Introduction
Bovine mastitis remains as the disease causing the largest economic losses to the dairy industry, despite the intensive research and prevention measures at the herd level carried out for decades (Hogeveen et al., Reference Hogeveen, Steeneveld and Wolf2019). Among bovine mastitis pathogens, Staphylococcus aureus is a major cause of mastitis and continues to be a significant issue for the dairy industry (Krishnamoorthy et al., Reference Krishnamoorthy, Suresh, Jayamma, Shome, Patil and Amachawadi2021). The chronic subclinical infection with sporadic clinical episodes is the most common form of the disease (Calvinho and Dallard, Reference Calvinho and Dallard2023a). Chronically infected cows respond poorly to classical antibiotic therapy and become the main reservoir of the organism in dairy herds (Barkema et al., Reference Barkema, Schukken and Zadoks2006). Persistence of S. aureus intramammary infection (IMI) has been associated with an impaired immune response that favours the establishment of infection and prevents clearance of the pathogen by the host (Lee et al., Reference Lee, Bannerman, Paape, Huang and Zhao2006; Calvinho and Dallard, Reference Calvinho and Dallard2023b), even though this response can be intense and sustained as seen in long-standing infections (Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a).
Involution is defined as the process through which the mammary gland (MG) returns to a non-lactating state. Although milking cessation is essential for proper cell renewal, during early involution, the cow MG is more susceptible to new IMI (Sordillo et al., Reference Sordillo, Nickerson, Akers and Oliver1987). Dry cow therapy plays a crucial role in mastitis control programmes, helping prevent new IMIs and eliminating existing ones. Although the efficacy of dry cow antibiotic therapy against S. aureus is higher than that observed during lactation, early research studies showed that 20–60% of S. aureus IMI can persist during the dry period despite the use of dry cow therapy (Eberhart, Reference Eberhart1986). Decreased cure rates during the dry period have been associated with several cow-level and pathogen factors (reviewed by Barkema et al., Reference Barkema, Schukken and Zadoks2006).
Active involution involves an increase in the permeability of tight junctions (TJs) between epithelial cells that favours an increase in the recruitment of immune cells to the MG (Kobayashi et al., Reference Kobayashi, Oyama, Numata, Rahman and Kumura2013) During the first 2–4 days after drying-off, neutrophils are the leukocytes found in greatest numbers in mammary secretions to be later replaced by macrophages that become the predominant cell type (Sordillo et al., Reference Sordillo, Nickerson, Akers and Oliver1987; Tatarczuch et al., Reference Tatarczuch, Philip, Bischof and Lee2000). Neutrophils and macrophages are thought to play a role in the clearance of casein micelles, lipid droplets and cellular debris (Monks et al., Reference Monks, Geske, Lehman and Fadok2002; Atabai et al., Reference Atabai, Sheppard and Werb2007). This partially explains the high susceptibility of the MG to infections during the early dry period due to interference from milk components (Paape et al., Reference Paape, Wergin, Guidry and Pearson1979) and the decline in the capacity of phagocytic cells (Tatarczuch et al., Reference Tatarczuch, Bischof, Philip and Seong Lee2002). Bacterial IMI causes pathogen-dependent damage of the blood–milk barrier and is accompanied by changes in the TJ permeability in bovine mammary epithelial cells (Xu et al., Reference Xu, Dong, Wang, Qi, Li, Cheng, Liu, Zhang and Qing Gao2018).
In previous studies, we observed sustained and exacerbated innate and adaptive immune responses during chronic S. aureus IMI at involution (Andreotti et al., Reference Andreotti, Pereyra, Baravalle, Renna, Ortega, Calvinho and Dallard2014; Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a, Reference Andreotti, Pereyra, Sacco, Baravalle, Renna, Ortega, Calvinho and Dallard2017b; Renna et al., Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019); however, this response was not associated with bacterial clearance from the MG, reflecting an equilibrium between pathogen and host that contributes to IMI persistence. Furthermore, in vivo studies using S. aureus strains with different adaptive capacities to the MG demonstrated a differential innate and adaptive immune response that influenced the course and severity of bovine mastitis (Niedziela et al., Reference Niedziela, Murphy, Grant, Keane and Leonard2020; Engler et al., Reference Engler, Renna, Beccaria, Silvestrini, Pirola, Pereyra, Baravalle, Camussone, Monecke, Calvinho and Dallard2022). Therefore, to characterize the kinetics of immune cell populations that migrate from the blood to the MG in S. aureus chronic infections during active involution is essential to understand the host defence capability and the impact of this response on mammary health. The aim of this study was to evaluate the kinetics of local and systemic immune cell populations in mammary secretions and blood samples from cows free of IMI and chronically infected with S. aureus during active involution.
