Journal of Medical Microbiology (2007), 56, 102–109
DOI 10.1099/jmm.0.46616-0
Amplified fragment length polymorphism of
Streptococcus suis strains correlates with their
profile of virulence-associated genes and clinical
background
Thomas Rehm,1 Christoph G. Baums,1 Birgit Strommenger,2
Martin Beyerbach,3 Peter Valentin-Weigand1 and Ralph Goethe1
Correspondence
Christoph G. Baums
christoph.baums@gmx.de
1
Institut für Mikrobiologie, Zentrum für Infektionsmedizin, Stiftung Tierärztliche Hochschule
Hannover, D-30173 Hannover, Germany
2
Robert Koch Institut, Wernigerode Branch, D-38855 Wernigerode, Germany
3
Institut für Biometrie, Epidemiologie und Informationsverarbeitung, Stiftung Tierärztliche
Hochschule Hannover, D-30173 Hannover, Germany
Received 8 March 2006
Accepted 29 September 2006
Amplified fragment length polymorphism (AFLP) typing was applied to 116 Streptococcus suis
isolates with different clinical backgrounds (invasive/pneumonia/carrier/human) and with known
profiles of virulence-associated genes (cps1, -2, -7 and -9, as well as mrp, epf and sly). A
dendrogram was generated that allowed identification of two clusters (A and C) with different
subclusters (A1, A2, C1 and C2) and two heterogeneous groups of strains (B and D). For
comparison, three strains from each AFLP subcluster and group were subjected to multilocus
sequence typing (MLST) analysis. The closest relationship and lowest diversity were found for
patterns clustering within AFLP subcluster A1, which corresponded with sequence type (ST)
complex 1. Strains within subcluster A1 were mainly invasive cps1 and mrp+ epf+ (or epf*) sly+
cps2+ strains of porcine or human origin. A new finding of this study was the clustering of invasive
mrp* cps9 isolates within subcluster A2. MLST analysis suggested that A2 correlates with a
single ST complex (ST87). In contrast to A1 and A2, subclusters C1 and C2 contained mainly
pneumonia isolates of genotype cps7 or cps2 and epf” sly”. In conclusion, this study demonstrates
that AFLP allows identification of clusters of S. suis strains with clinical relevance.
INTRODUCTION
Streptococcus suis causes invasive diseases such as meningitis,
septicaemia, arthritis and polyserositis in piglets (MacInnes
& Desrosiers, 1999). This pathogen is responsible for
substantial economic losses in the swine industry (Staats
et al., 1997). Clinically healthy pigs are carriers of the
pathogen and are responsible for spreading disease (CliftonHadley et al., 1984). S. suis has also been associated with
meningitis and other diseases in humans (Arends & Zanen,
1988). In 2005, an unusual outbreak of S. suis infections with
a high mortality occurred among 215 humans in Sichuan,
China (Yu et al., 2006), underlining that the zoonotic
potential of S. suis might have been underestimated.
strains are worldwide the most prevalent in association with
disease in pigs and humans (Wisselink et al., 2000; Arends &
Zanen, 1988). In addition, high prevalences of serotype 9
(cps9) and 1 (cps1) have been observed among porcine
invasive isolates in central Europe and Great Britain,
respectively. Serotype 7 strains have frequently been
associated with pneumonia in Scandinavia and central
Europe (Wisselink et al., 2000; Tian et al., 2004; Aarestrup
et al., 1998).
At present, 33 serotypes have been identified in S. suis.
Serotype is determined by the gene cluster cps, which is in
serotype 2 (cps2) strains so far the only gene locus identified
to be essential for virulence (Smith et al., 1999). Serotype 2
The muramidase-released protein (MRP, mrp), the extracellular protein factor (EF, epf) and the suilysin (SLY, sly) are
putative virulence factors of S. suis (Vecht et al., 1991; Staats
et al., 1999). These factors contribute further to the diversity
of S. suis strains, since various mrp and epf genotypes have
been described (Smith et al., 1993; Silva et al., 2006). Serotype
2 strains, which express the 136 kDa MRP and the 110 kDa
EF, are highly virulent, in contrast to European serotype 2
isolates that lack these factors (Wisselink et al., 2000).
Abbreviations: AFLP, amplified fragment length polymorphism; MLST,
multilocus sequence typing; ST, sequence type.
