| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Veterinary and Biomedical Sciences,
Department of Veterinary Population Medicine, and
Department of Animal Science, College of Veterinary Medicine, University of Minnesota, St. Paul 55108
1 Corresponding author: beyxx001{at}umn.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Key Words: bovine mastitis • Klebsiella pneumoniae • genetic diversity • repetitive DNA sequence polymerase chain reaction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Currently, the most effective control measures for coliform mastitis are prevention by implementing appropriate management strategies and immunization by using the core antigen bacterin of E. coli strain J5 or Salmonella Typhimurium Re17 (Hogan et al., 1995; Wilson and Gonzalez, 2003). Vaccination is beneficial in reducing the clinical severity of coliform mastitis; however, it is not efficacious in preventing new infections (Wilson and Gonzalez, 2003). Thus, mastitis caused by Klebsiella spp. continues to affect the dairy industry. A better understanding of how Klebsiella spp. is involved in mastitis and its epidemiology is needed. Molecular characterization of the K. pneumoniae responsible for causing bovine mastitis is lacking (Kikuchi et al., 1995). Understanding the molecular diversity would assist in the development of appropriate strategies for the prevention and control of udder infections in dairy cows. The economic losses, ineffectiveness of conventional control methods, and unsuccessful treatment rates of Klebsiella mastitis make this study timely and relevant. In our recent studies with a variety of K. pneumoniae isolates from clinical cases of mastitis in a single dairy herd, we identified a high degree of genetic diversity in K. pneumoniae (Paulin-Curlee et al., 2007). To determine whether a similar level of genetic diversity is present across dairy herds, we analyzed mastitis-associated K. pneumoniae isolates by repetitive DNA sequence PCR (rep-PCR) from 6 dairy herds from 4 different states.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
rep-PCR
Bacterial genomic DNA was extracted by PrepMan Ultra Reagent (Applied Biosystems, Foster City, CA). Briefly, frozen cultures were streaked onto MacConkey agar plates and incubated. Two individual colonies were harvested and suspended in 300 µL of PBS and pelleted by centrifugation for 3 min at 7,500 x g. The supernatant was discarded, 200 µL of PrepMan was added, and the tubes were placed in a heat block for 10 min at 100°C. After incubation, the solution was pelleted for 3 min at 7,500 x g, and the supernatant containing DNA was diluted 1:1 with sterile DNase-free water. Repetitive DNA sequence PCR fingerprints were obtained by using a boxA1R primer (5' CTACGGCAAGG-CGACGCTGACG 3'; Goldberg et al., 2006). The 25-µL PCR mixture contained 25 mM MgCl2, 10x Buffer II (Applied Biosystems), 25 pmol of boxA1R primer, 100 mM deoxynucleotide 5'-triphosphate mix and Ampli-Taq Gold DNA polymerase (Applied Biosystems), and 2 µL of DNA (150 ng) template. Polymerase chain reaction conditions were 95°C for 7 min, followed by 30 cycles consisting of 94°C for 1 min, 66°C for 8 min, 71°C for 1 min, and a final extension of 71°C for 15 min, followed by storage at 4°C (Goldberg et al., 2006). The DNA bands were separated by 2% agarose gel electrophoresis at 4°C for 5 to 6 h at 62 V (6 V/cm). Gels were stained in ethidium bromide-1x Tris-acetate-EDTA buffer, and the images were obtained with Labworks 4.0 Image Acquisition and Analysis Software (UVP Inc., Upland, CA) and saved as tagged image files. Reproducibility of PCR was verified as described previously (Paulin-Curlee et al., 2007) by repeating rep-PCR 4 times with 16 separate isolates and freshly extracted DNA each time. Analysis for reproducibility showed an average of 93.6% fingerprint similarity for the same isolate based on 4 iterations with 16 distinct isolates and freshly extracted DNA each time, and was used to establish the 90% similarity cutoff for genotype definition used in this study.
rep-PCR Fingerprint Analysis
Gel images were uploaded and analyzed by using BioNumerics software, version 4.0 (Applied Maths BVBA, Sint-Martens-Latem, Belgium). The positions of the bands in each gel were normalized by using a 1-kb DNA ladder (Hi-Lo DNA Marker, Minnesota Molecular Inc., Minneapolis, MN). Band matching and similarity of the fingerprints was accomplished by using the band-based Dice coefficient of similarity at 1.5% tolerance and 1.0% optimization.
