|MCRM algorithm; Genomic methylation; Helicobacter spp.; Bioinformatics; Restriction and modification;
|MCRM: Minimum Common Restriction and
Modification; REase: Restriction Endonuclease; MTase:
Methyltransferase; RM: Restriction and Modification.
Helicobacter pylori colonizes the stomach of about half
of the human population and is associated with several disease
outcomes, like gastritis, peptic ulcer and gastric cancer
(Dunn et al., 1997; Kusters et al., 2006). Other similar
spiral bacteria have been isolated from several animals, like
cat, dog or mice, among others. Helicobacter species can
be subdivided into gastric Helicobacter species and
enterohepatic (nongastric) Helicobacter species. The two
lineages demonstrate a high level of organ specificity, such
that gastric Helicobacter spp. in general does not colonize
the intestine or liver, and vice versa (Kusters et al., 2006).
The genomic DNA of H. pylori is characterised by the
presence of a high unusual number of restriction and modification
(R-M) system (Nobusato et al., 2000; Lin et al.,
2001; Takata et al., 2002). The type II R-M systems are
composed by least two genes: one coding for a restriction
endonuclease (REase) that recognizes a specific DNA sequence
and cuts both strands; and other gene coding for a
cognate MTase that methylates the same DNA sequence,
thus protecting the genomic DNA from being cleaved by
the companion REase (Roberts et al., 2003). Type II R-M
systems have been referred as selfish genetic elements,
because the descendants of cells that had lost these genes
appeared unable to modify a sufficient number of recognition
sites in their chromosomes to protect them from lethal
attack by the remaining restriction enzyme molecules (Naito
et al., 1995).
|Recently we have demonstrated that the diversity of RM
systems among H. pylori strains is high enough to be
used as a typing method (Vale and Vitor, 2007). Cluster
analysis by conventional methods does not consider the
propensity for R-M systems conservation after acquisition,
due to the “selfish behavior” (Naito et al., 1995). Considering
this we have recently developed a new clustering algorithm
[Minimum Common Restriction Modification (MCRM)
algorithm] that takes into account the pressure of REases
on MTases, and that is based on the hypothesis that each
strain evolves by acquiring new RM systems without loosing
acquired RM systems (Vale et al., 2008). In this algorithm it
is considered that: i) the strain with less RM systems is the
one that has the core set of the most abundant R-M systems
expressed among the typed strains; ii) these core set of RM
systems was the first to be acquired by H. pylori, so that
they exhibit a large dissemination (expression) among several
(Vale et al., 2008). MCRM analysis of the
genomic methylation data from H. pylori strains isolated
from different geographic revealed a clustering according
to strain’s continent of origin (Vale et al., 2008), which is in
agreement with H. pylori coevolution with its human host
(Covacci et al., 1999; Linz et al., 2007; Vale et al., 2008).
This observation led to the suggestion that R-M systems
may trace H. pylori geographic distribution and, by default
also its human host migrations (Vale et al., 2008).
In this study it was investigated if non-pylori Helicobacter strains also have a high number expressed
MTases and, if the genomic methylation status followed be
MCRM algorithm analysis permitted to discriminate between
H. pylori and non-pylori Helicobacter spp. (H.
canadensis, H. canis, H. felis, H. mustelae and H.
pullorum). Following, these results were compared with
the phylogenetic analysis of 16S rRNA gene sequences for
the same species. To our knowledge this is the first study
that systematically analysis the diversity of expression of
R-M systems in H. canadensis, H. canis, H. felis, H.
mustelae and H. pullorum.
|Material and Methods
H. pylori strains (26695 and J99) were cultured on H. pylori selective agar (Wilkins-Chalgren agar supplemented with
10% horse blood, vancomycin [10 mg liter-1], cefsulodin [5
mg liter-1], trimethoprim [5 mg liter-1], and cycloheximide
[100 mg liter-1] [Biogerm, Porto, Portugal]) and incubated
at 37°C for 48 h in an anaerobic jar (Oxoid, UK; or BBL, USA) with a gas generator system (CampyGen; Oxoid, UK)
(Megraud, 1996). Non-pylori Helicobacter spp. were
cultured on Muller Hinton agar (Oxoid, UK) supplemented
with 10% (v/v) defibrinated horse blood (Probiologica,
Portugal) and incubated in similar conditions. Genomic DNA
was extracted by standard methods.