In a study by King et al. (2002), 92 different multilocus
sequence types (STs) were identified for S. suis. The vast
102
Downloaded from www.microbiologyresearch.org by
46616 G 2007 SGM
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
Printed in Great Britain
AFLP of S. suis strains
majority of invasive isolates belonged to the ST1 complex. In
contrast, the complexes of ST27 and ST87 were found to
contain a higher proportion of lung isolates (King et al.,
2002). There was a high prevalence of serotype 2 strains
within the ST1 complex and of serotype 7 strains within the
ST27 complex. However, single serotypes were found to be
associated with multiple STs, indicating that serotypes are
not reliable as phylogenetic markers.
In the present study, clusters of S. suis strains in amplified
fragment length polymorphism (AFLP) typing were correlated with clinical background, profiles of virulenceassociated genes as well as multilocus sequence typing
(MLST) results.
moment correlation coefficient, and the dendrogram was calculated
by the unweighted pair-group method using average linkages
(UPGMA).
MLST analysis of S. suis strains. MLST was performed as
described by King et al. (2002), with the following modifications.
For a few samples, the PCR did not yield a product with the
described mutS primers. For these, primers mutS forward and mutS
reverse were replaced by mutS forwardnew (59-AAGCAGGCAGTCGGCGTGGT-39) and mutS reversenew (59-AGTACAAACTACCATGCTTC-39). The primer thrA forward was used as forward primer
as well as for subsequent sequencing reactions. Sequencing reactions
were performed using the DYEnamic ET Terminator kit and the
MegaBase capillary sequencer (Amersham Biosciences). MLST alleles
and resulting STs were assigned using BioNumerics software 4.0
based on the MLST scheme of King et al. (2002). Analysis of ST
complexes was performed with eBURST (www.mlst.net) (Feil et al.,
2004).
METHODS
Allele and ST assignment. Novel alleles and STs were assigned
Bacterial strains. A total of 106 porcine and 10 human S. suis iso-
through submission of the respective data to the S. suis MLST database (http://ssuis.mlst.net).
lates were analysed in this study. The majority of the porcine isolates
originated from northern Germany and had been characterized previously (Allgaier et al., 2001; Silva et al., 2006). Six of the human
strains were from Germany and four from Canada. Strains isolated
from affected organs of pigs or humans with meningitis, arthritis or
septicaemia were designated invasive isolates (n=58), isolates from
the lower respiratory tract of diseased animals were grouped as
pneumonia isolates (n=32), and isolates from the tonsils, and nasal
and vaginal swabs of healthy pigs were considered carrier isolates
(n=26). Reference strains for differentiation of cps types and virulence-associated factors as described by Silva et al. (2006) were also
included (Table 1). Bacteria were grown on sheep blood Columbia
agar or in Todd–Hewitt broth for 24 h at 37 uC.
Virulence-associated gene profiling of S. suis. DNA extraction
and typing of virulence-associated genes was done as described previously (Silva et al., 2006).
AFLP typing of S. suis. A modification of the AFLP protocol of
Gibson et al. (1998) was used. One microgram of HindIII-digested
streptococcal DNA was used in the ligation reaction containing
30 mM adapter oligonucleotides ADH1 (59-ACGGTATGCGACAG39) and ADH2 (59-AGCTCTGTCGCATACCGTGAG-39), 16 T4
DNA ligase buffer and 1 U T4 DNA ligase (both Promega) in a final
volume of 20 ml. Subsequently, PCR amplification was performed in
a final reaction volume of 50 ml containing the following components: 5 ml of a 1 : 5 dilution of ligated DNA (50 ng DNA), 1.5 mM
MgCl2, 1 mM primer HI-G (59-GGTATGCGACAGAGCTTG-39),
0.2 mM dNTPs (Invitrogen), 2.5 U Taq DNA polymerase
(Invitrogen), and PCR buffer provided by the manufacturer. Initial
denaturating was carried out for 4 min at 94 uC and followed by 33
cycles including denaturation for 1 min at 94 uC, annealing for
1 min at 60 uC and elongation for 2.5 min at 72 uC (final extension
for 5 min). The amplified fragments were separated by 2.5 % (w/v)
agarose gel electrophoresis. The primer HI-G was chosen in preliminary experiments because, in contrast to the other three primers
described by Gibson et al. (1998), amplification of at least eight
bands for each strain and appropriate differentiation of all tested
isolates was observed. Reproducibility of band patterns independent
of the DNA preparation was demonstrated. Reference strains were
included in every gel to verify inter-gel reproducibility. The inter-gel
similarity of the external reference strain was at least 88 % (including
different DNA preparations). AFLP patterns were analysed using
BioNumerics software 4.0 (Applied Maths). The pairwise comparison of band patterns was performed using the Pearson producthttp://jmm.sgmjournals.org
Statistical evaluation. Statistical analysis was performed to analyse
relationships between AFLP cluster and clinical background as well
as the profile of virulence-associated genes. The chi square test of
homogenicity (n >5) or the Fischer’s exact test (n <5) were used
for these comparisons. Probabilities lower than 0.05 were considered
significant.
