I
PB
AQ1
Pa
b
c
d
a
ARRA
KGGcc�
I
nepbnbi
H
h1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
ARTICLE IN PRESSG ModelJMM 50826 1–8
International Journal of Medical Microbiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Medical Microbiology
j ourna l h o mepage: www.elsev ier .com/ locate / i jmm
henotypic and molecular characterization of hyperpigmented group Streptococci
gnese Lupoa, Corinne Ruppena, Andrew Hemphill b, Barbara Spellerbergc,arham Sendia,d,∗
Institute for Infectious Diseases, University of Bern, Bern, SwitzerlandInstitute of Parasitology, University of Bern, Bern, SwitzerlandInstitute of Medical Microbiology and Hygiene, University Hospital Ulm, Ulm, GermanyDepartment of Infectious Diseases, Bern University Hospital, Bern, Switzerland
r t i c l e i n f o
rticle history:eceived 12 November 2013eceived in revised form 5 May 2014ccepted 11 May 2014
eywords:roup B streptococciranadaene pigmentyl operonovS/R-Hemolysin
a b s t r a c t
Group B Streptococcus (GBS) causes invasive infections in neonates, older adults and patients with comor-bidities. �-hemolysin/cytolysin is an important GBS virulence factor. It is encoded by the cyl operonand confers GBS hemolytic activity. Isolates displaying hyperpigmentation are typically hyperhemolytic.Comparison of clonally identical isolates displaying different levels of pigmentation has shown tran-scriptional dysregulation due to mutations in components of the control of the virulence S/R (CovS/R)regulatory system. In addition, hyperpigmented isolates show decreased CAMP factor and decreasedcapsule thickness. In analogy to findings in group A Streptococcus, a pivotal role of CovS/R has been pro-posed in the host-pathogen interaction of invasive GBS infection. However, corresponding investigationson multiple clinical GBS isolates have not been performed. We prospectively collected hyperpigmentedisolates found in a diagnostic laboratory and performed phenotypic, molecular and transcriptional anal-yses. In the period from 2008 to 2012, we found 10 isolates obtained from 10 patients. The isolatesreflected both invasive pathogens and colonizers. In three cases, clonally identical but phenotypicallydifferent variants were also found. Hence, the analyses included 13 isolates. No capsular serotype wasfound to be significantly more frequent. Bacterial pigments were analyzed via spectrophotometry and fortheir hemolytic activity. Data obtained for typical absorbance spectra peaks correlated significantly with
hemolytic activity. Molecular analysis of the cyl operon showed that it was conserved in all isolates. ThecovR sequence displayed mutations in five isolates; in one isolate, the CovR binding site to cylX was abro-gated. Our results on clinical isolates support previous findings on CovR-deficient isogenic mutants, butsuggest that – at least in some clinical isolates – for �-hemolysin/cytolysin and CAMP factor production,other molecular pathways may be involved.33
34
35
36
37
38
39
ntroduction
Group B Streptococcus (GBS) is a major cause of sepsis ineonates and pregnant women. The incidence of invasive GBS dis-ase in nonpregnant adults is growing, in particular in elderlyersons and in those with chronic underlying conditions (e.g. dia-etes mellitus). Cases of severe, life-threatening syndromes of
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
ecrotizing fasciitis and toxic shock syndrome due to GBS have alsoeen reported (Sendi et al., 2008). The development of GBS disease
s based on bacterial colonization (e.g. of the vaginal epithelium),
∗ Corresponding author at: Department of Infectious Diseases, Bern Universityospital, Bern, Switzerland. Tel.: +41 31 632 99 99; fax: +41 31 632 87 66.
E-mail address: [email protected] (P. Sendi).
ttp://dx.doi.org/10.1016/j.ijmm.2014.05.003438-4221/© 2014 Published by Elsevier GmbH.
40
41
42
43
© 2014 Published by Elsevier GmbH.
penetration of epithelial barriers and invasion into sterile com-partments. GBS expresses a diverse array of virulence factors thatmediate specific host–cell interactions and interfere with innateimmune clearance mechanisms (Liu and Nizet, 2004; Maisey et al.,2008).
