Vaccination strategies for reducing clinical severity of coliform mastitis

Vet Clin Food Anim 19 (2003) 187 197 Vaccination strategies for reducing clinical severity of coliform mastitis David J. Wilson, DVM, MS*, Rubén N. Gonza lez, DVM, MPVM, PhD Quality Milk Production Services, College of Veterinary Medicine, Cornell University, 22 Thornwood Drive, Ithaca, NY 14850, USA The history of vaccination against coliform mastitis is largely inseparable from the story of the search for protection against gram-negative mastitis pathogens. Gram-negative organisms that cause mastitis are coliform bacteria (genera Escherichia, Klebsiella, and Enterobacter), as well as Serratia, Citrobacter, Proteus, and Pseudomonas [1]. These bacteria are found in large numbers in the dairy environment (manure, bedding material, soil, slurry, and water [2,3]), and their passage to the inside of the udder may occur between milkings [4], at milking, or during the dry period [5]. Most cases of coliform mastitis are caused by Escherichia coli and Klebsiella spp [6 10]. Coliform intramammary infections are often without any clinical signs; however, coliform mastitis has also been reported to cause abnormal appearance of milk (often described as yellow or serous), hard mammary quarters, depressed appetite, reduced milk production, decreased milk-fat percentage, numerous biochemical changes in systemic bloodstream circulation evaluated as measures of severity, dehydration, diarrhea, recumbency, and death. [8,9,11,12]. Since 1910, a prime objective among researchers has been to provide bacterins to combat agents that cause bovine mastitis [13]. For the next 80 years, however, finding an effective vaccine proved to be difficult [14,15]. The main obstacles encountered by researchers were the disparity of bacterial species and strains, incomplete knowledge of the specific immunogenic factors, failure to maintain high immunoglobulin levels in milk, and lack of appropriate immunization schedules [16]. * Corresponding author. E-mail address: djw11@cornell.edu (D.J. Wilson). 0749-0720/03/$ - see front matter Ó 2003, Elsevier Science (USA). All rights reserved. PII: S 0 7 4 9-0 7 2 0 ( 0 2 ) 0 0 0 7 0-1

188 D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 One study [17] demonstrated that the local or general immune system of the cow was unable to acquire immunity from Escherichia coli mastitis. Later, researchers in Australia observed that cows do not produce a memory type of immune response following a mastitis infection [18]. Despite the information provided by those studies, the failure of postmilking teat dipping and nonlactating cow therapy to reduce coliform mastitis [4,19] continued to stimulate extensive research on immunization against coliform mastitis. In more recent times, it is often speculated that the proportion of total mastitis cases or the proportion of clinical mastitis cases caused by coliform mastitis has increased. Nevertheless, when one searches the published literature for evidence of this, it is confined to reports of one to a few dairy herds, usually large dairy herds in proximity to a university or veterinary practice. The authors have also observed this phenomenon of high levels of coliform mastitis in individual dairy herds. Reports of prevalence of mastitis among large numbers of dairy herds and dairy cattle sampled (including clinical mastitis cases), however, do not demonstrate an increase in the percentage of coliform mastitis cases [10,20,21]. Nevertheless, the financial importance of coliform mastitis is considerable [10], and its cost to the dairy industry may be increasing. This is at least partly because as somatic cell counts have been reduced in the milk of dairy cattle, particularly as more cows and quarters have somatic cell counts below 250,000/mL, increased susceptibility to clinical mastitis has been observed [22,23]. Precisely because of increased risk of clinical mastitis with lower somatic cell counts and the possibility of culling or death of affected cows, there continues to be considerable interest in immunization against coliform mastitis. Early experimental studies in immunization against Escherichia coli mastitis were performed using the heterogeneous oligosaccharide ( O or somatic) antigens. In one study, the bacterin was infused in the mammary gland during the nonlactating period [24,25]. Researchers observed that the bacterin stimulated a local antibody response in the subsequent lactation. In another experiment, cows were administered the first bacterin dose subcutaneously at the beginning of the nonlactation period and a booster intramammarily 5 weeks later [26,27]. It was found that immunization did not modify noticeably the mode of transfer of blood proteins into milk. In vitro tests, however, suggested that immunization with the Escherichia coli bacterin enhanced the recruitment of phagocytic cells and established preinflammatory opsonic activity [26]. The variety of genera, species, and serotypes of coliform bacteria as classified by surface antigens identified in individual dairy farms [6], the complexity of the problem involved, and the specificity of the vaccines used was thought to make impractical the prophylaxis of coliform mastitis by immunizing cows [28]. Most gram-negative bacteria, however, share a common inner lipopolysaccharide structure that is pathogenic and highly antigenic [29] and that cross-reacts with lipopolysaccharide derived from a variety of

