| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Animal Health Centre, Morrinsville, New Zealand
2 Institute of Veterinary, Animal and Biomedical Sciences, Massey University, Palmerston North, New Zealand
3 School of Veterinary Clinical Science, Division of Health Sciences, Murdoch University, Perth, Western Australia, 6150, Australia
Corresponding author: S. McDougall; e-mail: smcdoug{at}ahc.co.nz.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The prevalence of S. uberis was 12% for cows around parturition, and the median duration of infection was 16 d. Cows >2 yr old had a longer duration of infection than 2 yr old cows, and duration varied among herds. A total of 173 different PFGE types were identified from a total of 234 S. uberis isolates. Each farm had a unique set of PFGE types. Only 3 PFGE types were common to each of 3 pairs of cows, and these occurred on the same farm. Where S. uberis was isolated on more than one occasion from a gland, only 55% of the PFGE types were the same across time. For cows with multiple glands infected, only one-half (9 of 18) had the same PFGE type in more than one gland. No predominant PFGE type was identified in any herd.
It is concluded that there was wide heterogeneity of PFGE types, that the environment rather than other cows was the likely source of S. uberis infections, and that glands may be infected with multiple S. uberis PFGE types over a lactation.
Key Words: Streptococcus uberis • pulsed-field gel electrophoresis • mastitis • bovine
Abbreviation key: PFGE = pulsed-field gel electrophoresis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Streptococcus uberis can be isolated from a number of sites on the cow including the vagina, tonsils, and escutcheon (Cullen, 1966) as well as from bedding and pasture (Bramley and Dodd, 1984; Todhunter et al., 1995). The main route of transmission appears to be from environmental sources. However, recent epidemiological and molecular typing studies have demonstrated that cow-to-cow transmission is likely to be occurring in some herds (Phuektes et al., 2001; Zadoks et al., 2001).
Variations in S. uberis strain type have been reported using restriction fragment length polymorphism (Hill and Leigh, 1989; Jayarao et al., 1992), randomly amplified polymorphic DNA (Gillespie et al., 1997; Gillespie et al., 1998; Zadoks, 2002), repetitive extragenic palindromic-polymerase chain reaction (Wieliczko et al., 2002), and pulsed-field gel electrophoresis (PFGE) procedures (Baseggio et al., 1997; Douglas et al., 2000; Phuektes et al., 2001). A large number of S. uberis types have been observed with all methods (Hill and Leigh, 1989; Jayarao et al., 1992; Douglas et al., 2000; Phuektes et al., 2001; Wieliczko et al., 2002; Zadoks et al., 2003). All S. uberis isolates from clinical cases of mastitis were in 2 of the 8 clusters defined from a group of subclinical and clinical S. uberis isolates when strains were analyzed with RFLP (Jayarao et al., 1992). Additionally, all S. uberis isolates from clinical cases had DNA fragments of >21 kb; only 21% of subclinical isolates had such large DNA fragments (Jayarao et al., 1992). Although multiple genotypes are invariably present, a predominant type has been found in some herds (Phuektes et al., 2001; Zadoks, 2002). The degree of homogeneity of S. uberis types isolated from a herd has been used to infer the predominant route of transmission. Where there is a predominant type present, it has been suggested that cow-to-cow transmission is likely; if a large number of types are present and no type predominates, then an environmental source of the pathogen is likely (Zadoks, 2002).
Variations in the pathogenicity of phenotypically different isolates of S. uberis have been reported following intramammary challenge (Hill, 1988). Additionally, one PFGE type was shown to be more likely associated with clinical mastitis than other types present in the same herd (Phuektes et al., 2001). Hence, genotypic differences, identifiable by molecular typing techniques, may be associated with differences in pathogenicity.
This study aimed to describe the duration of S. uberis infection under New Zealand dairy production systems and to re-assess duration with knowledge of the PFGE type.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
), and the cows were fed predominantly on rye grass (Lolium perene) and white clover (Trifolium repens) pasture.
|
Any gland from which S. uberis was isolated (excluding contaminated samples and those with <3 colony forming units/10 µL) was enrolled in the trial. Single milk samples were then collected from each enrolled gland at approximately 28-d intervals. Sampling ceased from a cow when S. uberis was not isolated from 2 sequential samples from any enrolled gland or when lactation ended. Cows detected with clinical mastitis by the herd owner using the criteria of swelling and/or heat of a gland or visible change in the milk (e.g., clots, strings, etc.) had a single milk sample collected from all glands at the time of diagnosis and 21 d later. These samples were labeled and frozen at –20°C until collection at the next routine farm visit. The glands with clinical signs were treated 3 times at 12-h intervals by intramammary infusions of 1 g of penicillin (Penalone, Schering-Plough, Upper Hutt, New Zealand), and the milk was discarded for 144 h following the last treatment.