Statistical Analysis
The MacAnova statistical analysis program (School of Statistics, University of Minnesota, Minneapolis, MN) was used to calculate the correlation between finger-print similarity matrices and matrices of temporal distances (sample interval). The Spearman rank coefficient was used to calculate the temporal correlation to determine whether isolates collected at shorter sampling intervals were more likely to have similar finger-prints. MacAnova was also used to analyze the association between genotypes and herds, genotypes and the type of bedding used by each dairy herd, and genotypes and the dairy location. A P-value of 
0.05 was considered significant for the temporal correlation and association analysis. Simpson’s index of diversity was used for the measurement of genetic diversity and discriminatory power of the typing techniques (Hunter and Gaston, 1988). Simpson’s index of diversity was calculated by using BioNumerics software, version 4.0 (Applied Maths BVBA), and it was based on the total number of isolates, the total number of genotypes described, and the number of isolates belonging to each genotype. Dendrograms were generated by using the unweighted pair group method with arithmetic means. Because identical DNA fingerprints obtained from the same clinical mastitis case could bias our analysis (Johnson et al., 2004), all of the duplicate DNA fingerprints originating from the same mastitis case were eliminated in the final analysis. Thus, isolates originating from the same sample having exactly the same DNA fingerprint were represented singly in the analysis.
In addition to the descriptive analysis, analysis of molecular variance (AMOVA) was used to partition the variation among isolates within herds, among herds within states, and among states. A matrix was constructed based on the presence or absence of rep-PCR bands for each isolate to compute the genetic distance for each pair of isolates. The analysis was performed by using Genalex 6 (Peakall and Smouse, 2006).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Dairy A (WI, Recycled Manure Solids).
A total of 428 K. pneumoniae isolates originating from 141 clinical mastitis cases were collected over 18 mo (June 2003 to December 2004) and the genetic diversity was determined by rep-PCR fingerprinting. The 406 isolates originated from up to 3 separate colonies per mastitis sample. After eliminating replicate isolates with identical DNA fingerprints, a total of 167 isolates were used for the cluster analysis. Simpson’s index of diversity was 91.0% for the 30 distinct banding patterns identified, indicating wide genetic diversity and a high degree of discrimination by rep-PCR. The dendogram was arbitrarily divided into 2 major groups, I and II. All the genotypes comprising 
10 isolates belonged to group I. Group II consisted of all the genotypes comprising 1 to 9 isolates. Genotypes detected a single time were termed unique. Six genotype patterns contained between 11 and 37 isolates each, which accounted for 64.7% of the isolates (group I; Table 1
) and were detected throughout the study. The remaining 24 genotypes (35.3%) contained 1 to 7 isolates in each cluster (group II), and 7 unique genotypes were present (R3, R11, R14, R15, R21, R26, and R27). More than one genotype per mastitis case was present 12.1% of the time.
|
). Nineteen distinct genotypes were detected in 24 cases of reinfection among 16 cows. In 25% (6/24) of the cows that became reinfected, more than one genotype per mastitis case was detected. Genotypes R5, R6, R7 (group I), and R8 (group II) were the most frequently (53.2%, 25/47) isolated genotypes in reinfected cows. Reinfections occurred throughout the study and were the most frequent during winter (45.8%, 11/24) and summer (29.2%, 7/24). Some cows (11%, 10/91) were culled because of persistent K. pneumoniae mastitis. Genotype R5 was isolated the most frequently (30%, 3/10) among the 8 distinct genotypes from the 10 culled cows. Some of the genotypes isolated from reinfected cows were also detected in the culled cows (R2, R5, R7, R16, R21, R23, and R24). Genotype R5 was the most frequently isolated genotype in reinfected and in culled cows (22.8%, 13/57), whereas the other genotypes were isolated from reinfected and culled cows 1.8% (1/57) to 10.5% (6/57) of the time.