|Helicobacter spp. R-M Systems Diversity
|To evaluate the expression of the cognate
methyltransferase, the genomic DNA was digested with 27
REases [AciI, AseI, BseRI, BssHII, BstUI, DdeI, DpnI,
DpnII, DraI, EagI, FauI, Fnu4HI, FokI, HaeIII, HhaI,
Hpy188I, Hpy188III, Hpy99I, HpyCH4III, HpyCH4IV,
HpyCH4V, MspI, NaeI, NlaIII, Sau96I, ScrFI, and TaqI
(New England Biolabs, USA)]. The results were coded as“0” for digestion observed (DNA is unmethylated), and “1”
for absence of digestion, suggesting an active
methyltransferase (Vale and Vitor, 2007).
|Genomic Methylation Status Comparison
|The mean number of active MTases on non-pylori Helicobacter spp. was compared with: i) the mean number
of expressed MTases of 221 H. pylori tested by us; ii) the
mean number of M genes predicted by REBASE for all
(Roberts et al., 2007). The Kruskal-
Wallis test was performed using the statistical package SPSS
v.15 (SPSS Inc., Chicago, IL).
|Ribosomal RNA Alignment
|16S rRNA sequences available on public data bases from
the Helicobacter spp. (table 1) were aligned using ClustalW,
producing a cladogram (Chenna et al., 2003).
Several dendrograms were produced using MCRM
algorithm after genomic methylation analysis of 7 Helicobacter spp. Most of the dendrograms produced by
MCRM algorithm are indeed similar, but it is possible that
distinct dendrograms are generated as different choices of
strain or R-M system at ties may result in different clustering.
Thus, 10 different dendrograms were produced from the
same data in order to increase the confidence on the
|Helicobacter spp. Genomic Methylation
After genomic DNA hydrolysis with the selected REases
it was observed that among non-pylori Helicobacter spp.
the mean number of expressed MTases was 8 (SD=2.4). A similar analysis for 221 H. pylori strains (Vale et al.
unpublished results) revealed a mean of 17 active MTases
(SD=3.4). The mean number of genes coding for
methyltransferases for the overall 862 sequenced genomes
is 4.2 (SD=5.0) [data from REBASE (Roberts et al., 2007)].
Table 1 resumes the number of active methyltransferases
in tested Helicobacter spp.
A significant statistical difference between mean number
of active MTases from non-pylori Helicobacter spp. and H. pylori (p<0,001) and also between non-pylori Helicobacter spp. and the overall sequenced genomes
(p=0,005) was verified. Present study results showed that
the number of expressed MTases in a decreasing order by
organism, or group of organisms, was: H. pylori, non-pylori Helicobacter and all sequenced bacteria available at
REBASE (Roberts et al., 2007).
|Helicobacter spp. 16S rRNA Cladogram
Construction of a cladogram after multiple sequence
alignment using ClustalW (Chenna et al., 2003) 16S rRNA
sequences (from Helicobacter spp.) available on public
databases clearly isolated H. pylori species (Figure 1). As
expected the H. pylori sequenced strains (26695, J99,
HPAG1 and Shi470) presented a similarity level =98%, and
clustered together (>97% defines a species). Moreover,
Helicobacter gastric and enterohepatic species appear to
be in different clusters, according to previous work
(Dewhirst et al., 2005). The similarity levels among
Helicobacter spp. based on 16S rRNA is presented in table 2.