Materials and methods
Experimental design
Sixteen non-pregnant Holstein cows in late lactation (weeks 31–36) from the dairy herd of the School of Agriculture and Livestock of Universidad Nacional del Litoral (UNL) were included in this study. Cows were from parity 3 to 4, milked twice daily and produced an average of 25 kg/day of milk during lactation and 16 kg of milk/day before interruption of lactation. The animals were dried-off abruptly after the last milking of the day, and no antibiotic therapy was administered. During lactation, cows were kept under grazing conditions and received concentrates in the milking parlour. After drying-off, animals grazed on alfalfa and/or ryegrass pastures in a rotational system (stocking density: 8 animals/ha). Cows with similar lactation numbers were included in each experimental group (uninfected and S. aureus-infected). All procedures with animals were conducted under protocols approved by the Ethics and Security Committee (Protocol No 721/22) of the Facultad de Ciencias Veterinarias – UNL and were in accordance with the Guide for the Care and Use of Agricultural Animals in Agricultural/Research and Teaching (Federation of Animal Science Societies, 2010).
The infection status of the mammary quarters was determined within 6 months before the beginning of the experiment, 20 and 3 days before cessation of milking for confirmation, immediately before drying-off (day 0) and 7, 14 and 21 days post drying-off (Supplementary Figure S1). Staphylococcus aureus-infected cows were selected based on results of monthly somatic cell counts (SCC > 250 × 103 cells/mL) and subsequent testing by bacteriological analysis of quarter milk samples. A quarter was considered to be chronically infected if S. aureus was isolated at least three times from consecutive or interspersed monthly milk samples over a period of six months (Pereyra et al., Reference Pereyra, Picech, Renna, Baravalle, Andreotti, Russi, Calvinho, Diez and Dallard2016; Renna et al., Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019). Only S. aureus-infected quarters were included in the study and were randomly selected from cows showing at least one quarter infected with S. aureus. Any quarter co-infected with another pathogen at any sampling time was excluded from the study.
Uninfected and non-inflamed control quarters were selected from cows with negative bacteriological analysis and SCC less than 250 × 103 cells/mL during three consecutive monthly milk samples. Uninfected (n = 8) and S. aureus-infected (n = 8) animals were included. According to the eligibility criteria used for the selection of the experimental units, 16 cows were included in the study (8 cows with one S. aureus-infected quarter and 8 uninfected cows).
Sampling schedule
Peripheral blood samples (∼50 mL) were collected from the cranial mammary vein, in tubes containing ethylenediaminetetraacetic acid (EDTA; 46005 GT Lab, SF, Argentina) as an anticoagulant. Also, mammary secretion samples (∼50–100 mL) were aseptically collected using standard procedures (Oliver et al., Reference Oliver, Gillespie, Headrick, Moorehead, Lunn, Dowlen, Johnson, Lamar, Chester and Moseley2004). Sampling was performed immediately before drying-off (day 0) (blood samples) and at 7, 14 and 21 days after drying-off (both blood and mammary secretion samples) from uninfected and S. aureus-infected animals (Supplementary Figure S1). Milk from lactating cows was obtained before milking. To obtain mammary secretion samples during the involution period, teat ends were scrubbed vigorously with cotton buds moistened with 70% alcohol. The first two streams of mammary secretion from each teat were discarded, the next 5 mL were collected in sterile plastic vials for bacteriological analysis and a volume between 50 and 100 mL was obtained in sterile plastic containers for mammary secretion analysis. After obtaining mammary secretion from dry cows, the teats were dipped in a 0.5% iodophor solution. The procedure was carried out on each mammary quarter at all evaluated times.