Analysis of isolates collected on the same date as the bedding and the bulk tank milk samples identified 6 genotypes among 12 isolates (Figure 1). A common genotype was detected among isolates from mastitis, bedding, and bulk tank milk. The percentage of K. pneumoniae mastitis was similar during winter (36.2%, 51/141) and summer (32.6%, 46/141). The spring and fall seasons had a lower percentage (15.6%, 22/141) of K. pneumoniae mastitis cases. There was a small negative association between the fingerprint similarities matrix and the temporal distances matrix (R = –0.083, P = 0.005), indicating that isolates that were collected closer in time had higher similarity than those collected over a greater interval of time (Table 2).
|
|
).
Dairy C (WI, Sawdust and Wood Shavings).
Nine mastitis cases occurred over 6 mo (June through December 2004). Simpson’s index of diversity was 91.7% for the 6 distinct genotypes identified among the 27 isolates evaluated. The 3 separate isolates originating from the same clinical mastitis case showed the same genotype. As identified for other dairies, summer (44.4%, 4/9) and winter (33.3%, 3/9) had the highest number of cases of K. pneumoniae mastitis, with the remaining 22.2% (2/9) of mastitis cases occurring during fall. A negative association between the fingerprint similarities matrix and the temporal distances matrix was observed (R = –0.162, P = 0.147; Table 2).
Dairy D (MN, Sawdust).
Eighteen mastitis cases occurring during November 2004 yielded 54 isolates. Twenty-one isolates were included in the analysis and 18 distinct genotypes were identified. Among the isolates that originated from the same clinical case, a unique genotypic pattern was observed 83.3% of the time. Simpson’s index of diversity was 88.6%, indicating the higher degree of genetic diversity of K. pneumoniae occurring within this herd. Despite the fact that the majority of mastitis samples (60.0%) were collected at a single time point, a considerably high degree of genetic diversity was observed. There was a negative association between matrices of fingerprint similarities and temporal distances (R = –0.30, P = 0.003; Table 2).
Dairy E (PA, Sawdust and Wood Shavings).
Fifteen isolates were obtained over 30 d (September to October 2004), 6 of which were included in the cluster analysis. Simpson’s index of diversity was 53.3%, indicating relatively lower genetic diversity than those observed for other herds. The fact that the mastitis cases were identified during a 1-mo period may have limited the identification of genetic diversity. Only 2 unique genotypes were identified among 6 isolates and, as previously observed, a single genotype per mastitis case was detected most of the time (80%).
Dairy F (WI, Wood Shavings).
Six subclinical K. pneumoniae mastitis cases were identified in cows during the dry period (July and August 2004). Five unique genotypes were identified among 8 isolates; Simpson’s index of diversity was 64.3%, and a single genotype pattern was isolated from the majority (66.7%) of the mastitis cases. Sixty percent of the genotypes detected among the subclinical isolates were also detected among the clinical isolates. The remaining genotypes (40.0%) with arbitrarily assigned alleles, 20, 25, and 32 (Table 3), were unique to the subclinical mastitis cases. Genetic diversity was lower when compared with other dairies. As in dairy E, it is possible that the short period of time in which samples were collected (1 mo) may have limited identification of the true diversity in the population. To confirm this hypothesis, we analyzed the genetic diversity of other dairies for the period of approximately 1 mo and, whenever possible, at the same point in time as in dairies E and F. A decrease from 28.0% (dairy D) to 8.4% (dairy C) in Simpson’s index was observed when other dairies were analyzed for isolates obtained in a 1-mo period.