D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 189 gram-negative bacteria [29,30]. Lipopolysaccharide also protects invading organisms from cellular and humoral elements of the infected host s immune system while exhibiting a number of biologic activities directly or indirectly involved in the pathogenesis of sepsis and septic shock [31]. Moreover, lipopolysaccharide has been implicated in bacterial resistance to complement-mediated killing [32]. For immunization of dairy cows against coliform mastitis, two rough ( R ) bacterial mutants have been employed in commercial preparations of core antigen bacterins. They are Escherichia coli O111:B4 (strain J5) and Salmonella typhimurium Re-17. The rough mutant of Escherichia coli called J5 was found to produce antiserum that was protective against subsequent infective doses of the organism in mice in 1985 [33]. After observations that dairy cows with naturally occurring IgG1 ELISA titers to Escherichia coli J5 <1:240 were associated with five times the rate of clinical coliform mastitis compared to animals with titers >1:240 [34], the potential value of immunization using core antigen bacterins became apparent. Most of the commercially available vaccines against coliform mastitis are Escherichia coli J5 vaccines. The J5 strain has a relatively exposed core antigen, sometimes called the J5 core antigen. This core antigen is a lipopolysaccharide structure made up of lipid A and some common core polysaccharides [35,36]. These core polysaccharides are present in many kinds of gram-negative bacteria. This may at least partially explain why J5 antigen results in immunity associated with protection against many genera and strains of bacteria; however, the mechanisms for how J5 immunization actually works have not been explained [35,37]. Early evidence that the Re-17 mutant strain of Samonella typhimurium could produce a bacterin protective against coliforms and other gramnegative bacteria came in a study with calves. The Re-17 strain has no O side chains (somatic antigens), presenting only core antigens to the immune system. A bacterin made from killed bacteria and adjuvant was administered in cervical muscle to 12 calves aged 3 to 4 months, and a booster injection was given 14 days later. Placebo (adjuvant and saline) was administered at the same times to 12 control calves. Two weeks after the second injection of either bacterin or placebo, 10 calves (5 vaccinates, 5 controls) were challenged with Escherichia coli O55:B5 endotoxin intravenously and 14 calves (7 vaccinates, 7 controls) were challenged with Pasteurella multocida endotoxin intravenously. Vaccinated calves had significantly reduced signs of colic, dyspnea, and anorexia on a standardized scale of clinical severity (P<0.001) and higher levels of total IgG specific for the endotoxin [38]. Coliform mastitis core antigen bacterin efficacy In the late 1980s, evidence began to emerge that vaccination with J5 bacterins had some efficacy against clinical coliform mastitis in cows. In two

190 D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 commercial dairy herds in California, a regimen of two doses of J5 during the dry period and one dose after calving was administered to 246 cows (240 control cows were unvaccinated). Route of administration was subcutaneous injection, cranial to the scapula. Clinical coliform mastitis, mostly caused by Escherichia coli and Klebsiella spp, was diagnosed in 6 of 246 (2.4%) vaccinated cows and 29 of 240 (12.1%) unvaccinated cows. The estimated risk ratio for having clinical gram-negative mastitis for vaccinated cows compared with controls was 0.20. These results suggested effectiveness of J5 bacterin vaccination against naturally occurring clinical coliform mastitis [39]. In another study, either J5 bacterin or a placebo was administered subcutaneously at approximately 6 months, 7 months, and 8 months of gestation to Holsteins on one commercial dairy. Cows were randomly assigned to receive the vaccine (n ¼ 212) or placebo (n ¼ 229). The percentage of immunized cows and control cows contracting clinical coliform mastitis was 3.3% (7/212) and 10.9% (25/229), respectively. This reduction in clinical coliform cases was statistically significant [40]. A multiyear study of a 225-cow commercial dairy herd was conducted using subcutaneously administered J5 bacterin (caudal to the scapula). Doses of J5 were administered at dry-off, 30 days later, and at calving; there was also an unvaccinated control cow group. It was found that prevalence of gram-negative intramammary infections at calving was not different among vaccinates and controls. The rate of gram-negative bacterial clinical mastitis cases, however, was 0.014/100 cow-days in vaccinates compared with 0.057/100 cow-days in control cows during the first 90 days in milk (P<0.05) [41]. Two commercial dairy herds with 4800 and 1100 lactating Holstein cows were studied for the effects of immunization with Re-17 mutant Salmonella typhimurium bacterin toxoid. Cows were randomly assigned to control or vaccinate groups and later were assigned to pairs for analysis based on herd, parity, calving date, and projected 305-day milk yield. Pluriparous cow vaccinates were injected twice intramuscularly with Re-17 toxoid at dry-off and 2 to 3 weeks before