1
Comparison of Virulence Gene Identification, Ribosomal Spacer PCR, and Pulsed Field Gel 1
Electrophoresis for Strain-Typing Staphylococcus aureus Isolated from Cases of Subclinical 2
Bovine Mastitis in the USA 3
4
Pamela R.F. Adkins, a# John R. Middleton,a Lawrence K. Foxb 5
6
Department of Veterinary Medicine and Surgery, University of Missouri, Columbia 65211a; 7
Department of Veterinary Clinical Sciences, Washington State University, Pullman 99164b 8
9
Running Head: Strain-Typing S. aureus of Bovine Origin 10
11
#Address correspondence to Pamela R. F. Adkins, [email protected] 12
JCM Accepted Manuscript Posted Online 18 May 2016J. Clin. Microbiol. doi:10.1128/JCM.03282-15Copyright © 2016, American Society for Microbiology. All Rights Reserved.
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
2
ABSTRACT 13
Staphylococcus aureus is one of the most important pathogens causing contagious mastitis in 14
dairy cattle worldwide. The objectives of this study were to determine if the recently described 15
Staphylococcus aureus Genotype B was present among previously characterized isolates from 16
cases of bovine intramammary infection in the USA and to compare pulsed-field gel 17
electrophoresis (PFGE) to the combination of ribosomal spacer PCR (RS-PCR) and virulence 18
gene identification for strain-typing of S. aureus. The hypothesis was that isolates that were 19
previously characterized as contagious would be identified as Genotype B and that the two 20
strain-typing methods would be comparable. Isolates were selected from a collection of S. 21
aureus isolates from eight dairy farms. Mammary quarter milk somatic cell count (SCC) and N-22
acetyl-β-D-glucosaminidase (NAGase) activity was known and used to evaluate strain 23
pathogenicity. RS-PCR with conventional gel electrophoresis was performed and PCR was used 24
for toxin gene identification. RS-PCR patterns were associated with a specific virulence gene 25
pattern, as previously reported. Five RS-PCR banding patterns were identified. None of the 26
isolates were characterized as Genotype B. No association between RS-PCR types and milk SCC 27
was found, however milk NAGase activity was significantly higher in milk from mammary 28
glands infected with RSP-type-1 than those infected with RSP-type-2. The discriminatory 29
powers were 1.0 and 0.46 for PFGE and RS-PCR, respectively. These data suggest that 30
Genotype B may have a limited geographic distribution and PFGE is more discriminatory than 31
RS-PCR performed with conventional gel electrophoresis for strain-typing S. aureus isolates of 32
bovine origin. 33
34
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
3
INTRODUCTION 35
Staphylococcus aureus is one of the most important pathogens causing contagious mastitis in 36
dairy cattle worldwide (1-3). Although S. aureus is considered a contagious intramamammary 37
pathogen and spread from cow-to-cow during milk, strains of this agent are most often farm 38
specific and there is generally a predominant strain in a herd (4). Several methods have been 39
used to characterize and determine genetic subtypes or strain-types of this pathogen, allowing 40
identification of the predominant contagious strains. Methods commonly used to strain-type S. 41
aureus include pulsed-field gel electrophoresis (PFGE) (5), spa-typing (6), and multilocus 42
sequence typing (MLST) (7). Although labor intensive, PFGE is considered the gold standard 43
strain-typing method for this genus and species (8). Currently, an accurate, fast, and inexpensive 44
tool enabling one to determine which strains are important and likely to be a significant problem 45
for a farm is not readily available. 46
Recent studies from Switzerland used ribosomal spacer (RS)-PCR to characterize S. aureus 47
isolated from bovine milk samples and found various strain types, classified as genotypes, of S. 48
aureus possessing different virulence and pathogenicity factors (9-11). Several genotypes were 49
identified, with Genotype B (GTB) and Genotype C (GTC) being identified commonly, while 50
other genotypes (GTOG) were found rarely (9). Overall, Genotype B isolates were found to be 51
highly contagious, as GTB was the sole strain-type observed in herds having significant S. 52
aureus problems where prevalence of S. aureus intramammary infection (IMI) ranged from 65% 53
-87% of the herd (9,12). Genotype B isolates were also considered to have increased 54
pathogenicity, as SCC for this type was significantly higher than for GTC or GTOG. Further, 55
GTC and GTOG only caused infections in one to three cows per farm and were therefore not 56
considered very contagious (9,12). It was also shown that there was an association between RS-57
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
4
