.0 billion years ago [29,30]. The most
closely related organism to the eukaryotic mitochondrion
is Rickettsia prowazekii, a gram negative Alpha-proteobac-
terium that causes epidemic typhus. R. Prowazekii has a
1.1 Mb genome that is undergoing substantial reduction
and has a coding capacity of only 76%. Reduced coding
capacities are common in many of the Rickettsiae, which
represent a taxonomically diverse group of obligate intra-
cellular bacteria existing in both symbiotic and patho-
genic relationships with their hosts, and therefore
represent a useful model for understanding pseudogene
content across environments [3,31]. The 49 Rickettsia
genomes available in Ensembl Bacteria (release 22)
[32] have small chromosome sizes (range 0.86–1.82 Mb;
mean 1.23Mb), low %GC contents (mean 31.7%) and
reduced coding capacities (mean 76%). Low %GC con-
tents are important in this context, as alternative-frame
premature stop-codons are more prevalent in AT-rich
genomes, which are hypothesised to serve as error-cor-
recting sequences to minimise the effect of deleterious
Current Opinion in Microbiology 2015, 23:102–109
frameshift mutations [33]. The genomes of both M. leprae
[13] and S. glossinidius [24,34] have coding capacities of
only 50%, suggesting there has not been sufficient
evolutionary time or selective pressure for non-coding
regions to be removed.
Pseudogene accumulation is a hallmark of recent adap-
tation to a new host, wherein genes are being selectively
silenced as the organism adapts to its new environmental
niche (Figure 2). Salmonella genomes serve as a useful
example in this instance: Salmonella is a broad host-range
Gram-negative facultative anaerobe that exists in a num-
ber of different hosts, and for which six subspecies and
over 2500 serovars have been identified and several
genomes have been sequenced (Reviewed [35]). Further-
more, the genomes are clonal, around 4.5 Mb in size and
reflect evolutionary changes, such as genome reduction
and horizontal gene acquisition, to adapt to different
niches. Generally, Salmonella enterica serovars with a
broad host-range, such as S. Enteritidis PT4 (113 pseu-
dogenes) or S. Typhimurium SL1344 (63) have fewer
pseudogenes than host-specific serovars like S. Galli-
narum (309), which causes Fowl Typhoid, or the
human-restricted S. Typhi CT18 (204) [35]. Expanding
this to more bacterial genera, the same pattern of
increased pseudogenisation in narrow host range species
can be seen, for example in Yersinia [36], Escherichia and
Shigella [37] and Rickettsia [38].
The mystery of persistence
Some bacteria, particularly younger symbionts such as
Sodalis glossinidius or Baumannia cicadellinicola, maintain
high numbers of pseudogenes, and reduced coding
capacities. Comparisons have shown that pseudogenes
are generally species-specific, and that they are rarely
shared between closely related species [39–44]. A study in
Cicadas reported that two recently diverged bacterial
symbiont species within the same host have non-over-
lapping pseudogenes, and complementary gene contents
that, together, ameliorate individually silenced pathways
[45]. A lack of shared pseudogenes between species also
suggests that redundant genes are either eliminated
rapidly from genomes, or that they are subject to a high
degree of degradation that they are no longer recognised
by standard annotation pipelines [44].
Due to the efficiency of genome architecture in bacteria,
it is unclear why prokaryotic genomes would maintain
reduced coding-capacities. It is tempting to hypothesise
that persistent pseudogenes maintain some function, and
therefore there is some positive selective pressure to
maintain their presence. If there is a selective advantage
to the removal of extraneous DNA, however, due to, for
instance, reducing unnecessary energy expenditure on
DNA replication, transcriptional or translational mechan-
isms, or even to remove toxic genes [35], why should it
take so long for such regions to be deleted? Pseudogenes
www.sciencedirect.com

can be surprisingly persistent: Throughout its evolution,
the mean half-life of pseudogenes in Buchnera is approxi-
mately 24 million years, [46]. There are two hypotheses to
explain such a high level of persistence: Either there is
some residual function associated with pseudogenes
resulting in a small selective pressure to maintain the
pseudogenised sequence [47], or perhaps there is a bias
towards silencing regulatory sequences thereby removing
the need to delete the pseudogene itself [46].
