Nocardia cyriacigeogica from Bovine Mastitis Induced In vitro Apoptosis of Bovine Mammary Epithelial Cells via Activation of Mitochondrial-Caspase Pathway

Nocardia is one of the causing agents of bovine mastitis and increasing prevalence of nocardial mastitis in shape of serious outbreaks has been reported from many countries. However, the mechanisms by which this pathogen damages the bovine mammary epithelial cells (bMECs) is not yet studied. Therefore, this study was designed with the aim to evaluate the apoptotic effects elicited by Nocardia and to investigate the pathway by which the Nocardia induce apoptosis in bMECs. Clinical Nocardia cyriacigeorgica strain from bovine mastitis was used to infect the bMECs for different time intervals, viz. 1, 3, 6, 12, and 18 h, and then the induced effects on bMECs were studied using adhesion and invasion assays, release of lactate dehydrogenase (LDH), apoptosis analysis by annexin V and propidium iodide (PI) double staining, morphological, and ultrastructural observations under scanning electron microscope (SEM) and transmission electron microscope (TEM), mitochondrial transmembrane potential (ΔΨm) assay using flow cytometry, and the protein quantification of mitochondrial cytochrome c and caspase-9 and caspase-3 by western blotting. The results of this study showed that N. cyriacigeorgica possessed the abilities of adhesion and invasion to bMECs. N. cyriacigeorgica was found to collapse mitochondrial transmembrane potential, significantly (p < 0.05) release mitochondrial cytochrome c and ultimately induce cell apoptosis. Additionally, it promoted casepase-9 (p < 0.01) and casepase-3 (p < 0.05) levels, significantly (p < 0.01) increased the release of LDH and promoted DNA fragmentation which further confirmed the apoptosis. Furthermore, N. cyriacigeorgica induced apoptosis/necrosis manifested specific ultrastructure features under TEM, such as swollen endoplasmic reticulum, cristae degeneration, and swelling of mitochondria, vesicle formation on the cell surface, rupturing of cell membrane and nuclear membrane, clumping, fragmentation, and margination of chromatin. The present study is the first comprehensive insight into patho-morphological ultrastructural features of apoptosis/necrosis induced by N. cyriacigeorgica, which concluded that the clinical N. cyriacigeorgica induced apoptotic changes in the bMECs through mitochondrial-caspase dependent apoptotic pathway.

Bacterial adhesion and invasion are considered as important pathogenetic and virulence factors in the infection processes (Dego et al., 2002). Several in vivo and in vitro experiments demonstrated that Nocardia possessed the abilities to adhere to and invade into various types of cells, inducing cellular and tissue damages (Beaman and Beaman, 1998;Chapman et al., 2003;Beaman and Tam, 2008;Kohbata et al., 2009). When Nocardia attached to and rapidly penetrated through capillary endothelial cells (Beaman and Ogata, 1993), then entered the brain parenchyma, eliciting Lewy body inclusion in brain and Parkinson's symptoms in experimental animals (Chapman et al., 2003;Beaman and Tam, 2008). A previous study reported that Nocardia infection may induce macrophages and dendritic cells to differentiate into foamy cells (Meester et al., 2014). Furthermore, the invasion of Nocardia can even lead to the prevention of phagosome-lysosome fusion), inhibition of proteasome activity (Barry and Beaman, 2007), resistance to oxidative killing, blockage of phagosomal acidification, and alteration of lysosomal enzyme activity in macrophages (Beaman and Beaman, 1994). Staphylococcus aureus adhesion and invasion to bovine mammary epithelial cells (bMECs) has been proven to be the key events in the pathogenesis of bovine mastitis and the infected cells exhibited apoptotic morphology (Bayles et al., 1998;Dego et al., 2002); but for Nocardia, the adhesion and invasion ability to bMECs and the cell death effects are still unclear.
Cell death, the ultimate consequence of injury in host cells infection mainly includes apoptosis and necrosis. Necrosis is characterized by loss of cell membrane integrity, release of cellular contents and motivating inflammatory reaction; whereas, apoptosis is generally developed with cell membrane integrity, internucleosomal DNA fragmentation, formation of apoptotic bodies, no inflammatory reaction, and mediated through the intrinsic and extrinsic apoptosis pathway (Lamkanfi and Dixit, 2010). The extrinsic apoptosis pathways involve death receptors and caspase-8 signaling; whereas the intrinsic apoptotic pathway mainly targets the mitochondria (Lamkanfi and Dixit, 2010;Galluzzi et al., 2012).
In many bacterial pathogens, mitochondria-dependent apoptotic pathways have been well-recognized as a major pathogenic strategy (Ashida et al., 2011). During the process, apoptotic factors opens the mitochondrial permeability transition pore that ultimately result in loss of membrane potential and activation of cytochrome c (Yang et al., 2015;Xu et al., 2016). Then cytochrome c binds to apoptotic protease-activating factor I (Apaf-I) and tempts oligomerization. Consequently, apoptosome complex form by assembling of Apaf-1 oligomers with procaspase-9. Caspase-9 subsequently triggers apoptosis by caspase-3 activation (Elmore, 2007;Lamkanfi and Dixit, 2010). However, the specific mechanism through which capsase-9 activation triggers apoptosis in response to N. cyriacigeorgica in bMECs is unclear. Nocardia was shown to induce apoptotic death in dopaminergic cells, PC12 cells and HeLa cells; meanwhile, disruption of the mitochondrial membrane potential and caspase activation were involved in the apoptosis of HeLa cells (Barry and Beaman, 2007). However, the cell death effect of Nocardia on bMECs and the specific mechanisms involved in response to nocardial infection remain unknown.
Although, most of studies on Nocardia infections in various cells and laboratory animals were performed to demonstrate the pathogenicity and pathogenic mechanisms in central nervous system, respiratory system, and skin or cutaneous tissues (Barry and Beaman, 2007;Beaman and Tam, 2008;Meester et al., 2014;Lira et al., 2016). Nevertheless, there are rare studies focused on pathogenicity and mechanism underlying bovine mastitis caused by N. cyriacigeorgica. Therefore, the current study was designed with hypothesis that the Nocardia could adhere to and invade into bMECs, inducing apoptotic and necrotic cell death; in addition, Nocardia may regulate the cell apoptosis via mitochondrial-caspase pathway.