PCR genotype and virulence gene patterns. Specifically, GTB was characterized by the presence 58
of S. aureus enterotoxin A (sea) and D (sed), and by a polymorphism within the leukotoxin E 59
gene (lukE) caused by a point mutation, classified as lukEB (9). 60
Ribosomal spacer PCR has been described as a technique used to rapidly identify bacteria and, in 61
the case of Staphylococcus species, has been found to show significant intraspecies variation 62
(13). The method is completed by using a set of primers to amplify the spacer regions between 63
the 16S and 23S genes of the rRNA genetic loci. The spacer region exhibits a large degree of 64
sequence and length variations at the level of genus and species. Also, within a single genome, 65
there frequently are multiple rRNA genetic loci, and regions found within these loci also show 66
variations in length and sequence. This diversity is due, in part, to variations in the number and 67
type of tRNA sequences found within the spacers (14, 15). Ribosomal spacer PCR has 68
previously been compared to PFGE (16), and, when applied to the investigation of a methicillin 69
resistant S. aureus outbreak in humans, RS-PCR was found to be as discriminatory as PFGE. 70
However, it was noted that RS-PCR failed to distinguish isolates with minor genetic differences. 71
Research comparing RS-PCR and PFGE to characterize S. aureus isolated from bovine 72
intramammary infections (IMI) is limited, including one study done on a group of S. aureus 73
isolates collected from Holstein-Friesian cows from Northern Italy that found PFGE to have a 74
slightly higher discriminatory value than RS-PCR, 0.8236 and 0.8943, respectively (17). 75
However, in that study, a miniature electrophoresis system was used, rather than traditional gel 76
electrophoresis. Therefore, the objectives of this study were to determine if S. aureus GTB was 77
present among previously characterized isolates from cases of bovine IMI and to compare 78
pulsed-field gel electrophoresis (PFGE) to the combination of ribosomal spacer PCR (RS-PCR) 79
using conventional gel electrophoresis and virulence gene identification for strain-typing S. 80
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
5
aureus. RS-PCR with conventional gel electrophoresis was chosen over the method described 81
by Fournier et al. (9) because the laboratory equipment needed for this technique is available in 82
many diagnostic laboratories around the world, making the method less expensive and more 83
universally available to dairy farmers and researchers. The hypothesis was that isolates that were 84
previously characterized as contagious would be identified as Genotype B and that the two 85
strain-typing methods would be comparable. 86
MATERIALS AND METHODS 87
Isolates were selected from a cryopreserved collection of S. aureus strains isolated from the milk 88
of cows on eight dairy farms in the USA (5). These isolates were used because of the large 89
amount of information that had previously been collected pertaining to these specific isolates, 90
including strain-type based on PFGE, prevalence of each strain within herd, and mean milk 91
SCC and N-acetyl-β-D-glucosaminidase (NAGase) activity in mammary quarters infected with 92
each strain. The isolate collection included 35 PFGE strain-types of S. aureus, some of which 93
were isolated from multiple cows within a herd while others were isolated from only one cow. At 94
least one isolate within each PFGE strain-type was selected for characterization. If more than one 95
cow was affected with a single PFGE strain, up to three representative isolates were randomly 96
selected from each farm within a given strain-type. Selected isolates were sub-cultured onto 97
brain heart infusion agar slants (Remel, Lenexa, KS, USA) at Washington State University and 98
shipped chilled to the University of Missouri for further characterization. 99
At the University of Missouri, isolates were grown on Columbia Blood Agar (CBA) (Remel, 100
Lenexa, KS, USA) and confirmed to be coagulase positive staphylococci based on colony 101
morphology, a positive catalase test, and a positive coagulase test (Remel, Lenexa, KS, USA) . 102
Genomic DNA was extracted using DNAeasy Tissue Kit (Qiagen, Valencia, CA, USA). 103
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
6
Standard PCR was performed to identify the presence or absence of the nuc gene in order to 104
confirm the isolates as S. aureus (18). Standard PCR was also used to confirm the presence or 105
absence of the leukotoxin E (lukE) gene and enterotoxin genes, including sea, sed, seg, sei, and 106
sej. All primers and PCR running conditions were used as previously described (9,10). For 107
strain-typing by RS-PCR, previously described methods were used (13). DNA from known GTB 108