Bacteriological examination
Milk and mammary secretion samples (10 μL) were streaked onto blood agar plates (B0211106 Britania, CABA, Argentina) supplemented with 5% bovine blood and incubated for 48 h aerobically at 37°C. Plates were examined for bacterial growth at 24 and 48 h. Staphylococcus aureus was identified based on the haemolytic pattern on blood agar, catalase and coagulase tests and differentiated from other coagulase-positive Staphylococci by acetoin production and selective growth on P agar with 7 μg/mL acriflavine (Roberson et al., Reference Roberson, Fox, Hancock and Besser1992). Other mastitis pathogens were identified based on standard methodology (Oliver et al., Reference Oliver, Gillespie, Headrick, Moorehead, Lunn, Dowlen, Johnson, Lamar, Chester and Moseley2004). The presence of one colony of S. aureus on blood agar was considered a positive identification; therefore, the detection limit was 100 colony-forming units/mL). Coagulase-positive Staphylococci were presumptively identified as S. aureus and subjected to polymerase chain reaction for S. aureus identification, as described previously (Martineau et al., Reference Martineau, Picard, Roy, Ouellette and Bergeron1998). To confirm persistency, clonality of S. aureus strains isolated in the last sampling during lactation and 21 days after involution was compared by pulsed field gel electrophoresis (PFGE) of SmaI-digested chromosomal DNA fragments using a CHEF-DR II apparatus (Bio-Rad Laboratories, CA, USA), as described previously (Camussone et al., Reference Camussone, Molineri, Signorini, Neder, Vitulich and Calvinho2020).
Isolation of peripheral blood mononuclear cells
Peripheral blood mononuclear cells (PBMC) were purified following the methodology described previously by Silvestrini et al. (Reference Silvestrini, Beccaria, Renna, Engler, Simonutti, Cellone, Calvinho, Dallard and Baravalle2023). Briefly, 25 mL of anticoagulated blood samples were centrifuged at 680×g for 15 min at 4°C. The plasma from the upper layer was removed. Then, the fraction containing the total leukocytes and red blood cells was resuspended in a lysing solution (0.15 M NH4Cl, 1 mM NaHCO3 and 0.1 mM disodium EDTA) and incubated for 10 min at 4°C to achieve complete lysis of red blood cells. Then, the samples were centrifuged at 680×g for 10 min at 4°C, and a second lysis was performed by resuspending the cell pellet in 10 mL of lysis buffer for 5 min at 4°C. At the end of the incubation time, cells were washed with PBS 1× and finally resuspended in 10 mL of RPMI (21870-076 Gibco, MA, USA)/10% bovine foetal serum (BFS) (Internegocios, BA, Argentina). Using the trypan blue (15250061 Thermo Fisher Scientific, MA, USA) exclusion test, viable cells were counted in a Neubauer haemocytometer chamber and then resuspended to the desired concentration for flow cytometry (1 × 106 cells/tube).
Isolation of MG secretion mononuclear cells
Mammary secretion mononuclear cells (MSMC) were isolated as described previously by Renna et al. (Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019). Briefly, 25 mL of mammary secretion was diluted 1/2 with PBS and centrifuged at 400×g for 10 min at 4°C using a high-capacity centrifuge (RC6 plus, Sorvall; Thermo Fisher Scientific, MA, USA). Then, the fat was carefully removed; the pellet was washed twice with PBS and resuspended in 40 mL of PBS. Then, 5 mL of Histopaque 1077 (17144003 Cytyva, MA, USA) were placed on the bottom of the tube and centrifuged at 400×g for 30 min at room temperature. The mononuclear layer was collected, and cells were washed three times with PBS (400×g, 10 min at 4°C). As PBMC, viable MSMC were counted using the trypan blue exclusion test and resuspended to the desired concentration for flow cytometry (1 × 106 cells/tube).