The surface-associated toxin �-hemolysin/cytolysin (�-h/c) isa crucial GBS virulence factor. Production of �-h/c is associatedwith direct lysis of a variety of eukaryotic cell types (Gibsonet al., 1999), inflammatory activation (Doran et al., 2002) and vir-ulence in animal models (Ring et al., 2002). GBS isolates can behyperpigmented when they express an orange-reddish pigment, a
ecular characterization of hyperpigmented group B Streptococci.5.003
phenomenon that has also been linked to �-h/c expression. Hyper-pigmented GBS have been reported to be hyperhemolytic (Nizetet al., 1996). Hence, in invasive GBS diseases, a significant patho-genetic role has been attributed to the pigment (Liu et al., 2004;
44
45
46
47
ING ModelI
2 f Med
Sg2otsa2heoasm(
vCoo�J
u(eaadhnath
M
B
miiumvfrat(iaLdpdtlN
Aa
d
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
ARTICLEJMM 50826 1–8
A. Lupo et al. / International Journal o
endi et al., 2009; Whidbey et al., 2013). �-h/c is encoded by theenes of the cyl operon (Spellerberg et al., 1999; Pritzlaff et al.,001). Biosynthesis of the pigment also requires the 12-gene cylperon (Spellerberg et al., 1999; Liu et al., 2004), with the linko pigment expression having recently been proposed to lie oneveral genes within this operon. The genes encode enzymes cat-lyzing different steps in fatty acid biosynthesis (Whidbey et al.,013). Targeted mutations of different cyl genes result in a non-emolytic and non-hyperpigmented GBS phenotype (Spellerbergt al., 1999; Liu et al., 2004; Gottschalk et al., 2006). We previ-usly discovered two phenotypically different (i.e. hyperpigmentednd non-hyperpigmented) but clonally identical GBS isolates thathowed no mutation in the cylE and cylA genes. In the hyperpig-ented variant, the covR sequence had a three-base-pair deletion
Sendi et al., 2009).The two-component regulatory system CovS/R (Control of
irulence S/R) consists of CovS (sensor), which acts as a kinase, andovR (regulator), which activates the response. A complete deletionf covS/R results in up-regulation of cyl genes and downregulationf genes in the cps operon for capsule production. Additionally,covS/R mutants show reduced CAMP activity (Lamy et al., 2004;
iang et al., 2005).Interestingly, while previous studies on the regulation and vir-
lence of the GBS pigment were performed on laboratory strainsi.e. isogenic mutants) (Lamy et al., 2004; Liu et al., 2004; Jiangt al., 2005; Whidbey et al., 2013), our findings were discovered in
clinical isolate obtained from a patient with necrotizing fasciitisnd toxic shock syndrome (Sendi et al., 2009). To further eluci-ate these findings in clinical isolates, we prospectively collectedyperpigmented GBS isolates that were identified in a routine diag-ostic microbiology laboratory. We then performed phenotypicnd genetic analyses of the isolates. We thereby aimed to iden-ify the spectrum of variations within the CovS/R system in clinicalyperpigmented GBS isolates.
aterials and methods
acterial isolates
GBS species was determined in all isolates by selective agaredia, agglutination test and matrix-assisted laser desorption/
onization/time-of-flight mass spectrometry. Hyperpigmented GBSsolates were looked for and prospectively collected from July 2008ntil December 2012 during routine diagnostic procedures in theicrobiology laboratory of the Institute of Infectious Diseases, Uni-
ersity of Bern, Switzerland. To delineate hyperpigmented isolatesrom non-hyperpigmented isolates, we required four criteria: (1)eddish pigmentation in the center of a colony on a Columbia bloodgar plate; (2) reddish pellet of a Todd Hewitt Broth (THB) cul-ure after centrifuging and washing with phosphate buffer solutionPBS); (3) a reddish pellet after washing with dimethyl sulfox-de (DMSO); and (4) a typical absorbance spectrum with peakst 455, 485 and 520 nm (Merritt and Jacobs, 1978; Tapsall, 1986;iu et al., 2004). Because we aimed to unambiguously support theelineation of hyperpigmented from non-hyperpigmented isolates,igments from 22 randomly selected GBS isolates from a routineiagnostic laboratory were purified and subjected to spectropho-ometric analysis. If a hyperpigmented isolate was selected, weooked for phenotypic variants in the same specimen. We usedEM316 and ATC13813 as control strains.
ntimicrobial susceptibility, serotype determination, clonality
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
nalyses
The minimum inhibitory concentrations of penicillin, clin-amycin and erythromycin were determined twice with Etests
PRESSical Microbiology xxx (2014) xxx–xxx
(bioMérieux, Marcy l’Etoile, France), performed and interpreted asrecommended by the Clinical and Laboratory Standards Institutefor broth microdilution (CLSI, 2010). Serotyping was performed byusing a rapid latex agglutination test (Strep-B-Latex kit, StatensSerum Institut, Copenhagen, Denmark) (Slotved et al., 2003) andconfirmed by multiplex PCR (Imperi et al., 2010).
The clonal complex of the isolates was determined by multi-locus sequence typing (MLST), as described previously (Jones et al.,2003), using eBURSTv3 software (http://eburst.mlst.net/v3).
If various GBS phenotypes were observed within the same spec-imen, clonal relatedness was evaluated. This was investigated bymacrorestriction using SacII enzyme (Promega, Madison, WI, USA).Fragments were resolved by pulsed field gel electrophoresis (PFGE)for 21 h, with 1 s and 35 s as the initial and final impulse time,respectively, at 6 V/cm and 14 ◦C.
Analysis of pigment, hemolytic activity and capsule production
Pigment was extracted from bacterial cultures as describedpreviously (Rosa-Fraile et al., 2006; Whidbey et al., 2013). Theabsorbance of the purified pigment was measured with a spec-trophotometer (Cary 300 Bio, UV–vis, Varian, USA) from 410 to610 nm. To estimate the absorbance intensity of the pigment inselected hyperpigmented isolates in comparison to that of NEM316,we added a mathematically calculated borderline to each spectragraph. This borderline was defined to be more than at least onelogarithmic unit above the absorbance spectra of NEM316 at everymeasurement point.