isolates were used as positive controls for both standard PCR and RS-PCR, including S. aureus 109
54365/2, 133783/58, 187173/86, 175994/259, and 170649/92 (17). All RS-PCR DNA products 110
were separated using 1% agarose gel electrophoresis and bands were visualized and 111
photographed with UV transillumination. 112
For interpretation of the RS-PCR results, two patterns were considered different when one or 113
more band difference was noted. A unique name (RSP-type) was given to each RS-PCR banding 114
that was identified. For interpretation of PFGE results, isolates that had indistinguishable 115
banding patterns, based on having the same number of bands and corresponding bands of the 116
same apparent molecular weight, were considered to be the same strain (19). A unique strain 117
name was given to each PFGE strain that was identified. 118
In order to compare the discriminatory power of these two methods, an index of discrimination 119
based on Simpson’s index of diversity was used according to the formula,120
)1()1(
111
−−
−= =
j
s
jj xx
NND , where N is the total number of strains in the sample population, s 121
is the total number of types described, and nj is the number of strains belonging to the jth type 122
(20). To make this calculation, one isolate from each farm was used (n = 8), as the calculation is 123
based on the ability to discriminate between epidemiologically unrelated isolates. Isolates were 124
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
7
selected from each farm for the calculation using Microsoft Excel (2013) random number 125
generator. 126
Toxin gene prevalence, genotype, and PFGE strain-type data were expressed as absolute 127
numbers and percentages. For all statistical analyses, a commercially available software package 128
was used (Sigmaplot 13.0, SyStat Software, Inc., San Jose, CA, USA). Prevalence data were 129
compared using the Chi-square or Fisher’s Exact test as appropriate. For the purposes of 130
statistical analysis of SCC, the linear score somatic cell count (LSCC) was used to achieve 131
normality. A one-way analysis of variance (ANOVA) for normally distributed data (LSCC) and 132
a Kruskal-Wallis ANOVA on ranks (NAGase), respectively, were used to compare differences in 133
SCC and NAGase between RS-PCR genotypes. Significance was declared at P < 0.05. 134
RESULTS 135
Overall, staphylococcal enterotoxin (SET) genes were uncommonly identified among these 136
isolates, with 82% (53/65) of the isolates testing negative for all five of the tested SET genes 137
(Table 1). The most common toxin genes identified were lukE (100%) and sei (10.7%; 7/65). 138
Enterotoxin genes sed and sej were not identified in any of the isolates. The seg gene was only 139
identified in isolates from Farm 2, while the sea gene was only detected in isolates from Farms 5, 140
6, and 7. Only 12/65 (18.5%) isolates were found to have more than one toxin gene, with five 141
isolates being positive for lukE and sea, four isolates being positive for lukE and sei, and three 142
isolates being positive for lukE, sei, and seg. None of the isolates had the enterotoxin gene 143
profile (sed, sej, and sea) commonly found among GTB isolates (9). 144
Based on RS-PCR banding patterns, isolates were characterized into five different RSP-types. 145
(Figure 1). All identified RSP-types were found to have consistent toxin profiles. RSP-type-1 146
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
8
was identified on the majority of farms (6/8) and all isolates were only positive for lukE. RSP-147
type-2 was found on 3 different farms and had the toxin profile of lukE and sea. RSP-type-3 was 148
found on three different farms and had the toxin profile of lukE and sei. RSP-type-4 was 149
identified three times and all three isolates were from the same farm and had identical toxin 150
profiles, including lukE, sei, and seg. RSP-type-5 was also only found on one farm, farm 3, and 151
all isolates were only positive for lukE, which is the same toxin profile as RSP-type-1. The 152
majority of isolates were characterized as RSP-type-1 (50/65; 76.9%). The prevalence of RSP-153
types 2, 3, 4, and 5 did not differ from each other (P = 0.68). The prevalence of RSP-type-1 was 154
significantly higher than all other RSP-types (P < 0.01). Based on RS-PCR banding patterns and 155
toxin gene profiles, none of the isolates were characterized as GTB (9). Milk somatic cell count 156
was not statistically significantly different between the different RSP-types (P ≥ 0.08; Table 1). 157
Milk NAGase activity was significantly higher for RSP-type-1 when compared to RSP-type-2 (P 158
=0.03). 159
The discriminatory power of PFGE was 1.0 and for RS-PCR it was 0.46. A total of 35 different 160