calving. Primiparous heifer vaccinates were given vaccine during the last trimester of pregnancy and again approximately 3 weeks before calving. Final analysis was done for 646 controls and 646 vaccinates with both cows in each pair managed identically within each herd. Clinical cases of mastitis caused by coliform bacteria (96% of those were Escherichia coli) were compared among the treatment groups for the first 5 months of lactation. Control cows contracted 0.093 cases/100 cow days and vaccinates contracted 0.054 cases/100 cow days (P<0.05). Mortality during the first 5 months of lactation was 0.6% for controls and 0.2% for vaccinates; rate of removal from the herd (from death and culling combined) was 2.3% for controls and 0.9% for vaccinates. Case fatality rate for clinical coliform cases was also lower for vaccinates: 2.0% versus 5.1% for controls (P<0.05). The benefits of reduced clinical mastitis

D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 191 cases and increased survival in the herd were limited to cases caused by coliform bacteria [42]. Another field study of Re-17 mutant vaccine was performed in Canada. Dairy cattle from 82 herds in three provinces were randomly assigned to receive either vaccine or placebo, administered twice 3 weeks apart. Cows were randomized to one of the two treatment groups from within three blocks: 120 to 160 days before calving, 80 to 120 days before calving, or 40 to 80 days before calving. (Regardless of the time interval before calving that the cows were in at time of treatment, each cow was given only two doses of vaccine or placebo 3 weeks apart). Final analysis of 1882 cows showed limited differences between vaccinates and controls in coliform or clinical mastitis; however, overall death rate (from undifferentiated causes) was 1.9% per lactation for vaccinates and 3.4% for controls (P<0.05). When the authors combined cases of clinical mastitis with coliform bacteria isolated from milk together with clinical cases with no growth of bacteria, the lactational incidence rate was 10.8% in the vaccinate group and 17.7% in the placebo group. This difference was nearly significant at a ¼ 0.05 (P ¼ 0.06) [43]. It was of interest to one of the authors (RG) to see how well supramammary lymph node region administration of J5 bacterin would work in terms of safety and efficacy. A commercial herd of 280 lactating Holstein cows was studied for 2.5 years. Cows that had completed at least one lactation were randomly assigned at dry-off to a vaccinate group (n ¼ 180) or a control group (n ¼ 60). Vaccinates were administered J5 vaccine plus metabolizable oil adjuvant, and controls were administered placebo of 0.9% NaCl plus the same adjuvant in the supramammary region. There were 50 total cases of clinical coliform mastitis isolated: 36 Escherichia coli, 8 Klebsiella pneumoniae, 5 Enterobacter aerogenes, and 1 Enterobacter agglomerans. The percentage of immunized cows and control cows that contracted clinical coliform mastitis was 12.8% (23/180) and 45.0% (27/60), respectively, (P<0.005). There were no adverse effects from any supramammary injections of either vaccine or placebo [6]. A study to investigate short-term milk production effects following immunization with J5 was performed. A 7% decrease in milk production was observed at the second and third milkings following vaccination. This was the only significant production loss over the first 10 milkings (5 days) after vaccination [44]. Financial return from use of J5 immunization was investigated in the early 1990s. Partial budget analysis showed increased profits of $57.00 per cow lactation associated with use of J5. A computer spreadsheet analysis predicted that when >1% of cow lactations were affected by clinical coliform mastitis, a dairy herd vaccination program using J5 should be profitable [45]. The amount of genetic diversity present among strains of coliform bacteria that might be affected by J5 was studied in 1996. Using an

192 D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 automated ribotyping system (RiboPrinter Microbial Characterization System, E.I. dupont De Nemours and Co., Wilmington, DE), 25 coliform mastitis isolates were evaluated to identify genetically different strains. Genetic diversity was considerable; the number of different ribotypes found from the number of isolates (single isolated bacterial colonies) tested was as follows: eight from 16 Escherichia coli, six from 6 K pneumoniae, and two from 3 Enterobacter agglomerans. Together with the efficacy results from this study as reported earlier (clinical coliform mastitis in vaccinates 12.8% versus 45.0% for controls), these data suggest that J5 is indeed effective against a genetically varied population of bacteria [6]. Comparison of coliform mastitis bacterin regimens A major practical question regarding coliform mastitis immunization is whether to use a regimen of two or three doses of vaccine. This is partly a matter of the cost of vaccine and labor, but mostly because many dairy producers still find it more convenient to handle cattle at the time of dryoff and again at calving and not necessarily during the middle of the nonlactating dry period. The greatest reduction in coliform clinical mastitis cases has been observed when three immunizations have been administered. When three immunizations were used, the associated decrease in