RESULTS AND DISCUSSION
Cluster analysis based on AFLP pattern
Well-characterized S. suis strains were typed by AFLP to
generate a dendrogram that allowed correlation of AFLP
clusters with clinical background and the profile of
virulence-associated genes (Fig. 1, Table 1). In general,
substantial diversity of AFLP band patterns was observed
among S. suis strains (<15 % overall similarity level). At the
55 % similarity level, two main AFLP clusters, A and C, were
identified. The other strains were assigned either to group B
or group D (Fig. 1). The dendrogram suggested division of
cluster A into subclusters A1 and A2. Thirty isolates
clustered at a comparable high linkage level of 70 % within
subcluster A1. The patterns classified as group B were related
to those of cluster A with a linkage level between 42 and
54 %. Patterns within cluster C (n=28) were distinct, as
their relation to any of the remaining patterns was less
than 45 % similarity. Cluster C was further divided into
subclusters C1 and C2 at a linkage level of 57 %. An
extremely high diversity was found for the remaining strains
(group D). In this group, some similarity indices were
almost equal to those with the Streptococcus pyogenes
reference strain (Fig. 1).
Congruence between AFLP clusters and ST
complexes
To evaluate the AFLP-based clustering results, three isolates
each were selected randomly from A1, A2, B, C1, C2 and D,
and were subjected to MLST analysis (Table 2). One
additional reference strain (D282) investigated in this
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
103
T. Rehm and others
Table 1. Profile of virulence-associated genes and origin of strains investigated in AFLP
AFLP
cluster/group
A1
A2
B
104
Strain
AC 3940
I9841/3, 10203
A6169/1/96
Sw 132/B 2543
V 2154/3
I4627 C
I 9841/1
H P1/7
B139/97
A6165/2/96
A370/2/97
H6388
DSM 9683
P204
DSM 9682
AC 2947
D282
MAC 724
A2321/2/97
P25-348
AC585
P24-330
P182
A3313/3/98
B422/97
A2195/1/97
B418/97
B631/97
A5683/93
B2663/96
A3286/94
A5263/93
A3313/1/98
Sw 189/B3727
S1088-04
S1125-04
S1126-04
13749 SS
13893 SS
P202
A 7103/3
A10/94
A5439/2/96
3025/1 CA
B415/97
B2631/96
T15
A2883/2/97
Sw 286/B6347
A 7304/4
Sw 245/B5186
A2163/97
B2647/96
NumberD
1
2
1
1
1
4
1
1
1
2
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
Virulence-associated genes
Origin§
cps
mrpd
epf d
sly
1
2
2
2
1
2
2
2
2
s
+
+
+
+
+
+
+
s
s
s
s
2
+
+
+
+
+
+
+
+
+
+
*
*
+
*
*
*
*
*
s
*
*
*
+
+
+
+
+
2
2
2
2
2
2
2
2
2
2
*
2
2
+
+
+
*(3291)
*(3291)
+
+
+
*(24)
*(24)
*(24)
+
2
+
*(1890)
*(1890)
+
*(1890)
+
*(3291)
*(1890)
*(3291)
*(1890)
2
2
2
2
2
2
2
2
*(24)
2
2
2
+
*(1890)
*(193)
*(193)
*(1890)
*(24)
2
2
2
2
2
2
2
2
2
2
2
2
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
2
+
+
+
+
+
+
+
2
+
2
2
NTI
NT
1
1
2
2
1
2
2
2
2
2
2
2
9
9
2
9
9
9
9
9
9
9
9
9
2
2
2
2
2
NT
NT
NT
NT
NT
7
2
NT
7
NT
9
NT
NT
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
h-G
i
i
i
p
i
i
i
p
i
c
i
i
i
i
h-G
i
h-G
i
i
h-G
i
i
i
p
i
p
p
i
i
i
i
i
i
i
i
i
c
c
i
c
i
c
c
p
c
c
c
p
c
i
i
p
Journal of Medical Microbiology 56
AFLP of S. suis strains
Table 1. cont.
