1

Re-Identification of Streptococcus bovis Isolates Causing Bacteraemia 1

According with the New Taxonomy Criteria: Still an Issue? 2

3

Beatriz Romero1,2

, María-Isabel Morosini1,2

, Elena Loza1,2

, 4

Mercedes Rodríguez-Baños1, Enrique Navas

3, Rafael Cantón

1,2 and Rosa del Campo

1,2* 5

6

7

1Servicio de Microbiología, Hospital Universitario Ramón y Cajal and IRYCIS; Unidad 8

de Resistencia a Antibióticos y Virulencia Bacteriana asociada al Consejo Superior de 9

Investigaciones Científicas (CSIC). 2CIBER en Epidemiología y Salud Pública 10

(CIBERESP). 3Servicio de Enfermedades Infecciosas, Hospital Universitario Ramón y 11

Cajal. Madrid, Spain 12

13

*Corresponding author: Mailing address: Servicio de Microbiología, Hospital 14

Universitario Ramón y Cajal. Ctra. Colmenar, Km 9,1. 28034 Madrid. Spain. 15

Phone: 34-91 3368542; Fax: 34-91 3368809; e-mail: rosacampo@yahoo.com 16

17

Short title: S. bovis re-identification. 18

19

Key words: Streptococcus bovis, bacteraemia, 16S rDNA, sodA 20

Copyright © 2011, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.00524-11 JCM Accepts, published online ahead of print on 13 July 2011

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

2

ABSTRACT 21

All Streptococcus bovis blood culture isolates recovered from Jan-2003 to Jan-2010 22

(n=52) in our institution were re-identified using new taxonomic criteria based upon 23

their genetic traits. Initial identification was performed by the semiautomatic WIDER 24

System (Fco. Soria-Melguizo, Spain) and the API 20 Strep (BioMérieux, France). All 25

isolates were re-identified by PCR amplification and sequencing of both 16S rDNA and 26

sodA genes and by mass spectrometry using MALDI-TOF MS (Bruker, Germany). 27

Results of 16s rDNA/sodA gene sequencing were as follows: Streptococcus gallolyticus 28

subsp. gallolyticus 14/14, Streptococcus gallolyticus subsp. pasteurianus 24/24, 29

Streptococcus spp. 7/0, Streptococcus infantarius subsp. infantarius 0/2, Streptococcus 30

lutetiensis 0/5, Leuconostoc mesenteroides 4/0, and Lactococcus lactis 3/3. The 31

MALDI-TOF MS identified 27 S. gallolyticus without sub-speciation, 4 L. 32

mesenteroides, 3 L. lactis, 6 S. lutetiensis whereas 12 isolates rendered a “non reliable 33

identification” result. Pulsed-field gel electrophoresis grouped all S. gallolyticus subsp. 34

gallolyticus into 3 major clusters clearly different from those of the subsp. pasteurianus 35

which, in turn, exhibited no clonal relationship. Resistance percentages of tested 36

antimicrobials were: 38% erythromycin, 23% fosfomycin, 10% levofloxacin, 6% 37

tetracycline, and 4% cotrimoxazole. The most frequent underlying diseases were 38

hepatobiliar disorders (53%), endocarditis (17%) and malignancies (12%). We conclude 39

that sequencing of sodA gene was the most discriminatory method, and that S. 40

gallolyticus subsp. pasteurianus appear to have a higher genetic diversity than S. 41

gallolyticus subsp. gallolyticus. 42

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

3

INTRODUCTION 43

Streptoccocus bovis, a non-enterococcal group D Streptococcus, can be found as part of 44

human gastrointestinal microbiota in 5-16% of individuals (17). However, it causes 45

bacteraemia and endocarditis, particularly in men and in the elderly (7-8). S. bovis 46

bacteraemia and colon tumours association was established in the late 1970s (14). 47