|MCRM Clustering Analysis
After genomic DNA hydrolysis with the selected REases
the codified data were analysed using MCRM algorithm
(Vale et al., 2008). The Simpson index of diversity, which
reflects the capacity of the method to distinguish unrelated
strains, was 100%
(Hunter and Gaston, 1988). The produced
dendrogram is present in Figure 2. Surprisingly, this analysis
based on the genomic methylation status clearly isolated H.
pylori species from non-pylori Helicobacter spp., as it was
observed when the analysis focuses on the 16S rRNA. Out
of 10 produced dendrograms, 60% cluster H. pylori together
and 40% of the dendrograms also discriminate between H.
pylori and non-pylori Helicobacter spp. All of these last
mentioned dendrograms gathered H. pylori strains (data
not shown). H. pylori was discriminated from non-pylori Helicobacter spp. at k/nM=0.04 (where, k=1 and nM=27,
i.e. one MTase common to all Helicobacter species used).
This MTase common to all tested Helicobacter spp. was
M.NaeI (table 3).
When comparing the number of MTases expressed in
each Helicobacter spp. it was verified that H. pylori expresses a number of MTases higher than tested non-pylori Helicobacter spp. (p<0,001) and, that non-pylori Helicobacter spp. expresses a number higher than all
sequenced bacteria analysed by REBASE (p=0,005). This
analysis reveals that the increased number of MTases genes expressed is probably a characteristic of Helicobacter genus
and not only of H. pylori. To our knowledge the evaluation
of the expressed MTases in non-pylori Helicobacter spp.
has only been referred for the sequenced genomes of H.
acinonychis Sheeba (Eppinger et al., 2006) and H.
hepaticus ATCC 51449 (Suerbaum et al., 2003) with 29
and 8 M genes, according to REBASE (Roberts et al., 2007),
respectively. Besides this analysis only the sequence GATC
has been screened for methylation in H. mustelae (Edmonds
et al., 1992). Present study and Edmonds et al. study (Edmonds et al., 1992) found that the GATC methylation is
absent in H. mustelae (table3).
|The only MTase that it was found to be expressed among
all tested Helicobacter spp. is M.NaeI (table 3). Previously
we have reported that this MTase is probably conserved in
all H. pylori strains (Vale and Vitor, 2007; Vale et al., 2008).
In order to confirm if this MTase is common to the Helicobacter genus, other non-pylori Helicobacter spp.
should be included in the study, and for each species several
stains should be characterized.
Analysis of 16S rRNA gene sequences has become the
primary method for determining prokaryotic phylogeny,
which is the current basis for prokaryotic systematics.
Although it has been described that Helicobacter is
susceptible to horizontal transfer of 16S rRNA gene so that it can be misleading in Helicobacter spp. identification
(Vandamme et al., 2000; Dewhirst et al., 2005), the 16S
rRNA cladogram clearly discriminate the Helicobacter spp.
used in the present study as expected. Moreover, figure 1
presents gastric Helicobacter spp. and enterohepatic Helicobacter spp. in different clusters, as described
(Dewhirst et al., 2005). However, the cladogram
obtained from 60 kDa heat-shock protein (HSP60), referred
as better marker for Helicobacter species phylogeny
(Mikkonen et al., 2004) was similar to the one obtained with
16S rRNA (data not shown).
A surprising result was the capacity of the genomic
methylation and of MCRM algorithm to cluster separately
gastric Helicobacter spp. and enterohepatic Helicobacter spp. in 40% of the produced dendrograms. Also this
methodology discriminate H. pylori from non-pylori Helicobacter spp. in 60% of the produced dendrograms.
When genomic methylation and MCRM analysis is
compared with the current phylogeny approach the results
are remarkably similar. The genomic methylation status
appears to be an interesting new tool to characterize
genomes with a high number of MTases expressed. It has
been described that R-M systems are subject to horizontal
transfer (Jeltsch and Pingoud, 1996; Gressmann et al., 2005).