Hemolytic activity was determined by performing 10-fold serialdilutions of the pigment extract in PBS containing 0.2% glucose andincubating with 1% sheep blood at 37 ◦C for 1 h (Whidbey et al.,2013). The release of hemoglobin was estimated by optical densityat 420 nm. The values obtained at a dilution of 10−4 were com-pared with those obtained from 1% sheep blood treated with a 0.1%sodium dodecyl sulfate solution.
To evaluate the association between the absorbance intensity ofthe pigment and hemolytic activity, we correlated data on pigmentabsorption peaks at 455, 485 and 520 nm with results from thehemolytic activity assay.
The presence of the capsule was determined by using the fol-lowing two methods:
1. Buoyance density gradient stratification (Buchanan et al., 2005).Bacterial cultures were harvested by centrifugation and the pel-lets washed with PBS, followed by centrifugation. The pelletswere resuspended in 1 ml of PBS and loaded on a centrifuga-tion gradient obtained by stratification of Ficoll 400 (Pharmacia)solutions at 70, 60 and 50%.
2. Transmission electron microscopy (TEM). Bacterial isolates wereincubated in THB to an OD of 0.8 (absorbance 600 nm). Iso-lates were then washed and fixed with Karnovsky solution.After polymerization, samples were sectioned with an ultra-microtome (Ultracut E ultramicrotome, Reichert-Jung, Vienna,Austria) before TEM imaging (Phillips EM 400 operating at60 kV).
Molecular analysis
Genomic DNA was obtained according to the procedure ofMurray et al. (1990). Fifty nanograms was used in the PCRmixture reactions. The amplifications were performed with Q5High-Fidelity Master Mix (New England BioLab, Ipswich, MA, USA).
ecular characterization of hyperpigmented group B Streptococci.5.003
Denaturation and amplification steps were performed at the condi-tions suggested by the manufacturer. All primers used in the studyand melting temperatures are listed in Table S1 (supplementaryfiles). Amplified targets for covS/R and part of the cyl operon were
166
167
168
169
ING ModelI
f Medi
sBtasnn
o
R
(Dkarq(abtp(cd
S
tt
R
B
1Iat(oB((HaifwvfcpolSbw
o
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
ARTICLEJMM 50826 1–8
A. Lupo et al. / International Journal o
equenced, and results were analyzed with the NCBI nucleotidelast program (http://blast.ncbi.nlm.nih.gov). The sequences rela-ive to covS/R were compared with those deposited in GenBank withccession number AL766852.1 by using ClustalW. In addition, weequenced the CovR binding site to the cyl operon, which is inter-al to fragment 660346-660364 of nucleotide sequence accessionumber NC 004368.1.
Supplementary Table related to this article can be found, in thenline version, at http://dx.doi.org/10.1016/j.ijmm.2014.05.003.
NA isolation and quantification
RNA was extracted from cultures in THB at absorbance 0.8600 nm) as previously described (Sendi et al., 2009). ContaminantNA was eliminated by treating extracts with the TURBO DNA freeit (Applied Biosystem Inc., Foster City, CA, USA). To confirm thebsence of DNA, we used the RNA extracts as template in a PCReaction. Retro-transcription and real-time quantitative PCR (RT-PCR) were performed using the GoTaq 2-Step RT-qPCR SystemPromega, USA). Expression of selected virulence genes (cfb, cpsGnd cylE) in strains harboring a covS/R modification was evaluatedy absolute quantification of the mRNA transcripts normalized tohe expression level of the gyrA housekeeping gene. For this aim, weerformed a calibration curve for each gene as previously describedFlorindo et al., 2012). In addition, transcriptional analyses betweenlonally identical but phenotypically different variants were con-ucted with the 2−��Ct method (Livak and Schmittgen, 2001).
tatistical analysis
GraphPad Prism 5.0 was used for building graphs and for statis-ical analysis (Pearson r and Student’s t-test where appropriate). Awo-tailed p value of ≤0.05 was considered significant.
esults
acterial isolates
We identified 10 GBS hyperpigmented isolates obtained from0 individuals. In seven patients, there was only one phenotype.n three patients, the hyperpigmented isolate was accompanied by
phenotypically different variant (i.e. variant A and B). In two ofhem (010, 015), one variant was considered non-hyperpigmentedvariant A), whereas in one isolate (014) both variants showed anrange-reddish pigment. However, in one variant of 014 (variant), the pigment was more reddish and the absorbance more intenseFig. 1). PFGE analysis suggested that variants of the three isolates010, 014, 015) were indistinguishable (supplementary file, Fig. S1).ence, 13 isolates in total were included in this study. The sourcend characterization of the isolates is presented in Table 1. Threesolates were obtained from patients who had no disease mani-estation (i.e. colonization) and seven isolates from sites in patientsith a clinical syndrome. Disease manifestation varied from nonse-
ere (conjunctivitis, tonsillopharyngitis) to very severe (necrotizingasciitis). None of the capsular serotypes was significantly asso-iated with hyperpigmentation. The results of MSLT analysis areresented in Table 1. Three isolates were single locus variants ofther sequence types (i.e. isolate 008 ST335 = variant of ST19, iso-ates 010A/B ST255 = variants of ST6, isolate 016 ST9 = variant ofT10). Apart from isolate 015A/B (singleton ST130), 10 isolates
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
elonged to clonal complex (CC) 19, and one to CC17. All isolatesere susceptible to penicillin, erythromycin and clindamycin.
