Mastitis is the inflammation of the mammary gland and most commonly occurs in dairy cows as a result of a bacterial infection. This disease has a high prevalence and remains among the most frequently cited reasons for the culling of cows (Esslemont & Kossaibati, Reference Esslemont and Kossaibati1997; Bascom & Young, Reference Bascom and Young1998). Mastitis is also one of the most costly diseases to the dairy industry and is estimated to result in worldwide economic losses of $35 billion annually (Wellenberg et al. Reference Wellenberg, van der Poel and Van Oirschot2002).
Staphylococcus aureus is one of the most prevalent bacteria to cause mastitis (Fox et al. Reference Fox, Chester, Hallberg, Nickerson, Pankey and Weaver1995; Wilson et al. Reference Wilson, Gonzalez and Das1997). Although Staph. aureus commonly establishes a chronic subclinical infection, this pathogen is also a leading cause of clinical mastitis (Wilesmith et al. Reference Wilesmith, Francis and Wilson1986; Barkema et al. Reference Barkema, Schukken, Lam, Beiboer, Wilmink, Benedictus and Brand1998). Staph. aureus intramammary infections are often refractory to antibiotic therapy (Deluyker et al. Reference Deluyker, Michanek, Wuyts, VanOye and Chester2001; Gillespie et al. Reference Gillespie, Moorehead, Lunn, Dowlen, Johnson, Lamar, Lewis, Ivey, Hallberg, Chester and Oliver2002) and can persist throughout the lifetime of the animal (Sutra & Poutrel, Reference Sutra and Poutrel1994).
Breed-dependent genetic differences are known to influence the disease resistance of food producing animals (Kelm et al. Reference Kelm, Freeman and Kehrli Jr2001). In regard to two popular dairy breeds, Jersey cows have been reported to have a lower prevalence of, and risk for, mastitis than Holstein cows (Erb & Martin, Reference Erb and Martin1978; Motie et al. Reference Motie, Ramudit and Mohabir1985; Morse et al. Reference Morse, DeLorenzo, Wilcox, Natzke and Bray1987; Washburn et al. Reference Washburn, White, Green Jr and Benson2002; Dego & Tareke, Reference Dego and Tareke2003; Youngerman et al. Reference Youngerman, Saxton, Oliver and Pighetti2004; Biffa et al. Reference Biffa, Debela and Beyene2005; Berry et al. Reference Berry, Lee, Macdonald, Stafford, Matthews and Roche2007). Large-scale surveys have also reported that Jersey cows tend to have higher milk somatic cell counts (SCC) than Holstein cows (Sewalem et al. Reference Sewalem, Miglior, Kistemaker and Van Doormaal2006; Paape et al. Reference Paape, Wiggans, Bannerman, Thomas, Sanders, Contreras, Moroni and Miller2007). Because milk SCC increase dramatically in response to infection and play a protective role, it remains unclear whether the breed differences in SCC reported in these latter studies reflect a differential prevalence of underlying intramammary infection or influence the nature of the response to the infection.
Previous studies have established that differential innate immune responses elicited to bacteria influence the outcome of intramammary infection. Whereas Escherichia coli elicits a heightened inflammatory response characteristically leading to its rapid elimination in mid- and late-lactation cows, Staph. aureus evokes a more modest response and establishes a chronic infection (Riollet et al. Reference Riollet, Rainard and Poutrel2000; Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004). Induction of a heightened inflammatory response early after Staph. aureus infection results in a ten-fold reduction in bacterial recovery from infected glands (Kauf et al. Reference Kauf, Vinyard and Bannerman2007). Thus, differences in the nature of the inflammatory response influence the ability of the host to control intramammary infections. Because breed-dependent differences in the prevalence of mastitis between Holstein and Jersey cows have been reported, we studied whether there were differences in the innate immune response of these breeds to Staph. aureus intramammary infection that may explain the differential prevalence of mastitis.
