The bovine innate immune response during experimentally-induced Pseudomonas aeruginosa mastitis

Almost half of all clinical cases of mastitis are caused by Gram-negative bacteria. Among these bacteria, intramammary infection by Pseudomonas aeruginosa remains one of the most refractory to antibiotic therapy. The ability to recognize potentially harmful pathogens whether previously encountered or not, as well as the induction of an initial pro-inflammatory response to these pathogens, are critical components of host innate immunity. Although the innate immune response to another Gram-negative mastitis-causing pathogen, Escherichia coli, has been well-characterized, little is known about the response to other Gram-negative bacteria, including P. aeruginosa. The objective of the current study was to characterize the systemic and localized bovine innate immune response to intramammary infection with P. aeruginosa. The contralateral quarters of ten mid-lactating Holstein cows were challenged with either saline or P. aeruginosa. Following the establishment of infection, milk samples were collected and assayed for changes in cytokine and growth factor concentrations, complement activation, and changes in the levels of soluble CD14 (sCD14) and lipopolysaccharide (LPS)-binding protein (LBP), two accessory molecules involved in host recognition of Gram-negative bacteria. Initial increases in milk somatic cell counts were evident within 12h of experimental challenge and remained elevated for >or=3 weeks. Increased permeability of the mammary gland vasculature, as evidenced by elevated milk levels of BSA, was initially observed 20 h post-infection and persisted for 2 weeks. Within 32 h of challenge, increased levels of IL-8, TNF-alpha, IL-10, and IL-12 were detected, however, the elevated levels of these cytokines were not sustained for longer than a 24h period. In contrast, elevations in IL-1beta, IFN-gamma, TGF-alpha, TGF-beta1, TGF-beta2, sCD14, LBP, and activated complement factor 5 (C5a) were sustained for periods of >48 h. Systemic changes were characterized by elevated body temperature, induction of the acute phase protein synthesis of serum amyloid A and LBP, and a transient decrease in circulating neutrophils and lymphocytes. Together, these data demonstrate the capability of the mammary gland to mount a robust innate immune response to P. aeruginosa that is characterized by the induction of pro-inflammatory cytokines, complement activation, and increased levels of accessory molecules involved in Gram-negative bacterial recognition.     

Bacterial lipopolysaccharide stimulates bovine neutrophil production of TNF-alpha, IL-1beta, IL-12 and IFN-gamma

After intramammary infection, polymorphonuclear neutrophil leukocytes (PMN) are the first cells recruited into the mammary gland. Rapid recruitment of and bacterial phagocytosis and killing by PMN are the most effective defenses against establishment of bacterial infection. In addition to their phagocytic and bactericidal properties, PMN may play a key supportive role through secretion of cytokines during the innate immune response. We sought to determine whether bovine PMN produce cytokines in response to stimulation by lipopolysaccharide (LPS). To investigate the effects of LPS on the expression of cytokines secreted by bovine PMN, we measured the expression of tumor necrosis factor (TNF)-alpha, interleukin (IL)-1beta, IL-12, and interferon (IFN)-gamma by ELISA after stimulation with different concentrations of LPS, and secretion of IL-8 after co-stimulation with LPS and either TNF-alpha or IL-1beta. Bovine PMN were shown to secrete TNF-alpha , IL-1beta, IL-12, IL-8 and IFN-gamma in response to LPS. Co-incubation of PMN with LPS and TNF-alpha increased secretion of IL-8 when compared to LPS alone. It was concluded that LPS stimulation up-regulates the secretion of cytokines by bovine PMN, and that co-incubation of LPS with TNF-alpha had an additive effect on the secretion of IL-8. These data show that bovine PMN, in addition to their phagocytic and bactericidal properties, may play a supportive role in the innate immune response to infection by Gram-negative bacteria through their ability to produce immuno-regulating cytokines.     