Cell Culture
The bMECs line MAC-T was used in this study which was purchased from Shanghai Jingma Biological Technology Co., Ltd. China. Cells were cultured in DMEM/F-12 (HyClone, USA) supplemented with 10% heat-inactivated Gibco R Fetal Bovine Serum (FBS; HyClone, USA), 100 U/mL penicillin and 100 µg/mL streptomycin at 37 • C with 5% CO 2 . Cells for adhesion and invasion assay were cultured in DMEM/F12 medium without antibiotics and while for other assays bMECs were cultured in DMEM/F12 medium with 4% FBS without antibiotics.

Bacterial Culture
N. cyriacigeorgica isolated previously from bovine mastitis was activated from frozen stocks by culture on tryptose soya agar (Difco TM , Becton Dickison, Sparks, MD USA) supplemented with 5% defibrinated sheep blood and incubated at 37 • C for 72 h, then sub-cultured in 7H9 broth to mid-log phase for the following experiments.

Adhesion Assay
Adhesion assay of clinical N. cyriacigeorgica to bMECs was performed according to previously described protocols (Pöhlmann-Dietze et al., 2000;Pereyra et al., 2016), with slight modifications as bMECs were infected with N. cyriacigeorgica at a multiplicity of infection (MOI, ratio of N. cyriacigeorgica to cells) of 50:1 for 10 min, 1, 2, and 3 h at 37 • C with 5% CO 2 . After incubation, the supernatants of infected cells were removed and cells were washed three times with phosphate buffer saline (PBS, pH 7.4) to remove non-adherent bacteria. Subsequently, cells were lysed by 1 mL PBS and 1 mL 1% Triton X-100 (0.5% v/v) to release adherent Nocardia. In control groups, both Nocardia suspensions (1 mL) and the cells were also treated with 1 mL Triton X-100. Finally, cell lysates and treated Nocardia supernatants were 10-folds serially diluted, plated onto sheep blood agar plates and incubated at 37 • C for 48 h for enumeration of colony forming units (CFU). Adhesion rate of total Nocardia was expressed as: Adhesion rate of total Nocardia = Lysate of infected cells (CFU/mL) Lysate of Nocardia supernatant and infected cells (CFU/mL) × 100 The adhesion assay was repeated four times and each experiment was performed in quadruplicate.