strain-types were identified based on PFGE, as compared to five with RS-PCR (Table 2). With 161
PFGE, 4/35 (11.4%) strains were identified on more than one farm. With RS-PCR, 3/5 (60%) 162
RSP-types were found on more than one farm and this included the predominant type, RSP-type-163
1. 164
DISCUSSION 165
In this study, no isolates were characterized as the previously described GTB. Overall, very few 166
RS-PCR patterns were identified with the majority of isolates being categorized as RSP-type-1, 167
and RSP-type-1 was found in 75% of the herds evaluated. The finding of dominant RSP-types is 168
similar to previous reports that used RS-PCR and toxin profiles to strain-type isolates (9). This 169
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
9
finding is in contrast to other studies that used PFGE strain-typing to differentiate isolates and 170
demonstrated that the majority of strain-types are unique to one herd (4), which is similar to the 171
PFGE data in this study (5). Two RSP-types, RSP-type-4 and RSP-type-2 , could be classified 172
as a contagious strain based on the findings of this study. All three RSP-type-4 isolates were 173
from the same farm, had the same toxin profile and same PFGE pattern, while all five RSP-type-174
2 isolates had the same toxin profile and same PFGE pattern, but were found on three different 175
farms. Based on RS-PCR alone, RSP-type-1 would be considered highly contagious, being found 176
very commonly on 6 of the 8 farms included in the study. However, when including PFGE data, 177
some of the RSP-type-1 isolates would be considered contagious, such as the 63 isolates that 178
were found to be PFGE strain-type 32 from farm 7. However, there are other PFGE strain-types 179
that were characterized as RSP-type-1 that were only found once on a farm, such as strain types 180
34, 35, 37, and 38 that were also found on farm 7. Using RS-PCR alone would have grouped all 181
of these isolates as one type, which could lead to the presumption of the majority of cows being 182
infected with a highly contagious strain potentially leading to excessive culling of cattle. Pulsed-183
field gel electrophoresis, while more time-consuming to perform, provided greater discrimination 184
among strains and thus the potential for culling cattle with sporadic (non-host-adapted or less 185
contagious) strain-types could be prevented. 186
Unlike previous reports, this study was unable to find an association between RSP-types and 187
strain pathogenicity as measured by milk SCC. However, NAGase activity was significantly 188
higher in mammary glands infected with RSP-type-1 than those infected with RSP-type-2. Milk 189
NAGase activity is an indicator of tissue damage, as it is released into the milk from mammary 190
epithelial cells and, to a lesser extent, from milk somatic cells. (21, 22) and SCC is a measure of 191
mammary gland inflammation. Previous reports have indicated that GTB is not only highly 192
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
10
contagious but also associated with increased pathogenicity based on SCC being significantly 193
higher than for GTC or GTOG (9). Milk from glands infected with GTB isolates have been 194
found to have a geometric mean somatic cell count of 7.85 x 105 cells/ml (9), corresponding to a 195
LSCC of 6.1, which is lower than the median LSCC for all RSP-types found in this study except 196
RSP-type-2 and RSP-type-3. Milk NAGase activity was not assessed in previous studies using 197
RS-PCR, but did provide an indication of differences in pathogenicity between strain-types in the 198
present study. 199
Very few enterotoxin genes were identified in this study, which is similar to two previous studies 200
conducted in the USA (23, 24) and contrary to a third study (25). Previous work has shown that 201
sea negative results were observed in 100% of non-GTB strains and sea positive results were 202
observed in 77% of GTB isolates (9). Given that the detection of the gene encoding for SEA is 203
an important determinant for identifying GTB and this gene was not detected in three prior 204
studies from different geographical locations in the USA and at a low prevalence in the present 205
study, suggests that GTB is not prevalent on U.S. farms. Other studies have found sec to be the 206
most commonly identified enterotoxin gene in the northern USA (23, 24). In contrast, a study 207
conducted in southern USA found sem and sen to be the most common enterotoxin genes 208
identified (25). These specific enterotoxin genes were not tested for in this study due to a lack of 209
association with the previously described GTB (9). Studies from other parts of the world have 210
identified different prevalences of enterotoxin genes, suggesting possible geographic differences 211