Helenalin reduces Staphylococcus aureus infection in vitro and in vivo

Staphylococcus (S.) aureus is a major udder pathogen causing bovine mastitis. Some pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-alpha), enhance extracellular and intracellular growth of S. aureus, indicating that the inflammatory process favors S. aureus infection. Helenalin is a sesquiterpene lactone with potent anti-inflammatory properties. This study was designed to evaluate the effects of helenalin on S. aureus infection. First, in vitro experiments were conducted. These studies revealed that proliferation of S. aureus in bovine mammary epithelial MAC-T cells treated in the presence or absence of TNF-alpha was markedly reduced in the presence of helenalin. Secondly, in vivo effects of helenalin were investigated. Lactating mice treated in the presence or absence of helenalin were challenged by the intramammary route with S. aureus and the bacteria in the mammary glands were counted 12 h after infection. Significantly less numbers of bacteria were recovered from the infected glands of helenalin-treated mice compared with untreated mice. Moreover, histological examination of mammary tissue from helenalin-treated mice that were challenged with S. aureus indicated that helenalin is able to significantly reduce leukocyte infiltration in the mammary gland following S. aureus inoculation. Our results show that helenalin reduces S. aureus intracellular growth and experimental S. aureus infection. We conclude that helenalin may be of potential interest in the treatment of S. aureus-induced mastitis in the bovine species.     

Dynamics of leukocytes and cytokines during experimentally induced Streptococcus uberis mastitis

Streptococcus uberis causes a significant proportion of clinical and subclinical intramammary infections (IMI) in lactating and non-lactating dairy cows. In spite of this, its pathogenesis is incompletely understood. A study was conducted to determine leukocyte and cytokine dynamics during experimentally induced S. uberis mastitis. Five Jersey and five Holstein cows were challenged via intramammary inoculation of S. uberis into two uninfected mammary glands. Sixteen of 20 challenged mammary glands developed clinical mastitis with peak clinical signs observed at 144 h. The number of S. uberis in milk increased (P<0.05) 48 h after challenge, in spite of an increase in milk somatic cells that began at 18 h (P<0.001) and remained elevated throughout the study. Increased tumor necrosis factor-alpha (TNF-alpha), interleukin-1beta (IL-1beta) and interleukin-8 (IL-8) in milk were detected 66 h after challenge (P<0.05). Peak TNF-alpha and IL-8 concentrations occurred 120 h after challenge and preceded peak clinical signs. Experimental S. uberis IMI induced local production of TNF-alpha, IL-1beta and IL-8, which may play a role in the pathogenesis of S. uberis mastitis. Other mediators may be involved in initial leukocyte recruitment to the mammary gland, since increases in milk somatic cells occurred earlier than cytokine production.     

Morphometric analysis of proinflammatory cytokines in mammary glands of sows suggests an association between clinical mastitis and local production of IL-1beta, IL-6 and TNF-alpha

Twelve healthy primiparous sows received intramammary inoculation with Escherichia coli (serotype O127) during the 24-h period preceding parturition. Mammary gland biopsy samples were taken immediately before inoculation (0 h) and from the inoculated and the contralateral non-inoculated glands 24 h after inoculation. The analyses of interleukin-1 beta (IL-1beta), IL-6, IL-8, and tumor necrosis factor alpha (TNF-alpha) by immunohistochemistry revealed that the production of these proinflammatory cytokines significantly increased in the inoculated mammary glands of sows that developed clinical signs of mastitis (affected group, n=4) 24 h after inoculation. This was also true for IL-8 in the inoculated mammary glands of sows that did not develop clinical signs of mastitis (nonaffected group, n=8). Sows that developed clinical signs of mastitis displayed significantly lower constitutive production of IL-1beta than did sows that remained clinically healthy. The data indicate that the development of clinical symptoms of coliform mastitis in the sow is associated with a locally increased proinflammatory cytokine production in response to intramammary E. coli infection.     