Moreover, recent studies have described a frequent association between its isolation 48

with chronic liver and biliary tract disorders (10). 49

Streptococcal taxonomy has progressively changed according to the description of 50

new species originally grouped as S. bovis. During the late 1990s and at the beginning 51

of this decade, several authors re-named the S. bovis biotype I as Streptococcus 52

gallolyticus subsp. gallolyticus (21); the biotype II/1 as Streptococcus lutetiensis (18) 53

and finally the biotype II/2 as S. gallolyticus subsp. pasteurianus (20). 54

Clinicians still remain unfamiliar with the new taxonomy of S. bovis species, 55

mostly due to the complexity of current nomenclature and specific identification 56

requirements based on molecular microbiology not available in routine clinical 57

laboratories. Nevertheless, due to their specific disease association and their 58

microbiology features, a proper identification of the S. bovis isolates is needed. The aim 59

of this study was to review all S. bovis bacteraemic episodes documented along the last 60

seven years in our hospital, focusing on the new taxonomy and the probable association 61

of different subspecies and pathologies. 62

63

PATIENTS AND METHODS 64

Bacterial identification. All S. bovis isolates (n= 52, from 51 patients) causing 65

bacteraemia recovered from blood cultures between January 2003 and January 2010 66

were studied. Initially, S. bovis identification was routinely performed using the 67

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

4

commercial API 20 Strep gallery (bioMérieux, Marcy l´Etoile, France) and the 68

semiautomated WIDER system (Fco. Soria Melguizo, Madrid, Spain) (5). The ability to 69

grow on bile esculin agar (BD, New York, US) was determined after 24 h of incubation 70

at 37ºC. Isolates were tested for the presence of Lancefield streptococcal antigen D by 71

agglutination using the SLIDEX® Strepto Plus Kit (BioMérieux, Marcy l'Etoile, 72

Francia). 73

Subsequently, both 16S rDNA PCR with universal primers (16-F: 5’-AGGAT 74

TAGATACCCTGGTAGTCCA-3’; and 16-R: 5’-AGGCCCGGGAACGTATTCAC-3’) 75

and sodA PCR with degenerate primers (dl: 5´-CCITAYICITAYGAYG 76

CIYTIGARCC-3´; and d2: 5´-ARRTARTAIGCRTGYTCCCAIACRTC-3´) were 77

performed (18). Amplicons (500 bp for 16S rDNA and 609 bp for sodA, respectively) 78

were sequenced using an ABI prism 377 automated sequencer (PE, Norwalk, Conn, US) 79

after ExoSAP-IT (Amersham, Bucks, UK) purification. Sequences were aligned using 80

the ClustalW tool from the website www.ebi.ac.uk, and phylogenetic trees were 81

constructed with the TreeView software (Michael Eisen, Stanford University, 82

California, USA). 83

Mass spectrometry using the Bruker Biotyper MALDI78 TOF MS system (Bruker 84

Daltonics, Germany) was also performed as part of the bacterial re-identification 85

scheme (3). 86

Antimicrobial susceptibility testing. Susceptibility (microdilution method) was 87

performed using the WIDER panels for Gram-positive organisms, and interpreted 88

according to Clinical and Laboratory Standards Institute criteria (5-6).

89

Genetic diversity. Clonal relatedness was determined by pulsed-field gel 90

electrophoresis (PFGE) using a protocol initially described for Streptococcus suis 91

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

5

serotype 2 (15). A dendrogram was constructed using the Phoretix 5.0 software 92

(Nonlinear Dynamics Ltd., United Kingdom) based on the Dice’s coefficient. 93

Patients. Clinical charts of all patients were reviewed with the approval of the Ethic 94

Committee of our institution to assess both demographic and clinical data as well as 95

diagnostics and treatments. All patients in whom colonic pathologies were suspected 96

were submitted to colonoscopy examination. A possible biliary source of bacteraemia 97

was assigned if there was a clinical or surgical diagnosis of acute cholecystitis or 98

cholangitis, after the exclusion of other possible focus of infection. 99

100

RESULTS 101

Isolates previously identified as Streptococcus bovis were re-identified using the current 102

nomenclature. All 52 isolates, including the non-streptococcal ones, had a clear positive 103

reaction with latex agglutination test for group D Streptococcus. API 20 Strep initial 104

data were not available, and contemporary profiles corresponded to 29 S. bovis II/2 105

(which included 24 S. gallolyticus subsp. pasteurianus and 5 S. lutetiensis isolates, 106

according with the sodA identification), 14 S. bovis I, 2 S. bovis II/4, 3 L. lactis and 4 107