clinical coliform mastitis ranged from 70% to 80%, with a median of 75%; with two doses, the reduction was 39% and 42% [6,39 43]. Keeping in mind that although the mechanism or mechanisms of how both J5 and Re-17 vaccines work is unclear, it seems logical that more than one booster immunization may indeed be beneficial. Certainly, the above data support the administration of three vaccinations. Based on the data available, the authors recommend to dairy producers that they use a regimen of three vaccinations. One of the most frequent questions that dairy producers pose regarding coliform mastitis vaccination is which specific dosage schedule has been shown to work the best. Because of the multifactorial nature of mastitis and the influence of the herd effect in dairy cattle disease, this question cannot be definitively answered for every farm. In direct comparison of three regimens of J5 vaccination, however, the schedule of J5 bacterin administered at dry-off, 4 weeks later (regardless of due date or anticipated length of dry period), and 1 to 7 days after calving (on the same day each week to all cows fresh during the previous week) resulted in a 93% reduction in clinical coliform mastitis [6]. Because of this and from empiric experience, the authors recommend the use of three doses of J5 according to the above schedule. Nevertheless, some dairy producers find it easier to administer the second vaccination when cows are moved to a prefresh facility 3 weeks before they are due to calve. The authors do not consider this a critical difference from the schedule of immunizations at dry-off, 4 weeks later, and at calving; however, it should be noted that when one compares the above

D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 193 two schedules (the latter of which was associated with 93% reduction of clinical coliform mastitis cases), they result in substantially different timing of the second vaccination in cases of long dry periods (such as greater than 90 days). Dairy producers also often ask about using additional doses of coliform mastitis vaccine at various times during early lactation between 30 and 90 days in milk. There are few refereed publications of research results concerning this important question; therefore, there is currently no basis for recommending additional boosters of coliform mastitis bacterin during lactation. New information may soon be available, however, regarding this possible new application, particularly of J5 bacterin (J. Burton and R. Erskine, East Lansing, Mich, personal communication, July 2002). Other considerations in practical application on dairy farms Is the use of coliform mastitis vaccination indicated and cost effective for every dairy herd? In the experience of the authors, it probably is not. An important consideration in the adoption of any preventive health practice is that it is not always known whether it is working: a major outbreak of disease may be prevented or greatly reduced in magnitude and yet never be apparent. In herds using coliform mastitis vaccination on a regular schedule, how many cows might be saved from severe or fatal cases of mastitis that would otherwise have occurred? With seasonal and herd variation, this question cannot be answered with certainty. Therefore, the authors do not discourage the use of coliform mastitis vaccination by any dairy producers who have recently determined to adopt it or who have been using it in their herds. When one considers the relative costs of three vaccinations per year and of replacing a culled or dead dairy cow, it still seems quite reasonable to assume that if approximately 1 of every 100 cows lactating for a year can be saved from a severe case of gram-negative mastitis, then coliform mastitis vaccination makes financial sense, just as it was first reported 10 years ago [45]. The practical question on this subject facing most dairy herd veterinarians is whether a farm that is not currently using a coliform mastitis bacterin should start doing so. The authors recommend adoption of coliform mastitis vaccination in herds when a review of culture of milk of the entire lactating herd reveals 1% or more of the cows to be positive for Escherichia coli, Klebsiella sp, or Enterobacter spp, and/or the rate of clinical mastitis is greater than 4% per month, with the predominance of mastitis in the herd caused by gram-negative agents. During the year from July 2001 to June 2002, data was available regarding coliform mastitis bacterin use in 752 dairy herds in New York and northern Pennsylvania (unpublished data). Coliform mastitis vaccination was used on 21.3% of the farms utilizing a variety of trade names and