Supplementary Figure related to this article can be found, in thenline version, at http://dx.doi.org/10.1016/j.ijmm.2014.05.003.
PRESScal Microbiology xxx (2014) xxx–xxx 3
Pigment detection and hemolytic activity
Photos of the colonies, the DMSO-washed pellets and theabsorbance spectra of the GBS isolates are displayed in Fig. 1.The NEM316 pigment showed a typical spectrum with peaks at455, 485 and 520 nm, whereas neither ATCC13813 nor one of the22 randomly selected isolates showed absorbance at these wave-lengths (supplementary file, Fig. S2A and S2C). The mathematicallycalculated borderline (i.e. reflecting data that are at least one log-arithmic unit above the absorbance spectra of NEM316 at everymeasurement point) was set at 0.15 (supplementary file, Fig. S2B).The absorbance ranges of hyperpigmented isolates were 0.17–1.61,0.23–2.19 and 0.17–1.91 at 455, 485 and 525 nm, respectively. Theresults of the hemolytic activity assays are represented in Fig. 2A.We observed a significant correlation between hemolytic activity(Fig. 2B) and pigment absorbance values obtained at 455 nm (Pear-son r = 0.81, 95% CI 0.46–0.94, p = 0.0008), 485 nm (Pearson r = 0.80,95% CI 0.43–0.94, p = 0.0012) and 520 nm (Pearson r = 0.79, 95% CI0.41–0.93, p = 0.0015).
Supplementary Figure related to this article can be found, in theonline version, at http://dx.doi.org/10.1016/j.ijmm.2014.05.003.
Molecular analysis of the hyperpigmented GBS isolates
Molecular alterations within the CovS/R sites are presented inFig. 3A. The alterations included a three-base-pair deletion (Val31)(isolate 001) and two nucleotide substitutions leading to Ala90Asp(isolate 013) and Arg119His (isolate 016), respectively. Further, inisolate 014 (both variants A and B), there was a nucleotide inser-tion leading to a truncated CovR at amino acid 206. In variant B ofisolate 010, there was a 13-nucleotide deletion causing an abortiveCovR (stop at amino acid 11) and an IS1381 inserted between theribosome-binding site and the CovR binding site upstream of thecovR gene. In isolate 012, the analysis of the promotor region of thecyl operon revealed a 19-base deletion from −295 to −277, fromthe cylX start codon, overlapping the CovR binding site (−281 to−252) (Fig. 3B). No covS/R variations were found in hyperpigmentedisolates 002, 003, 008 and 015B.
Transcriptional analysis and phenotypic tests of virulence factors
We then quantified the transcripts of selected virulence genesin isolates harboring covS/R modifications and their variants (iso-lates 001, 010A and B, 013, 014A and B, 016) as well as that ofNEM316 (control). Transcription of the cpsG gene was low in allisolates. The cylE gene was overexpressed in all isolates, except invariant A of isolate 010, which had a functional CovR. It was alsosignificantly overexpressed when compared with cfb and cpsG inall isolates but 010A and 014A (Fig. 4). The expression patterns of010A and its hyperpigmented variant 010B, which has a truncatedCovR, were in agreement with the observed phenotypes of the iso-lates (Fig. 4 and supplementary file, Fig. S3), confirming the role ofthe CovS/R system as observed previously (Sendi et al., 2009). Vari-ant B of isolate 014, which harbored the same CovR mutation asthat of 014A, showed an overexpression of the cylE gene comparedwith the cfb and cpsG genes. Surprisingly, the cfb gene was over-expressed in variant 014B (Fig. 4). These analyzed transcript levelswere in agreement with phenotypic profiles (supplementary file,Fig. S3).
Supplementary Figure related to this article can be found, in theonline version, at http://dx.doi.org/10.1016/j.ijmm.2014.05.003.
ecular characterization of hyperpigmented group B Streptococci.5.003
Discussion
In this study, we report the molecular and phenotypic character-ization of a collection of clinical GBS isolates showing pronounced
283
284
285
ARTICLE IN PRESSG ModelIJMM 50826 1–8
4 A. Lupo et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
pellet
pmisil(yol
aot
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
Fig. 1. Photos of the colonies, the DMSO-washed
igmentation. To the best of our knowledge, the term hyperpig-entation is not precisely defined. The selection criteria applied
n this study were based on phenotypic characteristics and analy-es of pigment absorbance spectra. Each so-called hyperpigmentedsolate included in this study had a pigment absorbance that was ateast more than one log higher than those obtained from NEM316Fig. 1). Although more in-depth (e.g. chemical and physical) anal-sis is needed to define ‘hyperpigmented’ isolates, we believe thatur approach appears reasonable and practicable for other researchaboratories.
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
The hyperpigmented isolates were collected from differentnatomical sites and various patients. They were found as col-nizers of the urogenital tract; in nonsevere diseases, such asonsillopharyngitis and conjunctivitis; and in life-threatening
s and the absorbance spectra of the GBS isolates.
infections, such as necrotizing fasciitis. Thus, hyperpigmentationin clinical isolates is not uniformly associated with severe invasivedisease.