|
). The 49 genotypes defined among all dairies included the genotypes originating from mastitis cases that were also found on dairy A (Table 1 and Figure 1). Genotypes from dairy A were redefined to the cluster analysis including isolates from all dairy herds. Genotype 1 comprised 25.2% (n = 62) of all isolates, whereas the remaining genotypes comprised no more than 0.4% (n = 1) to 6.9% (n = 17) of the isolates. Genotype 1 was also identified in a dairy analyzed previously (Paulin-Curlee et al., 2007). Association analysis was performed to identify any relationship between genotypes and herds, genotypes and the type of bedding used by each dairy herd, and genotypes and dairy location (state). Of the 49 genotypes identified among the 6 dairy herds, 55.1% were unique to the dairy of origin (P = 0.015), suggesting that specific genotypes occurred in a particular dairy herd (Table 3). Among the genotypes detected in more than one herd, 34.7, 8.2, and 2.0% occurred in 2, 3, and 5 dairy herds, respectively. Genotype 1 was the most common across all dairies in the study. An association was also observed between genotypes and the types of bedding used by each dairy. Sawdust and wood shavings were combined into a single bedding category, termed wood by-products. A total of 55.1% of the genotypes were unique to each bedding type used (P = 0.011), indicating an association between genotypes and particular types of bedding material. Among the genotypes associated with a particular bedding type, 34.7% occurred with the use of recycled manure solids, 14.3% occurred with the use of wood by-products, and 6.1% occurred with sand. A total of 44.9% of the genotypes were associated with more than one bedding type. Most of those (18.4%) occurred with the use of recycled manure solids and wood by-products, followed by recycled manure and sand (14.3%), recycled manure, wood by-products, and sand (10.2%), and wood by-products and sand (2.0%).
The 6 dairies were located in 4 different states (Wisconsin, Minnesota, Indiana, and Pennsylvania), and an association between genotypes and dairy location was also observed. More than 63% of the genotypes were unique to the state of origin (P = 0.017), whereas 36.7% occurred in more than one state. Among the genotypes detected in more than one state, 18.4% occurred in Wisconsin and Indiana, 8.2% occurred in Wisconsin and Minnesota, and 4.1% occurred in Wisconsin and Pennsylvania. Two genotypes (4.1%) were detected in Wisconsin, Indiana, and Minnesota, and only one (2.0%) was detected in all 4 states. The results of AMOVA showed that none of the genetic diversity was due to variation among the states. Eighty-two percent (P = 0.001) of the genetic diversity was accounted for by variation among genotypes within herds. Herds within the same state accounted for 18% (P = 0.001) of the variation.
The genotypes isolated from reinfected and culled cows in dairy A were also detected in other dairies. Twenty-three genotypes matched those of reinfected or culled cows in dairy A and 52.2% were isolated from infected cows on other dairies. Genotype 1 was the most common among dairies A, B, C, D, and E and was among the genotypes isolated from reinfected and culled cows; it was identical to genotype R5, which was predominant in dairy A (Table 1). Two genotypes from subclinical cases of mastitis (19 and 23) were also found in cases of reinfection.
Overall, the incidence of K. pneumoniae mastitis was highest during winter for all of the dairies in this study except one (Figure 2). One dairy had a higher incidence of K. pneumoniae during summer. Dairy A had the highest incidence of K. pneumoniae mastitis overall.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Differences among dairies could have been due to variations in the genotyping technique used in this study. Repetitive DNA sequence PCR is not as discriminatory as other typing techniques such as pulsed-field gel electrophoresis and multilocus sequence typing and can lack interlaboratory reproducibility. We have addressed the reproducibility of rep-PCR by typing the same isolates several times and by comparing its discriminatory power with pulsed-field gel electrophoresis and multilocus sequence typing (Paulin-Curlee, 2007). On the basis of these studies, rep-PCR provides valuable and reproducible information regarding genetic diversity among K. pneumonia isolates.
If fecal shedding served as the source of K. pneumoniae on the dairy (Munoz et al., 2006), then a smaller number of animals may suggest lower overall environmental contamination and possibly a less diverse K. pneumoniae population. In this situation, only a few genotypes would have access to the mammary gland. Individual susceptibility of the host (age, stage of lactation, genetic resistance, and immune, nutritional, or metabolic status) may have played a role in the diversity of genotypes infecting the udder (Burvenich et al., 2003). Seasonality, topography, wind conditions, and the introduction of fewer cows into the milking rotation could also have played a role in the lower genetic diversity. Although all dairies in this study were not adding new animals, it does not rule out the possibility that a less diverse K. pneumoniae population could be present on dairies E and F.