It is important to understand that ‘pseudogenes’ are, by
definition, a result of bioinformatic analyses that follow a
set of rules for what a gene should look like. Therefore, if a
hypothetical ‘pseudogene’ is processed by cellular
machinery and produces RNA, or a protein (albeit
possibly altered relative to an ancestral orthologue), is
it actually a ‘pseudo’ gene? Unless the regulatory machin-
ery for a given gene has been altered, it is likely that
pseudogene-derived RNA is transcribed. There is
increasing evidence for pseudogene transcription in
eukaryotes: The ENCODE project predicting that 1/
5th of pseudogenes are being actively transcribed [48],
and that some pseudogenes have an active role in regu-
lation through siRNA pathways [49], although evidence
remains elusive in prokaryotes. Microarray-based expres-
sion analyses suggested that a large proportion of the M.
leprae pseudogenes are transcribed [50,51], although it is
unlikely that these transcripts are translated into func-
tional protein. No pseudogene-derived products were
observed from over 300 proteins identified in extracts
from M. leprae isolated from Armadillos [52] and few
peptides were detected matching the nine expressed
pseudogenes identified in S. Typhi Ty2 [53]. Pseudo-
gene-derived protein products are viable, if perhaps rare:
Cloning the M. leprae pyrR pseudogene into E. coli yielded
an appropriately sized protein product [51]. The pyrR
pseudogene contains a premature stop codon, which
corresponds to approximately 75% of its original coding
Figure 3
Homology
Proteome
Transcriptome
Genome
CDS unannotated
ORF
Building evidence for gene models and putatively functional pseudogenes. Coding sequences (CDS), can be predicted on the genome by one or
more of: (1) Homology to closely related genomes. (2) Overlapping proteomic data. (3) Overlapping transcriptome data. (4) Genomic data, that is
an open reading frame with a valid start codon, stop codon, ribosomal binding site and functional domain. Some functionality may exist for
pseudogenised open reading frames (Venn diagram, right).
www.sciencedirect.com
The open problem of pseudogenes Goodhead and Darby 105
sequence relative to its M. tuberculosis orthologue, and an
otherwise intact translational architecture. In a study of
three Yersiniae, between one and forty annotated ‘pseu-
dogenes’ showed evidence of translation [54]. The
concept of pseudogenes being silent in a translational
sense seems to be incorrect, although further evidence for
active protein products being produced from degraded
sequences will help elucidate the extent to which this is
the case (Figure 3).
The size of the problem
Automatic annotation pipelines for prokaryotic genomes
have vastly improved alongside this speed increase:
Prokka, for instance, can annotate the E. coli K12 genome
in only six minutes using a desktop computer [55],
however, the ability to predict and annotate pseudogenes
has lagged behind. Prokka does provide a list of suspi-
cious genes that are potentially pseudogenes, but these
need to be checked carefully.