(26, 27). Because of these differences, typing isolates by virulence gene identification in 212
combination with RS-PCR likely does not represent a universally applicable rapid method for 213
identifying contagious and pathogenic S. aureus. 214
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
11
There was not a sufficient number of enterotoxin producing strains in this study to test the 215
association between the presence of toxin genes and RS-PCR types or PFGE strain-type. 216
However, previous research done in the USA found no correlation between PFGE strain-types 217
and enterotoxin profiles (25). Overall, the lack of detection of enterotoxin genes meant there 218
was insufficient data to determine whether there was a statistically significant association 219
between toxin gene presence and SCC or NAGase activity. Based on the previously published 220
data not demonstrating a statistically significant relationship between PFGE strain-type and SCC 221
or NAGase activity (5), and placing that information in the context of these strains infrequently 222
having toxin genes, suggests that the toxin genes screened for are likely not important 223
determinants of udder inflammation for these strain-types. Other reports have found the 224
presence of virulence genes, specifically spa and sej, to be risk factors for subclinical mastitis 225
and suggest that strain characteristics and the expression of specific combinations of virulence 226
factors could result in the development of subclinical mastitis (28). However, in that study all 227
isolates harbored at least one enterotoxin gene and most had two or more, which is in contrast to 228
the findings of this study. 229
The discriminatory power of PFGE for strain-typing S. aureus has been previously described, 230
and ranges from 0.83 to 0.8887 (27, 29, 30). When comparing several strain-typing methods to 231
characterize S. aureus isolates of bovine origin, PFGE had the highest discriminatory power, 232
followed by multi-locus sequence typing (MLST), multiple locus variable number tandem repeat 233
analysis (MLVA), and then spa-typing (29). Furthermore, PCR based DNA fingerprinting 234
methods have been previously shown to cluster large groups of unrelated isolates together. For 235
example, in one study, PCR based DNA fingerprinting was used and found isolates from several 236
herds in several distinct regions could be categorized as the same type with approximately half of 237
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
12
the isolates in the study characterized as one of two types (31). Conversely, PFGE tends to find 238
strains to be unique to a herd (4, 5, 32). In the present study, the discriminatory power of PFGE 239
was higher than previously described. This was likely due to the small number of herds in the 240
comparison, because a very stringent cut off value (100% similarity) was used to define a strain-241
type, and because PFGE strain-type was part of the selection criteria for this study. Other studies 242
have used a 70-84% similarity as strain cut-off values (30, 33), which would decrease the 243
discriminatory power while allowing grouping of isolates that may not be epidemiologically 244
related. Isolates with genetically indistinguishable restriction patterns can be considered the 245
same strain-type, while isolates that are closely related may have a single genetic alteration 246
(point mutation, insertion, or deletion) which can result in two to three band differences (19). 247
One recent study found RS-PCR to have a similar discriminatory power to PFGE, with values of 248
0.8236 and 0.8943, respectively (17). A higher discriminatory power for RS-PCR may have been 249
identified in that study because RS-PCR results were used as part of the selection criteria for 250
isolates that were compared using different strain typing methods. Furthermore, in that study, a 251
chip electrophoresis system was used, which can determine smaller differences in banding 252
patterns than conventional gel electrophoresis. Conventional gel electrophoresis was used herein 253
to determine the utility of the technique as applied in a reasonably well-equipped veterinary 254
diagnostic laboratory thus making the results relevant to every day usage in aiding farmers to 255
make management decisions. 256
A few limitations should be noted in this study. One is related to the equipment used in this 257
study. It has been recommended to use a miniaturized electrophoresis system (Agilent 258
Technologies) for best resolution, data evaluation, and data management of RS-PCR results (34). 259
The suggested system can be expensive and not necessarily available in all labs conducting PCR, 260
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
13