Materials and Methods
Animals
Clinically healthy Holstein and Jersey cows, all of which were in their first lactation, were selected from the USDA National Animal Disease Center herd. The mean (±se) number of days in milk was comparable between the Holstein (132±12 d) and Jersey (152±10 d) cows (P=0·2145). None of the cows were pregnant at the time of the study. Experimental infection was induced in the right front quarter of each animal so as to eliminate potential quarter-dependent effects that could arise if different quarters on the various cows were infected. Milk SCC of <200 000 cells/ml and the absence of specific bacterial growth in two independent milk samples collected from these quarters within the 1-week period prior to the study established that these quarters were free of pre-existing infection and mastitis. The use and care of all animals in this study was approved by the USDA National Animal Disease Center's Animal Care and Use Committee.
Experimental intramammary infection
Staph. aureus strain 305 (American Type Culture Collection, Manassas VA, USA), which was originally isolated from a clinical case of mastitis (Newbould, Reference Newbould1977), was used to induce experimental intramammary infection. Preparation of the bacterial inoculum was performed as previously described (Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004). On experimental day 0, immediately following the morning milking, the right front quarters of ten Holstein and ten Jersey cows were infused with 3 ml of the inoculum. Plating of an aliquot of the inoculum on blood agar confirmed that each quarter was infused with 68 colony forming units (cfu) of Staph. aureus.
Quantification of intramammary Staph. aureus growth
Milk samples were aseptically collected from all infused quarters at various time points throughout the study and serially diluted in sterile phosphate-buffered saline. One-hundred microlitres of the resulting dilutions were spread on blood agar plates, the latter of which were subsequently incubated for 24 h at 37°C. Following incubation, the plates were examined for bacterial growth and colonies enumerated.
Quantification of milk SCC and circulating white blood cells (WBC)
To quantify somatic cells, milk samples were heated to 60°C for 15 min and subsequently maintained at 40°C until counted on an automated cell counter (Bentley Somacount 150; Bentley Instruments, Inc., Chaska MN, USA). For the determination of total and differential WBC counts, blood was collected from the jugular vein into Vacutainer® glass tubes containing EDTA (Becton-Dickinson Corp, Franklin Lakes, NJ, USA). The tubes were inverted ten times, and analysed using a HEMAVET® 1500 multi-species haematology system (CDC Technologies/Drew Scientific, Inc., Oxford CT, USA).
Whey and plasma preparation
Milk samples were centrifuged at 44 000 g at 4°C for 30 min and the fat layer removed with a spatula. The skimmed milk was decanted into a clean tube, re-centrifuged for 30 min, and the translucent whey supernatant collected and stored at −70°C. For the preparation of plasma, blood samples were collected as above, inverted ten times, centrifuged at 1500 g for 15 min, and the clear plasma supernatant collected and stored at −70°C.
Enzyme-linked immunosorbent assays (ELISAs)
ELISAs for bovine serum albumin (BSA), interferon (IFN)-γ, interleukin (IL)-12, LPS-binding protein (LBP), serum amyloid A (SAA), transforming growth factor (TGF)-α, and TGF-β1 were all performed as previously described (Bannerman et al. Reference Bannerman, Paape, Hare and Sohn2003, Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004, Reference Bannerman, Chockalingam, Paape and Hope2005, Reference Bannerman, Paape and Chockalingam2006).
Milk N-acetyl-beta-d-glucosaminidase (NAGase) activity assay
The NAGase activity assay was performed as previously described (Mattila & Sandholm, Reference Mattila and Sandholm1985) with slight modification. Isolated milk whey samples were diluted 1:2 in deionized H2O, and 10 μl of the resulting dilutions were added to the wells of a black, clear bottom 96-well plate. Forty microlitres of substrate solution (2·25 mm-4-methylumbelliferyl N-acetyl-β-d-glucosaminide, 0·25 m-sodium citrate, pH 4·6; Sigma Chemical Co., St. Louis MO, USA) were added to each well and the plate was shaken in the dark at 150 rpm for 15 min at 25°C. Immediately at the end of the incubation period, 150 μl of stop buffer (0·2 m-glycine, pH 10·8; Sigma Chemical Co.) were added to each well, and the fluorescence measured on a Synergy HT™ multi-modal plate reader (Bio-Tec Instruments, Inc., Winooski VT, USA) at an excitation wavelength of 360 nm and an emission wavelength of 460 nm. NAGase activity of the samples was calculated by extrapolation from a standard curve consisting of a series of 1:2 dilutions of 200 μm−4-methylumbelliferone (Sigma Chemical Co.) dissolved in stop buffer. To determine the amount of activity (i.e., nmoles/ml per min) the extrapolated values were multiplied by the dilution factor (i.e., 2), multiplied by a factor of 100 to calculate the activity in 1 ml of sample, and the resulting product divided by 15 to calculate the activity per min.