Leuconostoc spp. 108

Nucleotide sequences of 16S rDNA amplicons (500 bp) allowed the classification 109

of the isolates as Streptococcus gallolyticus subsp. gallolyticus (n=14); Streptococcus 110

gallolyticus subsp. pasteurianus (n=24); Streptococcus spp. (n=7); Lactococcus lactis 111

(n=3) and Leuconostoc mesenteroides (n=4) (Figure 1). 112

Isolates were also re-classified based on the nucleotide sequence of an internal 113

fragment of the sodA gene as: S. gallolyticus subsp. gallolyticus (n=14), S. gallolyticus 114

subsp. pasteurianus (n=24), Streptococcus infantarius subsp. infantarius (n=2); 115

Streptococcus lutetiensis (n=5), and L. lactis (n=3). The seven Streptococcus spp. 116

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

6

isolates, not identified at species level by the 16S rDNA sequence analysis, were 117

characterized as S. infantarius subsp. infantarius and S. lutetiensis using the sodA gene 118

amplification and sequencing. Consistently negative results for sodA amplification were 119

obtained in the 4 isolates previously identified as L. mesenteroides by 16s rDNA. The 120

sodA nucleotide phylogenetic analysis discriminated the S. gallolyticus subsp. 121

pasteurianus isolates in two clusters (Figure 2), although aminoacidic sequences 122

remained identical (Figure 3). However, several aminoacidic differences were detected 123

in the subsp. gallolyticus group. 124

Finally, the MALDI TOF MS identified 27 S. gallolyticus without no sub-125

speciation, 4 L. mesenteroides, 3 L. lactis, 6 S. lutetiensis whereas in 12 isolates a “non 126

reliable identification” result was obtained. 127

PFGE-SmaI analysis grouped all S. gallolyticus subsp. gallolyticus into 3 major 128

clusters clearly different from those of the subsp. pasteurianus which, in turn, exhibited 129

no PFGE related patterns among them (Figure 4). 130

Antimicrobial susceptibility results are shown in Table 1. All isolates remained 131

susceptible to penicillin, ampicillin, amoxicillin/clavulanate, oxacillin, 132

quinupristin/dalfopristin, linezolid, and rifampicin. Glycopeptides resistance, 133

vancomycin and teicoplanin, was only observed on L. mesenteroides isolates. As 134

expected, all streptococcal isolates (n=45) showed low level resistance to 135

aminoglycosides. Additionally, 16 isolates exhibited high level resistance to 136

streptomycin (>1,000 mg/l), whereas high level of resistance to gentamicin was not 137

observed. Resistance percentages for other antimicrobial tested varied according with 138

bacterial species (Table 1). 139

When considering clinical data from all 51 patients with documented S. bovis 140

bacteraemia, patient’s median age was 73.5±15 years (range from 27 to 98 years, mode 141

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

7

82 years). Gender distribution was 28 male (54.9%) and 23 female (45.1%). A mean of 142

6.5 episodes per year was documented, being the maximum (n=10 episodes) in 2005. 143

An accumulation of cases between April-June (n=21) were observed, being October-144

December the months with lower incidence (n=7). Blood cultures were obtained in 145

different wards, being the Emergency Unit the most frequent (51%) followed by 146

General Medicine (16%), ICU (10%), Gastroenterology (8%) and others (15%). The 147

most frequent symptoms referred at hospital admission were fever and abdominal pain. 148

Patient’s underlying diseases are showed in Table 2. The most common feature 149

detected was hepatobiliary disorder (53%), followed by endocarditis (17.6%), 150

cardiovascular disease (11.7%), diverticulitis and/or colon polyps (11.7%), digestive 151

tract carcinomas (8%), ferropenic anemia (6%), and other malignancies (4%). Only one 152

patient died during the bacteraemic episode. Interestingly, a female had two different 153

bacteraemic episodes separated by 3 years and caused by genetically different S. 154

gallolyticus subsp. pasteurianus isolates. 155

156

DISCUSSION 157

The former S. bovis group is, at present, divided in four major subspecies (12), being 158

two of them genetically close, S. gallolyticus subsp. gallolyticus, formerly S. bovis 159

biotype I, and S. gallolyticus subsp. pasteurianus, formerly S. bovis biotype II/2, and the 160

other two more distantly related, S. infantarius subsp. infantarius, and S. lutetiensis 161

