INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Mastitis is an inflammation of the mammary gland which, in most cases, is caused by infection with contagious pathogens, such as Staphylococcus aureus, or environmental pathogens, such as Streptococcus uberis and coliform bacteria. S. uberis is responsible for 12% (7) to 14% (29) of clinical mastitis in lactating cows. Postmilking teat disinfection and antibiotic therapy, successful in reducing the incidence of clinical mastitis caused by contagious pathogens, are relatively ineffective against S. uberis infection (22, 23). In cattle, polymorphonuclear neutrophils (PMN) play an important role in defense against Escherichia coli (20) and S. aureus (42) mastitis. The influx of PMN into the mammary gland following challenge with E. coli was shown to be more rapid in cows that developed only moderate cases of mastitis than in cows that developed severe cases (11). This result suggested that the timely influx of PMN into the mammary gland was responsible for the decrease in the severity of infection.

Numerous studies have been conducted to determine the dynamics of PMN migration (diapedesis) across the epithelial lining into the infected lumen of various organs in several species (1, 2, 26, 34). Because of the complexity of these organ systems, monolayers of epithelial cells and isolated PMN have been used to more closely determine the factors affecting PMN diapedesis. In vitro studies with epithelial cell culture monolayers showed that ₂-integrins (such as CD11b/CD18) on the surface of PMN bind to intercellular adherence molecule 1 (ICAM-1) on epithelial cells to effect PMN diapedesis (1, 3, 6, 26, 34, 36). Viral (47) and bacterial (3, 34) infections of human epithelial cell cultures enhanced epithelial cell ICAM-1 expression to induce ICAM-1- and CD11b/CD18-dependent transepithelial neutrophil migration. The inability of PMN to undergo diapedesis in calves with bovine leukocyte adhesion deficiency (17) has been attributed to a deficiency in CD18 (30, 31). Also, treatment of PMN from normal calves with monoclonal antibodies to CD18 decreased PMN migration to the same level as that of PMN from animals with bovine leukocyte adhesion deficiency (32).

The purpose of the present study was to investigate the expression of CD11b/CD18 adhesion receptors and diapedesis by PMN before and after experimentally induced S. uberis mastitis in cows.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Cows. Five midlactation cows of the East Flemish Red Pied breed were used. The results of bacteriological examination of all quarters were negative, and the somatic cell count (SCC) was below 250,000 cells/ml. Cows were permitted to adjust to the housing facilities and were given a daily ration of 8 kg of concentrates and free access to hay and water.

Bacterial suspension and experimental infection. S. uberis O140J (J. Leigh, Compton, United Kingdom) was maintained in lyophilization medium at 20°C. For experimental use, the organisms were cultured in Todd-Hewitt broth (LAB M, Amersham, United Kingdom) at 37°C for 18 h, washed, resuspended, and diluted in phosphate-buffered saline (PBS). At 1 h after the morning milking, the teats were aseptically prepared and both quarters of the left half of the udder were inoculated with a suspension containing approximately 500 CFU of S. uberis by use of a sterile teat cannula. Following inoculation, each gland was massaged for 30 s to distribute the organisms.

Clinical signs. Clinical measurements and observations, i.e., rectal temperature, heart rate, and pain in and swelling of the mammary gland, were carried out as described earlier (50).

Bacterial counts, blood leukocyte counts, and milk SCCs. S. uberis bacteria were counted by the plate count method. Leukocytes in whole blood were counted with a Coulter Counter (model ZF; Coulter Electronics Ltd., Luton, England). Smears were prepared from whole blood and stained with Hemacolor (Merck Diagnostics, Darmstadt, Germany). Differential microscopic counts were determined by counting 100 cells. The SCC of milk was measured with a Fossomatic cell counter (Foss Electric, Hillerod, Denmark).