194 D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 vaccination regimens; the other 78.7% of farms did not use coliform mastitis vaccine. The following are some characteristics of those dairy herds using coliform bacterin and those not using it, respectively: lactating herd size 202 cows, 67 cows; somatic cell counts in bulk milk 359,500/mL, 439,800/mL; 305-day milk production per cow 18,392 lb, 15,055 lb; percentage of farms using freestall housing 46%, 12%. Prevalence of coliform mastitis among both categories of herds was <1%, lower than the prevalence of gram-positive mastitis in both types of herds (unpublished data). These data support the perception that coliform mastitis vaccination is used more on farms with larger, higher producing dairy herds with better control of major contagious mastitis. One scenario can lead to great disappointment on the part of dairy producers regarding the use of coliform mastitis bacterins. When a dairy herd is known to be free of Streptococcus agalactiae and mycoplasmal mastitis and to be free of or to have low levels of Staphylococcus aureus intramammary infections in the herd, usually accompanied by bulk-tank somatic cell counts <250,000/mL, it may be a reasonable assumption that much of the clinical mastitis in the herd is environmental in origin [23]. However, regardless of appearance of milk, rectal temperatures of cows with clinical mastitis, and systemic signs of disease such as diarrhea, dehydration, recumbency, or death, it cannot be assumed what proportion of the clinical mastitis cases are indeed caused by gram-negative bacteria including coliforms. Sufficient milk microbiologic culture results from clinical mastitis cases are necessary to confirm this proportion [46]. On many farms, grampositive agents of mastitis such as environmental streptococci, coagulasenegative staphylococci, and Staphylococcus aureus are still much more common agents of mastitis than gram-negative agents [10,20]. Dairy producers should not be given the expectation that adoption of a vaccination program with J5 or Re-17 bacterins will drastically reduce (or even appreciably reduce) the rate or severity of clinical mastitis cases unless and until there is sufficient evidence from microbiologic culture results that gram-negative bacteria are important causative agents in their herds. Initiating a coliform mastitis bacterin immunization program in a herd where the majority of clinical mastitis cases are caused by gram-positive agents will not produce favorable results. There are limited data concerning direct comparison of the routes of administration of coliform mastitis bacterins. Jersey cows were given a commercial J5 bacterin at dry-off and 2 weeks before they were due to calve by either subcutaneous injection in the neck region or injection in the supramammary lymph node region. Route of administration was not associated with differences in severity of clinical coliform mastitis [47]. Administration to Holstein cattle of J5 bacterin subcutaneously (cranial to the scapula) was compared with supramammary injection (in the depression between the escutcheon and the rear leg on either the right or left side) by the authors (unpublished data). There was no difference in efficacy

D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 195 (unpublished data) and the authors found that supramammary injection is easier on cows and on people. Subcutaneous injection can be very difficult even when cows are restrained in lockup headgates or treatment chutes. In contrast, cows in tiestalls, stanchions, freestalls, lockup gates, or treatment chutes can usually be conveniently administered the supramammary injections. The authors have never observed or been informed of problems with injection site reactions or lesions from supramammary injections. As reported earlier, the highest reported reduction in clinical coliform mastitis cases associated with J5 vaccination (93%) was observed using supramammary injections [6]. Dairy producers should be encouraged to discuss this with their herd veterinarian and with technical experts from the manufacturer of the coliform mastitis vaccine they are using; however, the authors have found excellent acceptance of administration of coliform mastitis bacterins in the supramammary region by dairy producers. Reservations about practicality of administering the mid dry period dose of the vaccine are often eliminated when producers and other farm employees become used to the relative ease of supramammary vaccination in dry-cow housing areas. Possible future developments in coliform mastitis vaccination strategies An important practical question regarding future use of coliform mastitis immunizations is whether more boosters may further enhance the protection against clinical signs. New evidence is emerging that as many as 12 consecutive monthly J5 bacterin injections during lactation may provide additional immunity against clinical coliform mastitis; however, further studies concerning whether this confers additional resistance to clinical signs are needed (J. Burton and R. Erskine, East Lansing, Mich, personal communication July, 2002). References [1] Hogan J, González R, Harmon R, et al. Laboratory handbook on bovine mastitis. Madison (WI): National Mastitis Council; 1999. p. 85 111. [2] Carroll E. Environmental factors in bovine mastitis. J Am Vet Assoc 1977;170:1143 8. [3] Natzke R, LeClair B. Coliform contaminated bedding and new infections. J Dairy Sci 1975;59:2152 4. [4] Bramley A, Neave F. Studies on the control of coliform mastitis in dairy cows. Br Vet J 1975;131:160 9. [5] Eberhart RJ. Coliform mastitis. Vet Clin Food Anim 1984;6:287 99. [6] González R, Wilson D, Mohammed H, et al. A placebo-controlled trial of an Escherichia coli J5 bacterin and the ribotyping-based asessment of coliform bacteria diversity on a dairy farm. Proceedings of the 19th World Buiatrics Congress. Edinburgh: British Cattle Veterinary Association; 1996. p. 277 80. [7] Jasper D, Dellinger J, Bushnell R. Herd studies on coliform mastitis. J Am Vet Med Assoc 1975;166(8):778 80.