(named as S. infantarius subsp coli by some authors) (2), formerly S. bovis biotype II/1. 162

Streptococcus gallolyticus subsp. macedonicus, is also member of the group but it is 163

generally consider nonpathogenic for humans. However, and mainly due to 164

discrepancies still remaining, a deeply genetically-based classification is awaited. 165

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

8

It has been reported a clear association between S. gallolyticus subsp. gallolyticus 166

causing bactearemia and/or endocarditis and presence of colon cancer (1,4,11). 167

Nevertheless, the basis of this association is still not completely understood. In this 168

study, half of the patients had hepatobiliary disorders, whereas endocarditis or 169

cardiovascular diseases were only observed in 17.6 and 11.7% of the patients, 170

respectively. Colonic polyps and carcinomas of gastrointestinal tract were diagnosed in 171

11.7 and 8% of the patients, respectively. When considering only those patients with S. 172

gallolyticus subsp. gallolyticus bacteraemia (n=14), malignances were previously 173

diagnosed in 7 of them (50%) although some of them non-related to the gastrointestinal 174

tract (lymphoma, leukemia or prostate carcinoma), even though colonoscopy 175

exploration was not performed in all patients. 176

Different authors suggest a positive association between S. gallolyticus subsp. 177

gallolyticus isolation with male and elderly patients and endocarditis (7,18). In our case, 178

gender distribution was balanced (55% of male individuals), age of presentation was 179

around 70 years old, and endocarditis was only detected in 17% of patients. Our results 180

are in concordance with those recently published by other Spanish group (9). 181

During the studied period, 76,089 blood samples from patients with suspected 182

bacteraemia were processed in our laboratory. From these, 6,935 (9.1%) were positive, 183

and 52 of them rendered of S. bovis identification (0.74%), although considering the 184

present data of our work, only 45 isolates corresponded in fact to this specie (0.64%). In 185

the same period, 270 infective endocarditis were diagnosed in our center, mostly due to 186

Staphylococcus aureus (30.7%), coagulase-negative staphylococci (17.4%), and 187

Enterococcus (13.3%). “S. bovis” was identified in 7 cases (6 S. gallolyticus sbsp. 188

pasteurianus, and 1 S. gallolyticus sbsp gallolyticus); representing the 2.6% of the total 189

endocarditis (Navas et al. unpublished results). 190

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

9

Surprisingly, 4 Leuconostoc mesenteroides and 3 Lactococcus lactis causing 191

bacteraemia were initially identified as S. bovis by the semi-automated system used in 192

our laboratory. When re-identification was obtained by 16S rDNA sequence, bacterial 193

identification were retested using the same system, observing that the misleading 194

identification was due to their slow biochemical reactivity; at 24 h of incubation the 195

identification was again S. bovis whereas at 48 h, the identification was correct. 196

Moreover, vancomycin resistance in Leuconostoc isolates was only observed at 48 197

hours. Another important factor to be considered is the positive agglutination of our 198

Leuconostoc and Lactococcus isolates for the Lancefield antigen D. We were unable to 199

find similar references, and this fact has been only demonstrated for Pediococcus spp. 200

isolates (16). 201

As it has been described by other authors (13,19), we detected differences for the 202

isolates identification between the 16S rDNA and the sodA gene sequences methods, 203

being the sodA gene partial sequence considered as the most accurate identification. 204

Recently it has been described high concordance between MALDI-TOF MS and sodA 205

nucleotide sequenciation (12), although in our experience, Bruker technology is unable 206

to identified Streptococcus isolates yet, especially S. gallolyticus subsp. pasteurianus. 207

Contemporary API 20 Strep results agree with the soda classifications, although 5 S. 208

lutetiensis showed the same profile that S. gallolyticus subsp. pasteurianus. 209