Case Definitions
A gland was defined as being infected with S. uberis where 
3 more colony forming units/10 µL were isolated from both samples (at initial enrolment) or a single sample (at resampling and from clinical cases) and where only one other colony type was present (which occurred on 15 occasions where coagulase-negative staphylococci, Staphylococcus aureus, or a yeast was co-isolated with S. uberis).
Bacteriological cure was defined as having occurred when 2 sequential samples were found not to contain S. uberis.
The duration of infection (days) was defined in 4 ways:
Where a gland was still infected at the end of lactation, the duration was defined as the interval from the initial sampling to the last S. uberis-positive sample that was collected at the routine sampling visit. Clinical failure was defined as having occurred where the herd owner chose to retreat a gland previously treated for clinical mastitis within 21 d of initial treatment.
Microbiology
Milk samples (10 µL) were streaked onto one quarter of a 5% blood agar plate containing 0.1% esculin (Fort Richard, Auckland, New Zealand) and incubated at 37°C for 48 h. The bacteria were provisionally speciated based on colony morphology, hemolysis pattern, esculin reaction, and Gram stain. Gram-positive, catalase-negative cocci were defined as Streptococcus spp. A single colony of the presumptive Streptococcus spp isolate was streaked over the surface of a 0.1% esculin, 5% blood agar plate and grown for a further 24 h. A loopful of this growth was placed in each of 7 sugar reagents, and an isolate was defined as S. uberis, where the isolate did not ferment raffinose, but did ferment inulin, lactose, mannitol, salicin, sorbitol, and trehalose (Quinn et al., 1999). The remaining half plate of growth was collected onto a commercial transport swab (BBL Culture swab, Becton Dickinson, Sparks, MD) and held at 4°C for up to 30 d until transport to the Institute of Veterinary and Biomedical Sciences, Massey University for PFGE.
DNA Extraction
Genomic DNA for PFGE was prepared from the S. uberis strains according to the methods of Smith and Cantor (1997) and McEllistrem et al. (2000), with some modifications. Briefly, S. uberis isolates were grown on 5% sheep blood agar (Difco, Becton Dickinson) overnight at 37°C. The cells were harvested into 3 mL brain heart infusion broth (Difco) and the optical density of the broth was measured and adjusted to 1.4 at 610 nm. A 75-µL aliquot of cells was pelleted by centrifugation at 13,000 rpm for 5 min. Cells were washed once with 150 µL of Pett IV buffer (1 M NaCl, 10 mM Tris-HCl [pH 8.0], 10 mM EDTA; pH 8.0) then centrifuged and resuspended in 50 µL Pett IV buffer. The bacterial suspension was mixed with 100 µL of 1% low melt preparative grade agarose (Bio-Rad Laboratories (New Zealand) Pty Ltd, Auckland, New Zealand) and was dispensed into plug molds. After being solidified for 1 h on ice, the plugs were incubated overnight at 37°C in a lysis buffer solution (1M NaCl, 100 mM EDTA [pH 8.0] 6 mM Tris-HCl [pH 8.0], 0.5% Brij 58, 0.5% sodium deoxycholate, 0.5% sodium lauroyl sarcosine) containing 20 mg of lysozyme/mL (Roche Diagnostics, Auckland, New Zealand). The plugs were transferred into 1 mL ESP buffer (0.5 M EDTA, 1% sodium lauroyl sarcosine) containing 1 mg proteinase K/mL and were incubated overnight at 56°C. Following lysis, the plugs were washed 5 times for 1 h each time with 10 mL of TE buffer (10 mM Tris-HCl, 1 mM EDTA) on ice.
Restriction Endonuclease Digestion
A 3-mm slice of each plug was equilibrated in 100 µL of 1.2 x restriction buffer (New England Biolabs, Beverly, MA) for 45 min on ice. The restriction buffer was removed and replaced with 100 µL of fresh one x restriction buffer containing 30 units of SmaI (New England Biolabs). The plug slices were held on ice for an additional 45 min before an overnight incubation at 25°C.