In contrast, manual re-annotation projects have shown
some success in identifying originally unidentified pseu-
dogenes: For instance, Belda and colleagues used
BLASTX [56] searches of intergenic sequences >50 bp
against the E. coli K12 proteome, and found 529 additional
pseudogenes to the original 927 identified in Sodalis
glossinidius [34]. Similarly, the original E. coli K12 genome
contained only a single pseudogene [57], a subsequent
study found 207 [58]. Table 1 lists some basic statistics
for annotations held for a few bacterial genera taken from
two databases: the Prokaryotic Genome Analysis Tool
(PGAT, [59]) and the Integrated Microbial Genomes
(IMG) database [60]. The level of pseudogene annotation
can be seen to vary depending upon the source of the
data. Downstream analyses, such as gene expression,
comparative genomics or metabolomics or proteomic
studies, rely heavily upon comprehensive and accurate
genome annotations. In this context, it is important to
Pseudogenes
Expressed
Protein
CDS predicted
pseudogene
Current Opinion in Microbiology
Current Opinion in Microbiology 2015, 23:102–109
106 Genomics

Table 1

A list of seven genera with the number of genomes analysed, mean genome size, mean numbers of annotated coding sequences and
pseudogenes, and the maximum number of annotated pseudogenes

Source Genome size # coding Pseudogenes Coding capacity

(mean bp) sequences (mean) (mean %)
Mean Max
Buchnera IMG 620 134 539 2 18 84.7
Burkholderia PGAT 6 741 756 5831 295 493 68.1
IMG 6 788 019 6355 48 4011 84.9
Escherichia/Shigella PGAT 4 979 798 4573 229 858 82.2
IMG 5 167 476 5083 5 675 88.2
Francisella PGAT 1 914 213 1529 264 496 32.3
IMG 1 857 009 1902 25 317 88.0
Pseudomonas PGAT 6 467 936 5717 40 62 86.9
IMG 6 142 996 5814 16 1523 88.4
Rickettsia IMG 1 377 304 1373 10 416 78.4
Salmonella PGAT 4 790 558 4500 153 336 83.4
IMG 4 755 870 4677 13 660 87.2
NB 4 849 106 4460 162 311 n/a

Listed is the mean coding capacity (%) for all of the examined genomes within each genus. Data was taken from the PGAT and IMG databases, or
from the Nuccio and Bäumler (2014) reannotation of fifteen Salmonella genomes (‘NB’) [59,60,65]. For the latter, overall coding capacity could not
be calculated.

know whether or not gene fragments have been anno-
tated in the genome as discrepancies between annota-
tions may have biological relevance changing gene
function or active metabolic pathways.
An excellent study of fifteen S. enterica strains found that
comparative genomic analysis was hindered by the differ-
ent methods of annotation used on each genome, particu-
larly with reference to predicted pseudogenes [65]. The
authors therefore performed a substantial re-annotation
and identified 1004 novel ‘pseudogenes’. The 15 genomes
contained 471 intact CDS that had been misidentified as
pseudogenes. Analysis of genome degradation in the
corrected genome annotation revealed microbial path-
ways essential for Salmonella gastrointestinal success,
which were inactivated by pseudogenes in extra-intesti-
nal pathovars. Interestingly, the authors propose the term
‘hypothetically disrupted coding DNA sequences’ to
replace the ambiguous ‘pseudogene’ term.
The issue of properly annotating pseudogene-derived
sequences has been underestimated: an extensive survey
of 64 diverse prokaryotic genomes in 2004 estimated a
background level of between 1% and 5% of pseudogene-
derived sequences, and identified 7000 shared pseudo-
genes between genomes [20]. Release 22 of the Ensembl
Bacteria database now contains over 10 000 bacterial
genomes [32], and with the increase in speed of whole
genome sequencing, the number of available bacterial
genomes for comparison is rapidly expanding.
Clearly, high-throughput genomic data analyses are reliant
upon high quality, finished genomes, with correspondingly
accurate and thorough annotations. Disappointingly, there
is little support for submitting updated annotations to
public databases, although ‘wiki’ based solutions have
been suggested as a potential avenue to address this
problem [61]. Producing and curating comprehensive,
open-source databases of complete bacterial genomes,
their annotations — including pseudogenes — and
associated ‘omics data would serve as an unprecedented
resource for evolutionary biology. Without the ability to
constantly curate ‘re-annotations’ or provide ‘notes’ for
these genomes, we risk such information being lost or
sub-standard genomic information being used.
Systematic analyses combining genomics, transcriptomics
and proteomics can produce comprehensive annotations
and making available such datasets for wider comparison is
of fundamental importance of addressing various prokar-
yotic genomic and evolutionary questions. Initially, high-
quality, broad and deep bacterial genome sequencing can
be rapidly achieved by the latest DNA sequencing tech-
nologies (e.g. PacBio [4,5,62]). The presence of active
transcription and translation in annotated pseudogenes
could be determined by RNAseq [63], including transcrip-
tion start sites, operon structure, promoters, and antisense
regulation (e.g. [64]). Combining computational predic-
tions of gene content with deep proteomic profile using
mass spectrometry, for instance, has been shown to
improve the accuracy of annotations [54]. Further studies
could reveal the presence of any functional activity that
remains selectively important, implicating reasons for
pseudogene persistence, or conversely, any post-transcrip-
tional regulation mechanisms that remove any residual
pseudogene products that may be damaging.