Nevertheless, API 20 Strep can be considered a useful method to discriminate the two 210

main S. gallolyticus subspecies, but from a clinical and epidemiological point of view it 211

could be also of interest to further discriminate S. infantarius subsp. infantarius, and S. 212

lutetiensis using the sodA analysis. 213

In spite of many efforts have been direct to improve the S. bovis group 214

identification, data about genetic studies remain unknown. MLST tool has not yet 215

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

10

developed, and only one reference of high genetic diversity among Italian S. bovis 216

isolates by PFGE (22). In our experience, also high genetic diversity was observed 217

among the 45 streptococcal isolates. 218

Our study demonstrated that in a routine clinical microbiology laboratory, 219

incorporation of a MALDI TOF MS system will partially solve identification problems 220

of S. bovis group. Nevertheless, this system will advise the presence of S. gallolitycus 221

and further molecular testing can be performed if needed. Data entries in the data base 222

for different S. gallolitycus subspecies should be a future goal of MALDITOF MS 223

system as performed with other species. 224

225

226

ACKNOWLEDGMENTS 227

RdC has a “Miguel Servet” contract (CB05/137) from the Instituto de Salud Carlos III-228

FIS. MR-B has a contract from the Instituto de Salud Carlos III-FIS, project 229

AI07/90034. This work was partially funded by the European Project TROCAR 230

(HEALTH-F3-2008-223031). 231

232

233

234

235

236

237

238

239

240

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

11

REFERENCES 241

1. Abdulamir, A. S., R. R. Hafidh, and F. A. Bakar. 2011. The association of 242

Streptococcus bovis/gallolyticus with colorectal tumors: The nature and the 243

underlying mechanisms of its etiological role. J. Exp. Clin. Cancer Res. 30:11. 244

2. Beck, M., R. Frodl, and G. Funke. 2008. Comprehensive study of strains 245

previously designated Streptococcus bovis consecutively isolated from human 246

blood cultures and emended description of Streptococcus gallolyticus and 247

Streptococcus infantarius subsp. coli. J. Clin. Microbiol. 46:2966-2972. 248

3. Bizzini, A., C. Durussel, J. Bille, G. Greub, and G. Prod’hom. 2010. 249

Performance of matrix-assisted laser desorption ionization-time of flight mass 250

spectrometry for identification of bacterial strains routinely isolated in a clinical 251

microbiology laboratory. J. Clin. Microbiol. 48:1549–1554. 252

4. Boleij, A., R. M. Schaeps, and H. Tjalsma. 2009. Association between 253

Streptococcus bovis and colon cancer. J. Clin. Microbiol. 47:516. 254

5. Cantón, R., et al. 2000. Evaluation of the Wider system, a new computer-assisted 255

image-processing device for bacterial identification and susceptibility testing. J. 256

Clin. Microbiol. 38:1339-1346. 257

6. Clinical and Laboratory Standards Institute. 2007. Performance standards for 258

antimicrobial susceptibility testing; 17th

informational supplement. Document 259

M100-S17, CLSI, Wayne, PA. 260

7. Corredoira, J., et al. 2008. Characteristics of Streptococcus bovis endocarditis 261

and its differences with Streptococcus viridans endocarditis. Eur. J. Clin. 262

Microbiol. Infect. Dis. 27:285-291. 263

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

12

8. Durante-Mangoni, E., et al. 2008. Current features of infective endocarditis in 264

elderly patients: results of the International Collaboration on Endocarditis 265

Prospective Cohort Study. Arch. Intern. Med. 168:2095–2103. 266

9. Fernández-Ruiz, M., et al. 2010. Streptococcus bovis bacteraemia revisited: 267

clinical and microbiological correlates in a contemporary series of 59 patients. J. 268

