ABSTRACT
Staphylococcus aureus, an important cause of mastitis in mammals, is becoming increasingly problematic due to the development of resistance to conventional antibiotics. The ability of S. aureus to invade host cells is key to its propensity to evade immune defense and antibiotics. This study focuses on the functions of cathelicidins, small cationic peptides secreted by epithelial cells and leukocytes, in the pathogenesis of S. aureus mastitis in mice. We determined that endogenous murine cathelicidin (CRAMP; Camp) was important in controlling S. aureus infection, as cathelicidin knockout mice (Camp−/−) intramammarily challenged with S. aureus had higher bacterial burdens and more severe mastitis than did wild-type mice. The exogenous administration of both a synthetic human cathelicidin (LL-37) and a synthetic murine cathelicidin (CRAMP) (8 μM) reduced the invasion of S. aureus into the murine mammary epithelium. Additionally, this exogenous LL-37 was internalized into cultured mammary epithelial cells and impaired S. aureus growth in vitro. We conclude that cathelicidins may be potential therapeutic agents against mastitis; both endogenous and exogenous cathelicidins conferred protection against S. aureus infection by reducing bacterial internalization and potentially by directly killing this pathogen.
INTRODUCTION
Staphylococcus aureus is a common cause of mastitis, an inflammatory disease of the mammary gland (1–3). This condition is particularly common and problematic in dairy cows, where mastitis causes financial losses (4) due to decreased milk quality and yield as well as through increased culling and treatment costs (5, 6). Antibiotics are inconsistently effective, as S. aureus can survive inside leukocytes and mammary epithelial cells, leading to the development of chronic or persistent infections (7). Furthermore, antibiotic treatment of bovine mastitis is associated with the emergence of multidrug-resistant strains (8), including potentially methicillin-resistant S. aureus (MRSA) (9). The internalization and survival of S. aureus inside phagocytic and nonphagocytic cells are key factors in the pathogenesis of mastitis as strategies to evade both the immune system and the effects of antibiotics (10, 11).
Innate immune defenses in the mammary gland include physical epithelial barriers in the teat canal, the complement system, and, crucially, phagocytic functions (12). During acute bacterial inflammation, neutrophils and macrophages typically restrict bacterial proliferation and render infections self-limiting (13–15). For antibacterial defenses, neutrophils employ oxygen-dependent systems within their cytoplasmic phagolysosomes (NADPH oxidase in phagosome membranes and the production of bactericidal reactive oxygen species) (16) and oxygen-independent systems with the production of antimicrobial enzymes (17). A less-explored oxygen-independent mechanism that neutrophils and epithelial cells employ to fight infection is the production of small cationic host defense peptides (HDPs), also known as antimicrobial peptides (18), including cathelicidins (19). In humans, there is a single cathelicidin gene (cathelicidin antimicrobial peptide [CAMP]) that yields a C-terminal active peptide termed leucine-leucine with 37 amino acid residues (LL-37). Similarly, mice produce only one cathelicidin, the cathelicidin-related antimicrobial peptide (CRAMP) encoded by the Camp gene (20). Although LL-37 and CRAMP share <70% sequence identity in their active peptides, they are considered homologous based on their broadly similar functions (antimicrobial and immunomodulatory) and structural properties (α-helical and net charge of +6) (19). Indeed, human and murine cathelicidins have interspecies functional capacity: mice infected with influenza (H1N1) virus have enhanced survival and reduced viral titers after being nebulized with human cathelicidin LL-37 (21). Other mammals, including cattle, produce multiple cathelicidins, with bovine myeloid antimicrobial peptides (BMAPs) and bactenecins (Bacs) being most frequently described among at least seven types (19, 22).
Cathelicidins have been shown in previous studies to be protective in certain infectious epithelial diseases. Cathelicidin-deficient (Camp−/−) mice are more susceptible than wild-type mice to skin infections caused by group A streptococci (GAS) (20). Likewise, compared to wild-type mice, Camp−/− mice develop severe corneal keratitis due to Pseudomonas aeruginosa and have increased bacterial loads, neutrophil infiltration, and production of cytokines (CXCL-1, interleukin-1β [IL-1β], and tumor necrosis factor alpha [TNF-α]) (23). Similarly, compared to wild-type mice, Camp−/− mice develop severe Escherichia coli O157:H7 colitis (24). Even against protozoa such as Entamoeba histolytica, human intestinal epithelial cells have been shown to respond by producing cathelicidin (25, 26). The precise amounts of secreted cathelicidins have been closely measured under some human conditions. The plasma concentration of LL-37 decreased in healthy individuals compared with patients with sepsis (from 0.054 μM to 0.026 μM) (27) and increased (up to 280 μM) in inflamed skin of patients with psoriasis (28), but their inflammatory-immune function remains unclear. At the cellular level, cathelicidins seem to contribute to establishing inflammation. BMAPs upregulate the Tnf-α gene in bovine mammary epithelial cells (29), and S. aureus induces the production of Bac4 (30). Likewise, human cathelicidin LL-37 increases the transcription and synthesis of Toll-like receptor 4 (TLR4) in the presence of lipopolysaccharide (LPS), decreasing the invasion of enteric bacteria (E. coli and Salmonella spp.) in human (HT29) colonic epithelial cells (31, 32). However, cathelicidins can also have anti-inflammatory activity. For instance, LL-37 alleviates MRSA pneumonia by reducing lung inflammation and the release of cytokines (TNF-α and IL-6) in a murine model (33). Synthetic LL-37 also reduces TNF-α and IL-6 release through the inhibition of p38 mitogen-activated protein kinase (MAPK) and Akt phosphorylation in murine macrophages challenged with lipoteichoic acid (LTA) from S. aureus (34).
Cathelicidins may be equally important for innate mammary defense against mastitis in cattle. Cathelicidins are secreted by mammary glands into the milk of women, mice (35), and cows (36). In humans, cathelicidin synthesis is increased after 30 to 60 days of lactation (35). By dot blot analysis, the quantitative level of LL-37 in normal human milk was shown to be as high as ∼1 μmol/21.7 μg of total milk protein (milk protein concentration of 764 μg/ml) (i.e., 35 μM) (35). In cattle, cathelicidin genes are minimally expressed at the early stage of lactation but appear at midlactation (37). Cathelicidin concentrations in milk under physiological and pathological conditions are still uncertain and are likely dependent on factors such as age and lactational stage (35, 37). However, it is known that concentrations of bovine pancathelicidins correlate with high somatic cell counts (SCCs) (i.e., increased accumulation of cells, particularly leukocytes, in milk) (38) and that milk concentrations of cathelicidins are increased in cows with bacterial mastitis (36). Whether cathelicidins have any antimicrobial effects in milk is uncertain. A previous study demonstrated that bovine cathelicidins had killing activity against E. coli in culture medium but that only one particular cathelicidin, BMAP-27, had antibacterial activity in bovine serum or whey (the liquid remaining after milk is curdled and strained) (29). Whether cathelicidins exert such immune roles in the mammary gland and whether they play a role in mastitis are unknown.