Flow cytometry
The kinetics of immune cells populations: monocytes-macrophages, T CD4 lymphocytes, T CD8 lymphocytes and B lymphocytes in PBMC and MSMC were evaluated by flow cytometry, as described previously by Silvestrini et al. (Reference Silvestrini, Beccaria, Renna, Engler, Simonutti, Cellone, Calvinho, Dallard and Baravalle2023). Briefly, 1 × 106 cells were incubated with specific antibodies for the different surface markers in PBS–EDTA–BFS on ice and darkness for 30 min. The antibodies used are detailed in Supplementary Table 1. After washing, cells were resuspended in focusing fluid and acquired on a flow cytometer (Attune, NxT Acoustic Focusing Cytometer A24860, Life Technologies, CA, USA). Data were collected and analysed using FlowJo software (TreeStar Inc.).
The gating strategy employed for the analysis of PBMC and MSMC is shown in Supplementary Figure S2. To select the target populations (mononuclear cells), gating strategies were employed to exclude debris and doublet cells using forward scatter (FSC) high vs. area graph (R1). Mononuclear cells were then gated based on forward and side scatter characteristics (R2). Percentages of T CD4 and CD8 lymphocytes were obtained from a dot plot of different fluorescence channels within R3. Percentages of monocytes-macrophages and B lymphocytes were acquired from a SSC-A vs. the different fluorescence channels dot plot within R2. Only the low-granularity macrophage population expressing CD14 was included in the analysis. Fluorescence of Fluorescein isothiocyanate (FITC) was detected at the channel 530/30 nm in BL-1 detector, fluorescence of phycoerythrin (PE) was detected at the channel 574/26 nm in BL-2 detector, fluorescence of Allophycocyanin (APC) was detected at the channel 670/14 nm in RL-1 detector and fluorescence of Alexa Fluor 647 was detected at the channel 670/14 nm in RL-1 detector. Two staining mixes were prepared for each sample. One mix stained CD3, CD4 and CD8 populations, while the other mix stained CD14 and CD21 populations.
Flow cytometer performance was evaluated using Attune™ performance-tracking beads (Invitrogen). An unstained control sample was utilized to (i) adjust forward and side scatter to ensure cell populations of interest were on scale and (ii) adjust photomultiplier tube gain for each fluorochrome detector so that the peak mean fluorescence intensity of each channel was within 103 on a log scale. Appropriate single-colour controls were performed to determine the respective fluorescent peaks of individual stains. For the experiment, 100,000 events were recorded on R2. In the experimental design used, secondary controls were not required because all the antibodies used were directly conjugated to distinct fluorochromes.
Statistical analysis
Statistical analysis of data was performed using SPSS 25.0 Software (SPSS Inc.). To determine the effect of IMI on the evaluated variables, the data obtained were analysed using ANOVA of repeated measures (MRANOVA). ANOVA assumptions, such as normality of the distribution and homogeneity of variances, were verified by the Shapiro–Wilk and Levene tests, respectively. To evaluate the differences at each sampling time between healthy and chronically infected quarters, Student’s T test was performed. For the analysis of the percentages of immune cell populations during the different days of involution evaluated, one-way ANOVA followed by Duncan’s multiple comparison test were performed for each experimental group (IMI-free and chronically infected). Differences of P < 0.05 were considered significant. Data were presented as the mean ± standard error of the mean.
Results
Staphylococcus aureus IMI were naturally acquired either during the previous dry period or within the first two months of lactation preceding the beginning of the study. The selected S. aureus isolates obtained from cows during lactation (L) and the dry period (D) were characterized at the molecular level. Supplementary Figure S3 shows representative banding patterns obtained from SmaI-digested DNA and PFGE of 16 S. aureus strains isolated from milk in the last sampling during lactation and from mammary secretion 21 days after drying-off. Indistinguishable PFGE profiles were observed for isolates 1 (L,D), 2 (L,D), 3 (L,D), 5 (L,D), 7 (L,D) and 8 (L,D) recovered from the same animal during lactation and dry period. Isolates 4 (L,D) and 6 (L,D) showed a difference between their banding patterns obtained during lactation and dry period of 27% and 25%, respectively.