AFLP
cluster/group
C1
C2
D
Strain
A210/2/97
B2680/96
A386/94
Sc. suis K P45
A1147/94
2801 SS
A 5508/4
A5455/93
A 5373/4
2739 SS
Sw 125/B 2453
90-2741-7
B2795/96
B2801/96
V 831/1
V 7353/1
Sw 180/B 3377
Sw 123/B 2452
A896/98
A496/98
I4627 G
A41/2a/97
A217/1/97
A2155/2/97
B2441/96
P203
A6462/1/96
A 7169/3
Sw 257/B5878
B2858/97
S3401/02
S352/03
S 3385/02
I4627 D
99-734723
13607 CA
2746 SS
2CA
B72/97
B627A/97
2669/1
I4627 E
A5505/93
A425/1/97
13687
A305/3/97
NumberD
1
1
1
1
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
3
1
1
Virulence-associated genes
Origin§
cps
mrpd
epf d
sly
NT
2
+
+
2
2
2
**
2
**
2
***
***
***
***
*
****
***
***
*
+
+
*
+
+
+
+
+
***
+
***
+
+
+
*
***
+
+
+
2
2
2
+
+
+
+
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
+
2
+
2
2
2
2
2
2
2
*(24)
2
+
+
+
2
2
+
+
+
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
+
2
+
+
+
+
2
+
2
+
+
+
2
NT
NT
2
9
NT
NT
NT
NT
NT
2
2
7
NT
7
7
7
7
7
7
7
7
7
2
2
2
2
NT
2
NT
2
2
2
9
2
2
NT
2
NT
NT
NT
7
NT
NT
2
NT
p
p
i
i
i
c
c
c
c
c
p
h-C
p
p
p
p
i
p
i
p
i
i
p
c
p
i
p
p
i
p
c
c
c
i
h-C
c
c
c
p
i
c
i
c
p
c
p
DIf two or more strains with the same profile of virulence-associated genes generated the same AFLP band pattern, only one representative strain
is listed (see also Fig. 1).
dDefinition of mrp and epf variants is based on Silva et al. (2006).
§Porcine strains were classified based on their origin of isolation as invasive (i), pneumonia (p) or carrier (c). Isolates from humans (all invasive)
were from either Germany (h-G) or Canada (h-C).
IStrains that were negative for cps1, cps2, cps7 andcps9 were referred to as non-typable (NT).
http://jmm.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
105
T. Rehm and others
study had already been multilocus sequence-typed by King
et al. (2002). Using eBURST with the most stringent group
definition (Feil et al., 2004), six of the 18 sequence-typed
isolates were assigned to either ST complex 1 or ST complex
87 (Table 2). Exclusively strains of ST complex 1 belonged
to the distinct and relatively homogeneous AFLP subcluster
A1, while the three from subcluster A2 were part of ST
complex 87 (Table 2). The only discordance found between
AFLP and MLST analysis was that, in addition to the
investigated strains of A2, strain B2631/96 of AFLP group B
also belonged to ST complex 87. Analysis of the complete S.
suis MLST database revealed that ST1 and ST87 are the
primary founders of the two respective ST complexes
(bootstrap confidence values of 100 and 72 %, respectively).
The six strains from either AFLP subcluster C1 or AFLP
subcluster C2 belonged to the same ST complex (ST27), but
with a less-stringent group definition (identical alleles at five
or more of the seven loci; Table 2). Sequence typing of
strains of group B as well as those of group D made clear that
strains within these two groups might not be genetically
related, as the investigated strains did not share a single allele
in MLST analysis (Table 2). This is in agreement with the
low linkage level between strains in these AFLP groups. In
general, the results suggested congruence of AFLP and
MLST, which has also been found in other pathogens, e. g. in
Campylobacter jejuni (Schouls et al., 2003).
Correlation of AFLP-typing results with clinical
background and virulence-associated gene
profiles
The virulence-associated gene profile and the clinical
background of each strain is presented in Table 1. Cluster
A was significantly associated (P <0.0001) with strains that
were classified as invasive, based on their source of isolation.