In this study, we aim to evaluate potential roles of an endogenous murine cathelicidin (CRAMP; Camp) and synthetic human cathelicidin LL-37 in murine mammary gland defenses against S. aureus. First, we hypothesize that S. aureus infection will result in increased cathelicidin production by murine mammary cells. Second, we hypothesize that cathelicidins will reduce the severity of murine mastitis by reducing the internalization of S. aureus into mammary tissue and by modulating the expression of certain proinflammatory cytokines (including CXCL-1, IL-1β, and TNF-α). Finally, we hypothesize that the cellular mechanisms used to achieve these effects involve alterations in the expression of cell surface receptors (Toll-like receptors and CD36) and the MAPK p38 signaling pathway. To investigate these hypotheses, we use an in vivo mouse model, cultured murine mammary epithelial cells, and murine bone marrow-derived macrophages (BMMs).
RESULTS
Camp−/− mice develop more severe mastitis and higher mammary S. aureus burdens than do Camp+/+ mice.To address whether endogenous cathelicidins have any role in acute mastitis caused by S. aureus, wild-type Camp+/+ and cathelicidin knockout (Camp−/−) mice were intramammarily challenged with S. aureus isolated from a case of bovine mastitis. Histologically, all mice infected intramammarily with S. aureus developed mastitis, characterized by an influx of inflammatory cells, mostly neutrophils, into the alveolar lumen and dispersed throughout the parenchyma (Fig. 1A). Grades of mastitis, the severity of acute inflammation, and the extent of leukocyte infiltration were all higher in infected Camp−/− mice than in infected Camp+/+ mice (P < 0.05) (Fig. 1B). In terms of bacterial infection in the mammary gland, the S. aureus burden was higher in Camp−/− than in Camp+/+ mice (at 1 day; P < 0.05) (Fig. 1C).
Cathelicidin-deficient (Camp−/−) mice develop more severe mastitis and a higher mammary S. aureus burden than do Camp+/+ mice. (A) Representative histological sections from mammary glands of Camp+/+ and Camp−/− mice administered PBS or S. aureus. Photographs were taken using a 20× objective, and inset photographs were taken using a 40× objective. (B) Quantification of inflammatory cell infiltration in S. aureus-infected Camp+/+ and Camp−/− mammary glands (a P value of <0.05 [Mann-Whitney test] was considered significant). (C) Endogenous cathelicidin reduced S. aureus burdens in murine mammary glands. Camp−/− mice had higher S. aureus counts than did Camp+/+mice. Data are presented as means ± SEM (n = 7 to 8 mice per group). A P value of <0.05 (Student’s t test) was considered significant.
Bovine mastitis caused by S. aureus is characterized by the release of proinflammatory cytokines (39, 40). Likewise, intramammary challenge with S. aureus in lactating mice induces the release of proinflammatory cytokines, including CXCL-1, IL-1β, and TNF-α (1 and 2 days postinfection) (41–43). In our study, local transcriptional gene expression and secreted protein levels of Cxcl-1 and Il-1β were higher in infected Camp−/− mice than in infected Camp+/+ mice (P < 0.05) (Fig. 2A to D). Conversely, however, levels of Tnf-α transcription (Fig. 2E) and secretion (Fig. 2F) were higher in infected Camp+/+ mice than in infected Camp−/− mice. S. aureus-challenged mammary glands in both Camp+/+ and Camp−/− mice had similar increases in Ifn-γ mRNA (see Fig. S1A in the supplemental material) and decreases in Tlr2 mRNA (Fig. S1B). S. aureus induced higher cathelicidin mRNA transcription levels in the mammary glands of Camp+/+ mice (Fig. 2G), and as expected, no cathelicidin mRNA expression was observed in Camp−/− mice. Thus, we propose that murine cathelicidin aids in controlling S. aureus infection and in reducing local mastitis and the synthesis of the proinflammatory cytokines Cxcl-1 and Il-1β.
Inflammatory marker and cathelicidin expression differ between S. aureus-infected Camp+/+ and Camp−/− mammary glands. (A to D) Cxcl-1 gene expression (A), CXCL-1 protein (B), Il-1β gene expression (C), and IL-1β protein (D) levels were increased in S. aureus-infected Camp−/− mammary glands. (E and F) Conversely, Tnf-α gene expression (E) and TNF-α protein (F) levels were increased in S. aureus-infected Camp+/+ mammary glands. Data are shown as means ± SEM (n = 3 to 5 mice per group). A P value of <0.05 (Student’s t test) was considered significant. Mammary glands were harvested at 1 day postinfection. (G) S. aureus infection stimulated the transcription of cathelicidin in the mammary tissues of Camp+/+ mice (1 day postinfection). Data are presented as means ± SEM (n = 3 to 4 mice per group). A P value of <0.05 (Student’s t test) was considered significant.
Murine cathelicidin CRAMP promotes CXCL-1 in cultured murine mammary epithelial cells.Mammary epithelia are a source of host defense peptides, e.g., CRAMP in mice and the β-defensin LAP (lingual antimicrobial peptide) in cattle (35, 44). To determine whether cathelicidins promote defensive mechanisms in the mammary epithelium, we used murine mammary epithelial (HC11) cells, a functional model of mammary epithelial cells capable of producing milk casein (45). First, we determined whether the exposure of HC11 cells to S. aureus resulted in a proinflammatory response to the pathogen. Next, we investigated whether increasing amounts of cathelicidin peptide modified the immunomodulatory effects against S. aureus. HC11 cells exposed to S. aureus (up to a multiplicity of infection [MOI] of 10 for up to 4 h) increased transcriptional Camp (MOI of 5 at 30 min and 2 h) (Fig. 3A) and Cxcl-1 (MOI of 10 at 30 min and MOI of 5 at 2 h) (Fig. 3B) gene levels and CXCL-1 secreted protein levels (at 2 and 4 h, in a dose-dependent manner) (Fig. 3C). Secreted TNF-α was detected in HC11 cells but with no differences seen following challenge with S. aureus (MOI of 10 for 2 h) and/or CRAMP (Fig. S1C). Next, we ascertained whether increasing amounts of cathelicidin peptide modify immunomodulatory effects against S. aureus. The addition of synthetic CRAMP alone induced CXCL-1 secretion (P < 0.05) (Fig. 3D).