Escherichia coli inoculation of porcine mammary glands affects local mRNA expression of Toll-like receptors and regulatory cytokines

The expression of mRNAs for the Toll-like receptors (TLRs) TLR2 and TLR4, pro- and anti inflammatory cytokines and their receptors was evaluated in mammary gland biopsy material collected from sows intramammarily inoculated with Escherichia coli strain O127 at parturition. Quantitative real-time RT-PCR analysis showed increased mRNA levels for TLR2, the proinflammatory cytokines interleukin IL-1beta and tumor necrosis factor-alpha TNF-alpha, and the anti-inflammatory cytokine IL-10 in the inoculated mammary glands 24h after inoculation. Increased mRNA levels of the proinflammatory cytokine IL-6 were only observed in the inoculated mammary glands of sows that developed clinical signs of mastitis. In contrast, the expression of the anti-inflammatory cytokine, transforming growth factor-beta 1 (TGF-beta1) mRNA was unaltered, as was mRNA expression for the IL-1 receptor type I (IL-1R1). Furthermore, IL-1beta and IL-10 mRNA expression was higher in the inoculated mammary glands of sows that developed clinical signs of mastitis compared with sows that remained clinically healthy. Notably, sows that developed clinical signs of mastitis had significantly lower pre-inoculation levels of IL-1beta mRNA than sows that remained clinically healthy. These findings suggest that development of coliform mastitis is associated with the level of local expression of regulatory cytokines in response to intramammary E. coli inoculation and infection.     

Innate immune response to intramammary infection with Serratia marcescens and Streptococcus uberis

Streptococcus uberis and Serratia marcescens are Gram-positive and Gram-negative bacteria, respectively, that induce clinical mastitis. Once initial host barrier systems have been breached by these pathogens, the innate immune system provides the next level of defense against these infectious agents. The innate immune response is characterized by the induction of pro-inflammatory cytokines, as well as increases in other accessory proteins that facilitate host recognition and elimination of the pathogens. The objective of the current study was to characterize the innate immune response during clinical mastitis elicited by these two important, yet less well-studied, Gram-positive and Gram-negative organisms. The pro-inflammatory cytokine response and changes in the levels of the innate immune accessory recognition proteins, soluble CD14 (sCD14) and lipopolysaccharide (LPS)-binding protein (LBP), were studied. Decreased milk output, induction of a febrile response, and increased acute phase synthesis of LBP were all characteristic of the systemic response to intramammary infection with either organism. Infection with either bacteria similarly resulted in increased milk levels of IL-1 beta, IL-8, IL-10, IL-12, IFN-gamma, TNF-alpha, sCD14, LBP, and the complement component, C5a. However, the duration of and/or maximal changes in the increased levels of these inflammatory markers were significantly different for several of the inflammatory parameters assayed. In particular, S. uberis infection was characterized by the sustained elevation of higher milk levels of IL-1 beta, IL-10, IL-12, IFN-gamma, and C5a, relative to S. marcescens infection. Together, these data demonstrate the variability of the innate immune response to two distinct mastitis pathogens.     

Intramammary infusion of beta1,3-glucan for prevention and treatment of Staphylococcus aureus mastitis

Udder health problems associated with Staphylococcus aureus infections in dairy cows are difficult to control and antibiotics have limited effects. Lately, more interest has been directed towards ways to stimulate the innate immune mechanisms of the animal for better prevention and treatment of mastitis. The objectives of this study were to investigate if intramammary infusion at drying off with the immune modulator beta1,3-glucan can make the udder more resistant to experimental intra mammary S. aureus infection at this time, and to study if intramammary infusion of beta1,3-glucan into lactating udder quarters with chronic subclinical S. aureus infection can stimulate the clearing of the infection. Another aim was to evaluate the effect of beta1,3-glucan on the expression of major histocompatibility complex (MHC class II) on mammary leucocytes, measured by flow cytometry, during these circumstances. The results indicated a slight, but not statistically significant, positive effect of beta1,3-glucan at drying off on the clinical and anti-bacterial response to S. aureus infection, but no therapeutic effect of beta1,3-glucan treatment of udder quarters with chronic subclinical S. aureus mastitis. However, the proportion of MHCII+ milk lymphocytes tended to increase after glucan infusion in those udder quarters indicating a stimulation of the antigen presenting ability. To further evaluate a possible preventive effect of beta1,3-glucan infusion at drying off more studies are needed involving a larger number of animals.     