Statistical analysis
Repeated measures ANOVA was performed using SAS PROC MIXED® (SAS Version 9.1.3., SAS Institute, Cary NC, USA) to compare the mean responses of variables to control (time 0) values and to compare breed-dependent differences for a given variable. Milk bacterial counts, SCC, and IFN-γ concentrations were transformed to log10 values to satisfy distributional requirements of ANOVA. Correlations among repeated measurements across time within cows and between breeds were modelled using appropriate covariance structures for each parameter analysed. The Spearman rank correlation coefficient (r s) for the relationship between mean log10 SCC and mean NAGase values was calculated using GraphPad InStat® (Version 3.06, GraphPad Software, Inc., San Diego CA, USA). A P-value of <0·05 was considered significant.
Results
Recovery of Staph. aureus from experimentally-infected quarters
The right front quarters of ten Holstein and ten Jersey cows were infused with 68 cfu of Staph. aureus. One of the Holstein cows developed a mixed infection in the infused quarter, which was characterized by the recovery of more than one genus of bacteria. The colonies that grew on plated milk samples collected from this cow differed morphologically from those of the plated Staph. aureus infusate, as well as the plated samples from the other experimentally infected cows that were confirmed to have a Staph. aureus intramammary infection. Further, none of the colonies isolated from the cow with the mixed infection demonstrated the characteristic haemolysis of Staph. aureus. Therefore, because an intramammary infection other than Staph. aureus developed, the cow was dropped from the study. For the cows that remained in the study, Staph. aureus was recovered from six of nine and seven of ten infused Holstein and Jersey quarters, respectively, within 6 h of infusion. From 18 to 120 h after infusion, Staph. aureus was recovered from 100% of the infused quarters of both breeds. Milk samples obtained from one Holstein and one Jersey cow at both 168 h and 240 h after infusion were free of Staph. aureus. Thus, only one cow of each breed was able to successfully clear Staph. aureus during the experimental period. Maximal mean (±se) milk concentrations of Staph. aureus were detected 18 h after infusion in both Holstein (4·97±0·23 log10 cfu/ml) and Jersey (5·14±0·27 log10 cfu/ml) cows, and overall concentrations in the infected quarters were equivalent (P=0·3741) between breeds throughout the study (Fig. 1).
Recovery of Staph. aureus from experimentally infected quarters. Sixty-eight colony forming units (cfu) of Staph. aureus were infused into one mammary quarter of 9 Holstein and 10 Jersey cows. Milk samples, which were aseptically collected immediately prior to (time 0) and at various times after infusion, were spread on blood agar plates. Following overnight incubation of the plates, bacterial colonies were enumerated. In those quarters where Staph. aureus were recovered, the mean (±se) log10 milk bacterial concentration is reported in cfu/ml.
Holstein and Jersey cow milk production following Staph. aureus intramammary infection
Daily milk production, defined as the sum of the morning and evening milk weights, was recorded prior to and for several days following intramammary infusion of Staph. aureus (Fig. 2). Mean (±se) pre-infection (day −1) milk production was higher (P=0·0007) in Holstein cows (26·56±1·42 kg) than in Jersey cows (19·87±0·84 kg), as was overall milk production throughout the experimental period (P=0·0003). Relative to pre-infection amounts, milk production was significantly lower in both breeds for the first 2 d after infection. Within 3 d of infection, the milk production of both breeds returned to levels comparable to those observed prior to infection. By day 10 and 11, respectively, decreased milk production was once again detected in the Jersey (17·69±0·62 kg; P=0·0005) and Holstein (22·78±1·06 kg; P=0·0150) cows relative to pre-infection (day −1) production levels.