The association between the genetic similarity matrix and the temporal distance matrix could provide more information on the dynamics of K. pneumoniae mastitis. The results showed a negative association between the 2 matrices in all dairies analyzed, indicating that isolates that were collected closer in time had higher similarity than the ones collected farther apart in time. Although the correlation was not very strong, it was significant for 2 dairy herds. The great genetic diversity may explain the low correlation observed. Similar results have been observed previously (Paulin-Curlee et al., 2007) and suggest that the environment may be changing over time, causing the K. pneumoniae population on the dairy also to change with time.
Nineteen genotypes were involved in 24 cases of reinfection in dairy A. Given the great genetic diversity observed, it is not surprising that most of the cows were infected with a different genotype each time. Similarly, the culled cows were infected with a different genotype each time. An interesting preliminary result, with a small sample size, is that similar K. pneumoniae genotypes were detected in clinical cases of mastitis, in the bedding, and in bulk tank milk. Although the directionality of pathogen spread cannot be determined conclusively from our study, this finding supports the contention that bedding may serve as one of the possible sources of mastitis-causing microorganisms, and it highlights the impact of bedding quality and management on animal health and milk quality. To confirm this speculation, a prospective sampling of bedding and mastitis cases will be required to monitor genotypic similarities and differences from isolates derived from the 2 sample types.
Five out of 6 dairies used organic material as bedding. The relatively higher concentration and type of bacteria in bedding has been associated with bacterial concentration on the teat end and with the incidence of clinical mastitis (Zdanowicz et al., 2004). In the current study, recycled manure solids, with a high level of moisture and OM, were used as bedding material on one dairy. A high concentration of Klebsiella spp. and other coliform bacteria has been found in composted manure solids (Mote et al., 1988). Even if properly processed, coliform bacteria can quickly multiply to high numbers in recycled manure under favorable conditions (Zehner et al., 1986). Cows lying on highly contaminated bedding for long periods of time (Hogan et al., 1989) may have contributed to the higher incidence of K. pneumoniae mastitis on this dairy. Recycled manure solids were most likely the source of K. pneumoniae mastitis on dairy A. Dairies C, D, E, and F used either sawdust, wood shavings, or both as bedding. The use of wood by-products, which retain moisture and support the growth of Klebsiella spp., has been associated with mastitis outbreaks (Zdanowicz et al., 2004). Zehner et al. (1986) studied organic bedding materials as growth media for environmental pathogens and showed that K. pneumoniae had the highest growth in all bedding types evaluated. Despite the fact that dairy B used inorganic bedding (sand), which tends to retain less moisture at the surface and has lower OM, Klebsiella mastitis remained a problem. Munoz et al. (2006) detected a greater than 80% prevalence of K. pneumoniae fecal shedding in healthy dairy cattle and speculated that freshly voided feces could add bacteria, OM, and nutrients that could support growth in sand bedding. Thus, when recycling sand bedding material, it is important to remove as much of the manure and other organic material as possible to limit the potential growth of K. pneumoniae. In addition to the choice of bedding (recycled manure solids) and bedding management, many other variables (premilking udder preparation, proper attachment and handling of the milking unit, hygiene of installations, and age, lactation stage, and milk production of cows) could have influenced the high rate of K. pneumoniae mastitis on dairy A.
Distinct genotypes were isolated from the same mastitis case on 5 of the 6 dairies in this study. Such findings have been confirmed previously (Paulin-Curlee et al., 2007) and indicate that more than one K. pneumoniae genotype can be isolated from the infected mammary gland within a herd at a given time point. It is possible that exposure of the udder to a diverse K. pneumoniae population enabled more than one genotype to enter the teat canal concurrently, colonizing and infecting the mammary gland. The anatomical condition of the teat canal and teat end are the first barrier to preventing the entrance of invading pathogens (Elbers et al., 1998) and may have played a role in the isolation of distinct genotypes from the same clinical case. Equally critical is the preparation of the udder before milking, because it reduces bacterial populations at the teat orifice, thus reducing the probability of infection of the mammary gland (Pankey et al., 1987). The fact that cows were rarely infected twice with the same genotype and that more than one genotype pattern was isolated from the same clinical case suggests that K. pneumoniae populations were not only greatly diverse but appeared very dynamic. The dairy environment is continuously changing because of the introduction of new animals, changes in management practices, and the introduction of new bedding types or batches and feed-stuffs, each with its own unique microbial populations and ability to support bacterial growth (Lynn et al., 1998).