Experimental Staphylococcus aureus infection of the mammary gland induces region-specific changes in innate immune gene expression

Staphylococcus aureus is a prolific mastitis-causing bacterium that resides naturally in the environment of the dairy cow. The aim of this study was to profile immune gene expression in tissue from the alveolar, ductal, gland cistern and teat canal regions of the bovine mammary gland following intramammary infection with S. aureus. Quantitative real-time PCR (qPCR) was used to profile expression of innate immune genes including pattern recognition receptors (PRRs), cytokines, antimicrobial peptides (AMPs) and acute phase proteins (APPs). Consistent expression of Toll-like receptors (TLRs) 1-10 and NOD-like receptors (NODs) 1-2 was detected in all four tissue regions. Pro-inflammatory cytokines (IL6, IL17A and IL8) and anti-inflammatory cytokine (IL10) were induced in all 4 tissues. APP (SAA3 and HP) and AMP (DEFB4 and DEFB5) genes showed the greatest induction throughout the mammary gland in response to S. aureus, with particularly high expression in alveolar tissue (SAA3 and HP >133- and >80-fold respectively, P<0.05; DEFB4 and DEFB5 >9- and >27-fold respectively, P<0.05). Collectively, our data show both sentinel and effector immune functions throughout the mammary gland in response to S. aureus challenge.     

The cytokines of bovine mammary gland: prospects for diagnosis and therapy

The lack of efficacy of conventional strategies for the maintenance of healthy udders in domestic cattle has prompted studies on the use of cytokines for this purpose. The adjuvant use of recombinant bovine cytokines, such as IL-2, IFN-gamma and TNF-alpha, in normal mammary gland, mobilizes innate and acquired immunity. However, stimulated immunity does not prevent or eradicate infection, particularly in the case of Staphylococcus aureus mastitis. Cytokines do, however, improve the bactericidal efficiency of certain antibiotics. The subtle and sensitive changes in the cytokine network of normal and mastitic bovine mammary gland may encourage the use of cytokines in the diagnosis and prognosis of udder health. Numerous studies support this hypothesis, and detection and monitoring of cytokines could become an important alternative management for udder health. The use of cytokines in the immunotherapy, diagnosis and prognosis of mastitis will grow with knowledge of the cytokine network in bovine mammary glands and the development of efficient cytokine diagnostic techniques.     

Mouse mastitis model of infection for antimicrobial compound efficacy studies against intracellular and extracellular forms of Staphylococcus aureus

Despite the general in vitro susceptibility of Staphylococcus aureus isolates that cause infectious bovine mastitis, the pathogen remains difficult to eradicate with the available antibiotics. The capacity to survive intracellularly has been proposed as a factor contributing to the persistence of S. aureus in the bovine mammary gland. The costs associated with the use of cows or goats to assess the in vivo efficacy of new antibacterial compounds constitute a major drawback. Therefore, in the present study, a mouse model of intramammary infection has been characterized for in vivo testing of new experimental drugs. An inoculum of 100 CFU of S. aureus per gland caused an important level of infection with minimal tissue damage as observed at 24 h post-inoculation. By microscopy, polymorphonuclear neutrophil cell infiltration of the infected mammary glands was observed to increase over time. At 12-24 h of infection, the pathogen was primarily found alive and dividing in neutrophils and occasionally within mammary epithelial cells. Intramuscular or intravenous injections of cephapirin at t = 0 and 10 h reduced the number of CFU/g of gland in a dose-dependent manner. In conclusion, the mouse model of infectious mastitis proposed here is suitable for primary evaluation of experimental drugs after parenteral treatment of intramammary infection with a pathogen such as S. aureus that presents both intracellular and extracellular phases of growth.     

The effect of experimental infectious mastitis on leukocyte subpopulations and cytokine production in non-lactating ewes.