Changes in milk production following Staph. aureus intramammary infection. Total daily milk weights were recorded prior to (day −1) and following (days 1–11) experimental Staph. aureus intramammary infection of Holstein and Jersey cows. Mean (±se) daily milk production is reported in kg.
*,#Decreased (P<0·05) compared with pre-infection (day −1) production amounts of Holstein or Jersey cows, respectively.
@Differences (P<0·05) in milk production between breeds at a given time point.
Systemic responses of Holstein and Jersey cows to Staph. aureus intramammary infection
To determine whether there were breed-dependent differences in the systemic response to Staph. aureus intramammary infection, changes in body temperature, acute phase protein synthesis and differential WBC counts were evaluated. Fever is generally defined as a >1–1·5°C increase in body temperature (Cotran et al. Reference Cotran, Kumar and Collins1999; Ryan & Levy, Reference Ryan and Levy2003), and in cattle, has been characteristically defined as body temperatures that exceed 39·5°C (Drillich et al. Reference Drillich, Reichert, Mahlstedt and Heuwieser2006). Throughout this study, the mean body temperatures of either breed did not fluctuate by ⩾0·6°C from basal (pre-infection) body temperatures nor did they exceed 39°C (data not shown). Thus, Staph. aureus failed to elicit a febrile response in either Holstein or Jersey cows.
In contrast to the absence of an effect on body temperature, intramammary infection with Staph. aureus did evoke the synthesis of the acute phase proteins, SAA and LBP. From 30 to 120 h after infection, increased circulating blood concentrations of SAA were detected in Holstein and Jersey cows (Fig. 3A). In both breeds, maximal increases in SAA were observed 60 h after infection when mean (±se) concentrations reached 265±15 and 225±30 μg/ml, respectively, in Holstein and Jersey cows. The overall SAA response, as well as the maximal concentrations detected, were comparable between the two breeds (P=0·5192). Similarly to SAA, increases in circulating LBP concentrations were detected within 30 h of infection (Fig. 3B). Maximal mean (±se) increases in LBP were detected in both Holstein (119±22 μg/ml) and Jersey (88±16 μg/ml) cows within 6 h of the initial increase. Both the maximal concentrations (P=0·2537) and the magnitude of the overall LBP response (P=0·2072) were comparable between the two breeds. Temporal differences existed, however, as LBP concentrations remained elevated, relative to pre-infection (time 0) levels, for 168 h longer in the Holstein cows.
Induction of acute phase protein synthesis following intramammary Staph. aureus infection. Blood samples were collected immediately before (time 0) and at various time points after intramammary infusion of Staph. aureus into Holstein and Jersey cows. Plasma derived from the blood samples was assayed for serum amyloid A (SAA; A) and lipopolysaccharide-binding protein (LBP; B). Mean (±se) concentrations of SAA and LBP are reported in μg/ml.
*,#Increased (P<0·05) compared with pre-infection (time 0) measurements in Holstein or Jersey cows, respectively.
Changes in the total and differential circulating WBC counts were evaluated as another aspect of the systemic response to Staph. aureus intramammary infection. For both breeds, a transient leucocytosis was observed within 30 h of infection (Fig. 4A). Relative to pre-infection (time 0) counts, the duration of leucocytosis was greater in Holstein cows, and the overall magnitude of the response observed in these cows approached a level (P=0·0921) that significantly differed from that of Jersey cows. Analysis of differential WBC counts revealed a sustained increase in circulating neutrophils (Fig. 4B) and highly variable fluctuations in circulating lymphocytes (Fig. 4C) following infection. Although there were no breed-dependent difference in pre-infection neutrophil counts, the magnitude of the overall neutrophil response that accompanied infection was higher in Holstein than in Jersey cows (P=0·016). In contrast, there were no breed-dependent differences in the overall lymphocyte response (P=0·881).
Alterations in circulating white blood cell (WBC) counts following intramammary Staph. aureus infection. Blood samples were collected immediately before (time 0) and at various time points following intramammary infusion of Staph. aureus into Holstein and Jersey cows. Blood was analysed for total (A) and differential WBC counts (B, C). Mean (±se) cell counts are reported in thousands/μl.