The separation between gastric Helicobacter spp. and
enterohepatic Helicobacter spp. suggests that probably theses species, which have a high level of organ specificity
(Kusters et al., 2006) may have access to different sets of
R-M systems through horizontal gene transfer. The horizontal
gene transfer might occur only in ideal conditions provided
by the specific tissue environment characteristic of each
species. This could justify the presence of gastric Helicobacter spp. and enterohepatic Helicobacter spp. in
different clusters. Similarly, H. pylori unique reservoir may justify the presence of the two tested strains in a different
cluster. Finally, the results suggest that possibly some R-M
systems are not as mobile as previously described, or are
not available for horizontal transfer due to the isolation
provided by the human (or animal) reservoir of each species,
because the genomic methylation analysis permits to
discriminate among Helicobacter spp. The R-M systems
do not appear to be spread in a miscellaneous manner.
Indeed, a blast analysis (Zhang et al., 2000) of all
methyltransferases predicted by REBASE (Roberts et al.,
2007) for the recently sequenced H. pylori strain Shi470
clearly shows that most methyltransferases are identical to
other H. pylori sequenced strains and to H. acinonychis strain Sheeba (table 4, supplementary material). It is clear
from analysis of table 4 that most of the methyltransferases
of the recent sequenced H. pylori strain Shi470 are identical
to the other H. pylori sequenced strains and H. acinonychis strain Sheeba, but none is observed in H. hepaticus sequenced strain. We postulate that similarly to H. pylori, non-pylori Helicobacter may also present a high diversity
of MTases expressed which could be used for strain typing,
but this still needs to be confirmed with further investigation.
R-M systems probably play an important role in Helicobacter genus biology that has not been ascertained, yet.
|In conclusion, it was observed a high number of MTases
expressed in non-pylori Helicobacter spp. as it was
previously determined for H. pylori. The discrimination of
Helicobacter species by the dendrogram produced with
MCRM algorithm and by 16S rRNA alignment performed
in a similar way. The results suggest that some R-M systems
do not appear to be spread in a miscellaneous manner, once
genomic methylation analysis allows discrimination among
Helicobacter spp. Future work should include increasing
the number of Helicobacter species analysed and also the
number of tested strains from each species, in order to
confirm present study results.
|We thank Lurdes Monteiro, Francis Mégraud, Jay Solnick
and, Nuno Azevedo for the Helicobacter spp. strains. This
work was partially supported by New England Biolabs, Inc.
Chenna R,Sugawara H, Koike T, Lopez R, Gibson TJ, et al. (2003) Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res 31: 3497-3500. » CrossRef » PubMed » Google Scholar
Covacci A, Telford JL, Del Giudice G, Parsonnet J,
Rappuoli R (1999) Helicobacter pylori virulence and genetic geography. Science 284: 1328-1333. » CrossRef » PubMed » Google Scholar
Dewhirst FE, Shen Z, Scimeca MS, Stokes LN,
Boumenna T, et al. (2005) Discordant 16S and 23S
rRNA gene phylogenies for the genus Helicobacter: implications for phylogenetic inference and systematics. J Bacteriol 187: 6106-6118. » CrossRef » PubMed » Google Scholar
Dunn BE, Cohen H, Blaser MJ (1997) Helicobacter pylori. Clin Microbiol Rev 10: 720-741. » CrossRef » PubMed » Google Scholar
Edmonds P, Hall BM, Edwards WR, Hartline KM (1992) Presence of methylated adenine in GATC sequences in chromosomal DNAs from Campylobacter species. J Bacteriol 174: 8156-8157. » CrossRef » PubMed » Google Scholar