Recent advances in the understanding of the link between pig-ment production and �-h/c activity in GBS have been achieved.Whidbey et al. (Whidbey et al., 2013) showed that the pigmentconsists of an ornithine rhamnolipid compound, and that this pig-ment has hemolytic and cytolytic activity. Consistent with theseand previous results (Nizet et al., 1996), we found an overallgood correlation between pigmentation and hemolytic activity
ecular characterization of hyperpigmented group B Streptococci.5.003
(Fig. 2B). This was in particular obvious with isolates that revealedhigh absorbance values (i.e. 001, 008, 012, 010B and 014B). How-ever, because we did not quantify the amount of the ornithinerhamnolipid compound, a direct correlation between pigment and
310
311
312
313
ARTICLE IN PRESSG ModelIJMM 50826 1–8
A. Lupo et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx 5
Table 1Characterization of selected clinical GBS isolates.
Isolate Specimen Patient Clinical manifestation Serotype ST MIC (E-test)
Age Sex P E CLI
S ≤ 0.25 R > 0.25 S ≤ 0.25 R > 0.5 S ≤ 0.5 R > 0.5
001 Biopsy 74 M Prosthetic joint infection V 12 0.094 0.064 0.094002 Cervix swab 22 F No disease, colonization III 19 0.064 0.064 0.094003 Hip joint fluid 88 M Septic arthritis Ib 8 0.125 0.032 0.094008 Cervix swab 20 F No disease, colonization III 335 0.047 0.064 0.094010Aa Eye swab 1 month M Conjunctivitis Ib 255 0.064 0.047 0.25010Ba Ib 255 0.064 0.047 0.25012 Urine 64 F Urinary tract infectionb III 28 0.047 0.047 0.064013 Biopsy 77 M Necrotizing fasciitis III 17 0.064 0.032 0.064014Aa Vaginal swab 55 F Bleeding from uterus
myomatosusII 19 0.023 0.016 0.047
014Ba II 19 0.023 0.016 0.047015Aa Urine 65 M No disease, colonizationc IX 130 0.023 0.032 0.125015Ba IX 130 0.023 0.032 0.125016 Throat swab 4 M Tonsillopharnyngitis Ib 9 0.064 0.032 0.094
M, male; F, female; ST, sequence type determined by MLST; MIC, minimal inhibitory concentration determined by E test; P, penicillin; E, erythromycin; CLI, clindamycin., vari
hcommpnsaumfiip
aIpTe
Fd
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
a A and B correspond to clonally identical but phenotypically different variants: Ab Polymicrobial infection.c Urine after bladder reconstruction and radiation after cancer.
emolytic activity cannot be made. In addition, we found reddisholonies (e.g. 003, 015B) that did not reveal high absorbance valuesf the pigment extract or hemolytic activity. These discrepanciesay be explained by variations in the chromophore part of theolecule, which would not affect its hemolytic property. Thus,
igment and hemolytic activity are often coupled, but this phe-omenon cannot be applied to all clinical isolates. Moreover, resultsuggesting uniform correlation between pigment and hemolyticctivity should be interpreted with caution because �-h/c is annstable cell-wall-associated toxin, and pure extraction of both pig-ent and �-h/c is technically challenging. However, our study is the
rst to characterize pigmentation and hemolytic activity in a clin-cal collection of isolates with different genetic backgrounds andhysiological features.
In three specimens, hyperpigmented isolates (010B, 014Bnd 015B) were found together with an accompanying variant.
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
nterestingly, in isolate 014, both variants were considered ‘hyper-igmented’, though they showed different levels of pigmentation.his phenomenon of clonally identical but phenotypically differ-nt GBS variants has been described previously in isolates obtained
ig. 2. (a) Hemolytic activity of GBS isolates. The values obtained at a dilution of 10−4 weodecyl sulphate solution. (b) Association between absorbance intensity of the pigment a
ant; B, hyperpigmented.
from patients with severe disease (Sigge et al., 2008; Sendi et al.,2009). The hyperpigmented variants showed decreased CAMP andcapsule production and carried mutations in the regulatory genecovR. However, in this study, the previously described character-istics were found in only one (010A and B) of the three pairedvariants. While the capsule was less expressed in all three hyperpig-mented variants, the CAMP reaction was not decreased in isolates014B and 015B (supplementary file, Fig. S3). Thus, although pheno-typic characterization was performed in only three paired variants,it appears that in hyperpigmented isolates, capsule production isdecreased. In contrast, an association between hyperpigmentationand a decrease in the CAMP reaction was observed in only one ofthe three isolates.
We investigated a possible role of the covS/R system in hyperpig-mented isolates and found that in five isolates, the covR sequencewas mutated. These mutations caused truncation of CovR in two
ecular characterization of hyperpigmented group B Streptococci.5.003
isolates (variant B of isolate 010 and both variants [A and B] of iso-late 014) and a single amino acid deletion and substitution in theremaining three isolates (�Val31, Ala90Asp and Arg119His in iso-lates 001, 013 and 016, respectively) (Fig. 3A). In addition, a partial
re compared with those obtained from 1% sheep blood treated with a 0.1% sodiumt 455, 485 and 520 nm and hemolytic activity.