*,#Significantly different (P<0·05) than pre-infection (time 0) cell counts of Holstein or Jersey cows, respectively.
@Differences (P<0·05) in cell counts between breeds at time points where cell counts differed from those at time 0 in one or both breeds.
Localized inflammatory responses of Holstein and Jersey cows to Staph. aureus intramammary infection
As a local indicator of inflammation, milk SCC were enumerated prior to and after infection (Fig. 5). Pre-infection (time 0) milk SCC were equivalent (P=0·4877) between Holstein (11 333±1869 cells/ml) and Jersey cows (14 250±2923 cells/ml). In both breeds, transient increases in milk SCC were observed 6 h after infection and sustained increases detected from 18 h after infection until the end of the study. Maximal mean (±se) SCC were detected in both the Holstein (28·23×106±5·43×106 cells/ml) and Jersey cows (34·06×106±6·33×106 cells/ml) at 42 h after infection and did not statistically differ (P=0·5813). The overall SCC responses of the two breeds were similar (P=0·2116) although SCC in Holstein cows exceeded those in Jersey cows at 96 h and 240 h after infection.
Changes in milk somatic cell counts (SCC) following intramammary Staph. aureus infection. SCC were enumerated in milk samples collected immediately before (time 0) and at various time points following intramammary infusion of Staph. aureus into Holstein and Jersey cows. Mean (±se) milk SCC are reported in millions of cells/ml (A) and log10 cells/ml (B).
*,#Increased (P<0·05) compared with pre-infection (time 0) concentrations in Holstein or Jersey cows, respectively.
@Differences (P<0·05) in SCC between breeds at time points where SCC differed from those at time 0 in one or both breeds.
Two other indicators of local inflammation, increases in mammary vascular permeability and NAGase activity, were also evaluated. Initial increases in mammary vascular permeability, as evidenced by increased milk concentrations of BSA, were detected in both breeds at 30 h after Staph. aureus intramammary infection (Fig. 6A). Similarly, initial increases in NAGase activity over pre-infection (time 0) levels were detected in both breeds at 30 h after infection (Fig. 6B). Both parameters of inflammation remained elevated, relative to pre-infection levels, for longer durations in Jersey cows. The overall change in milk concentrations of BSA in response to infection were comparable (P=0·5830) between the two breeds, whereas, overall NAGase activity was higher (P=0·0082) in Jersey cows than in Holstein cows. A significant correlation between SCC and NAGase was identified in both Jersey (r=0·8821; P<0·0001) and Holstein (r s=0·9642; P<0·0001) cows consistent with reports that NAGase is an indicator of cell injury, and that high SCC are associated with mammary tissue injury (Kitchen et al. Reference Kitchen, Middleton and Salmon1978, Reference Kitchen, Middleton, Durward, Andrews and Salmon1980; Capuco et al. Reference Capuco, Paape and Nickerson1986).
Effect of Staph. aureus intramammary infection on milk bovine serum albumin (BSA) concentrations and N-acetyl-beta-d-glucosaminidase (NAGase) activity. Milk samples were collected immediately before (time 0) and at various time points after experimental Staph. aureus intramammary infection of Holstein and Jersey cows. Milk whey was assayed for BSA (A) and NAGase activity (B). Mean (±se) BSA concentrations are indicated in μg/ml and mean (±se) NAGase activity reported in nmol/ml/min.
*,#Increased (P<0·05) compared with pre-infection (time 0) concentrations in Holstein or Jersey cows, respectively.
Cytokine response of Holstein and Jersey cows to Staph. aureus intramammary infection
To evaluate breed-dependent differences in the cytokine response to Staph. aureus intramammary infection, cytokines known to be up-regulated in response to this pathogen (Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004, Reference Bannerman, Paape and Chockalingam2006) were evaluated in milk samples collected prior to and following infection. Relative to pre-infection (time 0) concentrations, initial increases in the type 1 helper T cell (Th1-Type) cytokines, IFN-γ and IL-12, were detected at 30 h after infection in both breeds (Fig. 7A and B). Maximal mean (±se) concentrations of IFN-γ were detected in Jersey (961±260 pg/ml) and Holstein (1830±903 pg/ml) cows at 42 h after infection and were higher in the latter (P=0·0165). In contrast, the maximal concentrations of IL-12, which were detected 30 h after infection, were comparable between the two breeds (P=0·5081). The duration of the increase of each cytokine was similar between breeds as was the overall magnitude of the IFN-γ (P=0·8585) and IL-12 (P=0·3369) response.