196 D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 [8] Jones G, Ward G. Cause, occurrence, and clinical signs of mastitis and anorexia in cows in a Wisconsin study. J Am Vet Med Assoc 1989;195(8):1108 13. [9] Wenz J, Barrington G, Garry F, et al. Bacteremia associated with naturally occurring acute coliform mastitis in dairy cows. J Am Vet Med Assoc 2001;219(7):976 81. [10] Wilson D, González R, Das H. Bovine mastitis pathogens in New York and Pennsylvania: prevalence and effects on somatic cell count and milk production. J Dairy Sci 1997; 80(10):2592 8. [11] Fuquay J, Zook A, Poe W. Metabolic and physiologic response of dairy cattle to coliform mastitis. J Dairy Sci 1975;58(5):751 2. [12] Jackson E, Bramley J. Coliform mastitis. In Practice 1983;5:135 46. [13] Munch-Petersen E. Bovine mastitis. In: Survey of the literature to the end of 1935. Weybridge, England: Imperial Bureau of Animal Health; 1938. p. 219 33. [14] Anderson JC. The problems of immunization against staphylococcal mastitis. Br Vet J 1978;134:412 20. [15] Poutrel B. Susceptibility to mastitis: a review of factors related to the cow. Ann Rech Vet 1982;13:85 9. [16] Vecht U. Immunological approach to mastitis control. Kieler Milch Forsch 1985;37:515 23. [17] Bramley A, Godinho K, Grindal R. Evidence of penetration of the bovine teat duct by Escherichia coli in the interval between milkings. J Dairy Res 1981;48:379 86. [18] Colditz I, Watson D. The immunophysiological basis for vaccinating ruminants against mastitis. Aust Vet J 1985;62:145 53. [19] Dodd FH. Progress in mastitis control. Kieler Milch Forsch 1985;37:216 23. [20] Myllys V, Asplund K, Brofeldt E, et al. Bovine mastitis in Finland in 1988 and 1995 changes in prevalence and antimicrobial resistance. Acta Vet Scand 1998;39(1):119 26. [21] Myllys V, Honkanen-Buzalski P, Huovinen M, et al. Association of changes in the bacterial ecology of bovine mastitis with changes in the use of milking machines and antibacterial drugs. Acta Vet Scand 1994;35(4):363 9. [22] Beaudeau F, Fourichon C, Seegers H, et al. Risk of clinical mastitis in dairy herds with a high proportion of low individual milk somatic-cell counts. Prev Vet Med 2002;53(1 2):43 54. [23] Green M, Green L, Cripps P. Low bulk milk somatic cell counts and toxic mastitis: an association. Proceedings of the 19th World Buiatrics Congress. Edinburgh: British Cattle Veterinary Association; 1996. p. 312 5. [24] Wilson M. The influence of preparturient intramammary vaccination on bovine mammary secretions. Immunology 1972;23:947 55. [25] Wilson M, Duncan J, Heistand F, et al. The influence of preparturient intramammary vaccination on immunoglobulin levels in bovine mammary secretions. Immunology 1972;23:313 20. [26] Rainard P. Experimental mastitis with Escherichia coli: sequential response of leukocytes and opsonic activity in milk of immunized and unimmunized cows. Ann Rech Vet 1983;14:281 6. [27] Rainard P, Caffin J. Sequential changes in serum albumin, immunoglobulin (IgG1, IgG2, IgM) and lactoferrin concentrations in milk following infusion of Escherichia coli into the udder of immunized and unimmunized cows. Ann Rech Vet 1983;14:271 9. [28] Smith K. Mastitis control: a discussion. J Dairy Sci 1983;66:1790 4. [29] Braude A, Ziegler E, Douglas H, et al. Antibody to cell wall glycolipid of gram-negative bacteria: induction of immunity to bacteremia and endotoxemia. J Infect Dis 1977; 136:S167 72. [30] Nells M, Niswander C. Mouse monoclonal antibodies reactive with J5 lipopolysaccharide exhibit extensive serological cross-reactivity with a variety of gram-negative bacteria. Infect Immun 1984;46:677 81. [31] Oishi K, Koles N, Guelde G, et al. Antibacterial and protective properties of monoclonal antibodies reactive with Escherichia coli O111:B4 lipopolysaccharide: relation to antibody isotype and complement-fixing activity. J Infect Dis 1992;165:34 45.