Current Opinion in Microbiology 2015, 23:102–109 www.sciencedirect.com


Acknowledgements
The authors are grateful to the BBSRC for funding to ACD (Grant
BBJ017698/1) and to the reviewers and editors for useful comments for
improving this manuscript.
References and recommended reading
Papers of particular interest, published within the period of review,
have been highlighted as:
 of special interest
 of outstanding interest
1. Hou Y, Lin S: Distinct gene number-genome size relationships
for eukaryotes and non-eukaryotes: gene content estimation
for dinoflagellate genomes. PLoS ONE 2009, 4:e6978 http://
dx.doi.org/10.1371/journal.pone.0006978.
2. Mira A, Ochman H, Moran NA: Deletional bias and the evolution
of bacterial genomes. Trends Genet 2001, 17:589-596 http://
dx.doi.org/10.1016/S0168-9525(01)02447-7.
3. Andersson JO, Andersson SG: Insights into the evolutionary
process of genome degradation. Curr Opin Genet Dev 1999,
9:664-671 http://dx.doi.org/10.1016/S0959-437X(99)00024-6.
4. Koren S, Phillippy AM: One chromosome, one contig: complete
microbial genomes from long-read sequencing and assembly.
Curr Opin Microbiol 2015, 23 in this issue.
5. Loman NJ, Constantinidou C, Chan JZM, Halachev M,
 Sergeant M, Penn CW et al.: High-throughput bacterial genome
sequencing: an embarrassment of choice, a world of
opportunity. Nat Rev Microbiol 2012, 10:599-606 http://
dx.doi.org/10.1038/nrmicro2850.
Second-generation and third-generation sequencing technologies are
revolutionising microbial genomics. We are now in the position where
novel bacterial strains can be fully sequenced in hours or days. This
review excellently summarises the current state-of-play and the future
direction of NGS in relation to microbiology.
6. Ochman H, Lawrence JG, Groisman EA: Lateral gene transfer
and the nature of bacterial innovation. Nature 2000, 405:299-
304 http://dx.doi.org/10.1038/35012500.
7. Baker S, Hardy J, Sanderson KE, Quail M, Goodhead I,
Kingsley RA et al.: A novel linear plasmid mediates flagellar
variation in Salmonella Typhi. PLoS Pathog 2007, 3:e59.
8. Stokes HW, Hall RM: A novel family of potentially mobile DNA
elements encoding site-specific gene-integration functions:
integrons. Mol Microbiol 1989, 3:1669-1683 http://dx.doi.org/
10.1111/j.1365-2958.1989.tb00153.x.
9. Ramage HR, Connolly LE, Cox JS: Comprehensive functional
analysis of Mycobacterium tuberculosis toxin–antitoxin
systems: implications for pathogenesis, stress responses,
and evolution. PLoS Genet 2009, 5:e1000767 http://dx.doi.org/
10.1371/journal.pgen.1000767.
10. Piel J, Hofer I, Hui D: Evidence for a symbiosis island involved in
horizontal acquisition of pederin biosynthetic capabilities by
the bacterial symbiont of Paederus fuscipes Beetles. J
Bacteriol 2004, 186:1280-1286 http://dx.doi.org/10.1128/
JB.186.5.1280-1286.2004.
11. Darmon E, Leach DRF: Bacterial genome instability. Microbiol
Mol Biol Rev 2014, 78:1-39 http://dx.doi.org/10.1128/
MMBR00035-13.
12. Mizrahi V, Dawes SS, Rubin H: DNA replication. Mol Genet
Mycobact 2000:159-172.
13. Cole ST, Eiglmeier K, Parkhill J, James KD, Thomson NR,
Wheeler PR et al.: Massive gene decay in the leprosy bacillus.
Nature 2001, 409:1007-1011 http://dx.doi.org/10.1038/35059006.
14. Moran NA, Mira A: The process of genome shrinkage in the
obligate symbiont Buchnera aphidicola. Genome Biol 2001,
2:1-12 http://dx.doi.org/10.1186/gb-2001-2-12-research0054.