Cytokine response following intramammary Staph. aureus infection. Milk samples were collected immediately before (time 0) and at various time points following intramammary infusion of Staph. aureus into Holstein and Jersey cows. Milk whey was assayed for interferon (IFN)-γ (A), interleukin (IL)-12 (B), transforming growth factor (TGF)-α (C), and TGF-β1 (D). Mean (±se) concentrations of IFN-γ and TGF-α are expressed in pg/ml; mean (±se) concentrations of IL-12 are expressed in units (U)/ml; and mean (±se) TGF-β1 concentrations are reported in ng/ml.
*,#Increased (P<0·05) compared with pre-infection (time 0) concentrations in Holstein or Jersey cows, respectively.
@Differences (P<0·05) in concentrations between breeds at time points where concentrations differed from those at time 0 in one or both breeds.
Within 6 h of infection, transient increases in milk concentrations of TGF-α and TGF-β1 were detected in Jersey, but not in Holstein cows (Fig. 7C and D). Initial sustained increases in TGF-α were observed in both breeds at 30 h after infection and increases were detected as late as 60 h post-infection. Maximal mean (±se) concentrations were detected at 36 h post-infection and were comparable between Holstein (130±27 pg/ml) and Jersey (149±25 pg/ml) cows. Similarly to TGF-α, elevated TGF-β1 concentrations were detected in both breeds as late as 60 h after infection and the maximum concentrations detected were comparable. There was no overall difference in the magnitude of the TGF-α (P=0·1850) or TGF-β1 (P=0·1779) response between the two breeds.
Discussion
The innate immune responses of Holstein and Jersey cows during Staph. aureus-induced mastitis were evaluated in the current study. In contrast to other parameters (e.g, milk production, milk composition, longevity, conception rates, etc.), there are few published reports comparing breed-dependent immune responses to intramammary infection in a controlled manner. A previous study investigated the response of Holstein and Jersey cows to Escherichia coli intramammary infection following immunization, however, because only three Jersey cows were included in that study, breed-dependent differences were not examined (Todhunter et al. Reference Todhunter, Smith, Hogan and Nelson1991). Another study investigated the innate immune response of a group of five Holstein and five Jersey cows during experimentally induced Streptococcus uberis mastitis (Rambeaud et al. Reference Rambeaud, Almeida, Pighetti and Oliver2003). The main objective of that study did not focus on investigating breed-dependent differences and the investigators presented combined data for all cows. In a subanalysis of the data, the investigators reported no significant differences between the two breeds in the innate immune parameters evaluated. Several factors were not controlled in that study, however, including: (1) the bacterial inoculum preparation used to infect the two breeds differed by ~60%; (2) each breed was housed on a different farm and subjected to different management practices; and (3) the breeds were infected, sampled, and monitored at different times of the year. In contrast, the current study was designed to control for these factors by housing all cows in a single barn and subjecting them to the same management practices, infecting the same quarter on each cow with the same preparation and amount of bacterial challenge inoculum, and infecting and sampling all cows in parallel. A larger sample size was used in the current study and several additional innate immune responses were assessed, including: induction of acute phase protein synthesis; changes in total and circulating WBC; and induction of the cytokines IFN-γ, IL-12, TGF-α, and TGF-β1. Because innate immune responses to intramammary infection can be influenced by parity (Gilbert et al. Reference Gilbert, Grohn, Miller and Hoffman1993; Van Werven et al. Reference Van Werven, Noordhuizen-Stassen, Daemen, Schukken, Brand and Burvenich1997; Mehrzad et al. Reference Mehrzad, Duchateau, Pyörälä and Burvenich2002) and stage of lactation (Lescourret & Coulon, Reference Lescourret and Coulon1994; Shuster et al. Reference Shuster, Lee and Kehrli Jr1996; Vangroenweghe et al. Reference Vangroenweghe, Lamote and Burvenich2005), only cows of the same parity and in similar stages of lactation were evaluated.