Adhesion receptors. Blood was collected aseptically in a 10-ml syringe containing 4 ml of Alsever's solution (0.42% NaCl, 0.8% trisodium citrate, 2.05% glucose; adjusted to pH 6.1 with 10% citric acid solution; GIBCO BRL, Life Technologies, Grand Island, N.Y.). Blood samples were incubated for 15 min at room temperature with 50 µl of RPMI 1640 medium (Life Technologies) (control) or 50 µl of monoclonal antibodies to bovine CD11a (IL-A99), bovine CD11b (IL-A15), and bovine CD18 (MF14B4). After incubation, erythrocytes were lysed by the addition of 300 µl of an ice-cold sterile buffer solution (2.6 g of Tris, 100 ml of water, 7.4 g of NH₄Cl/900 ml of water) for 6 min. After centrifugation (200 × g, 10 min, 4°C), leukocytes were washed twice in RPMI 1640 medium. A second incubation was done on ice and in the dark for 30 min with 50 µl of secondary antibody solution (goat anti-mouse immunoglobulin G conjugated to fluorescein isothiocyanate [FITC] [Sigma Chemical Co., Bornem, Belgium] and diluted 1/500 with RPMI 1640 medium). The cells were collected by centrifugation (200 × g, 10 min, 4°C) and washed twice with PBS. The cell pellet was fixed in 0.5 ml of 1% paraformaldehyde in PBS (pH 7.5). Samples were stored at 4°C in the dark until assayed on the flow cytometer as described below.

Flow cytometric analysis. Fluorescence was measured with a FACScan flow cytometer (Becton Dickinson Immunocytometry Systems, San Jose, Calif.). The excitation wavelength was 488 nm, and emitted fluorescence was measured at between 530 and 560 nm. Dot plots were gated for polymorphonuclear leukocytes. The cells were assayed for size by forward scattering and for granularity by side scattering. The expression of cell adhesion molecules, i.e., mean fluorescence intensity (MFI), was calculated after plotting of the fluorescence of the histograms. Fluorescence associated with PMN incubated with the FITC-goat anti-mouse immunoglobulin G secondary antibody in the absence of anti-CD18, anti-CD11a, and anti-CD11b was considered the control for nonspecific fluorescence. Data were expressed as relative fluorescence intensity units after subtraction of nonspecific (control) fluorescence.

Determination of cortisol. Plasma cortisol levels were measured by a commercially available dissociation-enhanced lanthanide fluoroimmunoassay (Delfia cortisol kit; Wallac Oy, Turku, Finland).

Cell culture system. Epithelial cell monolayers were used to measure diapedesis of PMN as described by Smits et al. (45) with minor modifications. Briefly, membrane inserts (12-mm diameter, 12-µm pore size; Millicell-PCF; Millipore, Bedford, Mass.) were coated with collagen by immersion in wells of 24-well plates containing 1% calfskin collagen (type I) solution (Sigma) and incubated at 4°C for 48 h. The collagen-coated inserts were inverted in six-well plates containing culture medium. Bovine secretory epithelial cells (3 × 10⁵ per insert) were added, and the inserts were incubated at 37°C in 95% air-5% CO₂ for 7 days. The culture medium was changed every 2 days. To test for confluence of the epithelial cell monolayer, transepithelial electrical resistance was measured for each filter with an ohmmeter (Millipore). To ensure confluence, only monolayers with resistances of >1,000 /cm² were used.

Diapedesis assay. The PMN diapedesis assay was previously described (45). Bovine PMN, isolated as previously described (12), were suspended in Hanks balanced salt solution-0.1% bovine serum albumin at a concentration of 10⁷/ml. The epithelial cell monolayers were rinsed extensively in Hanks balanced salt solution to remove media and residual serum components. PMN were added to the upper chamber (basal), and 50 ng of purified bovine complement factor C5a-des-Arg (courtesy of P. Rainard, Institut National de la Recherche Agronomique, Nouzilly, France) per ml was added to the lower chamber (apical); the inserts were incubated for 5 h at 37°C in 95% air-5% CO₂. At 5 h, 50 µl of the apical medium was removed, and the PMN undergoing diapedesis were counted in duplicate with a hemocytometer. Blood PMN from each cow were assayed in duplicate, and the results were averaged.