The interactions between leukocytes and cytokines during the acute response to intramammary infections in the dry mammary gland of sheep were studied. Dry ewes were experimentally infected in one udder half with either Staphylococcus aureus or Escherichia coli, or infused with saline as control. Udder secretion samples, blood samples and udder tissue samples were collected before and 4, 8 and 24 h after infections/infusions. Total and differential leukocyte counts were calculated in both blood and mammary secretions, and flow cytometry was used to detect the presence of CD4+, CD8+, WC1+, IL-2R+, CD18+ or L-selectin + lymphocytes, CD18+ or L-selectin + neutrophils, and CD14+ leukocytes. Moreover, the concentrations of interleukin-1 beta (IL-1 beta), interleukin-8 (IL-8) and granulocyte-macrophage colony stimulating factor (GM-CSF) in udder secretions were measured using ELISA, and RT-PCR was used to detect the presence of corresponding cytokine mRNA in udder tissue biopsies. The results suggest an association between the concentrations of IL-1 beta, IL-8 and the intensity of neutrophil infiltration of the infected gland. Immunologically relevant changes in proportions of lymphocyte subpopulations might also occur in the acute phase of the inflammatory reaction of the udder. Greater cellular and cytokine responses to E. coli infection may have contributed to the milder clinical picture and more rapid resolution of infection than that seen for S. aureus. Enhancing the production of pro-inflammatory cytokines may improve defence against bacterial mastitis.     

Staphylococcus aureus lipoteichoic acid triggers inflammation in the lactating bovine mammary gland

The response of the bovine mammary gland to lipoteichoic acid (LTA), which is a major pathogen-associated molecular pattern of Gram-positive bacteria, was investigated by infusing purified Staphylococcus aureus LTA in the lumen of the gland. LTA was able to induce clinical mastitis at the dose of 100 microg/quarter, and a subclinical inflammatory response at 10 microg/quarter. The induced inflammation was characterized by a prompt and massive influx of neutrophils in milk. LTA proved to induce strongly the secretion of the chemokines CXCL1, CXCL2, CXCL3 and CXCL8, which target mainly neutrophils. The complement-derived chemoattractant C5a was generated in milk only with the highest dose of LTA (100 microg). The pro-inflammatory cytokine IL-1beta was induced in milk, but there was very little if any TNF-alpha and no IFN-gamma. The re-assessment of CXCL8 concentrations in milk whey of quarters previously challenged with S. aureus, by using an ELISA designed for bovine CXCL8, showed that this chemokine was induced in milk, contradicting previous reports. Overall, S. aureus LTA elicited mammary inflammatory responses that shared several attributes with S. aureus mastitis. Purified LTA looks promising as a convenient tool to investigate the inflammatory and immune responses of the mammary gland to S. aureus.     

Neutrophil apoptosis during experimentally induced Staphylococcus aureus mastitis

The objective of this study was to determine whether neutrophil apoptosis and their consequent elimination by macrophages from the mammary gland is modulated by an infection caused by Staphylococcus aureus (S. aureus). The study was performed on twenty mammary glands of 5 virgin heifers. A buffered physiological solution (PBS) was administered as a means of control into the mammary glands of the heifers and after 168 h, the glands were inoculated with S. aureus. The samples of cell populations were obtained by lavages of the mammary glands in 4 intervals (24, 48, 72 and 168 h) after the experimental infection. Flow cytometry was used for determination of Annexin-V positivity and propidium iodide (PI) negativity of neutrophils. Light microscopy was used for determination of neutrophil karyopyknosis. Cytochemistry was used for the detection of myeloperoxidase-positive (MPO+) macrophages. Instillation of S. aureus resulted in an intramammary infection which persisted during the following experimental period. The total number of both Annexin-V-positive and PI negative neutrophils and karyopyknotic neutrophils peaked at 24 h after both of PBS and S. aureus administration. The highest percentages of Annexin-V-positive and PI negative neutrophils and karyopyknotic neutrophils were detected 48 and 168 h after PBS and S. aureus administration, respectively. The total number of MPO+ macrophages was the highest 24 h and 48 h after PBS and S. aureus administration, respectively; the percentage of MPO+ macrophages was the highest at 72 h in both cases. The dynamics of resolution of mastitis caused by S. aureus was very similar to the resolution of inflammatory response of the mammary gland after PBS administration. Mechanisms of cell pathogen elimination as well as inflammation resolution were very intensively involved; nevertheless, the mammary gland infection persisted. An early inclusion of the mechanisms of an acute inflammatory resolution thus paradoxically led to chronic infection.     