With the exception of changes in circulating neutrophils and NAGase activity, the overall magnitude of all other innate immune responses was comparable between the two breeds. Although Holstein cows demonstrated an increased number of circulating neutrophils after infection relative to Jersey cows, the overall milk SCC response was equivalent between the breeds. Because neutrophils compose >90% of milk somatic cells during acute mastitis (Saad & Ostensson, Reference Saad and Ostensson1990) it is interesting that the elevated number of circulating neutrophils did not correspond with an overall SCC response that was higher in Holstein cows. One may speculate that rate-limiting steps, such as adhesion molecule expression and transendothelial migration, limit the overall number of neutrophils that can be recruited to the mammary gland. Alternatively, cytokines and lipid mediators of inflammation with chemoattractant properties may have been produced in equivalent amounts in the two breeds, similarly to the other cytokines that were measured. The increased circulating numbers of neutrophils did not appear to confer an advantage to Holstein cows as the overall number of quarters infected and the bacterial concentrations within these quarters were comparable between breeds.
In contrast to cytokines such as TNF-α and IL-8, intramammary infection with Staph. aureus has been shown to induce the production of IL-12 and IFN-γ (Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004). These cytokines contribute to the immunological response of the host by activating neutrophils and macrophages and promoting a Th1-type immune response (Trinchieri, Reference Trinchieri1997). The finding that both IFN-γ and IL-12 increased at the same time (i.e., 30 h after infection) in both breeds is consistent with reports that these cytokines positively regulate the production of each other (Collins et al. Reference Collins, Camon, Chaplin and Howard1998; Munder et al. Reference Munder, Mallo, Eichmann and Modolell1998; Ma, Reference Ma2001). The overall concentrations of these cytokines in the milk of the two breeds after Staph. aureus intramammary infection were comparable with those reported in a previous study (Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004). Similarly to IFN-γ and IL-12, the concentrations of TGF-α and TGF-β1 have been reported to increase in milk following Staph. aureus intramammary infection. TGF-α and TGF-β1 are pleiotropic cytokines that are structurally and functionally distinct from one another. TGF-α has pro-inflammatory properties that are conferred by its ability to up-regulate the production of prostaglandins and synergistically enhance the effects of other pro-inflammatory cytokines (Bry, Reference Bry1993; Unemori et al. Reference Unemori, Ehsani, Wang, Lee, McGuire and Amento1994; Subauste & Proud, Reference Subauste and Proud2001). In contrast, TGF-β1 suppresses inflammation in a variety of disease states (Johns et al. Reference Johns, Flanders, Ranges and Sriram1991; Santambrogio et al. Reference Santambrogio, Hochwald, Saxena, Leu, Martz, Carlino, Ruddle, Palladino, Gold and Thorbecke1993; Neurath et al. Reference Neurath, Fuss, Kelsall, Presky, Waegell and Strober1996; Powrie et al. Reference Powrie, Carlino, Leach, Mauze and Coffman1996; Williams et al. Reference Williams, Humphreys, Cornere, Edwards, Rae and Hussell2005) and it is believed that its anti-inflammatory properties are partly governed by its ability to down-regulate the pro-inflammatory responses of macrophages and other cell types (Ayoub & Yang, Reference Ayoub and Yang1997; Letterio & Roberts, Reference Letterio and Roberts1998; Ashcroft, Reference Ashcroft1999). Both TGF-α and TGF-β1 are expressed in the healthy mammary gland and the relative increases in the expression of these cytokines following Staph. aureus intramammary infection are comparable with a previous report (Bannerman et al. Reference Bannerman, Paape and Chockalingam2006). Although it was beyond the scope of the this study to perform an exhaustive evaluation of all cytokines known to influence the inflammatory response, the finding that the overall production of these four cytokines were equivalent across the two breeds precludes that their differential induction is responsible for the previously reported differences in mastitis susceptibility between Holstein and Jersey cows.