15. Moran NA: Accelerated evolution and Muller’s rachet in
endosymbiotic bacteria. Proc Natl Acad Sci U S A 1996,
93:2873-2878 http://dx.doi.org/10.1073/pnas.93.7.2873.
www.sciencedirect.com
The open problem of pseudogenes Goodhead and Darby 107
16. Feavers IM, Maiden MC: A gonococcal porA pseudogene:
implications for understanding the evolution and
pathogenicity of Neisseria gonorrhoeae. Mol Microbiol 1998,
30:647-656.
17. Dunbar HE, Wilson ACC, Ferguson NR, Moran NA: Aphid thermal
tolerance is governed by a point mutation in bacterial
symbionts. PLoS Biol 2007, 5:e96 http://dx.doi.org/10.1371/
journal.pbio.0050096.
18. Halling SM, Peterson-Burch BD, Bricker BJ, Zuerner RL, Qing Z,
Li L-L et al.: Completion of the genome sequence of Brucella
abortus and comparison to the highly similar genomes of
Brucella melitensis and Brucella suis. J Bacteriol 2005,
187:2715-2726 http://dx.doi.org/10.1128/JB.187.8.2715-
2726.2005.
19. Jacq C, Miller JR, Brownlee GG: A pseudogene structure in 5S
DNA of Xenopus laevis. Cell 1977, 12:109-120 http://dx.doi.org/
10.1016/0092-8674(77)90189-1.
20. Liu Y, Harrison PM, Kunin V, Gerstein M: Comprehensive
 analysis of pseudogenes in prokaryotes: widespread gene
decay and failure of putative horizontally transferred genes.
Genome Biol 2004, 5:R64 http://dx.doi.org/10.1186/gb-2004-5-9-r64.
Pseudogenes were once thought of as being rare in bacteria, as their
genomes are generally efficient in their coding capacity. This study com-
prehensively analysed 64 annotated prokaryotic genomes, and suggest
that, actually, pseudogenes may occur at a background level of 1–5% of all
genes, the majority arising from failed horizontal gene transfer events.
21. Lawrence JG, Hendrix RW, Casjens S: Where are the
pseudogenes in bacterial genomes? Trends Microbiol 2001,
9:535-540 http://dx.doi.org/10.1016/S0966-842X(01)02198-9.
22. Mighell AJ, Smith NR, Robinson PA, Markham AF: Vertebrate
pseudogenes. FEBS Lett 2000, 468:109-114 http://dx.doi.org/
10.1016/S0014-5793(00)01199-6.
23. Khachane AN, Harrison PM: Strong association between
pseudogenization mechanisms and gene sequence length.
Biol Direct 2009, 4:38 http://dx.doi.org/10.1186/1745-6150-4-38.
24. Toh H, Weiss BL, Perkin SAH, Yamashita A, Oshima K, Hattori M
et al.: Massive genome erosion and functional adaptations
provide insights into the symbiotic lifestyle of Sodalis
glossinidius in the tsetse host. Genome Res 2006 http://
dx.doi.org/10.1101/gr.4106106.
25. Belda E, Silva FJ, Peretó J, Moya A: Metabolic networks of
Sodalis glossinidius: a systems biology approach to reductive
evolution. PLoS ONE 2012, 7:e30652 http://dx.doi.org/10.1371/
journal.pone.0030652.
26. Akman L, Yamashita A, Watanabe H, Oshima K, Shiba T, Hattori M
et al.: Genome sequence of the endocellular obligate symbiont
of tsetse flies, Wigglesworthia glossinidia. Nat Genet 2002,
32:402-407 http://dx.doi.org/10.1038/ng986.
27. Shigenobu S, Watanabe H, Hattori M, Sakaki Y, Ishikawa H:
Genome sequence of the endocellular bacterial symbiont of
aphids Buchnera sp. APS. Nature 2000, 407:81-86 http://
dx.doi.org/10.1038/35024074.