Of all invasive isolates, 69 % belonged to cluster A.
Furthermore, the two strains (D282 and P1/7) which have
been demonstrated to be highly virulent in experimental
infections (Vecht et al., 1989; Norton et al., 1999), were
grouped in this rather homogeneous cluster. In accordance
with this, PFGE patterns of invasive S. suis strains have also
been found to be less heterogeneous than those of S. suis
isolates from other sources (Berthelot-Herault et al., 2002;
Allgaier et al., 2001; Vela et al., 2003).
With the exception of four isolates, cluster A contained only
cps1 (n=5), cps2 (n=28) and cps9 (n=13) positive strains.
All strains within this cluster were positive for the suilysin
gene sly and, with the exception of two strains, also for mrp.
The genotype sly+ mrp+ epf+ cps2+ was significantly
associated with cluster A (P <0.0001) and with subcluster
A1 (P=0.0075). Furthermore, all investigated cps1 isolates
belonged to subcluster A1 (Table 1). The homogeneous
AFLP subcluster A1 represented by virulent cps2 and cps1
genotypes is in accordance with the described unique
ribotype profile of these pathotypes (Smith et al., 1997).
Subcluster A2 was significantly associated with mrp+ (all
variants) cps9 strains (P <0.0001), most of which (10 of 13)
had an invasive clinical background. Only two of the 15
mrp+ cps9 strains did not belong to A2. With the exception
of one strain (A5263/93), all cps9+ strains of subcluster A2
were positive for a specific large variant of mrp (mrp*,
GenBank accession no. DQ295197; Silva et al., 2006).
Interestingly, homogeneous clustering of invasive cps9 strains
has not been described so far. King et al. (2002) showed that
serotype 9 strains might belong to different STs. In our study,
MLST of three mrp* cps9 strains of subcluster A2 revealed that
they belonged to ST complex 87 (Table 2), and that they were
single-locus variants of the two invasive serotype 9 strains of
ST complex 87 investigated by King et al. (2002). The
virulence-associated mrp* cps9 type isolated frequently from
diseased pigs in central Europe (Wisselink et al., 2000; Silva et
al., 2006) might, thus, be represented in the MLST study of
King et al. (2002) by only these two strains. Furthermore, we
speculate that ST complex 87 might be associated with
invasive infections in central Europe to a greater extent than
suggested elsewhere (King et al., 2002).
In comparison with cluster A, the prevalence of the virulenceassociated factors sly, mrp and epf was significantly lower
among the other strains (not in cluster A: sly+, 39 %; mrp+
(all variants), 64 %; epf+, 3 %). The majority of the strains of
group B (17 of 23) and D (nine of 15) did not belong to types
cps1, -2, -7 or -9. Most carrier isolates were assigned to one of
these two groups (42 % in B and 23 % in D).
In contrast to groups B and D, the virulence-associated gene
profiling and clinical background of the strains suggested a
more specific composition of strains of cluster C. All strains
of cluster C were negative for sly and epf. Cluster C, and in
particular subcluster C1, was significantly associated with
cps7 strains (P <0.0001). These cps7 strains were all positive
for one of the different mrp variants. The genotype mrp+
(different variants) epf2 sly cps2+ was significantly
associated with subcluster C2 (P <0.0001). Clustering of
cps7 and cps2 strains has also been observed in MLST and
PFGE typing (King et al., 2002; Vela et al., 2003). However,
as these studies did not include profiling of virulenceassociated genes, our study showed for the first time that
Fig. 1. Dendrogram of 116 S. suis strains based on AFLP band patterns. At a 55 % similarity level, two main AFLP clusters, A
and C, were identified. Strains within subcluster A1 share a 70 % overall similarity. Subcluster A2 contains the remaining
strains of cluster A. A linkage level of 57 % determines division of cluster C into C1 and C2. Based on their position in the
dendrogram, strains that belong to neither cluster A nor cluster C were assigned to either group B or group D. If two or more
strains with the same profile of virulence-associated genes generated the same band pattern (similarity index >95 %), only
one of them was included in the dendrogram. The virulence-associated gene profiles and the origin of the strains are shown in
Table 1.