D.J. Wilson, R.N. González / Vet Clin Food Anim 19 (2003) 187 197 197 [32] Pluschke G, Mayden J, Achtman M, et al. Role of the capsule and the O antigen in resistance of O18:K1 Escherichia coli to complement-mediated killing. Infect Immun 1983;42:907 13. [33] Sakulramrung R, Domingue G. Cross-reactive immunoprotective antibodies to Escherichia coli O111 rough mutant J5. J Infect Dis 1985;151(6):995 1004. [34] Tyler J, Cullor J, Osburn B, et al. Relationship between serological recognition of Escherichia coli O111:B4 (J5) and clinical coliform mastitis in cattle. Am J Vet Res 1988;49:1950 4. [35] Dosogne H, Vangroenweghe F, Burvenich C. Potential mechanism of action of J5 vaccine in protection against severe bovine coliform mastitis. Vet Res 2002;33:1 12. [36] Tyler J, Spears H, Nelson R. Antigenic homology of endotoxin with a coliform mastitis vaccine strain, Escherichia coli O111:B4 (J5). J Dairy Sci 1992;75(7):1821 5. [37] Baumgartner J, Heumann D, Calandra T, et al. Antibodies to lipopolysaccharides after immunization of humans with the rough mutant Escherichia coli J5. J Infect Dis 1991;163(4):769 72. [38] Sprouse R, Garner H, Lager K. Cross-protection of calves from E. coli and P. multocida endotoxin challenges via S. typhimurium mutant bacterin-toxoid. Agri Pract 1990;11(2): 29 34. [39] González R, Cullor J, Jasper D, et al. Prevention of clinical coliform mastitis in dairy cows by a mutant Escherichia coli vaccine. Can J Vet Res 1989;53(3):301 5. [40] Cullor JS. The Escherichia coli J5 vaccine: investigating a new tool to combat mastitis. Vet Med 1991;86(8):836 44. [41] Hogan J, Todhunter D, Tomita G, et al. Field trial to determine efficacy of an Escherichia coli J5 vaccine on mammary health. J Dairy Sci 1991;74(Suppl 1):169. [42] McClure A, Christopher E, Wolff W, et al. Effect of Re-17 mutant Salmonella typhimurium bacterin toxoid on clinical coliform mastitis. J Dairy Sci 1994;77(8):2272 80. [43] Scott H, Sargeant J, Ireland M, et al. Efficacy of an Re-17 mutant Salmonella typhimurium core antigen vaccine under field conditions. Proceedings of the 19th World Buiatrics Congress. Edinburgh: British Cattle Veterinary Association; 1996. p. 281 4. [44] Musser J, Anderson K. Effect of vaccination with an Escherichia coli bacterin-toxoid on milk production in dairy cattle. J Am Vet Med Assoc 1996;209(7):1291 3. [45] DeGraves F, Fetrow J. Partial budget analysis of vaccinating dairy cattle against coliform mastitis with an Escherichia coli J5 vaccine. J Am Vet Med Assoc 1991;199(4):451 5. [46] White M, Glickman L, Montgomery M, et al. Analysis of the clinical findings used to diagnose coliform mastitis in dairy cows, and comparison to a prediction model. Cornell Vet 1987;77:13 20. [47] Tomita G, Nickerson S, Owens W, et al. Influence of route of vaccine administration against experimental intramammary infection caused by Escherichia coli. J Dairy Sci 1998;81(8):2159 64.