Eppinger M, Baar C, Linz B, Raddatz G, Lanz C, et al. (2006) Who ate whom? Adaptive
Helicobacter genomic changes that accompanied a host jump from early humans to large felines. PLoS Genet 2: e120. » CrossRef » PubMed » Google Scholar
Gressmann H, Linz B, Ghai R, Pleissner KP, Schlapbach
R, et al. (2005) Gain and loss of multiple genes during the evolution of Helicobacter pylori. PLoS Genet 1: e43. » CrossRef » PubMed » Google Scholar
Hunter PR, Gaston MA (1988) Numerical index
of the discriminatory ability of typing systems: an application
of Simpson’s index of diversity. J Clin Microbiol 26: 2465-2466. » CrossRef » PubMed » Google Scholar
Jeltsch A, Pingoud A (1996) Horizontal gene
transfer contributes to the wide distribution and evolution of type II restriction-modification systems. J Mol Evol 42: 91-96. » CrossRef » PubMed » Google Scholar
Kusters JG, van Vliet AH, Kuipers EJ (2006) Pathogenesis of Helicobacter pylori infection. Clin Microbiol Rev 19: 449-490. » CrossRef » PubMed » Google Scholar
Lin LF, Posfai J, Roberts RJ, Kong H (2001) Comparative genomics of the restriction-modification systems in Helicobacter pylori. Proc Natl Acad Sci USA 98: 2740-2745. » CrossRef » PubMed » Google Scholar
Linz B, Balloux F, Moodley Y, Manica A, Liu H, et al. (2007) An African origin for the intimate association between humans and Helicobacter pylori. Nature 445: 915-918. » CrossRef » PubMed » Google Scholar
Megraud F (1996) Diagnostic bactériologique standart
de l’infection à Helicobacter pylori. In Megraud F. and Lamouliatte H. (eds), Helicobacter pylori. Elsevier Amsterdam pp 249-266.
Mikkonen TP, Karenlampi RI, Hanninen ML
(2004) Phylogenetic analysis of gastric and enterohepatic Helicobacter species based on partial HSP60 gene sequences. Int J Syst Evol Microbiol 54: 753-758. » CrossRef » PubMed » Google Scholar
Naito T, Kusano K , Kobayashi I (1995) Selfish behavior of restriction-modification systems. Science 267: 897-899. » CrossRef » PubMed » Google Scholar
Nobusato A, Uchiyama I, Kobayashi I (2000) Diversity of restriction-modification gene homologues in Helicobacter pylori. Gene 259: 89-98. » CrossRef » PubMed » Google Scholar
Roberts RJ, Belfort M, Bestor T, Bhagwat AS, Bickle
TA, et al. (2003) A nomenclature for restriction enzymes, DNA methyltransferases, homing endonucleases and their genes. Nucleic Acids Res 31: 1805-1812. » CrossRef » PubMed » Google Scholar
Roberts RJ, Vincze T, Posfai J, Macelis D (2007) REBASE—enzymes and genes for DNA restriction and modification. Nucleic Acids Res 35: D269-D270. » CrossRef » PubMed » Google Scholar
Suerbaum S, Josenhans C, Sterzenbach T, Drescher B,
Brandt P, et al. (2003) The complete genome sequence of the carcinogenic bacterium Helicobacter hepaticus. Proc Natl Acad Sci USA 100: 7901-7906. » CrossRef » PubMed » Google Scholar
Takata T, Aras R, Tavakoli D, Ando T, Olivares AZ, et al. (2002) Phenotypic and genotypic variation
in methylases involved in type II restriction-modification systems in Helicobacter pylori. Nucleic Acids Res 30: 2444-2452. » CrossRef » PubMed » Google Scholar
Vale FF, Encarnacao P, Vitor JM (2008) A new algorithm for cluster analysis of genomic methylation: the Helicobacter pylori case. Bioinformatics 24: 383-388. » CrossRef » PubMed » Google Scholar
Vale FF, Vitor JM (2007) Genomic Methylation: a Tool for Typing Helicobacter pylori Isolates. Appl Environ Microbiol 73: 4243-4249. » CrossRef » PubMed » Google Scholar
Vandamme P, Harrington CS, Jalava K, On SL (2000) Misidentifying helicobacters: the Helicobacter cinaedi example. J Clin Microbiol 38: 2261-2266. » CrossRef » PubMed » Google Scholar
Zhang Z, Schwartz S, Wagner L, Miller W (2000) A greedy algorithm for aligning DNA sequences. J Comput Biol 7: 203-214. » CrossRef » PubMed » Google Scholar