which may be considered a limitation of this technique. Because this system was not used, the 261
isolates in this study could not be classified using the previously described genotyping system 262
(9). However, to allow proper comparison of the genotype of interest, a GTB control was used. 263
Another limitation in this study is sample size. Increasing the number of farms and number of 264
isolates would allow for further assessment of prevalent genotypes found within the USA and 265
determine if GTB is truly absent from USA dairy farms. This limitation is also likely a factor in 266
some of the calculations, such as discriminatory power, and these values would likely change 267
with further isolate characterization from a larger number of farms. 268
In conclusion, GTB was not identified in this group of isolates. It is possible that GTB may be 269
localized to some European countries, as previous studies have identified this genotype in 270
Austria, Belgium, France, Germany, Italy, and Switzerland but did not find it in Ireland, 271
Slovenia, Macedonia, Norway, the Netherlands, or Sweden (35). Further work on the 272
geographical distribution of this virulent strain of S. aureus is needed. In this study, RS-PCR 273
patterns were associated with a specific virulence gene pattern, as previously reported (9). Also, 274
these data suggest that PFGE is more discriminatory than traditional methods of RS-PCR for 275
strain-typing S. aureus isolates of bovine origin. Although PFGE is more time consuming and 276
expensive than PCR-based techniques, a higher discriminatory power is important to help 277
determine the contagiousness of a strain and to define the epidemiology of S. aureus within a 278
given herd. Few enterotoxin genes were identified in the isolates studied herein, which 279
corroborates the findings from other studies conducted in the USA. Further studies are needed to 280
identify a simple method that could be applied universally to allow identification of virulent or 281
contagious S. aureus strains. 282
Funding Information 283
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
14
This study was funded by a grant from University of Missouri Phi Zeta Research Fund to Pamela 284
R. F. Adkins. The funders had no role in study design, data collection and interpretation, or the 285
decision to submit the work for publication. 286
Acknowledgements 287
This work is part P.R.F.A.’s PhD dissertation. J.R.M. and L.K.F. contributed equally to this 288
work. Special thanks to Mario Luini for sharing DNA from his collection of S. aureus Genotype 289
B isolates. 290
291
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
15
References 292
1. Piepers S, De Meulemeester L, de Kruif A, Opsomer G, Barkema HW, De Vliegher 293
S. 2007. Prevalence and distribution of mastitis pathogens in subclinically infected dairy 294
cows in Flanders, Belgium. J Dairy Res 74:478-83. 295
2. USDA-APHIS. 2008. Prevalence of Contagious Mastitis Pathogens on U.S. Dairy 296
Operations, 2007, p. 1-2, APHIS Info Sheet, Fort Collins, CO. 297
3. Abera M, Habte T, Aragaw K, Asmare K, Sheferaw D. 2012. Major causes of mastitis 298
and associated risk factors in smallholder dairy farms in and around Hawassa, Southern 299
Ethiopia. Trop Anim Health Prod 44:1175-9. 300
4. Joo YS, Fox LK, Davis WC, Bohach GA, Park YH. 2001. Staphylococcus aureus 301
associated with mammary glands of cows: genotyping to distinguish different strains. Vet 302
Microbiol. 80: 131-8. 303
5. Middleton JR, Fox LK, Gay JM, Tyler JW, Besser TE. 2002. Influence of 304
Staphylococcus aureus strain-type on mammary quarter milk somatic cell count and N-305
acetyl-beta-D-glucosaminidase activity in cattle from eight dairies. J Dairy Sci 85:1133-306
40. 307
6. Shopsin B, Gomez M, Montgomery SO, Smith DH, Waddington M, Dodge DE, Bost 308
DA, Riehman M, Naidich S, Kreiswirth BN. 1999. Evaluation of protein A gene 309
polymorphic region DNA sequencing for typing of Staphylococcus aureus strains. J Clin 310
Microbiol 37:3556-63. 311
7. Smith EM, Green LE, Medley GF, Bird HE, Fox LK, Schukken YH, Kruze JV, 312
Bradley AJ, Zadoks RN, Dowson CG. 2005. Multilocus sequence typing of 313
intercontinental bovine Staphylococcus aureus isolates. J Clin Microbiol 43:4737-43. 314
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
16
8. Bannerman TL, Hancock GA, Tenover FC, Miller JM. 1995. Pulsed-field gel 315
electrophoresis as a replacement for bacteriophage typing of Staphylococcus aureus. J 316
Clin Microbiol 33:551-5. 317
9. Fournier C, Kuhnert P, Frey J, Miserez R, Kirchhofer M, Kaufmann T, Steiner A, 318
Graber HU. 2008. Bovine Staphylococcus aureus: association of virulence genes, 319
genotypes and clinical outcome. Res Vet Sci 85:439-48. 320