Infect. 61:307-313. 269

10. González-Quintela, A., C. Martínez-Rey, J. C. Castroagudin, M. C. Rajo-270

Iglesias, and M. J. Domínguez-Santalla. 2001. Prevalence of liver disease in 271

patients with Streptococcus bovis bacteraemia. J. Infect. 42:116–119. 272

11. Gupta, A., R. Madani, and H. Mukhtar. 2010. Streptococcus bovis 273

endocarditis, a silent sign for colonic tumour. Colorectal Dis. 12:164-171. 274

12. Hinse, D., et al. 2011. Differentiation of species of the Streptococcus 275

bovis/equinus-complex by MALDI-TOF Mass Spectrometry in comparison to 276

sodA sequence analyses. Syst. Appl. Microbiol. 34:52-57. 277

13. Hoshino, T., T. Fujiwara, and M. Kilian. 2005. Use of phylogenetic and 278

phenotypic analyses to identify nonhemolytic streptococci isolated from 279

bacteremic patients. J. Clin. Microbiol. 43:6073-6085. 280

14. Klein, R., et al. 1977. Association of Streptococcus bovis with carcinoma of the 281

colon. N. Engl. J. Med. 297:800-802. 282

15. Luey, C.K., et al. 2007. Rapid pulsed-field gel electrophoresis protocol for 283

subtyping of Streptococcus suis serotype 2. J. Microbiol. Methods. 68:648-650. 284

16. Murray, P. R., E. J. Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken 285

(ed.). 2003. Manual of Clinical Microbiology. 8th

ed. American Society for 286

Microbiology, Washington, D.C. 287

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

13

17. Noble, C. J. 1978. Carriage of group D streptococci in the human bowel. J. Clin. 288

Pathol. 31:1182–1186. 289

18. Poyart, C., G. Quesne, and P. Trieu-Cuot. 2002. Taxonomic dissection of the 290

Streptococcus bovis group by analysis of manganese-dependent superoxide 291

dismutase gene (sodA) sequences: reclassification of 'Streptococcus infantarius 292

subsp. coli as Streptococcus lutetiensis sp. nov. and of Streptococcus bovis 293

biotype II.2 as Streptococcus pasteurianus sp. nov. Int. J. Syst. Evol. Microbiol. 294

52:1247-1255. 295

19. Sasaki, E., R. Osawa, Y. Nishitani, and R. A. Whiley. 2004. Development of a 296

diagnostic PCR assay targeting the Mn-dependent superoxide dismutase gene 297

(sodA) for identification of Streptococcus gallolyticus. J. Clin. Microbiol. 298

42:1360-1362. 299

20. Schlegel, L., et al. 2000. Streptococcus infantarius sp. nov., Streptococcus 300

infantarius subsp. infantarius subsp. nov. and Streptococcus infantarius subsp. 301

coli subsp. nov., isolated from humans and food. Int. J. Syst. Evol. Microbiol. 302

50:1425–1434. 303

21. Schlegel, L., F. Grimont, E. Ageron, P. A. D. Grimont, and A. Bouvet. 2003. 304

Reappraisal of the taxonomy of the Streptococcus bovis/Streptococcus equinus 305

complex and related species: description of Streptococcus gallolyticus subsp. 306

gallolyticus subsp. nov., S. gallolyticus subsp. macedonicus subsp. nov. and S. 307

gallolyticus subsp. pasteurianus subsp. nov. Int. J. Syst. Evol. Microbiol. 53:631–308

45. 309

22. Tripodi, M. F., R. Fortunato, R. Utili, M. Triassi, and R. Zarrilli. 2005. 310

Molecular epidemiology of Streptococcus bovis causing endocarditis and 311

bacteraemia in Italian patients. Clin. Microbiol. Infect. 11:814-819. 312

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

Table 1. Antibiotic susceptibility for the different species studied. 1

Antibiotic Resistance Percentage (MIC50)

Stra Lev Ery Cln Fos SxT

S. gallolyticus

subsp. gallolyticus n=14

42.8

(≤1000)

14.2

(2)

35.7

(≤0.5)

35.7

(≤0.5)

35.7

(≤32)

35.7

(≤1/19)

S. gallolyticus

subsp. pasteurianus n=24

20.8

(≤1000)

8.3

(≤1)

37.5

(≤0.5)

25.0

(≤0.5)

29.1

(≤32)

25.0

(≤1/19)

S. infantarius

subsp. infantarius n=2

0

(≤1000)

0

(≤1)

0

(≤0.5)

0

(≤0.5)

0

(≤32)

0

(≤1/19)

S. lutetiensis n=5

60.0

(>1000)

40.0

(2)

60.0

(>2)

60.0

(>2)

20.0

(≤32)

20.0

(≤1/19)

L. lactis n=3

66.6

(>1000)

0

(2)

33.3

(≤0.5)

33.3

(≤0.5)

66.6

(>64)

66.6

(>2/38)

L. mesenteroides n=4

0

(≤0.25)

0

(2)

50.0

(≤0.5)

25.0

(≤0.5)

25.0

(≤0.32)

25.0

(≤1/19)

a: High Level Resistance (>1000 mg/L). Str: streptomycin; Lev: levofloxacin; Ery: erythromycin; Cln: 2

clindamycin; Fos: fosfomycin; SxT: trimethropim sulfamethoxazole. 3

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

Table 2. Patient’s clinical features.