In addition to the evaluation of cytokine production, two markers of the systemic acute phase response, SAA and LBP, and two markers of localized inflammation, BSA and NAGase, were evaluated. Similarly to the cytokine responses, overall changes in the circulating concentrations of the acute phase proteins were comparable across breeds. The temporal expression of LBP from 30–72 h after Staph. aureus intrammary infection is consistent with a previous report (Bannerman et al. Reference Bannerman, Paape, Lee, Zhao, Hope and Rainard2004). Further, the concentrations of circulating SAA induced by Staph. aureus were comparable with other studies (Gronlund et al. Reference Gronlund, Hulten, Eckersall, Hogarth and Persson Waller2003; Eckersall et al. Reference Eckersall, Young, Nolan, Knight, McComb, Waterston, Hogarth, Scott and Fitzpatrick2006). The temporal change in milk BSA concentrations, as well as the magnitude of the change, were consistent with a previous report (Bannerman et al. Reference Bannerman, Paape and Chockalingam2006). Similarly to LBP and SAA, overall changes in BSA were comparable across breeds. In contrast, milk NAGase activity was significantly higher in the infected quarters of Jersey cows relative to those of Holstein cows. Increased NAGase activity is an indicator of mammary tissue injury and elevated SCC are associated with mammary tissue injury (Bogin & Ziv, Reference Bogin and Ziv1973; Kitchen et al. Reference Kitchen, Middleton and Salmon1978, Reference Kitchen, Middleton, Durward, Andrews and Salmon1980, Reference Kitchen, Kwee, Middleton and Andrews1984; Mattila et al. Reference Mattila, Pyörälä and Sandholm1986; Chagunda et al. Reference Chagunda, Larsen, Bjerring and Ingvartsen2006). Correspondingly, a significant correlation between SCC and NAGase was identified in both breeds. Interestingly, however, there were no overall differences in the SCC response between the two breeds. Thus, if the higher level of NAGase activity in the milk of Jersey cows is truly representative of greater mammary tissue injury, then factors other than elevations in SCC must be operative in the induction of this injury.
The risk of clinical mastitis is positively correlated with milk production (Oltenacu & Ekesbo, Reference Oltenacu and Ekesbo1994; Waage et al. Reference Waage, Sviland and Odegaard1998; Fleischer et al. Reference Fleischer, Metzner, Beyerbach, Hoedemaker and Klee2001). Despite the fact that milk production was higher in Holstein than in Jersey cows, all cows developed an intramammary infection with Staph. aureus that was sustained for several days. This finding is consistent with a previous study reporting that the severity of the inflammatory response to intramammary infection is not affected by differences in milk production within a given breed (Kornalijnslijper et al. Reference Kornalijnslijper, Beerda, Daemen, van der Werf, van Werven, Niewold, Rutten and Noordhuizen-Stassen2003). Thus, the correlation between risk of clinical mastitis and milk production may be related to differences in susceptibility to initial infection rather than differences in the response to infection.
To our knowledge, the present report is the first controlled study to compare the innate immune responses of Holstein and Jersey cows to Staph. aureus intramammary infection. Overall, the magnitude, temporal onset, and duration of these responses were highly conserved between the two breeds. Because innate immunity encompasses evolutionarily ancient and highly conserved host defence mechanisms (Hoffmann et al. Reference Hoffmann, Kafatos, Janeway and Ezekowitz1999; Janeway & Medzhitov, Reference Janeway Jr and Medzhitov2002) perhaps it is not surprising that Holstein and Jersey cow innate immune responses to intramammary infection are similar despite the genetic and phenotypic differences that exist between the two breeds.
The authors would like to acknowledge Jennifer Bilheimer, Mary Bowman, Amanda Hall, Derrel Hoy and Duane Zimmerman for their technical assistance. The authors would also like to acknowledge Creig Caruth, Jerri Grove, Andy Mosier and Norman Tjelmeland for their care and handling of the animals in this study. Funding for this research was supported, in part, by a grant from the American Jersey Cattle Association.