Murine cathelicidin CRAMP promotes CXCL-1 in cultured murine mammary epithelial cells. (A) S. aureus increased the gene expression of cathelicidin at 30 min and 2 h but not at 4 h postinfection. (B and C) Cxcl-1 transcription (B) and protein (C) were stimulated by S. aureus in a dose-dependent manner. (D) CRAMP (2 h) alone also induced the secretion of CXCL-1 in mammary epithelial cells. Data are presented as means ± SEM from three independent experiments in duplicate. A P value of <0.05 (one- and two-way ANOVAs and post hoc tests) was considered significant.
Cathelicidins decrease internalization of S. aureus into cultured murine mammary epithelial cells.Staphylococcus aureus is able to invade and survive in nonphagocytic cells, including mammary epithelia (46), fibroblasts (47), and endothelia (48). In our study, S. aureus was rapidly internalized into cultured mammary epithelial (HC11) cells. High S. aureus MOIs (from 10 to 50) were detected intracellularly at 1 h, with no major increases up to 4 h postchallenge (Fig. 4A); however, these high-MOI bacterial challenges were associated with host cell death (Fig. S1E). Therefore, lower infective doses of S. aureus (MOIs of 0.01 to 10) were tested to determine similar S. aureus internalization in HC11 cells, in a dose-dependent fashion during the first 2 h postinfection, and detectable internalization grades from challenges at an MOI of 1 (Fig. 4B). Based on these dose infection studies, we selected a challenge of S. aureus at an MOI of 1 for 2 h as the challenge dose that achieved maximum intracellular invasion while remaining noncytotoxic to cultured cells (Fig. 4B to D). We established that pretreatment of HC11 cells with either synthetic cathelicidin (human LL-37 or murine CRAMP) (0, 2, and 8 μM for 1 h) decreased the intracellular burden of S. aureus compared with nonpretreated cells (Fig. 4C and D).
Both human and murine cathelicidins reduce the internalization of S. aureus into cultured murine mammary epithelial cells. (A) S. aureus rapidly invaded the epithelium within 1 h postinfection at all MOIs tested and with similar rates after 4 h. (B) S. aureus at low MOIs (<1) had dose-dependent internalization into mammary epithelial cells after 2 h postinfection. (C and D) Synthetic human cathelicidin LL-37 (C) and murine cathelicidin CRAMP (D) decreased S. aureus (MOI of 1) internalization (at 8 μM). Data are presented as means ± SEM from at least three independent experiments in triplicate. A P value of <0.05 (one-way ANOVA and a post hoc test) was considered significant.
Human cathelicidin LL-37 inhibits S. aureus growth and is internalized into cultured murine mammary epithelial cells.We next investigated whether synthetic LL-37 and CRAMP doses (<35 μM [35]) had any direct killing effect on S. aureus. When cathelicidin peptides (up to 16 μM for up to 2 h) were incubated with S. aureus, high doses of LL-37 (4 to 16 μM) reduced bacterial growth (Fig. 5A). This microbicidal effect, however, was not observed with murine cathelicidin CRAMP at the doses used (Fig. 5B). As human cathelicidin LL-37 had some killing effect on S. aureus, we investigated whether this peptide was taken up by cultured murine mammary epithelial cells. 6-Carboxyfluorescein fluorescence-labeled LL-37 (FAM-LL-37) was detected inside cells, with distribution in the cytosol and around the nucleus (Fig. 5C).
Human cathelicidin LL-37 inhibits S. aureus growth and is internalized into cultured murine mammary epithelial cells. (A and B) Synthetic human cathelicidin LL-37 (A) impaired S. aureus proliferation in DMEM (2 h), whereas murine cathelicidin CRAMP (B) did not have any microbicidal activity. Data are presented as means ± SEM from at least two independent experiments in triplicate. A P value of <0.05 (one-way ANOVA and a post hoc test) was considered significant. (C) FAM-LL-37 was detected inside cultured murine mammary (HC11) epithelial cells, dispersed within the cytoplasm and around the cell nucleus.
Cathelicidin does not affect TLR2 signaling during internalization of S. aureus into cultured murine mammary epithelial cells.The internalization of S. aureus can be facilitated by TLRs, pattern recognition receptors that detect conserved pathogen molecules (pathogen-associated molecular patterns [PAMPs]). Among the TLRs, TLR2 senses S. aureus peptidoglycan (PGN), lipoproteins, and LTA (49) and facilitates the engulfment of S. aureus into the cytosol in leukocytes (50). The scavenger receptor type B CD36, a coreceptor of TLR2, has also been involved in S. aureus internalization, augmenting NF-κB activation and TNF production when combined with TLR2 and in response to LTA (51). In our study, the mammary epithelial Tlr2 gene (Fig. 6A) and TLR2 protein (Fig. 6B) were upregulated by S. aureus challenge (MOI of 5 for 2 h). However, the addition of exogenous LL-37 (8 μM) did not modify the cell surface expression of TLR2 (Fig. 6C) or CD36 (Fig. 6D) in cultured mammary epithelial cells, in either the presence or the absence of S. aureus challenge. Thus, although LL-37 reduced the invasion of S. aureus into mammary epithelia, this function was independent of TLR2 and CD36 cell surface expression.
Human cathelicidin LL-37 does not alter the surface expression of TLR2 or CD36 receptors in cultured murine mammary epithelial cells. (A and B) S. aureus infection upregulated Tlr2 gene expression (A) and total TLR2 protein (B) (2 h postinfection at an MOI of 5). (C and D) However, TLR2 (C) and CD36 (D) cell surface expression levels were similar in the presence of LL-37. Data are presented as means ± SEM from three independent experiments in triplicate. A P value of <0.05 (one-way ANOVA and a post hoc test) was considered significant.