28. Gray MW, Burger G, Lang BF: The origin and early evolution of
mitochondria. Genome Biol 2001, 2:1-5 http://dx.doi.org/
10.1186/gb-2001-2-6-reviews1018.
29. Andersson SG, Zomorodipour A, Andersson JO, Sicheritz-
Pontén T, Alsmark UC, Podowski RM et al.: The genome
sequence of Rickettsia prowazekii and the origin of
mitochondria. Nature 1998, 396:133-140 http://dx.doi.org/
10.1038/24094.
30. Emelyanov VV: Evolutionary relationship of Rickettsiae and
mitochondria. FEBS Lett 2001, 501:11-18 http://dx.doi.org/
10.1016/S0014-5793(01)02618-7.
31. Fuxelius H-H, Darby AC, Cho N-H, Andersson SGE: Visualization
of pseudogenes in intracellular bacteria reveals the different
tracks to gene destruction. Genome Biol 2008, 9:R42 http://
dx.doi.org/10.1186/gb-2008-9-2-r42.
32. Flicek P, Amode MR, Barrell D, Beal K, Billis K, Brent S et al.:
Ensembl 2014. Nucleic Acids Res 2014, 42:D749-D755 http://
dx.doi.org/10.1093/nar/gkt1196.
Current Opinion in Microbiology 2015, 23:102–109
108 Genomics
33. Wong T-Y, Fernandes S, Sankhon N, Leong PP, Kuo J, Liu J-K:
Role of premature stop codons in bacterial evolution. J
Bacteriol 2008, 190:6718-6725 http://dx.doi.org/10.1128/
JB00682-08.
34. Belda E, Moya A, Bentley S, Silva FJ: Mobile genetic element
proliferation and gene inactivation impact over the genome
structure and metabolic capabilities of Sodalis glossinidius,
the secondary endosymbiont of tsetse flies. BMC Genomics
2010:1-17.
35. Kuo C-H, Ochman H: The extinction dynamics of bacterial
pseudogenes. PLoS Genet 2010, 6:e1001050 http://dx.doi.org/
10.1371/journal.pgen.1001050.
36. Chain PSG, Carniel E, Larimer FW, Lamerdin J, Stoutland PO,
Regala WM et al.: Insights into the evolution of Yersinia pestis
through whole-genome comparison with Yersinia
pseudotuberculosis. Proc Natl Acad Sci U S A 2004, 101:13826-
13831 http://dx.doi.org/10.1073/pnas.0404012101.
37. Feng Y, Chen Z, Liu S-L: Gene decay in Shigella as an incipient
stage of host-adaptation. PLoS ONE 2011, 6:e27754 http://
dx.doi.org/10.1371/journal.pone.0027754.
38. Blanc G, Ogata H, Robert C, Audic S, Suhre K, Vestris G et al.:
Reductive genome evolution from the mother of Rickettsia.
PLoS Genet 2007, 3:e14 http://dx.doi.org/10.1371/
journal.pgen.0030014.
39. Baumler AJ, Tsolis RM, Ficht TA, Adams LG: Evolution of host
adaptation in Salmonella enterica. Infect Immun 1998, 66:
4579-4587.
40. Parkhill J, Dougan G, James KD, Thomson NR, Pickard D, Wain J
et al.: Complete genome sequence of a multiple drug resistant
Salmonella enterica serovar Typhi CT18. Nature 2001, 413:848-
852 http://dx.doi.org/10.1038/35101607.
41. Thomson NR, Clayton DJ, Windhorst D, Vernikos G, Davidson S,
Churcher C et al.: Comparative genome analysis of Salmonella
Enteritidis PT4 and Salmonella Gallinarum 287/91 provides
insights into evolutionary and host adaptation pathways.
Genome Res 2008, 18:1624-1637 http://dx.doi.org/10.1101/
gr.077404.108.
42. Holt KE, Thomson NR, Wain J, Langridge GC, Hasan R, Bhutta ZA
et al.: Pseudogene accumulation in the evolutionary histories
of Salmonella enterica serovars Paratyphi A and Typhi. BMC
Genomics 2009, 10:36 http://dx.doi.org/10.1186/1471-
2164-10-36.