Kinetics of systemic immune cells
Percentages of monocytes-macrophages (CD14+ cells), helper T lymphocytes (CD4+ cells), cytotoxic T lymphocytes (CD8+ cells) and B lymphocytes (CD21+ cells) were evaluated from blood samples of uninfected and S. aureus-infected animals at different days after drying-off (Fig. 1).
Percentages of A) CD14+, B) CD4+, C) CD8+ and D) CD21+ cells from PBMC of uninfected and S. aureus-infected animals at different days after drying-off. The data are presented as means ± standard error of mean (SEM). Results of a repeated measure ANOVA (RMANOVA) are indicated. At each sampling time, asterisks represent statistically significant differences between uninfected and S. aureus-infected animals (** P < 0.01; * P < 0.05). Different letters correspond to statistically significant differences between one group over time, capital letter for S. aureus-infected animals and lowercase for uninfected animals (P < 0.05).
The percentages of CD14+ cells in PBMC were affected by S. aureus infection during involution (P = 0.02; Fig. 1A), being higher in S. aureus-infected animals than in uninfected animals at D7 (P = 0.02) and D21 (P = 0.02) after drying-off. The percentages of CD14+ cells in uninfected animals were maintained over time (∼20%). Also, when the S. aureus-infected group was analysed over time, the percentages of CD14+ cells were higher at D7 and D21 compared with D0 (P = 0.02).
No effect of S. aureus infection was observed in the percentages of CD4+ cells in PBMC (P = 0.69; Fig. 1B). The percentages of CD4+ cells in both uninfected and S. aureus-infected animals were maintained over time (∼35%).
No effect of S. aureus infection was observed in the percentages of CD8+ cells in PBMC (P = 0.17; Fig. 1C). However, when the uninfected group was analysed over time, the percentage of CD8+ cells on D21 was higher than on D0 (P = 0.05). In S. aureus-infected animals, the highest percentages were also observed on D21, although this increase was not significant.
No effect of S. aureus infection was observed in the percentages of CD21+ cells in PBMC (P = 0.09; Fig. 1D). However, when the analysis was performed by time, the percentages of CD21+ cells decreased in S. aureus-infected animals compared with uninfected animals at D21 (P = 0.004).
Kinetics of local immune cells
Percentages of monocytes-macrophages (CD14+ cells), helper T lymphocytes (CD4+ cells), cytotoxic T lymphocytes (CD8+ cells) and B lymphocytes (CD21+ cells) were evaluated from secretion samples of uninfected and S. aureus-infected animals at different days after drying-off (Fig. 2).
Percentages of A) CD14+, B) CD4+, C) CD8+ and D) CD21+ cells from MSMC of uninfected and S. aureus-infected quarters at different days after drying-off. The data are presented as means ± standard error of mean (SEM). Results of a repeated measure ANOVA (RMANOVA) are indicated. At each sampling time, asterisks represent statistically significant differences between uninfected and S. aureus-infected quarters (*** P < 0.001, ** P < 0.01, * P < 0.05). Different letters correspond to statistically significant differences between one group over time, capital letter for S. aureus-infected animals and lowercase for uninfected animals (P < 0.05).
The percentages of CD14+ cells in MSMC were affected by S. aureus infection during involution (P < 0.001; Fig. 2A), being lower in mammary secretions of S. aureus-infected quarters compared with uninfected quarters at D7 (P < 0.001), D14 (P = 0.007) and D21 (P = 0.01) of involution. Also, when each experimental group was analysed over time, in both mammary secretion of S. aureus-infected and uninfected quarters, the percentages of CD14+ cells were higher at D7 and D14 compared with D21 (P = 0.01, P = 0.001, respectively).
No effect of S. aureus infection was observed in the percentages of CD4+ cells in MSMC (P = 0.35; Fig. 2B). However, when the uninfected group was analysed over time, the percentage of CD4+ cells on D7 was higher than on D14 and D21 (P < 0.001).