PFGE
The restriction fragments were separated by PFGE on a CHEF-DRII system in a 1% agarose gel (Pulsed Field Certified Agarose, Bio-Rad Laboratories (New Zealand) Pty Ltd.) in 0.5 x TBE buffer (45 mM Tris, 45 mM Boric acid, 1 mM EDTA; pH 8.0) at 14°C. The gel was run at 6 V/cm for 23 h with an initial pulse time of 1 s and a final pulse time of 40 s. Lambda Ladder PFG Marker and Low Range PFG Marker (New England Biolabs) were included as molecular size standards. Twenty unknown samples were run on each gel in date order without regard to farm of origin or cow. Thus, repeated S. uberis isolations from the same cow might have been run and, hence, analyzed on different gels.
Analysis of PFGE Patterns
Gels were stained with ethidium bromide, and images were captured under ultraviolet illumination by the Gel Doc 2000 (Bio-Rad Laboratories (New Zealand) Pty Ltd.). The PFGE patterns were analyzed by Diversity Database software (Bio-Rad Laboratories (New Zealand) Pty Ltd). Similarities between the PFGE patterns were calculated by the Dice correlation coefficient with a maximum position tolerance of 1%. Dendrograms were constructed by the unweighted pair group method with arithmetic mean (UPGMA). Where the level of similarity between isolates was >60% on the dendrograms, the PFGE patterns were compared visually. Multiple isolates from the same cow were also visually assessed. Where the level of similarity was unclear, isolates were rerun on the same gel. Minor differences i.e., <3 band differences were recorded as subtypes, but treated as one PFGE type for analysis.
Statistical Analysis
The duration of infection of glands diagnosed with subclinical S. uberis infections were not normally distributed (Table 2) with most intervals being <20 d, but with a few longer intervals. Thus, data are presented as the median and percentiles (Table 2). The duration was further examined using Kaplan-Meier or Cox’s proportion hazards models ("survival analysis"). Analysis was performed for each of the 4 definitions described previously. Initially, factors thought to affect the duration of infection (including age [coded as 2 yr old or >2 yr old], herd of origin, days postpartum at enrollment, whether the gland was a front or rear gland, number of glands infected within a cow, and the SCC [and log10 SCC] of the gland at the time of enrollment) were tested in univariate models. Factors associated (i.e., P < 0.2) were then offered to a reverse stepwise model using likelihood ratios as the factor exclusion method. Glands were defined as right-censored when infection persisted to the end of lactation. The validity of the models was assessed by examination of the hazards ratio and log minus log plots.
|
2). If these factors were associated with bacteriological cure proportion (P < 0.2), they were then offered to a reverse stepwise logistic regression model. The effect of age code (i.e., 2, 3, 4 to 8, and >8 yr old), herd of origin, and days postpartum at diagnosis (coded into quartiles as above) on the proportion of clinical cases from which S. uberis was isolated was analyzed by 2. The trial was performed following approval by the Animal Ethics Committee of AgResearch, Ruakura and the Massey University Animal Ethics Committee.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
). Streptococcus uberis was isolated from 12% of all cows and 4% of all glands (Table 1). A total of 59 cows and 71 glands were enrolled with subclinical S. uberis infection after this initial sampling. Forty-two cows had one S. uberis-infected gland, 13 had 2 S. uberis-infected glands, and one cow had 3 S. uberis-infected glands at enrolment.
Duration of Subclinical S. uberis Infection
The majority of S. uberis-infected glands underwent spontaneous bacteriological cure (55 of 71 [78%]), with a small number lost to follow up (2 of 71 [3%]), diagnosed with clinical mastitis (9 of 71 [13%]) or were still infected at the end of lactation (5 of 71 [7%]).
The median duration of infection was 16 d for all 4 definitions of duration (50th percentile; Table 2; Figure 1), but with numerically shorter 25th and 75th percentiles for the estimates that took account of changes in PFGE type (Table 2). Inclusion of the period of infection following diagnosis of clinical mastitis did not increase the estimated median duration of infection (Table 2). The duration of infection (Definition 1) was longer for cows than heifers (12 [95% CI = 7 to 17] vs. 11 [95% CI = 5 to 15] median d for cows and heifers, respectively; P = 0.001; Figure 2, hazards ratio = 6.0 [95% CI = 2.1 to 17.4]) and varied among herds (P = 0.01; Figure 3). Duration of infection was unaffected by whether a cow had only one or more than one gland infected with S. uberis (P = 0.79), the day postpartum at enrolment postpartum (P = 0.32), whether the gland was a front or rear gland (P = 0.79) or by the SCC or log10 SCC of the gland at enrolment (P = 0.81 and P = 0.47, respectively). The same variables were significant with similar hazards ratios when the other 3 methods of calculating duration were tested as the dependent variables (data not presented).