S. gallolyticus

subsp.

gallolyticus

S. gallolyticus subsp.

pasteurianus S. lutetiensis

S. infantarius

subsp.

infantarius

L. lactis L. mesenteroides

Cholecystitis / Cholangitis 3 5 1

Cirrhosis 2 10 1 1 1

HVC 5 1 1

Acute appendicitis 1

Colonic adenoma 1 4 1

Colonic polyps 2 7 1

Rectal carcinoma 1

Diabetes 2

Isquemic Cardiopathy 3 1 1 1

Endocarditis 1 6

Mitral stenosis 1

Renal Insufficiency 1

Urinary tract infection 1 1

Bladder cancer 1

Prostate cancer 1

Pulmonary cancer 1

MALT lymphoma 1

Leukemia 1

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

0.1

52

50

45

36

32

23

16

11

1

5

49

44

34

30

20

14

6

18

47

46

37

3

9

24

38

42

39

8

35

27

19

22

51

41

40

31

28

26

25

21

13

10

4

7

48

43

33

29

17

12

2

15

Group B. S. gallolyticus sbsp. gallolyticus/pasteurianus

Group C. S. gallolyticus sbsp. pasteurianus

Group D. S. lutetiensis/S. infantarium sbsp. infantarium

Group E. L. lactis

Group F. L. mesenteroides

Group A. S. gallolyticus sbsp. gallolyticus/pasteurianus

1

2

Figure 1. Clustering of 16S rDNA nucleotide sequences obtained fron all 52 isolates, 3

and constructed by the TreeView sowftare. 4

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

Group 1. S. gallolyticus sbsp. pasteurianus

Group 2. S. gallolyticus sbsp. pasteurianus

Group 3. S. gallolyticus sbsp. gallolyticus

Group 5. S. infantarium sbsp. infantarium

Group 4. S. lutetiensis

Group 6. L. lactis

0.1

41

51

40

31

28

25

21

10

7

4

52

30

20

15

1

12

14

36

6

11

29

50

23

43

37

47

46

3

38

9

24

19

22

27

18

34

32

17

5

16

2

49

48

45

44

33

13

26

5

6

Figure 2. Clustering of sodA nucleotide sequences obtained from non-Leuconostoc 7

isolates (n=48) and constructed by the TreeView sowftare.8

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

1 NVNAALEKHPEIGDDLEALLADVDSIPTDIRQAVINNGGGHLNHALFWELLSPEKQEPTTQVLAAIEEAFGSFDEFKAAFTQAATTRFGSGWAWLVVNENGKLEVLSTANQDTPISQGKAP

d E A A D

e F E A T A D

a E A A

c E A A

g E A A V

b EA N A A

f E A V

a A L VI L K A A S F ADV V K KDD I N E K

A L VI L K A A S F ADV V K KDD I N E K

4 A L VI EL K A A S AD ED T KD T T L E K

41

18

34

2

49

52

30

14

6

29

23

43

37

46

9

9

10

Figure 3. Aminoacidic sequences comparison for the SodA protein from the different sodA alleles detected. 11

12

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from

Figure 4. Dendrograme obtained for the 24 S. gallolyticus subsp. pasteurianus and 14

S. gallolyticus subsp. gallolyticus isolates using the UPGMA method and the Phoretix

5.0 sowftare.

S. gallolyticus sbsp. pasteurianus

S. gallolyticus sbsp. gallolyticus

S. gallolyticus sbsp. pasteurianus

S. gallolyticus sbsp. gallolyticus

on Septem

ber 27, 2020 by guesthttp://jcm

.asm.org/

Dow

nloaded from