The percentages of CD8+ cells in MSMC were affected by S. aureus infection during involution (P = 0.01; Fig. 2C), being lower in mammary secretions of S. aureus-infected quarters compared with uninfected quarters at D7 (P = 0.004), D14 (P = 0.01) and D21 (P < 0.001) of involution.
No effect of S. aureus infection was observed in the percentages of CD21+ cells in MSMC (P = 0.69; Fig. 2D). However, when the analysis was performed by time, the percentages of CD21+ cells were lower in mammary secretions of S. aureus-infected quarters compared with uninfected quarters at D7 (P = 0.008). Also, when each experimental group was analysed over time, in mammary secretions of uninfected quarters, the percentages of CD21+ cells at D21 were higher compared with D7 and D14 (P < 0.001); while, in mammary secretions of S. aureus-infected quarters, the percentages of CD21+ cells at D7 were lower compared with D14 and D21 (P = 0.05).
Discussion
This study focused on the kinetics of different immune cell populations in blood and mammary secretion samples from cows free of IMI and chronically infected with S. aureus during active involution to gain insight into the mechanisms allowing S. aureus to successfully persist in the MG. Using PFGE, we demonstrated the clonal persistence over time of S. aureus strains isolated during lactation and drying. Only two strains isolated during lactation showed differences of less than 30% in their PFGE patterns than those isolated during the dry period. Strains that differ by no more than three bands in their patterns are classified as closely related, which may have arisen from only one genetic event (Dendani Chadi and Arcangioli, Reference Dendani Chadi and Arcangioli2023). In the present study, both strains that showed differences between the lactation sample and the dry cow sample differed in two bands. This difference could be due to spontaneous mutations of the same strain that could increase genetic adaptability to survive in the MG environment, as observed in previous studies (Wang et al., Reference Wang, Wu, Shen, Wu and Wang2013).
Macrophages are essential phagocytic cells responsible for engulfing and processing cellular debris and pathogens. In the present study, we analysed CD14 expression only in the monocyte-macrophage population with low granularity. The vacuolated macrophage population (with high granularity) was not included in the analysis due to its low and variable CD14 expression. This phenomenon was observed in vacuolated macrophages from the bovine MG after both Gram-positive (Sladek et al., Reference Sládek and Rysánek2006) and Gram-negative infections (Sladek et al., Reference Sládek, Rysánek and Faldyna2002). In the present study, the percentages of CD14+ cells in blood samples were higher in S. aureus-infected animals than in uninfected animals at days 7 and 21 post-drying-off. However, the percentages of these cells in the mammary secretions from S. aureus-infected quarters were significantly lower compared with those of the uninfected quarters in all evaluated periods. In a previous study, in tissue sections from uninfected and chronically S. aureus-infected MG, we observed a high number of CD14+ cells in S. aureus-infected quarters compared with uninfected quarters at days 7 and 21 of involution (Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a). Moreover, in the aforementioned study, the tissue immunoexpression for IL-1α and IL-17 was higher in S. aureus-infected quarters compared with uninfected quarters at 7 and 14 and 7 and 21 days of involution, respectively. Both cytokines are capable of inducing the production of chemokines such as CCL2 (MCP-1), crucial for the recruitment of monocytes to the site of infection (Shahrara et al., Reference Shahrara, Pickens, Mandelin, Karpus, Huang, Kolls and Pope2010; Dinarello, Reference Dinarello2018). Taken together, these results could indicate that, although chronic S. aureus IMI induces a significant increase in the number of CD14+ cells in the blood circulation, these cells do not appear to migrate to the mammary secretion and remain retained in the tissue.