350
351
352
353
ARTICLE IN PRESSG ModelIJMM 50826 1–8
6 A. Lupo et al. / International Journal of Medical Microbiology xxx (2014) xxx–xxx
F g mu( ion of
( s in bo
ddlGfveC5ftlthtpr
l
Fot
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
ig. 3. (a) Alignment of the CovR amino acid sequences from the isolates harborinaccession number AL766852.1) (ClustalW). (b) Alignment of the cylX promoter reg−281, −252 from ATG starting codon) proposed by Lamy et al. (2004) (Clustal W) i
eletion in the CovR binding site (Pcyl region) in isolate 012 wasetected (Fig. 3B). All of these mutations, except �Val31 in iso-
ate 001 (Sendi et al., 2009), are here reported for the first time inBS. The functional consequences of the amino acid substitutions
ound in our study are unknown. They may be in line with the initro findings described by Lin et al. (2009) in GBS and Horstmannt al. (2011) in Group A Streptococcus (GAS). Promoter binding ofovR required phosphorylation of the aspartate residue at position3. Phosphorylation of threonine at position 65 prevented CovRrom binding to DNA (Lin et al., 2009). Expression of �-h/c mayhereby be de-repressed. Considering the homology and the simi-ar functions between GAS and GBS CovR, a parallelism between thewo structures can be speculated. The substitutions observed in theyperpigmented isolates most likely affect the receiver domain ofhe CovR protein, probably interfering with the efficiency of phos-horylation by CovS. In particular, Arg119 substitution has been
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
epeatedly described in GAS isolates (Horstmann et al., 2011).To confirm a role for the detected mutations in the regu-
atory pattern of our hyperpigmented isolates, we performed a
ig. 4. cfb, cpsG and cylE gene expression was analyzed in the isolates harboring mutationf the mRNA transcripts normalized to the expression level of the gyrA housekeeping gehe cpsG genes by t-test.
tations and highlights of the different residues compared with sequence NEM316isolate 012 and NEM316 (accession number AL766846.1). The CovR binding regionld and underlined.
transcription analysis of the cfb (CAMP), cpsG (capsule) and cylEgenes (hyperpigmentation). In hyperpigmented isolates (001, vari-ant B of 010, 013, variants A and B of 014, and 016), CovR mutationswere associated with the overexpression of the cylE gene andthe downregulation of the cpsG gene. The findings in isolate 014are consistent with the phenotypic observation that both variantsshowed an orange-reddish pigment. The downregulation of the cfbgene was not observed in variant B of isolate 014 or in isolate 016(supplementary file, Fig. S3). The cfb gene was upregulated in vari-ant A of isolate 010. Thus, in clinical hyperpigmented isolates witha mutated CovR, transcript analysis was uniformly consistent withcylE overexpression and cpsG downregulation, but downregulationof cfb (CAMP) was not found in all isolates. This parallels the above-described phenotypic observations in the three isolates with pairedvariants (010, 014, 015). Considering that �-h/c and the capsule aremajor virulence components, this investigation on clinical isolates
ecular characterization of hyperpigmented group B Streptococci.5.003
supports the role of the CovS/R system in the regulation of thesefactors involved in GBS pathogenesis. Of note, CAMP appears not tobe essential for systemic virulence in GBS (Hensler et al., 2008).
s in the covS/R sequence. For each isolate, we performed an absolute quantificationne. The level of expression of the cylE gene was compared with that of the cfb and
389
390
391
ING ModelI
f Medi
fa(PdaawpDsffisiuioirhitbCemiihs
wtawpotalcbo(2or[Tip
A
tTVEceC
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
ARTICLEJMM 50826 1–8
A. Lupo et al. / International Journal o
Our study has limitations. Hyperpigmented isolates were lookedor on the basis of phenotypic characteristics, but in contrast toccepted characteristics of phenotypic variants in other speciese.g. small colony variants of Staphylococcus aureus (Sendi androctor, 2009)), there are no uniformly accepted characteristics toiscriminate these GBS isolates. We focused on reddish isolates,nd (photo) documented applied criteria that appeared reason-ble to us. We also used several criteria because one criterionould be unreliable and no correlation can be made on visualarameters (e.g. phenotype on an agar plate versus product afterMSO treatment). However, these criteria may be unpractical for
creening in a routine diagnostic laboratory. Selecting colonies forurther investigation on the basis of phenotype through identi-cation by laboratory staff may be an insensitive method. Alsopecial media may be required (e.g. Columbia blood agar contain-ng proteose peptone 3 and starch). Thus, it is possible that annknown number of hyperpigmented isolates were missed dur-
ng the observation period. Consequently, no firm incidence raten hyperpigmented isolates can be given. The selection of 10solates during a 4.5-year period was found in an institute thategisters approximately 350–400 GBS isolates per year. Yet, weave presented the largest published series of clinical GBS show-
ng hyperpigmentation. Another limitation includes the fact thathe array of genes co-regulated by CovR is wider than the num-er of genes analyzed in this study. On the other hand, the coreovS/R regulon comprises the genes selected in this study (Jiangt al., 2005). Finally, hyperpigmentation may be associated withutations outside of the CovS/R pathways, since we found four
solates without mutations in the covS/R genes and their bind-ng sites. Analyzing all possible genes potentially responsible foryperpigmented phenotypes is, however, beyond the scope of thistudy.