Statistical analysis. Means, standard deviations, and standard errors of the mean were computed with a statistical analysis program (Statistics 4.0; NH Analytical Software, Tallahassee, Fla.). A mixed linear model was used to determine the effect of infection with S. uberis during a longitudinal study. The time relative to challenge with S. uberis was used as the fixed variable. The individual cows were treated as the random variable. The general model for analysis of variance (46) was Y = µ + T + C + Int + e, where Y is the leukocyte counts in blood, milk SCC, MFI for adhesion receptors, or percentage of PMN undergoing diapedesis; µ is the overall mean; T is time; C is cow; Int is interaction term (T × C); and e is experimental error term. Statistical testing of the observed differences in time relative to infection was done with general contrasts. Statistical significance was determined at P values of <0.05, <0.01, and <0.001.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Clinical symptoms and bacterial counts. Intramammary inoculation with S. uberis in both quarters of the left half of the udders of five midlactation cows resulted in clinical mastitis of all challenged quarters. Local symptoms, peaking between 24 and 60 h after challenge, were accompanied by moderate systemic disturbances, including fever, tachycardia, and mild depression. The maximum rectal temperature was 40.1 ± 0.4°C. Milk production was decreased at 24 h after challenge in all cows. Significant changes in milk composition were observed in all cows. A decrease in lactose and potassium levels and an increase in sodium, chloride, and serum albumin levels were observed, indicating that the blood-milk barrier was damaged by S. uberis.

Leukocyte counts in peripheral blood and SCCs in milk. Figure 1 shows the decrease in blood leukocyte counts and the increase in SCCs in the infected quarters of cows after inoculation with S. uberis. Baseline blood leukocyte counts were 8.0 × 10⁶ ± 0.6 × 10⁶/ml. The number of blood leukocytes was decreased significantly (P < 0.01) at 48 h after challenge. Minimum blood leukocyte counts were 5.1 × 10⁶ ± 0.7 × 10⁶/ml, and peak SCCs were 6.8 × 10⁶ ± 1.2 × 10⁶/ml.

View larger version (18K): [in this window] [in a new window]   FIG. 1.   Counts of leukocytes in blood () and milk SCCs () in the challenged quarters before and during experimentally induced S. uberis mastitis. Data are expressed as the mean ± standard error of the mean for five cows. Comparison with prechallenged levels: *, P < 0.05, **, P < 0.01, and ***, P < 0.001.

CD11a and CD11b adhesion receptors on polymorphonuclear cells. The expression of CD11a and CD11b, two  chains of the CD11/CD18 ₂-integrins, before and during experimentally induced S. uberis mastitis is shown in Fig. 2. The expression of CD11a was decreased at 36 and 48 h after S. uberis challenge, but CD11a has been shown to have no effect on PMN transepithelial migration (35). The MFI for the CD11b adhesion receptor was upregulated at 24 h after S. uberis inoculation and returned to the baseline after 60 h.

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Expression of CD11a and CD11b on polymorphonuclear leukocytes before and during experimentally induced S. uberis mastitis. Data are expressed as the mean ± standard error of the mean for five cows. Comparison with prechallenge levels: *, P < 0.05.

CD18 adhesion receptors on polymorphonuclear cells. Figure 3 illustrates the time course of MFI for CD18, the common  chain of the CD11/CD18 ₂-integrins. At 24 h after S. uberis challenge, the MFI was increased 18% (P < 0.001). The MFI was decreased at 36 and 48 h postchallenge and increased to 12% (P < 0.01) above prechallenge values at 60 h. The MFI returned to the baseline at 72 h.

View larger version (16K): [in this window] [in a new window]   FIG. 3.   Expression of CD18 on PMN () and plasma cortisol concentration (concn) () before and during experimentally induced S. uberis mastitis. Data are expressed as the mean ± standard error of the mean for five cows. Comparison with prechallenge levels: **, P < 0.01, and ***, P < 0.001.

Cortisol concentration. Prior to inoculation with S. uberis, the mean cortisol concentration was 18.6 ± 2.02 nM. The cortisol concentration was increased (P < 0.001) to 46.11 ± 11.59 nM at 24 h after infection and returned to normal levels at 36 h after challenge (Fig. 3).

Diapedesis of polymorphonuclear cells. The average percentage of PMN undergoing diapedesis during S. uberis mastitis is shown in Fig. 4. No significant day-to-day variation was observed. Before inoculation of S. uberis, complement component C5a (50 ng/ml) induced the diapedesis of blood PMN by an average of 16.3% ± 2.8%. Although there was some variation in the onset of the inflammatory response, PMN diapedesis in all cows was synchronously decreased 80% at 24 h after challenge (P < 0.001). PMN diapedesis returned to the baseline at 504 h postinoculation. The period of diminished PMN diapedesis was not significantly correlated with the appearance of immature neutrophils (Fig. 5).

View larger version (15K): [in this window] [in a new window]   FIG. 4.   Percentage of PMN undergoing diapedesis before and during experimentally induced S. uberis mastitis. Data are expressed as the mean ± standard error of the mean for five cows. Comparison with prechallenge levels: *, P < 0.05, **, P < 0.01, and ***, P < 0.001.