The percentages of CD14+ cells in blood increased as involution progressed only in S. aureus-infected animals. The percentages of CD14+ cells decreased over time in both S. aureus-infected and uninfected animals in secretion samples, suggesting a lower influx of phagocytic cells to the MG as involution progressed. These results coincided with those of Dallard et al. (Reference Dallard, Baravalle, Ortega, Tumini, Canavesio, Neder and Calvinho2009), who observed a greater number of CD14+ cells in mammary tissue of uninfected and S. aureus-infected quarters on day 7 of involution and a progressive decrease until day 21 of involution. In the aforementioned study, the number of CD14+ cells was significantly higher in S. aureus-infected than uninfected quarters during the three involution periods evaluated (7, 14 and 21 days), which reinforces the theory that macrophages do not migrate to the mammary secretion and remain retained in the mammary tissue. Although cell migration assays were not performed in this study, the results observed at the local level, in uninfected and S. aureus-infected quarters, could suggest that phagocytes that migrate to the tissues and cross the blood–milk barrier fulfil their phagocytic functions during the process of active remodelling to gradually decrease, as the process of mammary involution advances. Previous in vitro studies have shown that the presence of viable S. aureus within macrophages from mammary secretion could favour evasion of the host defences by the bacterium, contributing to its persistence within the MG (Sacco et al., Reference Sacco, Velázquez, Renna, Beccaria, Baravalle, Pereyra, Monecke, Calvinho and Dallard2020). Taken together, the lower percentages and the progressive decrease in the number of monocytes-macrophages in the mammary secretion of S. aureus-infected quarters compared with uninfected quarters could be related to bacterial intracellular persistence that stimulates a significant increase in blood macrophages that migrate and are retained in the mammary tissue, leading to less migration of this cell type to mammary secretion.
Lymphocytes are an important part of the adaptive immune defence against IMI; however, the kinetics of different subsets of these cells that migrate from the blood to the MG in S. aureus chronic infections have not been fully elucidated (Nagahata et al., Reference Nagahata, Kawai, Higuchi, Kawai, Yayou and Chang2011; Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a). To the best of our knowledge, there are no studies in which the different lymphocytes subsets in mammary secretion from S. aureus-infected quarters have been quantified up to 21 days after drying-off. The lymphocyte population in the bovine MG consists predominantly of T lymphocytes, with few B lymphocytes (Sordillo, Reference Sordillo2018). The T lymphocytes, including α/β (Davis and Splitter, Reference Davis and Splitter1991) and γ/δ cells (Asai et al., Reference Asai, Komine, Kozutsumi, Yamaguchi, KI and Kumagai2000; Faldyna et al., Reference Faldyna, Leva, Sladek, Rysanek and Toman2006), express CD2 molecules on their surface; α/β lymphocytes are made up of CD4+ (T-helper) and CD8+ (T-cytotoxic or T-suppressor) cells (Asai et al., Reference Asai, Kai, Rikiishi, Sugawara, Maruyama, Yamaguchi, Ohta and Kumagai1998). In the present study, the percentages of CD4+ cells were similar between uninfected and S. aureus-infected animals at all involution times evaluated in both blood and mammary secretion samples. Regarding CD8+ cells, we observed a significant decrease in this cell subset in mammary secretions from S. aureus-infected compared with uninfected quarters at all involution stages evaluated. These results could indicate that cytotoxic T cells do not cross the blood–milk barrier and are retained in the mammary tissue, fulfilling their functions of direct cytotoxicity against cells with intracellular bacteria (Sordillo, Reference Sordillo2018). In a previous study carried out during MG involution, we observed that the number of T lymphocytes (labelled with anti-CD2) in mammary tissue was significantly higher in S. aureus-infected quarters than in uninfected quarters at 7 and 14 days of involution (Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a). Taken together, these results could indicate that the cytotoxic T cell population migrates to the infected MG to a greater extent than the helper T cell population. Riollet et al. (Reference Riollet, Rainard and Poutrel2001), evaluating milk cells from naturally occurring S. aureus IMI established for at least two months prior to the initiation of the study, found that CD4+ and CD8+ lymphocytes were significantly increased in milk. Besides, S. aureus infection induced a larger recruitment of CD8+ lymphocytes compared with CD4+ lymphocytes (Riollet et al., Reference Riollet, Rainard and Poutrel2001). Differences between studies could have been due to the use of lactating cows by Riollet et al., Reference Riollet, Rainard and Poutrel2001. The results of the present study and those obtained previously (Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a) suggest that CD8+ cells play an important role during chronic S. aureus IMI through their cytotoxic or suppressive activity. In addition, the decrease of CD8+ lymphocytes in mammary secretion samples suggests that they are not released but retained in the MG as part of the inflammatory process aimed at eliminating the infection along with CD14+ cells.