In conclusion, clinical hyperpigmented isolates are associatedith increased hemolytic activity in vitro, although hyperpigmen-
ation in GBS isolates is not strictly related to invasive diseasesnd mutations within the covS/R genes. In three of four isolatesithout genetic alterations, no associated clinical syndrome wasresent. In 6 of 10 isolates, mutations were found, but only twof them were obtained from patients with acute severe infec-ions (periprosthetic joint infection and necrotizing fasciitis). Thelterations found within the CovS/R system of hyperpigmented iso-ates were consistently correlated to increased cylE and decreasedpsG (capsule) expression. These findings can suggest a correlationetween the alteration of the CovS/R system and the developmentf invasive disease, as previously postulated for isogenic mutantsLamy et al., 2004; Liu et al., 2004; Jiang et al., 2005; Whidbey et al.,013). However, covR mutations were also found in patients with-ut clinical syndromes. Adding to the puzzle are the unpredictedesults in the regulatory pattern of the analyzed genes (e.g. cfbCAMP]) compared with in vitro investigations of isogenic mutants.his highlights the necessity of performing additional clinical stud-es to further elucidate the complex GBS regulatory network and itsathogenesis.
cknowledgments
This study was presented in part at the 7th Interna-ional Conference on Gram-Positive Microorganisms, Montecatinierme, Tuscany, Italy (P165). This work is supported by theelux Stiftung, Zurich, Switzerland. We thank Drs. Andrea
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
ndimiani and Sara Droz, and Gabriela Reichen, for criti-al discussion and technical support. The manuscript wasdited by Barbara Every, BioMedical Editor, St. Albert, Alberta,anada.
PRESScal Microbiology xxx (2014) xxx–xxx 7
References
Buchanan, J.T., Stannard, J.A., Lauth, X., Ostland, V.E., Powell, H.C., Westerman, M.E.,Nizet, V., 2005. Streptococcus iniae phosphoglucomutase is a virulence factor anda target for vaccine development. Infection and Immunity 73, 6935–6944.
CLSI, 2010. Clinical and Laboratory Standards Institute, Performance Standardsfor Antimicrobial Susceptibility Testing Twentieth Informational SupplementM100-S20. Clinical and Laboratory Standards Institute, Wayne, PA.
Doran, K.S., Chang, J.C., Benoit, V.M., Eckmann, L., Nizet, V., 2002. Group B streptococ-cal beta-hemolysin/cytolysin promotes invasion of human lung epithelial cellsand the release of interleukin-8. Journal of Infectious Diseases 185, 196–203.
Florindo, C., Ferreira, R., Borges, V., Spellerberg, B., Gomes, J.P., Borrego, M.J., 2012.Selection of reference genes for real-time expression studies in Streptococcusagalactiae. Journal of Microbiological Methods 90, 220–227.
Gibson, R.L., Nizet, V., Rubens, C.E., 1999. Group B streptococcal beta-hemolysinpromotes injury of lung microvascular endothelial cells. Pediatric Research 45,626–634.
Gottschalk, B., Broker, G., Kuhn, M., Aymanns, S., Gleich-Theurer, U., Spellerberg, B.,2006. Transport of multidrug resistance substrates by the Streptococcus agalac-tiae hemolysin transporter. Journal of Bacteriology 188, 5984–5992.
Hensler, M.E., Quach, D., Hsieh, C.J., Doran, K.S., Nizet, V., 2008. CAMP factor is notessential for systemic virulence of Group B Streptococcus. Microbial Pathogenesis44, 84–88.
Horstmann, N., Sahasrabhojane, P., Suber, B., Kumaraswami, M., Olsen, R.J., Flores, A.,Musser, J.M., Brennan, R.G., Shelburne 3rd, S.A., 2011. Distinct single amino acidreplacements in the control of virulence regulator protein differentially impactstreptococcal pathogenesis. PLoS Pathogens 7, e1002311.
Imperi, M., Pataracchia, M., Alfarone, G., Baldassarri, L., Orefici, G., Creti, R., 2010. Amultiplex PCR assay for the direct identification of the capsular type (Ia to IX) ofStreptococcus agalactiae. Journal of Microbiological Methods 80, 212–214.
Jiang, S.M., Cieslewicz, M.J., Kasper, D.L., Wessels, M.R., 2005. Regulation of virulenceby a two-component system in group B streptococcus. Journal of Bacteriology187, 1105–1113.
Jones, N., Bohnsack, J.F., Takahashi, S., Oliver, K.A., Chan, M.S., Kunst, F., Glaser, P.,Rusniok, C., Crook, D.W., Harding, R.M., Bisharat, N., Spratt, B.G., 2003. Mul-tilocus sequence typing system for group B Streptococcus. Journal of ClinicalMicrobiology 41, 2530–2536.
Lamy, M.C., Zouine, M., Fert, J., Vergassola, M., Couve, E., Pellegrini, E., Glaser, P.,Kunst, F., Msadek, T., Trieu-Cuot, P., Poyart, C., 2004. CovS/CovR of group BStreptococcus: a two-component global regulatory system involved in virulence.Molecular Microbiology 54, 1250–1268.