106
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
Journal of Medical Microbiology 56
AFLP of S. suis strains
http://jmm.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
107
T. Rehm and others
Table 2. Multilocus STs for selected strains according to King et al. (2002) in comparison to AFLP cluster analysis
Strain
A6169/1/96
I9841/1
H P1/7
D282D
B631/97
A3286/94
A5683/94
B2631/96
A1147/94
B2647/96
Sw125
B2795/96
A896/98
B2441/96
A6462/1/96
S3385/02
A5505/93
B72/97
2 CA
ST
1
1
1
1
98d
99d
98d
89
93d
96d
25
29
29
28
100d
28
94d
95d
97d
Locus
aroA
cpn60
dpr
gki
mutS
recA
thrA
1
1
1
1
8
5
8
8
40§
25
9
8
8
2
2
2
8
41§
1
1
1
1
1
17
17
17
8
44§
46§
30
30
30
30
30
30
21
45§
47§
1
1
1
1
5
5
5
24
38§
40§
5
5
5
5
5
5
5
39§
1
1
1
1
1
12
12
12
12
48§
50§
34
34
34
34
47§
34
45
49§
51§
1
1
1
1
1
1
1
1
43§
5
30
30
30
31
31
31
44d
45§
46§
1
1
1
1
10
3
10
10
15
43§
3
3
3
3
10
3
22
42§
44§
1
1
1
1
4
4
4
4
33§
35§
25
25
25
25
25
25
4
34§
1
ST complex
AFLP subcluster
or group*
1
1
1
1
87
87
87
87
Unrelated
Unrelated
27||
27||
27||
27
27||
27
61
Unrelated
Unrelated
A1
A1
A1
A1
A2
A2
A2
B
B
B
C1
C1
C1
C2
C2
C2
D
D
D
*For assignment of strains to AFLP subclusters and groups see Fig. 1.
DStrain D282 was not sequence-typed in this study, but data were taken from King et al. (2002).
dNovel STs identified in this study.
§Novel alleles identified in this study.
||ST25, ST29 and ST100 belong to ST complex 27 only with a less-stringent approach that defines an ST complex by sharing of alleles at five or
more of the seven loci.
these cps2 strains are representative of a geno- or phenotype
(epf2, EF2, respectively) which has been shown to be
avirulent in experimental infections (Vecht et al., 1992).
The distribution of invasive, pneumonia and carrier isolates
suggested association of cluster C with pneumonia
(P=0.0017). Of the pneumonia group, 47 % belonged to
cluster C (28 and 19 % of the lung isolates within subclusters
C1 and C2, respectively). Among the three major ST
complexes described by King et al. (2002), ST complex 27
showed the highest fraction of lung isolates (28 %). This is in
accordance with the AFLP-typing results of this study, as
comparative MLST analysis of six cluster C strains provided
evidence of congruence between AFLP cluster C and the ST
complex 27 (Table 2). In Scandinavia and central Europe,
serotype 7 strains have been isolated frequently from piglets
with bronchopneumonia (Wisselink et al., 2000; Tian et al.,
2004; Aarestrup et al., 1998; Silva et al., 2006). Although
pneumonia has been elicited by experimental S. suis serotype
2 infection (Berthelot-Herault et al., 2001; Soerensen et al.,
2005), pathotypes causing mainly pneumonia have not been
identified in the respective experiments.
All investigated human isolates from Germany (n=6) were
grouped into subcluster A1, which was found to correlate
108
with ST complex 1 identified in MLST analysis (Table 2).
Accordingly, the vast majority (87 %) of the human isolates
investigated with MLST by King et al. (2002) belonged to ST
complex 1. Furthermore, the isolates of the outbreak among
humans in Sichuan in 2005 were all identified as ST7, a
single-locus variant of ST1 within the ST1 complex (Ye et al.,
2006). The human isolates assigned to AFLP subcluster A1
carried, except for one strain, a large variant of epf (epf *).
This genotype has been detected frequently in serotype 2
strains of moderate virulence for piglets and also in human
isolates from Europe (Smith et al., 1993). In contrast to
the human isolates from Germany, the four examined
human isolates from Canada belonged to cluster C or group
D, were negative for epf and sly, and contained a large variant
of the mrp gene. The different AFLP clustering of the few
human isolates from central Europe, North America and
Asia has to be further clarified in future experiments using
more strains.