10. Boss R, Naskova J, Steiner A, Graber HU. 2011. Mastitis diagnostics: quantitative 321
PCR for Staphylococcus aureus genotype B in bulk tank milk. J Dairy Sci 94:128-37. 322
11. Syring C, Boss R, Reist M, Bodmer M, Hummerjohann J, Gehrig P, Graber HU. 323
2012. Bovine mastitis: the diagnostic properties of a PCR-based assay to monitor the 324
Staphylococcus aureus genotype B status of a herd, using bulk tank milk. J Dairy Sci 325
95:3674-82. 326
12. Graber HU, Naskova J, Studer E, Kaufmann T, Kirchhofer M, Brechbühl M, 327
Schaeren W, Steiner A, Fournier C. 2009. Mastitis-related subtypes of bovine 328
Staphylococcus aureus are characterized by different clinical properties. J Dairy Sci. 92: 329
1442-51 330
13. Jensen MA., Webster JA, Straus N. 1993. Rapid identification of bacteria on the basis 331
of polymerase chain reaction-amplified ribosomal DNA spacer polymorphisms. Appl 332
Environ Microbiol 59:945-52 333
14. Brosius J, Dull TJ, Sleeter DD, Noller HF. 1981. Gene organization and primary 334
structure of a ribosomal RNA operon from Escherichia coli. J Mol Biol. 148: 107-27. 335
15. Loughney K, Lund E, Dahlberg JE. 1982. tRNA genes are found between 16S and 23S 336
rRNA genes in Bacillus subtilis. Nucleic Acids Res. 10:1607-24. 337
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
17
16. Kumari DN, Keer V, Hawkey PM, Parnell P, Joseph N, Richardson JF, Cookson B. 338
1997. Comparison and application of ribosome spacer DNA amplicon polymorphisms 339
and pulsed-field gel electrophoresis for differentiation of methicillin-resistant 340
Staphylococcus aureus strains. J Clin Microbiol 35:881-5. 341
17. Cremonesi P, Pozzi F, Raschetti M, Bignoli G, Capra E, Graber HU, Vezzoli F, 342
Piccinini R, Bertasi B, Biffani S, Castiglioni B, Luini M. 2015. Genomic 343
characteristics of Staphylococcus aureus strains associated with high within-herd 344
prevalence of intramammary infections in dairy cows. J Dairy Sci 98:6828-38. 345
18. Brakstad OG, Aasbakk K, Maeland JA. 1992. Detection of Staphylococcus aureus by 346
polymerase chain reaction amplification of the nuc gene. J Clin Microbiol 30:1654-60. 347
19. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, 348
Swaminathan B. 1995. Interpreting chromosomal DNA restriction patterns produced by 349
pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 350
33:2233-9. 351
20. Hunter PR, Gaston MA. 1988. Numerical index of the discriminatory ability of typing 352
systems: an application of Simpson's index of diversity. J Clin Microbiol 26:2465-6. 353
21. Kitchen BJ, Middleton G, Salmon M. 1978. Bovine milk N-acetyl-beta-D-354
glucosaminidase and its significance in the detection of abnormal udder secretions. J 355
Dairy Res 45:15-20. 356
22. Fox LK, Hancock DD, McDonald JS, Gaskins CT. 1988. N-acetyl-ß-D-357
glucosaminidase activity in whole milk and milk fractions. J. Dairy Sci. 71:2915-2922. 358
359
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
18
23. Oliveira L, Rodrigues AC, Hulland C, Ruegg PL. 2011. Enterotoxin production, 360
enterotoxin gene distribution, and genetic diversity of Staphylococcus aureus recovered 361
from milk of cows with subclinical mastitis. Am J Vet Res 72:1361-8. 362
24. Haran KP, Godden SM, Boxrud D, Jawahir S, Bender JB, Sreevatsan S. 2012. 363
Prevalence and characterization of Staphylococcus aureus, including methicillin-resistant 364
Staphylococcus aureus, isolated from bulk tank milk from Minnesota dairy farms. J Clin 365
Microbiol 50:688-95. 366
25. Srinivasan V, Sawant AA, Gillespie BE, Headrick SJ, Ceasaris L, Oliver SP. 2012. 367
Prevalence of enterotoxin and toxic shock syndrome toxin genes in Staphylococcus 368
aureus isolates from milk of cows with mastitis. . Foodborne Pathog Dis 3:274-283. 369
26. Zschock M, Kloppert B, Wolter W, Hamann HP, and Lammler C. 2005. Pattern of 370
enterotoxin genes seg, seh, sei and sej positive Staphylococcus aureus isolated from 371
bovine mastitis. Vet Microbiol 108:243-9. 372
27. Silveira-Filho VM, Luz IS, Campos AP, Silva WM, Barros MP, Medeiros ES, 373
Freitas MF, Mota RA, Sena MJ, Leal-Balbino TC. 2014. Antibiotic resistance and 374
molecular analysis of Staphylococcus aureus isolated from cow's milk and dairy products 375
in northeast Brazil. J Food Prot 77:583-91. 376
28. Zecconi A, Cesaris L, Liandris E, Dapra V, Piccinini R. 2006. Role of several 377
Staphylococcus aureus virulence factors on the inflammatory response in bovine 378
mammary gland. Microb Pathog 40:177-83. 379
29. Ikawaty R, Brouwer EC, Jansen MD, van Duijkeren E, Mevius D, Verhoef J, Fluit 380
AC. 2009. Characterization of Dutch Staphylococcus aureus from bovine mastitis using a 381
Multiple Locus Variable Number Tandem Repeat Analysis. Vet Microbiol 136:277-84. 382
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
19