In the present study, the percentages of CD21+ cells in blood samples decreased in S. aureus-infected animals compared with uninfected animals at day 21, suggesting a potential migration of these cells to the MG only at the end of the active involution period. Besides, in secretion samples, the percentages of CD21+ cells decreased in S. aureus-infected quarters at day 7 compared with non-infected quarters, suggesting that these cells do not pass the blood–milk barrier and are possibly retained in the mammary tissue. In a previous study carried out during involution in mammary tissue, we observed that the number of B lymphocytes (labelled with anti-CD21) was significantly higher in S. aureus-infected quarters than in uninfected quarters at 7, 14 and 21 days after drying-off, showing a progressive increase as involution advanced (Andreotti et al., Reference Andreotti, Baravalle, Sacco, Lovato, Pereyra, Renna, Ortega, Calvinho and Dallard2017a). Taken together, these results show the importance of this cell type in the immune response against S. aureus chronic infections at the local level. In earlier studies, the distribution of Ig-producing plasma cells in mammary tissue from uninfected and S. aureus naturally occurring chronic infected glands from lactating cows was determined, finding higher numbers of cells in infected quarters but only significantly different from uninfected quarters at the Furstenberg rosette area (Doymaz et al., Reference Doymaz, Sordillo, Oliver and Guidry1988). In addition, no significant differences in concentrations of Ig isotypes in milk from S. aureus-infected and uninfected quarters were observed, which the authors attributed to minimal stimulation of the humoral immune response (Doymaz et al., Reference Doymaz, Sordillo, Oliver and Guidry1988). In a previous study, we evaluated the antibody response in mammary secretions during MG involution of cows with S. aureus chronic infections. The levels of specific IgG and IgG1 antibodies in mammary secretions were lower in S. aureus-infected quarters compared with uninfected quarters at every involution time evaluated (7, 14 and 21 days), while specific IgG2 levels were not affected by the infectious status (Renna et al., Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019). These results are indicative of a diminished adaptive humoral response during naturally acquired chronic IMIs (Renna et al., Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019). The decrease in percentages of B lymphocytes in mammary secretions from infected animals at d 7 and 21 observed in this study, as well as the lower levels of some IgG subtypes in mammary secretion from S. aureus-infected MG during involution observed previously (Renna et al., Reference Renna, Silvestrini, Beccaria, Velázquez, Baravalle, Engler, Pereyra, Calvinho and Dallard2019) are indicative of a lower adaptive humoral immune response during chronic infection that would contribute to S. aureus persistence.
Conclusions
We demonstrated that chronic S. aureus IMI induces a significant increase in the number of CD14+ cells in the blood circulation. However, a significant decrease in CD14+ cells was observed in the mammary secretion of S. aureus-infected animals compared with uninfected ones in all evaluated periods. This suggests that these cells may not migrate to the mammary secretion and are potentially retained in the tissue. Further migration studies are needed to confirm this hypothesis.
In the blood circulation, the populations of CD4+ and CD8+ lymphocytes did not differ between S. aureus-infected and uninfected animals during involution. However, a significant decrease in CD8+ lymphocytes was observed in the mammary secretion of S. aureus-infected animals compared with uninfected ones in all evaluated periods. This suggests that CD8+ lymphocytes may be retained in the mammary tissue, carrying out their specific function of eliminating intracellularly infected cells in S. aureus-infected animals. Further characterization of CD8+ lymphocytes in mammary tissue is required to confirm this hypothesis.
Chronic S. aureus IMI induces a significant decrease in the number of CD21+ cells in blood circulation. However, their low numbers in the mammary secretions of infected animals may compromise the humoral defense capacity of the gland, potentially contributing to S. aureus persistence.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029925101064.
Acknowledgements
The authors express their appreciation to Mr Gaston Reibel for field technical assistance. This work was supported by Argentine National Agency for the Promotion of Science and Technology (ANPCyT) (PICT 2017-0479 and PICT-2021-1003).
Competing interests
Authors have no conflicts of interest.