43. Bronowski C, Fookes MC, Gilderthorp R, Ashelford KE, Harris SR,
Phiri A et al.: Genomic characterisation of invasive non-
typhoidal Salmonella enterica subspecies enterica Serovar
Bovismorbificans isolates from Malawi. PLoS Negl Trop Dis
2013, 7:e2557 http://dx.doi.org/10.1371/journal.pntd.0002557.
44. Lerat E, Ochman H: Recognizing the pseudogenes in bacterial
genomes. Nucleic Acids Res 2005, 33:3125-3132.
45. Van Leuven JT, Meister RC, Simon C, McCutcheon JP: Sympatric
speciation in a bacterial endosymbiont results in two genomes
with the functionality of one. Cell 2014, 158:1270-1280 http://
dx.doi.org/10.1016/j.cell.2014.07.047.
46. Mira A, Pushker R: The silencing of pseudogenes. Mol Biol Evol
2005, 22:2135-2138 http://dx.doi.org/10.1093/molbev/msi209.
47. Tamas I, Wernegreen JJ, Nystedt B, Kauppinen SN, Darby AC,
 Gomez-Valero L et al.: Endosymbiont gene functions impaired
and rescued by polymerase infidelity at poly(A) tracts. Proc
Natl Acad Sci U S A 2008, 105:14934-14939 http://dx.doi.org/
10.1073/pnas.0806554105.
Long-term obligate symbionts, with degraded %AT-rich genomes have
an over-representation of poly-A tracts within pseudogenes, silencing
their function. Despite this, this study shows that such function can be
rescued through polymerase infidelity. Such symbionts also express a
chaperonin at very high-levels which may counteract problems with
incorrect protein folding that results from polymerase slippage within
AT-rich genes.
48. The ENCODE Project Consortium: The ENCODE (ENCyclopedia
of DNA Elements) project. Science 2004, 306:636-640 http://
dx.doi.org/10.1126/science.1105136.
Current Opinion in Microbiology 2015, 23:102–109
49. Sasidharan R, Gerstein M: Genomics: protein fossils live on as
RNA. Nature 2008, 453:729-731 http://dx.doi.org/10.1038/
453729a.
50. Suzuki K, Nakata N, Bang PD, Ishii N, Makino M: High-level
expression of pseudogenes in Mycobacterium leprae. FEMS
Microbiol Lett 2006, 259:208-214 http://dx.doi.org/10.1111/
j.1574-6968.2006.00276.x.
51. Williams DL, Slayden RA, Amin A, Martinez AN, Pittman TL, Mira A
et al.: Implications of high level pseudogene transcription in
Mycobacterium leprae. BMC Genomics 2009, 10:397 http://
dx.doi.org/10.1186/1471-2164-10-397.
52. Marques MAM, Neves-Ferreira AGC, da Silveira EKX, Valente RH,
Chapeaurouge A, Perales J et al.: Deciphering the proteomic
profile of Mycobacterium leprae cell envelope. Proteomics
2008, 8:2477-2491 http://dx.doi.org/10.1002/pmic.200700971.
53. Perkins TT, a Kingsley R, Fookes MC, Gardner PP, James KD, Yu L
et al.: A strand-specific RNA-Seq analysis of the transcriptome
of the typhoid bacillus Salmonella typhi. PLoS Genet 2009,
5:e1000569 http://dx.doi.org/10.1371/journal.pgen.1000569.
54. Schrimpe-Rutledge AC, Jones MB, Chauhan S, Purvine SO,
 Sanford JA, Monroe ME et al.: Comparative omics-driven
genome annotation refinement: application across yersiniae.
PLoS ONE 2012, 7 http://dx.doi.org/10.1371/
journal.pone.0033903.
A study that uses a thorough proteomic profile to assess the full comple-
ment of coding genes within three Yersinia genomes. Not only do they find
a number of novel genes, but also proteins being translated from pre-
viously uncharacterised pseudogenes.
55. Seemann T: Prokka: rapid prokaryotic genome annotation.
Bioinformatics 2014:btu153 http://dx.doi.org/10.1093/
bioinformatics/btu153.