In conclusion, the described AFLP typing allowed identification of S. suis clusters of clinical relevance. Profiles of
virulence-associated genes (cps, mrp, epf and sly) and the
clinical background of strains showed homogeneous
clustering (A1) of European invasive cps2+ isolates from
pigs (typically mrp+ epf+ sly+) and humans (mrp + epf *
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
Journal of Medical Microbiology 56
AFLP of S. suis strains
sly+). Another very different cluster (C) was associated with
pneumonia and consisted of cps7 and mrp2 epf2 sly2 cps2
strains. Furthermore, this study suggests for the first time a
correlation of invasive mrp* cps9+ strains in central Europe
with a certain AFLP subcluster (A2) and a single ST complex
(ST87) in MLST analysis.
polymorphism, multilocus sequence typing, and short repeat
sequencing: strain diversity, host range, and recombination. J Clin
Microbiol 41, 15–26.
ACKNOWLEDGEMENTS
strains of Streptococcus suis type 2 are absent in pathogenic strains.
Infect Immun 61, 3318–3326.
We thank Tosso Leeb and Heike Klippert (Institut für Tierzucht und
Vererbungsforschung, Stiftung Tierärztliche Hochschule Hannover) for
support in sequencing. This study was supported by a grant from the
Deutsche Forschungsgemeinschaft (DFG), Bonn, Germany (SFB587).
Smith, H. E., Rijnsburger, M., Stockhofe-Zurwieden, N., Wisselink,
H. J., Vecht, U. & Smits, M. A. (1997). Virulent strains of
Silva, L., Baums, C. G., Rehm, T., Wisselink, H., Goethe, R. &
Valentin-Weigand, P. (2006). Virulence-associated gene profiling of
Streptococcus suis isolates by PCR. Vet Microbiol 115, 117–127.
Smith, H. E., Reek, F. H., Vecht, U., Gielkens, A. L. J. & Smits, M. A.
(1993). Repeats in an extracellular protein of weakly pathogenic
Streptococcus suis serotype 2 and highly virulent strains of
Streptococcus suis serotype 1 can be recognized by a unique ribotype
profile. J Clin Microbiol 35, 1049–1053.
Smith, H. E., Damman, M., van der Velde, J., Wagenaar, F.,
Wisselink, H. J., Stockhofe-Zurwieden, N. & Smits, M. A. (1999).
REFERENCES
Aarestrup, F. M., Jorsal, S. E. & Jensen, N. E. (1998). Serological
characterization and antimicrobial susceptibility of Streptococcus suis
isolates from diagnostic samples in Denmark during 1995 and 1996.
Vet Microbiol 60, 59–66.
Allgaier, A., Goethe, R., Wisselink, H. J., Smith, H. E. &
Valentin-Weigand, P. (2001). Relatedness of Streptococcus suis
isolates of various serotypes and clinical backgrounds as evaluated
by macrorestriction analysis and expression of potential virulence
traits. J Clin Microbiol 39, 445–453.
Arends, J. P. & Zanen, H. C. (1988). Meningitis caused by
Streptococcus suis in humans. Rev Infect Dis 10, 131–137.
Berthelot-Herault, F., Gottschalk, M., Labbe, A., Cariolet, R. &
Kobisch, M. (2001). Experimental airborne transmission of
Identification and characterization of the cps locus of Streptococcus
suis serotype 2: the capsule protects against phagocytosis and is an
important virulence factor. Infect Immun 67, 1750–1756.
Soerensen, C. M., Holmskov, U., Aalbaek, B., Boye, M., Heegaard,
P. M. & Nielsen, O. L. (2005). Pulmonary infections in swine induce
altered porcine surfactant protein D expression and localization to
dendritic cells in bronchial-associated lymphoid tissue. Immunology
115, 526–535.
Staats, J. J., Feder, I., Okwumabua, O. & Chengappa, M. M.
(1997). Streptococcus suis: past and present. Vet Res Commun 21,
381–407.
Staats, J. J., Plattner, B. L., Stewart, G. C. & Chengappa, M. M.
(1999). Presence of the Streptococcus suis suilysin gene and
Streptococcus suis capsular type 2 in pigs. Vet Microbiol 82, 69–80.
expression of MRP and EF correlates with high virulence in
Streptococcus suis type 2 isolates. Vet Microbiol 70, 201–211.
Berthelot-Herault, F., Marois, C., Gottschalk, M. & Kobisch, M.