Murine Camp+/+ BMMs release more TNF-α in response to S. aureus challenge than do Camp−/− BMMs.Macrophages are critical in mastitis; these cells are a major source of TNF-α (52), and mice depleted of macrophages are more susceptible to S. aureus infection (53, 54). In our study, bone marrow-derived macrophages (BMMs) from both Camp+/+ and Camp−/− mice secreted TNF-α in a dose-dependent manner after S. aureus challenge (2 h), with the highest increase present in Camp+/+ BMMs (Fig. 7A). It has been reported that S. aureus induces TNF-α through various signaling pathways, including MAPK p38 (52); however, in our study, the levels of phosphorylation of p38 were similar in Camp+/+ and Camp−/− infected BMMs (Fig. 7B). In another phagocytic cell model (murine [J774A.1] macrophages), S. aureus enhanced TNF-α secretion (2 h postchallenge) in a dose-dependent manner, but synthetic CRAMP (up to 2 μM) concurrent with S. aureus (up to an MOI of 10) did not have any effect on TNF-α production (Fig. S1D).
Murine Camp+/+ bone marrow-derived macrophages (BMMs) release more TNF-α in response to S. aureus challenge than do Camp−/− BMMs but with no differences in bacterial burdens. (A) Enhanced synthesis of the cytokine TNF-α was observed in infected Camp+/+ BMMs (2 h). (B) Phosphorylation of p38 was similar in Camp+/+ and Camp−/− infected BMMs. Data are presented as means ± SEM (n = 4 to 5 mice per group). A P value of <0.05 (Student’s t test) was considered significant. (C and D) BMMs isolated from Camp+/+ and Camp−/− mice challenged by S. aureus had similar intracellular (C) and extracellular (D) bacterial burdens (2 h postinfection). Data are means ± SEM (n = 4 to 5 mice per group). A P value of <0.05 (Student’s t test) was considered significant.
Levels of intracellular and extracellular S. aureus killing are similar in murine Camp−/− and Camp+/+ bone marrow-derived macrophages.Macrophages have several strategies to eliminate pathogens (e.g., phagocytosis, the production of reactive oxygen species, and the release of proinflammatory cytokines). Additionally, endogenous cathelicidins have been shown to improve mycobacterial killing by BMMs (55). In our study, however, BMMs isolated from Camp+/+ and Camp−/− mice challenged by S. aureus had similar intracellular (Fig. 7C) and extracellular (Fig. 7D) bacterial burdens (2 h postchallenge).
DISCUSSION
Using murine models, we demonstrated a role for cathelicidins in attenuating acute S. aureus mastitis. Experimental S. aureus infection induced cathelicidin gene transcription in both murine mammary glands and cultured mammary epithelial cells. Previously, in cows, cathelicidins have been detected in milk and were shown to be increased in association with elevated somatic cell counts after intramammary infection with Streptococcus uberis (38, 56). In a bovine mammary epithelial cell line, S. aureus infection has been shown to upregulate cathelicidins Bac1 and indolicidin (36). Cathelicidin expression has also been previously demonstrated in mammary lobules of lactating mice (35). We also showed that LL-37 and CRAMP reduced S. aureus internalization into the murine mammary epithelium and cultured mammary epithelial cells. A similar effect has been described for rat osteoblasts (57), human lung epithelial cells (58), and human macrophages (59), where LL-37 decreases intracellular S. aureus, Pseudomonas aeruginosa, and Mycobacterium paratuberculosis burdens, respectively.
Potential mechanisms for this reduction in bacterial burden have been investigated. Previously, LL-37 prevented Salmonella enterica serovar Typhimurium internalization in human colonic epithelial cells by upregulating TLR4 (31) and reduced the burden of different mycobacteria inside macrophages by promoting the fusion of phagosomes and lysosomes (55). Here, we investigated whether the cathelicidin-associated reduction in S. aureus internalization was caused by a direct bactericidal activity of cathelicidins. We showed that, when incubated with S. aureus, LL-37 killed S. aureus but that CRAMP had no effect on bacterial growth (i.e., killing activity). Consistent with these findings, LL-37 has previously been shown to be more effective than CRAMP in reducing S. aureus proliferation in vitro (60). We confirmed this lack of substantial direct bactericidal activity by CRAMP in our experiments by showing that Camp+/+ and Camp−/− BMMs had similar S. aureus burdens (at 2 h). Consistent with these results, Mycobacterium avium is recovered from both Camp+/+ and Camp−/− BMMs at similar levels (61), and E. coli phagocytosis is similar between peritoneal Camp+/+ and Camp−/− macrophages (62). Because LL-37 had some killing effect on S. aureus, we investigated whether this peptide was internalized by murine mammary epithelial cells after exogenous administration. Using fluorescence labeling, we showed that LL-37 was indeed rapidly taken up by cultured mammary epithelial cells, with distribution throughout the cytosol and around the nucleus. This finding suggests that after internalization, human cathelicidin might contribute to the prevention of the entry of S. aureus into the host cell and also to its intracellular killing.
Regarding CRAMP, we did not demonstrate direct microbicidal activity when it was incubated with S. aureus, in contrast to LL-37. However, we still showed a lower S. aureus burden and less severe mastitis in Camp+/+ mice than in Camp−/− (CRAMP-deficient) mice. This suggests that CRAMP attenuates S. aureus mastitis through some means other than direct extracellular killing of bacteria. A possible mechanism involves the uptake of secreted CRAMP by neutrophils, resulting in enhanced antimicrobial killing within neutrophils. It has been shown that Camp+/+ neutrophils, in which cathelicidins are highly concentrated within specific granules (63), kill intracellular S. aureus more efficiently than do Camp−/− neutrophils (64) and that the phagocytosis of IgG-opsonized S. aureus is impaired in Camp−/− mice (65). The killing effects of cathelicidins in neutrophils occur mainly in intracellular compartments (phagolysosomes) and are restricted in extracellular trap networks (64). The observed rapid internalization of LL-37 into murine epithelial cells could potentially also happen with CRAMP in murine neutrophils, augmenting their antibacterial activity. Enhanced antimicrobial activity by Camp+/+ neutrophils could explain the observed increased S. aureus burden in the mammary glands of cathelicidin-deficient mice. Additionally, a role for cathelicidins in enhancing intracellular killing by phagocytic cells could explain the comparatively reduced microbicidal effects of host defense peptides when secreted into physiological fluids (e.g., milk), where they are commonly blocked by anions, salts, or serum proteins. This supports an immunomodulatory role for murine cathelicidin rather than a direct bacterial killing action.