56. Altschul S: Gapped BLAST and PSI-BLAST: a new generation of
protein database search programs. Nucleic Acids Res 1997,
25:3389-3402 http://dx.doi.org/10.1093/nar/25.17.3389.
57. Blattner FR: The complete genome sequence of Escherichia
coli K-12. Science 1997, 80-.(277):1453-1462 http://dx.doi.org/
10.1126/science.277.5331.1453.
58. Ochman H, Davalos LM: The nature and dynamics of bacterial
 genomes. Science 2006, 311:1730-1733 http://dx.doi.org/
10.1126/science.1119966.
The original E. coli genome was annotated as having a single pseudo-
gene. This study shows that actually there were many more degraded
sequences than originally predicted. Understanding the provenance of
such degraded sequences is essential to elucidating biological and
evolutionary processes.
59. Brittnacher MJ, Fong C, Hayden HS, Jacobs MA, Radey M,
Rohmer L: PGAT: a multistrain analysis resource for microbial
genomes. Bioinformatics 2011, 27:2429-2430 http://dx.doi.org/
10.1093/bioinformatics/btr418.
60. Markowitz VM, Chen IMA, Palaniappan K, Chu K, Szeto E, Pillay M
et al.: IMG 4 version of the integrated microbial genomes
comparative analysis system. Nucleic Acids Res 2014, 42 http://
dx.doi.org/10.1093/nar/gkt963.
61. Salzberg SL: Genome re-annotation: a wiki solution? Genome
 Biol 2007, 8:102 http://dx.doi.org/10.1186/gb-2007-8-1-102.
Annotations are rarely updated, and the support for making such
resources available is lacking. This opinion article is an excellent summary
of the pitfalls on relying on GenBank (or similar) databases and empha-
sises the utility of updating, making available, and actively curating our
current knowledge of bacterial genomes.
62. Koren S, Harhay GP, Smith TPL, Bono JL, Harhay DM, Mcvey SD
et al.: Reducing assembly complexity of microbial genomes
with single-molecule sequencing. Genome Biol 2013, 14:R101
http://dx.doi.org/10.1186/gb-2013-14-9-r101.
63. Wilhelm BT, Marguerat S, Goodhead I, Bähler J: Defining
transcribed regions using RNA-seq. Nat Protoc 2010, 5:255-266
http://dx.doi.org/10.1038/nprot.2009.229.
64. Kröger C, Dillon SC, Cameron ADS, Papenfort K, Sivasankaran SK,
 Hokamp K et al.: The transcriptional landscape and small RNAs
of Salmonella enterica serovar Typhimurium. Proc Natl Acad
www.sciencedirect.com

Sci U S A 2012, 109:E1277-E1286 http://dx.doi.org/10.1073/
pnas.1201061109.
This study utilises RNAseq to fully characterise the intricate transcrip-
tional landscape of an important intracellular bacterium, including operon
structure, multiple transcriptional start sites, regulators and small RNAs.
65. Nuccio SP, Bäumler AJ: Comparative analysis of Salmonella
 genomes identifies a metabolic network for escalating growth
in the inflamed gut. MBio 2014, 5 http://dx.doi.org/10.1128/
mBio.00929-14.
This study found a novel signature of gastrointestinal pathogenesis in
Salmonella, that had been missed due to inconsistencies in the available
www.sciencedirect.com
The open problem of pseudogenes Goodhead and Darby 109
‘pseudogene’ annotations. It includes a thorough reannotation of fifteen
Salmonella genomes in the Supplementary Information.
66. Fraser CM, Gocayne JD, White O, Adams MD, Clayton RA,
Fleischmann RD et al.: The minimal gene complement of
Mycoplasma genitalium. Science (80–) 1995, 270:397-404 http://
dx.doi.org/10.1126/science.270.5235.397.
67. Gil R, Sabater-Muñoz B, Latorre A, Silva FJ, Moya A: Extreme
genome reduction in Buchnera spp.: toward the minimal
genome needed for symbiotic life. Proc Natl Acad Sci U S A
2002, 99:4454-4458 http://dx.doi.org/10.1073/pnas.062067299.
Current Opinion in Microbiology 2015, 23:102–109