(2002). Genetic diversity of Streptococcus suis strains isolated from
Tian, Y., Aarestrup, F. M. & Lu, C. P. (2004). Characterization of
pigs and humans as revealed by pulsed-field gel electrophoresis. J Clin
Microbiol 40, 615–619.
Clifton-Hadley, F. A., Alexander, T. J., Upton, I. & Duffus, W. P.
(1984). Further studies on the subclinical carrier state of
Streptococcus suis type 2 in pigs. Vet Rec 114, 513–518.
Feil, E. J., Li, B. C., Aanensen, D. M., Hanage, W. P. & Spratt, B. G.
(2004). eBURST: inferring patterns of evolutionary descent among
clusters of related bacterial genotypes from multilocus sequence
typing data. J Bacteriol 186, 1518–1530.
Gibson, J. R., Slater, E., Xerry, J., Tompkins, D. S. & Owen, R. J.
(1998). Use of an amplified-fragment length polymorphism
technique to fingerprint and differentiate isolates of Helicobacter
pylori. J Clin Microbiol 36, 2580–2585.
King, S. J., Leigh, J. A., Heath, P. J., Luque, I., Tarradas, C., Dowson,
C. G. & Whatmore, A. M. (2002). Development of a multilocus
sequence typing scheme for the pig pathogen Streptococcus suis:
identification of virulent clones and potential capsular serotype
exchange. J Clin Microbiol 40, 3671–3680.
MacInnes, J. I. & Desrosiers, R. (1999). Agents of the ‘suis-ide
diseases’ of swine: Actinobacillus suis, Haemophilus parasuis, and
Streptococcus suis. Can J Vet Res 63, 83–89.
Norton, P. M., Rolph, C., Ward, P. N., Bentley, R. W. & Leigh, J. A.
(1999). Epithelial invasion and cell lysis by virulent strains of
Streptococcus suis is enhanced by the presence of suilysin. FEMS
Immunol Med Microbiol 26, 25–35.
Streptococcus suis serotype 7 isolates from diseased pigs in Denmark.
Vet Microbiol 103, 55–62.
Vecht, U., Arends, J. P., van der Molen, E. J. & van Leengoed, L. A.
(1989). Differences in virulence between two strains of Streptococcus
suis type II after experimentally induced infection of newborn germfree pigs. Am J Vet Res 50, 1037–1043.
Vecht, U., Wisselink, H. J., Jellema, M. L. & Smith, H. E. (1991).
Identification of 2 proteins associated with virulence of Streptococcus
suis type 2. Infect Immun 59, 3156–3162.
Vecht, U., Wisselink, H. J., van Dijk, J. E. & Smith, H. E. (1992).
Virulence of Streptococcus suis type 2 strains in newborn germfree
pigs depends on phenotype. Infect Immun 60, 550–556.
Vela, A. I., Goyache, J., Tarradas, C., Luque, I., Mateos, A., Moreno, M. A.,
Borge, C., Perea, J. A., Dominguez, L. & Fernandez-Garayzabal, J. F.
(2003). Analysis of genetic diversity of Streptococcus suis clinical
isolates from pigs in Spain by pulsed-field gel electrophoresis. J Clin
Microbiol 41, 2498–2502.
Wisselink, H. J., Smith, H. E., Stockhofe-Zurwieden, N., Peperkamp, K.
& Vecht, U. (2000). Distribution of capsular types and production of
muramidase-released protein (MRP) and extracellular factor (EF) of
Streptococcus suis strains isolated from diseased pigs in seven European
countries. Vet Microbiol 74, 237–248.
Ye, C., Zhu, X., Jing, H., Du, H., Segura, M., Zheng, H., Kan, B.,
Wang, L., Bai, X. & other authors (2006). Streptococcus suis sequence
type 7 outbreak, Sichuan, China. Emerg Infect Dis 12, 1203–1208.
Schouls, L. M., Reulen, S., Duim, B., Wagenaar, J. A., Willems, R. J.,
Dingle, K. E., Colles, F. M. & Van Embden, J. D. (2003). Comparative
Yu, H., Jing, H., Chen, Z., Zheng, H., Zhu, X., Wang, H., Wang, S., Liu, L.,
Zu, R. & other authors (2006). Human Streptococcus suis outbreak,
genotyping of Campylobacter jejuni by amplified fragment length
Sichuan, China. Emerg Infect Dis 12, 914–920.
http://jmm.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 78.47.19.138
On: Sun, 02 Oct 2016 19:02:58
109