Tumor necrosis factor alpha (TNF-α) is a critical proinflammatory cytokine involved in the chemotaxis of neutrophils. We showed an expected increase in TNF-α production by mastitic mammary glands, in alignment with previous studies where TNF-α was increased during acute S. aureus mastitis in mice (41, 42, 54). Additionally, we demonstrated that cathelicidins enhance TNF-α production. The levels of transcription and secretion of TNF-α were higher in the mammary glands of S. aureus-infected Camp+/+ mice than in those of infected Camp−/− mice. Similarly, S. aureus-infected Camp+/+ BMMs released more TNF-α than did infected Camp−/− BMMs. Other types of cathelicidins also promote TNF-α synthesis; for example, bovine cathelicidins BMAP-27 and -28 induce TNF-α transcription in bovine mammary epithelial cells (29), and rat CRAMP increases TNF-α release from rat mast cells (66). The functions of TNF-α in mastitis could include improved neutrophil-based killing activity against S. aureus; TNF-α has been shown to enhance phagocytosis and oxidative bursts in bovine neutrophils (67).
We investigated whether the observed increase in TNF-α was mediated via the MAPK p38 downstream signaling pathway. Mechanistically, heat-killed S. aureus (HKSA) has been shown to promote TNF-α secretion by macrophages via this pathway (52). The phosphorylation of p38 has been reported to be enhanced in both Camp+/+ and Camp−/− BMMs challenged by LPS, although TNF-α secretion was only slightly increased in Camp+/+ BMMs (68). In our study, levels of phosphorylation of p38 were similar between Camp+/+ and Camp−/− BMMs. This apparent discrepancy between our findings and those of others may be due to undetected earlier p38 phosphorylation when modulating TNF-α secretion in BMMs (before 2 h). It may also be due to the use of different macrophage cell lines that respond differently to S. aureus; in a previous study, the level of secretion of TNF-α by peritoneal macrophages was higher than that by pulmonary alveolar macrophages when challenged by HKSA (52). Furthermore, S. aureus has been shown to induce alternate pathways, such as NF-κB, ERK (extracellular signal-regulated kinase), and JNK (c-Jun N-terminal kinase) phosphorylation, in RAW 264.7 macrophages (69). In all cases, NF-κB has been shown to be key in murine intramammary infections by either S. aureus or E. coli, although S. aureus produces comparatively low levels of proinflammatory TNF-α, monocyte chemoattractant protein 1 (MCP-1), and IL-1β and local neutrophil recruitment (70).
CXCL-1 is another cytokine involved in neutrophil chemotaxis. We hypothesized that cathelicidins upregulate CXCL-1 synthesis in mammary epithelial cells, thus recruiting neutrophils to sites of infection. Supporting this, LL-37 stimulates the homologue of CXCL-1, CXCL-8/IL-8, in human lung epithelial cells and gingival fibroblasts (71, 72). In our experiments, we showed that cultured mammary epithelial cells exposed to S. aureus increased Camp and Cxcl-1 gene transcription and CXCL-1 protein secretion in a dose-dependent manner. Unexpectedly, however, Camp−/− mice produced more CXCL-1 when infected intramammarily with S. aureus than did Camp+/+ mice. A possible explanation for this could be that Camp+/+ mice control mastitis earlier and more effectively by reducing the bacterial burden and inducing TNF-α production, thereby improving neutrophil efficacy at the onset of infection. A more rapid control of infection and a reduced need for further neutrophilic infiltration could thus reduce the secretion of CXCL-1. Another possible explanation is that the functions of cathelicidins may be tissue or infection specific due to their pleiotropic nature, driving either pro- or anti-inflammatory outcomes depending on the cell types and surrounding molecules. For example, LL-37 reduces TNF-α secretion in pneumonia caused by methicillin-resistant S. aureus in mice (33) and in macrophages challenged with LTA (34), and CRAMP inhibits interferon gamma (IFN-γ) production in spleens of Mycobacterium avium-infected mice (61). In contrast, the lack of cathelicidin promotes survival in a murine sepsis model by repressing the proinflammatory cytokines IL-1β and IL-6 (62).
We showed that cathelicidins reduced S. aureus internalization into the murine mammary epithelium and cultured mammary epithelial cells. As potential mechanisms for the cathelicidin-associated reduction in bacterial burdens, we investigated the effects of cathelicidins on the cell surface expression of TLR2 and CD36, receptors utilized by S. aureus to invade the mammary epithelium. Indeed, HeLa cells overexpressing CD36 have increased intracellular S. aureus burdens (73), and a blockage of CD36 in bovine mammary epithelial cells reduces the internalization of S. aureus (46). Moreover, LL-37 deactivates TLR2 in the presence of TRL2 agonists in dendritic cells (74) and inhibits CD36 receptors in adipocytes and hepatocytes (75). However, in our study, LL-37 did not alter the cell surface expression of TLR2 or CD36 in murine mammary epithelial cells exposed to S. aureus (MOI of 1). Therefore, LL-37 reduces the invasion of S. aureus into mammary epithelia, but this function is likely independent of TLR2 and CD36 cell surface expression. Alternative receptors may be involved in the cathelicidin-associated reduction in the bacterial burden. For example, S. aureus binds extracellular matrix (α5β1) integrins during invasion, and LL-37 is also a ligand of related integrins (αMβ2) in macrophages (76). Thus, bacteria and cathelicidins could compete for integrin receptors, thereby reducing bacterial internalization. A further understanding of udder immune-inflammatory responses and the function of cathelicidins in mastitis as they apply to dairy cows would require multiplex cytokine assays, studies using cathelicidins originating from cattle, and bovine primary mammary epithelial cells capable of producing milk proteins.
In conclusion, both human and murine cathelicidins inhibit S. aureus invasion into mammary epithelial cells, and human cathelicidin (LL-37) has direct bacterial killing ability. Cathelicidins could have a potential therapeutic role, alone or in combination with antibiotics, for controlling or preventing mastitis caused by S. aureus, particularly in dairy cattle. However, while human cathelicidin LL-37 is a valid cathelicidin prototype and has been previously tested in animal models, including murine lung diseases and rat mast cells (33), the effects of this peptide on preventing S. aureus internalization could be species specific; our findings should be applied cautiously to other cathelicidins, animal species, and disease conditions.