|
|
|
Streptococcus uberis was the most prevalent pathogen isolated from cases of clinical mastitis (63.2%; Table 3) with ‘others’ and coagulase-negative staphylococci the next most common isolates. The proportion of glands from which S. uberis was isolated did not differ among herds (P = 0.57), age codes (18 of 42 (43%), 2 of 8 (25%), 37 of 88 (42%), and 8 of 25 (32%) for 2, 3, 4 to 8, and >8 yr old cows, respectively; P = 0.64), or vary with time postpartum (20 of 43 [47%], 18 of 40 [45%], 13 of 30 [43%], and 13 of 48 [27%] for 0 to 1, 2 to 3, 4 to 25, and >25 d postpartum at diagnosis, respectively; P = 0.18). The proportion of glands that underwent bacteriological cure was 89.6%, and this proportion declined with increasing days postpartum at diagnosis (Table 4; P = 0.03).
|
|
). An example of a PFGE gel is presented in Figure 4, and the dendrogram constructed from isolates from one herd (Herd A) is presented in Figure 5.
|
|
|
).
For glands from which S. uberis was isolated on more than one occasion, 55% (16 of 29) had the same PFGE type (Table 5). However, there was a range among herds, with between 16% (Herd E) and 100% (Herds A and C) of glands having the same PFGE type on more than one occasion in the same gland (Table 5).
Only 14% of cows that had multiple glands infected with S. uberis simultaneously had the same PFGE type in 2 glands (Table 5).
Half of the cows (9 of 18) from which more than one S. uberis isolate was recovered over time had the same PFGE type within cow over time (Table 5).
Isolates from 2 glands had isolates showing minor variations in the PFGE banding pattern over time. For example, the PFGE pattern of isolates from one gland over a 4-mo period showed 3 closely related PFGE types (designated L2, L2a, and L2b; Figure 6).
|
). However, these changes were relatively minor as the PFGE types had 75 and 92% similarity with the original isolates (Table 6). In one case, the same PFGE type as that recovered pre-treatment was re-isolated following 2 samplings (at 28-d intervals) where no bacteria were isolated.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Multiple S. uberis PFGE types were identified both within and between farms in the present study. Previous studies have also identified a large number of S. uberis strains on farms in New Zealand (Douglas et al., 2000; Wieliczko et al., 2002) and elsewhere (Hill and Leigh, 1989; Jayarao et al., 1992; Phuektes et al., 2001; Zadoks, 2002). In the present study, the same PFGE type was only found in 2 different cows on 3 occasions, and all of these cows were in the same herd. Approximately only one-half of the cows and one-half of the glands with chronic infections had the same PFGE type over time. Additionally, in only 14% of cows in which multiple glands were infected did the same PFGE type occur in the same animal. These data suggest that in most cases infection occurs from a source other than an infected cow. In contrast to the current study, a predominant S. uberis type has been identified within some herds in other studies. For example, 50% (5 of 10) and 78% (22 of 28) of cows from which S. uberis were isolated had the same strain type within 2 herds in a Dutch study, respectively (Zadoks et al., 2003). The same strain type was found in 88% of infections from which multiple samples were collected (Zadoks, 2002). The same strain type was identified in each gland where a cow had multiple glands infected simultaneously (Zadoks et al., 2003). A predominant strain was also identified in 2 of 4 herds in an Australian study (Phuektes et al., 2001). It was concluded from these 2 studies that significant cow-to-cow transmission was occurring. The greater degree of heterogeneity of S. uberis types in the present study and a failure to find a dominant type may reflect differences in the epidemiology of S. uberis infection among production systems. Under the extensive pasture systems in New Zealand, it appears that a point source of S. uberis infection is uncommon and that transmission of S. uberis may be predominantly from environmental sources.
Where isolates differ by only 2 to 3 PFGE bands, it has been suggested that this may represent a single point change in the genome and that the types are closely related (Tenover et al., 1995). Two isolates differed by only 1 or 2 bands when isolated sequentially from the same glands over time in the present study. In one of these cases, the change occurred following treatment of clinical mastitis with an antibiotic that did not result in bacterial cure of the gland. It is plausible that treatment resulted in a genotype change that allowed the isolate to survive antibiotic treatment. However, no change in antibiotic sensitivity was demonstrated following testing using the Kirby Bauer disc diffusion method for these 2 PFGE types.