Klebsiella pneumoniae had the highest growth in sterile bedding materials when compared with other environmental bacterial species (Zehner et al., 1986). Studies have shown that Klebsiella spp. has extended survival rates in fresh water and marine environments and is more resistant to solar radiation than E. coli (McCambridge and McMeekin, 1981; Lopez-Torres et al., 1987, 1988). Environmental K. pneumoniae under nutrient starvation were able to grow rapidly when transferred to high-nutrient media (Lappin-Scott et al., 1988) and were as virulent as clinical isolates in a mouse infection model (Struve and Krogfelt, 2004). The dairy environment is complex and is influenced by many factors that are difficult to control. These factors, combined with the presence of a wide range of OM and nutrients (dripping milk, feces, urine, feed, bedding, water, etc.), make the dairy facility a complex and nutrient-rich microbial habitat. Thus, it is possible that, in addition to the ubiquitous and hearty nature of K. pneumoniae, the particular conditions encountered in dairy facilities contribute to the higher levels of genetic diversity observed.
Similar seasonal findings were observed previously in our laboratory (Paulin-Curlee et al., 2007), which is contrary to what others have reported for udder infections caused by gram-negative bacteria (Todhunter et al., 1990). Animal crowding, poorly ventilated confinement barns, and high moisture levels associated with the fluctuation of temperature during fall and winter could have supported the survival of bacteria. Such factors may have partially accounted for the increased rate of K. pneumoniae mastitis during the winter.
Although the sample size was limited, specific genotypes appeared to be associated with dairy herd, type of bedding, and state. The results of AMOVA showed that most of the genetic diversity (82%) was attributable to variation among genotypes within herds. Although the degree of difference given by the descriptive analysis and AMOVA varied, both methods consistently revealed that genetic diversity was associated with variation among herds within states and variation among genotypes within herds. Genetic diversity of fecal K. pneumoniae from various mammalian hosts has been reported to be partly associated with geographic location and taxonomic group of the host (Gordon and Lee, 1999). The exact reasons for the occurrence of unique genotypes in particular herds in the present study is unknown. Because specific genotypes also tended to occur with the use of a particular bedding material, it is possible that bedding was one of the risk factors. The characteristics of each dairy, such as the type of bedding and feed, the inherent intestinal microbial population of the animals in the herd, and the presence of other animals (domestic and wild) could have played a role in the observed associations.
Different management systems appeared to have an effect on the number and types of bacteria detected in clean and recycled sand bedding (Kristula et al., 2005). A similar effect may have been demonstrated regarding the genetic diversity of K. pneumoniae mastitis and the association of genotypes with herd and bedding type. Despite some common characteristics, each dairy is different in terms of herd size, geographic location, topography, installations, animal health status, and environment hygiene standards, all factors that could influence which genotypes of K. pneumoniae predominate in the herd. Conversely, when analyzing genetic diversity among all herds, genotype 1 was more nonrandomly distributed across dairies than were other genotypes. In addition, genotype 1 was associated with the largest number of reinfections and culling in dairy A. Whether this was due to the higher number of animals infected with this genotype remains unknown. Such findings raise the possibility of a prevalent genotype among herds or the existence of a common source of contamination. Alternately, it is possible that specific genotypes (e.g., genotype 1) may have converged into a pathogenic phenotype.
Molecular typing of K. pneumoniae, together with a detailed assessment of management practices, could be useful in identifying the major sources of Klebsiella within a herd. Detection of major genotypes causing mastitis can be useful to identify the source, which could then be targeted for management. This information could then be used to prevent an outbreak before it becomes a problem for the dairy. Considering the rapid progression and the high fatality rates of K. pneumoniae infection, the early detection of a contaminated source may prevent other cows from becoming ill and the dairy from sustaining major economic losses.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Received for publication June 25, 2007. Accepted for publication September 27, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