MATERIALS AND METHODS
Staphylococcus aureus.We used an S. aureus strain originally isolated from milk samples of cows with clinical mastitis (77) (Mastitis Pathogen Collection of the Canadian Bovine Mastitis and Milk Quality Research Network [CBMQRN]). This S. aureus strain was of particular interest as a true bovine mastitis pathogen. Whole-genome analysis of this S. aureus strain demonstrated the presence of virulence genes involved in the synthesis of extracellular adhesion protein (eap) (78), fibronectin-binding proteins (FnBPs) (fnbA and fnbB) (79), and an iron metabolism protein (iron-regulated surface determinant B [IsdB]) (80; J. De Buck, unpublished data) (Table 1). S. aureus was cultured on LB agar (Thermo Fisher Scientific) plates (18 to 24 h) and grown in LB broth (10 colonies in 10 ml at 37°C overnight) with shaking. An aliquot (1 ml) of the bacterial suspension was added to LB broth (3 ml) and cultured until an optical density measured at 600 nm (OD600) of 0.2 to 0.3 was attained. An aliquot of this culture was assessed for growth on agar plates to confirm CFU of 2 × 108/ml. Prior to cell infection, the bacterial suspension was centrifuged (3,000 × g for 10 min) and washed with phosphate-buffered saline (PBS).
Whole-genome analysis of the S. aureus strain used, originally isolated from milk samples of cows with clinical mastitis (Mastitis Pathogen Collection of the Canadian Bovine Mastitis and Milk Quality Research Network)a
Murine model of S. aureus-induced mastitis.Lactating female (8- to 10-week-old) C57BL/6 Camp+/+ and cathelicidin-null (Camp−/−) mice (B6.129X1-Camptm1Rlg/J; The Jackson Laboratory) were housed in a pathogen-free environment at the University of Calgary, with ad libitum access to feed and water. The cathelicidin gene Camp had exons 3 and 4 deleted, and there is no formation of mature peptide in these mice (20). Both Camp+/+ and Camp−/− mice were infected intramammarily (10 to 15 days after parturition) with either S. aureus (50 μl PBS containing 500 CFU) or an equal volume of PBS, instilled into the left and right fourth (L4 and R4) mammary glands (42). Prior to infection, an analgesic (0.05 ± 0.1 mg/kg of body weight of buprenorphine hydrochloride) was given subcutaneously to females, and pups were removed from the mothers. For inoculating bacteria, the distal end (1 to 2 mm) of the teat was excised, and the bacterial suspension was injected into ducts using a blunt smooth 31-gauge hypodermic needle. Mice were anesthetized with a combination of ketamine (100 mg/kg) and xylazine (10 mg/kg) injected intraperitoneally; they were subsequently euthanized at 1 day postinoculation. This short interval (1 day) was chosen because, in pilot experiments, mice had become lethargic and died 2 to 3 days after being challenged with S. aureus, irrespective of the dose used.
For collecting mammary tissues for culture and histological and immunological assessments (∼100 mg per piece), the area surrounding the 4th pair of mammary glands (L4 and R4) was disinfected with ethanol (70%), and the sample was excised across the length of the gland using a scalpel blade. For RNA extraction, the sample was weighed, and PBS (3 ml) was added in gentleMACS C tubes for further homogenization, using the adipose tissue program of the gentleMACS Octo dissociator (Miltenyi Biotec). Immediately after tissue dissociation, an undiluted homogenate (0.1 ml) and serial dilutions were plated onto LB agar and incubated at 37°C for 24 h. Residual homogenized samples were centrifuged (20,000 × g for 10 min at 4°C), and aliquots were stored at −80°C for later detection of cytokines.
Histological mastitis assessment.Mammary tissues were fixed in a formalin (10%) solution, embedded in paraffin wax, sectioned (5 μm), and stained with hematoxylin and eosin (H&E; Sigma-Aldrich). Inflammatory cell infiltrates, principally neutrophils, were graded in a blind manner by a veterinary pathologist based on the level of inflammatory cell infiltration and the percentage of tissue affected, as follows: 0 for no lesions and an absence of a cell infiltrate; 1 for a mild inflammatory infiltrate and undamaged tissue, with up to 30% of tissue affected; 2 for a moderate inflammatory infiltrate but little tissue injury, with 30% to 60% of tissue affected; and 3 for a severe inflammatory infiltrate and the presence of necrotic areas, with >60% of tissue affected, as previously described (42).
Culture of bone marrow-derived macrophages, mammary epithelial cells, and transformed macrophages.Studies on functional mammary epithelia were conducted on murine mammary epithelial (HC11) cells (provided by Carrie Shemanko, University of Calgary, Calgary, AB, Canada). These HC11 cells are capable of producing milk casein (45) and were tested for epithelial proliferation, signal transduction, and differentiation (81). HC11 cells were grown in RPMI 1640 medium containing fetal bovine serum (FBS) (10%), l-glutamine (2 mM), penicillin-streptomycin (100 U/ml), insulin (5 μg/ml), and epidermal growth factor (EGF) (0.01 μg/ml).
To determine the roles of cathelicidins in phagocytic cells under S. aureus infection, we utilized 2 types of macrophages. First, we used immortalized murine macrophages (J774A.1) (ATCC TIB-67). J774A.1 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) with high glucose, l-glutamine (2 mM), and penicillin-streptomycin (100 U/ml). RPMI 1640 medium, DMEM, antibiotics, FBS, insulin, and EGF were purchased from Thermo Fisher Scientific. Cells were maintained at 37°C with 90% humidity in an atmosphere containing 5% CO2. Second, we used bone marrow-derived macrophages (BMMs) from Camp+/+ and Camp−/− mice (obtained using a Cold Spring Harbor Laboratory protocol [82]). Briefly, bone marrow was collected from femurs and centrifuged (300 × g for 5 min). The resulting pellet was suspended in BMM medium, which is composed of RPMI 1640 medium supplemented with FBS (10%), antibiotics (1% penicillin and 0.1% streptomycin), β-2-mercaptoethanol (55 mM), and mouse fibroblast (L-929)-conditioned medium (10%) as a source of macrophage colony-stimulating factor required for macrophage lineage differentiation. BMMs were seeded in 12-well plates (4 × 105 cells/ml) with BMM medium for 6 days to promote differentiation into macrophages. Although we did not polarize the BMMs toward an M stage, the production of TNF-α by these macrophages when infected by S. aureus indicates possible M1 (proinflammatory) differentiation. This is based on the fact that induction to M1 occurs after incubation with LPS and certain proinflammatory cytokines. For instance, J774A.1 cells were stimulated to an M1 phenotype by incubation overnight with LPS (100 ng/ml) and IFN-γ (100 ng/ml) (83), and LPS and L-929-conditioned medium polarized murine BMMs (4 × 105 cells/ml) toward M1 (84).