View larger version (15K): [in this window] [in a new window]   FIG. 5.   Differential counts of circulating segmented () and band () PMN before and during experimentally induced S. uberis mastitis. Data are expressed as the mean ± standard error of the mean for five cows.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Experimental infection of the bovine mammary gland with virulent strains of S. uberis has been shown to result in an accumulation of large numbers of polymorphonuclear leukocytes in the secretory acini (49). In this study, an inflammatory reaction began approximately 24 h after challenge and was characterized by a breakdown in the blood-milk permeability barrier and an increase in milk SCCs.

After challenge with S. uberis, biphasic upregulation of adhesion receptor CD18 was observed. There was a first peak at 24 h postchallenge and a second peak at 60 h. Upregulation of CD11b/CD18 expression on peripheral blood PMN at 24 h postchallenge was coincident with the onset of mammary gland inflammation. The upregulation of adhesion receptors may have been caused by several mediators, i.e., tumor necrosis factor alpha and interleukins, which are formed during inflammation (37). Shuster et al. (44) reported no upregulation of CD18 until 12 h after challenge with E. coli, indicating a delay in neutrophil recruitment from an initial lack of bacterial recognition and inflammatory mediator production. At 24 h after S. uberis challenge, a release of cortisol occurred during the fever stage, as was observed in other in vivo mastitis models (28). The transient downregulation of the expression of CD18 at 36 and 48 h postchallenge might be explained by this endogenous release of cortisol at 24 h postchallenge. It was recently shown (15) that the glucocorticoid receptor attenuates the upregulation of CD11/CD18 adhesion molecules on human neutrophils in response to activation. Also, several studies on bovine neutrophils (9, 10, 18) revealed that one of the mechanisms of the action of glucocorticoids is to induce the downregulation of CD18 adhesion molecules. These observations are comparable with earlier work of Roth and Kaeberle (38) and Paape et al. (33), in which pharmacological doses of glucocorticoids reduced the phagocytosis of PMN, diminished PMN adhesiveness, and decreased PMN diapedesis to the inflammatory site.

Although upregulation of CD11b/CD18 adhesion receptors was observed, the sudden and marked decrease in the rate of PMN diapedesis during S. uberis infection agreed with reports on the migration of PMN during E. coli mastitis in cows (25) and sows (27). The decreased expression of other PMN adhesion receptors may be partly responsible for the reduced PMN exudation to extravascular sites seen in patients with systemic inflammatory response syndrome (4). An impaired PMN function has also been observed during viral infections in humans (13) and in cattle (8).

Kremer et al. (25) postulated that the increase in the number of circulating immature neutrophils in severely diseased cows partially accounted for the lower chemotactic response of circulating PMN. Chemotaxis, unlike the expression of adhesion receptors, is a manifestation of functional differentiation which occurs primarily in segmented cells (19). Segmented neutrophils have twice the chemotactic activity of band neutrophils. Immature neutrophils fail to deform normally following chemotactic stimulation (5). It has been postulated that the defect may be due to a decrease in the level of cellular ATP, which affects actin-myosin dissociation (21). During S. uberis infection, however, the lower proportion of circulating immature neutrophils was not correlated with a decrease in PMN diapedesis, indicating that other factors may be important.

In cattle persistently infected with bovine diarrhea virus (38-41), PMN function is suppressed. Bacterial infections as well (44) generally induce a nonspecific acute-phase response, which is initiated by the production and release of cytokines, i.e., tumor necrosis factor, interferons, and interleukins. Recent studies demonstrated that pretreatment of PMN with cytokines enhances adherence (16, 24) as well as phagocytosis and killing (14) but reduces the chemotactic activity of PMN (43, 48). In vitro incubation with recombinant bovine gamma interferon was shown to inhibit random migration under agarose (8). Therefore, cytokines represent an important part of the inflammatory response.

PMN in vivo continuously receive numerous regulatory signals influencing the recruitment of circulating PMN to inflammatory sites and the migration of cells across the epithelium. In this study, CD11b/CD18 adhesion receptors on blood PMN obtained after S. uberis challenge were upregulated, while diapedesis across secretory epithelial cells was depressed, suggesting that factors other than CD11b/CD18 are involved in PMN diapedesis.