Factors involved in the early pathogenesis of bovine Staphylococcus aureus mastitis with emphasis on bacterial adhesion and invasion. A review

Staphylococcus aureus is the most important and prevalent contagious mammary pathogen; it causes clinical and subclinical intramammary infection with serious economic loss and herd management problems in dairy cows. In vitro studies have shown that Staphylococcus aureus adheres to mammary epithelial cells and extracellular matrix components and invades into mammary epithelial as well as other mammary cells. Staphylococcus aureus strains from intramammary infection produce several cell surface-associated and extracellular secretory products. The exact pathogenic roles of most of the products and their effects on adhesion and invasion are not well evaluated. It is also known that mammary epithelial cell-associated molecules and extracellular matrix components interact with S. aureus during the pathogenesis of mastitis, but their roles on adhesion and invasion have not been characterized. The adhesion of S. aureus to epithelial cells may involve non-specific physicochemical interactions and/or specific interactions between bacterial cell-associated ligands and host cell surface receptors. In vitro adhesion depends on the S. aureus strain, the growth phase of the bacteria, the growth medium and the origin of the epithelial cells. Adhesion is hypothesized to be a prerequisite and crucial early step for mammary gland infection. Staphylococcus aureus invades mammary epithelial cells. It also invades other cells such as endothelial cells and fibroblasts. Bacteria are found enclosed in membrane bound vacuoles in the cytoplasm of mammary epithelial cells. Recent observations indicate that S. aureus escapes from the phagosome into the cytoplasm and induces apoptosis. The invasion into mammary epithelial cells may occur through an endocytic process that requires involvement of elements of the cytoskeleton or by direct binding of bacteria to epithelial cells through a process mediated by specific receptors that needs de novo protein synthesis by both cells. Thus, the recurrent subclinical infection may result from this intracellular existence of bacteria that are protected from host defenses and effects of antibiotics. This review emphasizes on recent findings on S. aureus adhesion to mammary epithelial cells and extracellular matrix components and invasion into mammary epithelial cells.     

Prolactin stimulates the internalization of Staphylococcus aureus and modulates the expression of inflammatory response genes in bovine mammary epithelial cells

The incidence of mastitis in dairy cattle is highest at the drying off period and parturition, which are characterized by high levels of the lactogenic hormone prolactin (PRL). One of the most frequently isolated contagious pathogens causing mastitis is Staphylococcus aureus. However, the role of PRL on S. aureus infection in mammary epithelium has not been studied. In this work we evaluated the effect of bovine PRL (bPRL) on S. aureus internalization in a primary culture of bovine mammary epithelial cells (bMEC) and on the expression of cytokine and innate immune response genes. Our data show that 5ng/mL bPRL enhances approximately 3-fold the internalization of S. aureus (ATCC 27543) into bMEC. By RT-PCR analysis, we showed that bPRL is able to up-regulate the expression of tumor necrosis factor-alpha (TNF-alpha), interleukin-1beta (IL-1beta) and inducible nitric oxide synthase (iNOS) mRNAs. However, bPRL together with S. aureus did not modify the expression of TNF-alpha and iNOS mRNAs, while it down-regulated the expression of beta-defensin and IL-1beta mRNAs, as well as nitric oxide production, suggesting that infection and bPRL together can inhibit elements of the host immune response. To our knowledge, this is the first report that shows a role of bPRL during the internalization of S. aureus into bMEC.     