S. aureus internalization into cultured mammary epithelial cells.To assess the roles of cathelicidins in preventing the invasion of S. aureus into cultured mammary epithelial cells, HC11 cells were seeded into 24-well plates (2.5 × 105 cells/ml) and grown in RPMI 1640 medium until reaching confluence (24 h). Cells were pretreated (1 h) with synthetic human cathelicidin LL-37 amide (H-Leu-Leu-Gly-Asp-Phe-Phe-Arg-Lys-Ser-Lys-Glu-Lys-Ile-Gly-Lys-Glu-Phe-Lys-Arg-Ile-Val-Gln-Arg-Ile-Lys-Asp-Phe-Leu-Arg-Asn-Leu-Val-Pro-Arg-Thr-Glu-Ser-NH2 trifluoroacetate salt) (catalogue number H-6224; Bachem) or murine cathelicidin CRAMP (H-Gly-Leu-Leu-Arg-Lys-Gly-Gly-Glu-Lys-Ile-Gly-Glu-Lys-Leu-Lys-Lys-Ile-Gly-Gln-Lys-Ile-Lys-Asn-Phe-Phe-Gln-Lys-Leu-Val-Pro-Gln-Pro-Glu-Gln-OH trifluoroacetate salt) (catalogue number 4056438; Bachem) (up to 8 μM). These cathelicidin doses (up to 8 μM) are considered physiologically relevant based on cathelicidin concentrations in human milk (35), plasma (27), and skin (28) and have been previously tested for studying the immunomodulatory roles of cathelicidins (57, 58). To eliminate residual peptides, HC11 cells were washed several times with PBS. HC11 cells were challenged with S. aureus (MOIs of up to 50 for up to 4 h at 37°C with 5% CO2), washed with PBS, and treated with RPMI 1640 medium supplemented with gentamicin (100 μg/ml for 1 h) to eliminate extracellular bacteria. Gentamicin was removed by repeated washes with PBS. For bacterial counting, distilled water was added to epithelial cells to release intracellular bacteria (20 min), and serial 10-fold dilutions of cell lysates were made and plated onto LB agar overnight to quantify CFU per milliliter. The bacterial burden in macrophages was measured over a relatively short time (2 h), consistent with other studies measuring S. aureus intracellular survival in macrophages (incubation from 90 to 180 min) (85, 86). This is because cathelicidins are cationic peptides that rapidly bind anionic microbial membranes (87), meaning that peptide killing should occur within the first few hours.
Immunofluorescence detection of human cathelicidin LL-37 in cultured mammary epithelial cells.A fluorescent human cathelicidin, FAM-LL-37 (catalogue number H-8284.0100; Bachem) was used to assess peptide internalization into host cells. HC11 cells were seeded in 8-well chambers (2.5 × 104 cells/well for 24 h) and exposed to FAM-LL-37 (8 μM for 1 h). Cells were washed three times with PBS and fixed with paraformaldehyde (VWR International) (4% for 15 min). Actin filaments were stained with Alexa Fluor 660 phalloidin (catalogue number A22285; Thermo Fisher Scientific), and nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole) (catalogue number A2248; Thermo Fisher Scientific). Slides were mounted using ProLong gold antifade mountant (Thermo Fisher Scientific) and visualized using an Olympus IX71 wide-field system.
Bactericidal activity of cathelicidins against S. aureus.The direct killing activity of human and murine cathelicidins against S. aureus was tested. An initial inoculum of S. aureus (optical densities of 0.15, measured at 600 nm) was incubated with synthetic LL-37 or CRAMP (up to 16 μM diluted in RPMI 1640 medium without FBS and antibiotics) for up to 2 h at 37°C. The remaining bacteria were calculated by the OD600 values.
Transcriptional expression of cathelicidins, TLRs, and proinflammatory cytokines in cultured murine mammary epithelial cells and in murine mammary glands.Gene expression levels of mouse cathelicidin (Camp), Tlr2, and the proinflammatory cytokines Cxcl-1, Il-1β, and Tnf-α were evaluated by quantitative real-time reverse transcriptase PCR (qRT-PCR). Protein-coding murine genes were named using Mouse Genomic Informatics guidelines (http://www.informatics.jax.org/mgihome/nomen/gene.shtml#kf). Total RNA was isolated from uninfected or infected HC11 cells or from mouse mammary tissue using TRIzol reagent (Thermo Fisher Scientific). Next, RNA (1 μg) was transcribed into cDNA using the qScript cDNA synthesis kit (Quantabio), and RNA and cDNA were measured (NanoVue spectrophotometer; GE Healthcare). Predesigned primers (RT2 qPCR [quantitative PCR] primer assay; Qiagen, Toronto, ON, Canada) specific for murine Camp (PPM25023A; GenBank accession number NM_009921.2), Tlr2 (PPM04220B; GenBank accession number NM_011905.3), Cxcl-1 (PPM03058C; GenBank accession number NM_008176.3), Il-1β (PPM03109F; GenBank accession number NM_008361), Tnf-α (PPM03113G; GenBank accession number NM_013693), and the reference gene Gapdh (glyceraldehyde-3-phosphate dehydrogenase) (catalog no. PPM02946E; GenBank accession number NM_008084.3) were used. These primers were experimentally verified for specificity and efficiency to ensure the amplification of a single product of the correct size with high PCR efficiency (>95%). The qPCR efficiency for each gene was calculated from the slope and determined by a linear regression model according to the equation efficiency = 10(−1/slope), as indicated in MIQE guidelines (88). The qRT-PCRs were performed in reaction mixtures containing cDNA (100 ng), SsoAdvanced universal SYBR green supermix (Bio-Rad) (1×), specific primers (0.5 μM each), and nuclease-free water (final volume of 10 μl). Each reaction was performed by an initial incubation step (95°C for 5 min), followed by a denaturation step (95°C for 5 s) and a combined annealing-extension step (60°C for 10 s), for a total of 40 cycles. Target gene mRNA values were corrected relative to the normalizer Gapdh. Data were analyzed using the
TLR2 and CD36 identification in cultured mammary epithelial cells.The protein expression of TLR2 in mammary HC11 cells was assessed by Western blotting. Cells were grown on 6-well plates (48 h until reaching 90% confluence) and challenged with S. aureus (up to an MOI of 10 for 2 h). Cells were scraped into PBS and centrifuged (5,000 × g for 5 min), and the pellet was incubated with denaturing cell extraction buffer (DCEB; Thermo Scientific) with protease inhibitors (1 μg) (catalogue number 11836153001; Roche) on ice. The suspension was mixed by vortexing and placed on ice (10 min). Following centrifugation (16,000 × g for 15 min at 4°C), the supernatant (protein) was collected. Protein concentrations were measured using a protein assay kit (Pierce bicinchoninic acid [BCA] protein assay kit; Thermo Scientific). Proteins were separated on an SDS-polyacrylamide gel and transferred to polyvinylidene difluoride (PVDF) membranes by electrophoresis. The PVDF membranes were blocked with bovine serum albumin (BSA) (5%) or skimmed milk (1 h) and incubated overnight with the following primary antibodies diluted in skimmed milk (5%) in Tris-buffered saline with Tween 20 (0.1%) (TBST): anti-TLR2 rabbit monoclonal IgG (1:1,000) (catalogue number 13744; Cell Signaling) and anti-GAPDH rabbit monoclonal IgG (1:1,000) (catalogue number 2118; Cell Signaling). Membranes were washed in TBST and incubated (1 h) with secondary peroxidase-AffiniPure goat anti-mouse IgG antibodies (catalogue number 115-035-146; Jackson ImmunoResearch Lab) diluted 1:10,000 in BSA (5%) in TBST. Chemiluminescence was detected using the Clarity Western ECL substrate (Bio-Rad), and bands were imaged using a ChemiDoc system (Bio-Rad). Densitometric quantification of bands was done with ImageLab 4.0.1 software (Bio-Rad).