In the small number of glands treated for clinical mastitis and in which subsequent sampling suggested that treatment failure had occurred, different PFGE types were isolated post-treatment compared with pre-treatment. However, it is not clear if bacteriological cure had indeed occurred and a new infection had subsequently become established, whether treatment itself resulted in a change in the PFGE type profile, or whether, in fact, multiple PFGE types were present pre-treatment but only one identified because of the laboratory techniques used (see subsequent).
In the present study, clinical mastitis was not associated with any particular strain type. In contrast, in a previous Australian study, one PFGE type was more commonly associated with clinical mastitis than other PFGE types (Phuektes et al., 2001). Those researchers suggested that the predominant stain had a virulence factor(s) not present in other strains endemic in the herd. In contrast, another study failed to find that different PFGE types had a different probability of being associated with clinical mastitis (Zadoks et al., 2001). Variations in virulence have been previously demonstrated among S. uberis strains (Hill, 1988). No one PFGE type or group of PFGE types were more likely to be associated with prolonged subclinical infection, diagnosis with clinical mastitis, or a failure to cure following treatment in the current study. Thus, cows in this population are likely to have been exposed to a range of S. uberis PFGE types, and the duration of infection and clinical outcome (i.e., diagnosed as clinical or not) may be related to cow factors or dose of S. uberis rather than to the specific PFGE type the gland was infected with.
A limitation of the present study, and many other studies, is that only one colony type from a number of colonies grown during initial culture was genotyped, because of the expense of PFGE. However, given the large number of S. uberis PFGE types identified, it is possible that multiple PFGE types might have been present concurrently within a single gland. Two PFGE types of S. uberis in the same gland at the same time have been previously reported as they were phenotypically different and readily distinguishable on the primary culture plate (Zadoks, 2002). Further studies are required to determine the prevalence of multiple S. uberis PFGE type infections within the same gland. This is important as where only one colony type is tested and where multiple PFGE types are concurrently present, it may be falsely assumed that new infections are occurring over time as the PFGE type appears to change with time, when in fact it is just chance that one or other of the infecting strains is actually selected for typing at sequential sampling times.
It was concluded that there are many different PFGE types of S. uberis present within the New Zealand environment, that cow-to-cow transmission of S. uberis is likely uncommon relative to transmission from environmental sources, that changes in PFGE type within chronically infected glands are relatively common, that PFGE type is not correlated with the diagnosis of clinical mastitis, and that the duration of infection may be overestimated where the PFGE type of the infection is not known.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication November 2, 2003. Accepted for publication February 12, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. S. Petersson-Wolfe, S. Adams, S. L. Wolf, and J. S. Hogan Genomic Typing of Enterococci Isolated from Bovine Mammary Glands and Environmental Sources J Dairy Sci, February 1, 2008; 91(2): 615 - 619. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tomita, B. Meehan, N. Wongkattiya, J. Malmo, G. Pullinger, J. Leigh, and M. Deighton Identification of Streptococcus uberis Multilocus Sequence Types Highly Associated with Mastitis Appl. Envir. Microbiol., January 1, 2008; 74(1): 114 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Munoz, F. L. Welcome, Y. H. Schukken, and R. N. Zadoks Molecular Epidemiology of Two Klebsiella pneumoniae Mastitis Outbreaks on a Dairy Farm in New York State J. Clin. Microbiol., December 1, 2007; 45(12): 3964 - 3971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. I. Parker, C. Compton, F. M. Anniss, A. Weir, C. Heuer, and S. McDougall Subclinical and Clinical Mastitis in Heifers Following the Use of a Teat Sealant Precalving J Dairy Sci, January 1, 2007; 90(1): 207 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Zadoks, Y. H. Schukken, and M. Wiedmann Multilocus Sequence Typing of Streptococcus uberis Provides Sensitive and Epidemiologically Relevant Subtype Information and Reveals Positive Selection in the Virulence Gene pauA J. Clin. Microbiol., May 1, 2005; 43(5): 2407 - 2417. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) infection and of PFGE-specific Streptococcus uberis pulsed-field gel electrophoresis type (•) of individual glands infected with Streptococcus uberis.