The production and characterization of anti-bovine CD14 monoclonal antibodies

To characterize further the chemical and biological properties of bovine soluble (bos) CD14, a panel of ten murine monoclonal antibodies (mAb) reactive with recombinant (r) bosCD14 were produced. A sandwich ELISA, using murine mAb and rabbit polyclonal antibodies reactive with rbosCD14 was developed. All the mAb were reactive by ELISA with baculovirus-derived rbosCD14 and they recognized rbosCD14 (40 kDa) by western blot analysis. The mAb also identified by western blot sCD14 (53 and 58 kDa) in milk and blood and sCD14 (47 kDa) in a lysate of macrophages obtained from involuted bovine mammary gland secretions. Analysis by ELISA of whey samples after intramammary injection of lipopolysaccharide (LPS) (10 micro g) revealed increased sCD14 levels between 8 to 48 h after injection. Flow cytometric analysis showed that the mAb bound to macrophages isolated from involuted mammary gland secretions and mouse macrophages but not to swine or horse monocytes. Addition of anti-rbosCD14 mAb to monocytes stimulated with LPS reduced in vitro production of TNF-alpha. The anti-rbosCD14 antibodies generated in this study will be useful in studying CD14, an accessory molecule that contributes to host innate recognition of bacterial cell wall components in mammary secretions produced during mastitis.     

Bactericidal activity of macrophages against Streptococcus uberis is different in mammary gland secretions of lactating and drying off cows

The aim of this study was to compare the ability of milk macrophages and macrophages from the mammary gland secretions during the mid-dry period for their interaction with the mastitis-causing Streptococcus uberis. We also aimed to determine if S. uberis induced the release of the cytokine tumour necrosis alpha (TNF-alpha) and the bactericidal moiety nitric oxide (NO) from milk macrophages of lactating cows and macrophages from the mammary gland secretions at the mid-dry period. Macrophages were isolated from the mammary gland secretions of cows during the mid-lactation or mid-dry period, and compared with blood monocytes for their interaction with the important mastitis-causing pathogen S. uberis. When infected in vitro with S. uberis, milk macrophages from lactating cows with S. uberis released modest amounts of the cytokine tumour necrosis factor alpha (TNF-alpha) (139 pg/ml) and the bactericidal moiety nitric oxide (NO) (3-4 microM of nitrite). Blood monocytes from lactating cows released significantly higher amounts of TNF-alpha (345 +/- 143 pg/ml) and NO (7 +/- 2 microM of nitrite) after interaction with S. uberis, compared to milk macrophages (P < 0.01 for both TNF-alpha and NO). Stimulation of blood monocytes with the cytokine interferon-gamma (IFN-gamma) enhanced significantly the release of NO and TNF-alpha, but IFN-gamma did not significantly enhance the production of NO and TNF-alpha by milk macrophages from lactating cows. Milk macrophages from all lactating cows failed to kill S. uberis efficiently, and this lack of killing was unaffected by prior treatment with gamma interferon (IFN-gamma) (P > 0.05). Rather, S. uberis multiplied significantly inside infected milk macrophages from lactating cows, with a two-fold increase in bacterial numbers at 2 h post-infection. Milk macrophages from lactating cows were able however, to kill a significant proportion (50-60%, P < 0.01) of phagocytosed Staphylococcus aureus. Blood monocytes from all cows were found to exert significant bactericidal activity against S. uberis. There were no significant differences in the bactericidal activity of milk macrophages obtained from lactating cows with low somatic cell counts (SCC; < 10(5) ml(-1)) compared with those with a mildly elevated SCC (> 10(5) ml(-1)) (P > 0.05). In contrast, mammary gland secretion macrophages isolated from the same cows in the mid-dry period killed a significant proportion of phagocytosed S. uberis (50-65% of ingested S. uberis killed, P < 0.01) although cytokine production in response to in vitro bacterial infection was low. We conclude that the bactericidal activity of mammary gland secretion macrophages against a virulent strain of S. uberis is low during the lactation period. In addition, our data indicate that S. uberis is not a strong inducer of NO and TNF-alpha in macrophages from the milk or mammary gland secretions of cows during the drying off period. Finally, IFN-gamma does not activate milk macrophages or macrophages from cows during the lactating period or mammary gland secretions during the drying off period.