The cell surface expression of TLR2 and CD36 receptors was assessed by flow cytometry with specific antibodies. HC11 cells (2.5 × 105 cells/well) were cultured until they reached 90% confluence on 24-well plates (24 h) and challenged with S. aureus (up to an MOI of 1 for 2 h). Cells were washed several times with PBS and detached with trypsin-EDTA. The cell pellet was recovered by centrifugation (5,646 × g for 5 min at 4°C) and washed twice with cold PBS (pH 7.4). Cells were fixed with paraformaldehyde (4% for 10 min at 4°C), followed by blocking with BSA (5%) in PBS (30 min at 4°C), with shaking. Cells were washed, centrifuged (300 × g for 5 min at 4°C), and incubated with BV421 rat anti-mouse CD282 (TLR2) (BD Biosciences) (1 μg/ml in PBS and 0.1% BSA) and allophycocyanin (APC) mouse anti-mouse CD36 (BD Biosciences) (2 μg/ml in PBS and 0.1% BSA overnight at 4°C). Cells were washed twice with stain buffer (BD Bioscience) and processed using a BD FACSCanto2 flow cytometer. Ten thousand events were collected and analyzed from individual cells.
Determination of p38 activation in BMMs.Total p38 and phosphorylated p38 from Camp+/+ and Camp−/− BMMs were detected using Western blotting. Cells were grown on 6-well plates for 6 to 7 days and challenged with S. aureus (MOI of 10 for 2 h). Protein extraction was performed as described above, and the primary anti-phospho-p38 MAPK rabbit monoclonal IgG antibody was diluted in BSA (5%) in TBST (1:1,000) (catalogue numbers 9212S and 4511S; Cell Signaling). Membranes were washed in TBST and then incubated (1 h) with secondary peroxidase-AffiniPure goat anti-rabbit IgG antibodies (catalogue number 111-035-144; Jackson ImmunoResearch Lab) diluted 1:10,000 in BSA (5%) in TBST. Chemiluminescence was detected using the Clarity Western ECL substrate (Bio-Rad), and bands were imaged using a ChemiDoc system (Bio-Rad). Densitometric quantification of bands was done with ImageLab 4.0.1 software (Bio-Rad).
Quantification of secreted inflammatory cytokines.The release of proinflammatory murine CXCL-1, TNF-α, and IL-1β cytokines in supernatants from cells and mammary tissues was determined by using sandwich enzyme-linked immunosorbent assays (ELISAs) (catalogue numbers DY453-05, DY410-05, and DY401-05; R&D Systems).
Statistical analyses.Experiments were conducted at least three times (independent experiments), and each experiment was repeated in duplicate. Data are presented as means ± standard errors of the means (SEM) to ascertain how far the sample mean of the data is likely to be from the true population mean. Differences between groups were determined using one-way analysis of variance (ANOVA) (followed by Tukey’s posttest) or unpaired Student’s t test, for normally distributed data. When data were not normally distributed, a nonparametric Kruskal-Wallis test, followed by a post hoc Mann-Whitney test, was used. All analyses were performed using GraphPad Prism (version 5.0; GraphPad). Differences were considered significant when the P value was <0.05.
Ethics statement.All studies in mice were conducted according to regulations specified by the Canadian Guidelines for Animal Welfare (CGAW) and were approved by the University of Calgary Health Sciences Animal Care Committee (approval number AC16-0061).
ACKNOWLEDGMENTS
This study was financially supported by a Natural Sciences and Engineering Research Council (NSERC) discovery grant (RGPAS-2017-507827), an Eyes High international collaborative grant for young researchers (10014539) (University of Calgary), and Dairy Research Cluster 3 (1049122) to E.R.C.
We thank John Kastelic for editing the manuscript and Jeroen De Buck and Carrie Shemanko from the University of Calgary for providing the S. aureus strain and the HC11 cell line, respectively.
We declare that the research was conducted in the absence of any commercial or financial relationship that could be construed as a potential conflict of interest.
P.A.C. performed the in vitro and in vivo experiments and acquisition and analysis of data and drafted the work. C.G.K. conducted the histopathological analysis and edited the manuscript. Y.-L.T. participated in performing cell culture and mouse experiments. A.P.A.M. participated in and was essential for the design of the intramammary infection. H.W.B. revised the manuscript critically for important intellectual content. P.A.C. and E.R.C. conceived this research, designed the experiments, and wrote the manuscript. All authors reviewed and approved the manuscript.
FOOTNOTES
- Received 15 April 2020.
- Accepted 17 April 2020.
- Accepted manuscript posted online 27 April 2020.
Supplemental material is available online only.
- Copyright © 2020 American Society for Microbiology.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