Lin, W.J., Walthers, D., Connelly, J.E., Burnside, K., Jewell, K.A., Kenney, L.J., Rajagopal,L., 2009. Threonine phosphorylation prevents promoter DNA binding of theGroup B Streptococcus response regulator CovR. Molecular Microbiology 71,1477–1495.
Liu, G.Y., Doran, K.S., Lawrence, T., Turkson, N., Puliti, M., Tissi, L., Nizet, V., 2004.Sword and shield: linked group B streptococcal beta-hemolysin/cytolysin andcarotenoid pigment function to subvert host phagocyte defense. Proceedingsof the National Academy of Sciences of the United States of America 101,14491–14496.
Liu, G.Y., Nizet, V., 2004. Extracellular virulence factors of group B Streptococci.Frontiers in bioscience: A Journal and Virtual Library 9, 1794–1802.
Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 25,402–408.
Maisey, H.C., Doran, K.S., Nizet, V., 2008. Recent advances in understanding themolecular basis of group B Streptococcus virulence. Expert Reviews in MolecularMedicine 10, e27.
Merritt, K., Jacobs, N.J., 1978. Characterization and incidence of pigment produc-tion by human clinical group B streptococci. Journal of Clinical Microbiology 8,105–107.
Murray, B.E., Singh, K.V., Heath, J.D., Sharma, B.R., Weinstock, G.M., 1990. Comparisonof genomic DNAs of different enterococcal isolates using restriction endonu-cleases with infrequent recognition sites. Journal of Clinical Microbiology 28,2059–2063.
Nizet, V., Gibson, R.L., Chi, E.Y., Framson, P.E., Hulse, M., Rubens, C.E., 1996. GroupB streptococcal beta-hemolysin expression is associated with injury of lungepithelial cells. Infection and Immunity 64, 3818–3826.
Pritzlaff, C.A., Chang, J.C., Kuo, S.P., Tamura, G.S., Rubens, C.E., Nizet, V., 2001. Geneticbasis for the beta-haemolytic/cytolytic activity of group B Streptococcus. Molec-ular Microbiology 39, 236–247.
Ring, A., Braun, J.S., Pohl, J., Nizet, V., Stremmel, W., Shenep, J.L., 2002. Group Bstreptococcal beta-hemolysin induces mortality and liver injury in experimentalsepsis. Journal of Infectious Diseases 185, 1745–1753.
Rosa-Fraile, M., Rodriguez-Granger, J., Haidour-Benamin, A., Cuerva, J.M., Sampedro,A., 2006. Granadaene: proposed structure of the group B Streptococcus polyenicpigment. Applied and Environmental Microbiology 72, 6367–6370.
Sendi, P., Johansson, L., Dahesh, S., Van-Sorge, N.M., Darenberg, J., Norgren, M., Sjolin,J., Nizet, V., Norrby-Teglund, A., 2009. Bacterial phenotype variants in group Bstreptococcal toxic shock syndrome. Emerging Infectious Diseases 15, 223–232.
Sendi, P., Johansson, L., Norrby-Teglund, A., 2008. Invasive group B Streptococcal
ecular characterization of hyperpigmented group B Streptococci.5.003
disease in non-pregnant adults: a review with emphasis on skin and soft-tissueinfections. Infection 36, 100–111.
Sendi, P., Proctor, R.A., 2009. Staphylococcus aureus as an intracellular pathogen: therole of small colony variants. Trends in Microbiology 17, 54–58.
534
535
536
537
ING ModelI
8 f Med
S
S
S
538
539
540
541
542
543
544
545
546
547
ARTICLEJMM 50826 1–8
A. Lupo et al. / International Journal o
igge, A., Schmid, M., Mauerer, S., Spellerberg, B., 2008. Heterogeneity of hemolysinexpression during neonatal Streptococcus agalactiae sepsis. Journal of ClinicalMicrobiology 46, 807–809.
Please cite this article in press as: Lupo, A., et al., Phenotypic and molInt. J. Med. Microbiol. (2014), http://dx.doi.org/10.1016/j.ijmm.2014.0
lotved, H.C., Elliott, J., Thompson, T., Konradsen, H.B., 2003. Latex assay for serotyp-ing of group B Streptococcus isolates. Journal of Clinical Microbiology 41,4445–4447.
pellerberg, B., Pohl, B., Haase, G., Martin, S., Weber-Heynemann, J., Lutticken,R., 1999. Identification of genetic determinants for the hemolytic activity of
PRESSical Microbiology xxx (2014) xxx–xxx
Streptococcus agalactiae by ISS1 transposition. Journal of Bacteriology 181,3212–3219.
Tapsall, J.W., 1986. Pigment production by Lancefield-group-B streptococci (Strep-
ecular characterization of hyperpigmented group B Streptococci.5.003
tococcus agalactiae). Journal of Medical Microbiology 21, 75–81.Whidbey, C., Harrell, M.I., Burnside, K., Ngo, L., Becraft, A.K., Iyer, L.M., Aravind, L.,
Hitti, J., Waldorf, K.M., Rajagopal, L., 2013. A hemolytic pigment of Group BStreptococcus allows bacterial penetration of human placenta. Journal of Exper-imental Medicine 210, 1265–1281.
548
549
550
551
552