30. Du XF, Xiao M, Liang HY, Sun Z, Jiang YH, Chen GY, Meng XY, Zou GL, Zhang 383
L, Liu YL, Zhang H, Sun HL, Jiang XF, Xu YC. 2014. An improved MLVF method 384
and its comparison with traditional MLVF, spa typing, MLST/SCCmec and PFGE for the 385
typing of methicillin-resistant Staphylococcus aureus. Int J Mol Sci 15:725-42. 386
31. Mathews KR, Kumar SJ, O'Conner SA, Harmon RJ, Pankey JW, Fox LK, Oliver 387
SP. 1994. Genomic fingerprints of Staphylococcus aureus of bovine origin by 388
polymerase chain reaction-based DNA fingerprinting. Epidemiol. Infect. 107: 177-185 389
32. Middleton JR, Fox LK, Gay JM, Tyler JW, Besser TE. 2002. Use of pulsed-field gel 390
electrophoresis for detecting differences in strain populations between dairy herds with 391
different cattle importation practives. Epidemiol. Infect. 129: 387-395. 392
33. Boerema JA, Clemens R, Brightwell G. 2006. Evaluation of molecular methods to 393
determine enterotoxigenic status and molecular genotype of bovine, ovine, human and 394
food isolates of Staphylococcus aureus. Int J Food Microbiol 107:192-201. 395
34. Boss R, Cosandey A, Luini M, Artursson K, Bardiau M, Breitenwieser F, 396
Hehenberger E, Lam T, Mansfeld M, Michel A, Mösslacher G, Naskova J, Nelson S, 397
Podpečan O, Raemy A, Ryan E, Salat O, Zangerl P, Steiner A, Graber HU. 2016. 398
Bovine Staphylococcus aureus: subtyping, evolution, and zoonotic transfer. J Dairy Sci. 399
99:515-28. 400
35. Cosandey A, Boss R, Luini M, Artursson K, Bardiau M, Breitenwieser F, 401
Hehenberger E, Lam T, Mansfeld M, Michel A, Mösslacher G, Naskova J, Nelson S, 402
Podpečan O, Raemy A, Ryan E, Salat O, Zangerl P, Steiner A, Graber HU. 2016. 403
Staphylococcus aureus genotype B and other genotypes isolated from cow milk in 404
European countries. J Dairy Sci. 99:529-40. 405
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
20
Table 1. Distribution of S. aureus enterotoxin genes, median (range) for NAGase activity, 406
median (range) linear score SCC (LSCC) identified within each RSP-type, and number (percent) 407
of isolates possessing the different enterotoxin genes within each RSP-type. 408
RSP-type (n=65)
Median NAGase (range)
Median LSCC (range)
sea sed sei sej Seg lukE
1 (n = 50)a 17.8 (2.3-363.6)a 8.1 (2.5-12.5) 0 0 0 0 0 50 (100%)
2 (n = 5)b 3.04 (2.6-10.4)b 4.8 (2.5-7.8) 5 (100%)* 0 0 0 0 5 (100%)
3 (n = 4)b 8.3 (4.3-18.3) 5.4 (4.8-5.8) 0 0 4 (100%) 0 0 4 (100%)
4 (n = 3)b 14.2 (8.0-46.5) 7.8 (5.0-11.9) 0 0 3 (100%) 0 3 (100%) 3 (100%)
5 (n = 3)b 23.3 (5.6-23.7) 6.8 (4.8-8.1) 0 0 0 0 0 3 (100%)
Total (n = 65) 5 (7.7%) 0 7 (10.8%) 0 3 (4.6%) 65 (100%)
409
a-b Values with different superscripts within column differ from each other (P < 0.05). 410
*Percentages of isolates possessing the different SET genes are the number of isolates with a 411
given SET gene detected divided by the row total, e.g., 5 of 5 isolates identified by RSP-type-2 412
as possessed the sea gene (100%). 413
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
21
414
Table 2. Distribution of S. aureus PFGE strain-types and RS-PCR types identified within each 415
farm. 416
Farm Number of
isolates 1PFGE stain ID
2Number of isolates identified on farm 3N= 4RSP-type
1 8 1 15 3 1 2 4 2 1
3 5 3 1 2 3 4 14 3 4 3 4 5 1 1 5
6 5 2 5 7 1 1 3 4 14 7 1 1 3
8 19 3 1 9 4 2 1 10 17 3 1 11 1 1 1 12 1 1 3 16 1 1 1 17 1 1 1
19 1 1 1 5 10 21 4 3 1
22 4 3 1 23 1 1 2 24 1 2 1
26 1 1 1 6 9 23 4 3 2
27 10 2 1 28 1 1 1 29 2 1 1 30 1 1 1
31 1 1 1 7 13 8 9 3 1
23 2 1 2 31 1 1 1 32 63 3 1 34 1 1 1 35 1 1 1 37 1 1 1
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
22
38 1 1 1 39 1 1 1 8 4 40 6 1 1
41 1 1 1 42 1 1 1
43 1 1 3
TOTAL 65 isolates 35 PFGE
strains 210 isolates 65
isolates 5 RSP-types
417
1PFGE strain-types displayed in bold font were found on more than one farm. 418 2Reference: Middleton JR, Fox LK, Gay JM, Tyler JW, Besser TE. 2002. Influence of 419 Staphylococcus aureus strain-type on mammary quarter milk somatic cell count and N-acetyl-420 beta-D-glucosaminidase activity in cattle from eight dairies. J Dairy Sci 85:1133-40. 421 3 Number of isolates characterized within PFGE strain-type; note that only one RSP-type was 422 found within each PFGE strain-type. 423 4RSP-type-1 was found on six farms, RSP-type-2 and RSP-type-3 were found on three farms, 424 RSP-type-4 and RSP-type-5 were each found on one farm. 425
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
23
Figure 1. Gel electrophoresis of all RSP-types. Lanes labelled O are 100 base pair (bp) ladders 426
with the 500 bp bands labelled. Lane 1 is RSP-type-4. Lane 2 is RSP-type-3. Lane 3 is RSP-type-427
1. Lane 4 is RSP-type-5. Lane 5 is RSP-type-2. Lane 6 is the GTB positive control. Lane 7 is the 428
negative control. 429
430
431
432
433
on April 4, 2018 by guest
http://jcm.asm
.org/D
ownloaded from