Invasion Assay
In our previous study, we tested several dilutions of amikacin (25, 50, 75, and 100 µg/mL) with high concentration of Nocardia (1 × 10 7 CFU/mL) and with different incubation time (1, 2, and 3 h). As a result, we found that 50 µg/mL of amikacin was enough to kill all Nocardia within 2 h; thus, this concentration of amikacin and incubation time was used for the following invasion assay. Invasion assay of N. cyriacigeorgica was performed according to the previous method with minor modifications (Beaman and Beaman, 1998;Pereyra et al., 2016). The bMECs were infected with N. cyriacigeorgica at a MOI of 50:1 for 10 min, 1, 2, and 3 h. Following incubation, cells were washed three times with PBS and treated with amikacin (50 µg/mL) for 2 h to kill extracellular Nocardia. The infected cells without amikacin treatment were used as control group. Then, cells were washed three times with PBS to remove non-adherent bacteria, further lysed with 0.5% Triton X-100 (v/v). Finally, cell lysates were 10folds serially diluted for CFU determination. Invasion rate of adhered Nocardia was expressed as: Invasion rate of adhered Nocardia = Lysate of infected cells with amikacin treatment (CFU/mL) Lysate of infected cells without amikacin treatment (CFU/mL) × 100 Invasion rate of total Nocardia was expressed as: Invasion rate of total Nocardia = Adhesion rate of total Nocardia ×

Invasion rate of adhered Nocardia
The invasion assay was repeated four times and each experiment was performed in quadruplicate.

Lactate Dehydrogenase (LDH) Release Assay
LDH, an enzyme, normally present in the cytoplasm and could release into the cell culture medium through damaged cell membrane after bacterial infection (Loeffler et al., 2004;Viguier et al., 2009). LDH release assay can be used to evaluate cytopathic effect of N. cyriacigeogica on bMECs. Cells were infected with N. cyriacigeorgica at a MOI of 5:1 for 1, 3, 6, 12, and 18 h at 37 • C with 5% CO 2 . In addition, Nocardia suspensions in DMEM/F12 with 4% FBS and noninfected cells incubated for the same time were used as control. After incubation, the supernatants were collected and centrifuged at 18,000 g for 15 min. The supernatants were collected again for measurement of LDH release by cytotoxicity LDH Assay Kit-WST R (Dojingdo Laboratories, Kumamoto, Japan). The absorbance at 490 nm was measured by a microplate reader.

DNA Ladder Analysis
As a characteristic feature for apoptotic cell death, DNA ladder was applied for the verification of cell apoptosis. The bMECs were infected with N. cyriacigeorgica in the same manner as described above. Moreover, cells without Nocardia treatment were used as control. Fragmented DNA was collected from cells according to the manufacturer's instructions of the DNA Ladder Extraction Kit (Beyotime, China). Moreover, this kit was also used to extract the DNA of N. cyriacigeorgica to determine the interference effect of N. cyriacigeorgica DNA. Finally, the DNA samples were electrophoresed on a 1% agarose gel, stained with ethidium bromide, and photographed by gel documentation system (Alpha R Imager EC, SAN LEADRO, USA).

Apoptosis/Necrosis Analysis by Annexin V/Propidium Iodide (PI) Double Staining
Cell death was detected by the FITC Annexin V Apoptosis Detection Kit I (BD, USA) according to the manufacturer's instructions. N. cyriacigeorgica were incubated with bMECs at a MOI of 5:1 for 1, 3, 6, 12, and 18 h at 37 • C with 5% CO 2 . Cells without Nocardia treatment were used as the control groups.
After incubation, cells were harvested, washed twice with cold PBS and re-suspended in binding buffer at a concentration of 1 × 10 6 cells/mL, then 100 µL of the cell suspension was stained by 5 µL FITC Annexin V and 5 µL PI. Following incubation for 15 min at room temperature, the samples were analyzed by the BD FacsCalibur flow cytometer (New Jersey, USA) within 1 h.

Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM)
The bMECs adhered on coverslips were washed three times with cold PBS and fixed with 2.5% glutaraldehyde at 4 • C for 1.5 h After washing, cells were dehydrated through a graded series of ethanol (30, 50, 70, 80, 90, 100, and 100% ethanol) for 15 min in each at room temperature. Subsequently, cells were immersed into tert-butyl alcohol for 30 min. After lyophilization and gold coating, these cell samples were observed on a scanning electron microscope (Hitachi S-3000N, Japan). For TEM, cells were harvested and the pretreatment of the cells before dehydration was similar to that for SEM. After dehydration by graded ethanol and acetone (three changes, for 10 min each), cells were sequentially embedded in epoxy resinacetone mixtures (2:1) for 2 h and in pure resin overnight at 37 • C. When the resin had polymerized, ultra-thin sections were cut by an ultramicrotome (Leica EM, Germany), stained with 1% uranyl acetate followed by lead citrate and viewed on a transmission electron microscope (Hitachi H-7650, Japan).

Mitochondrial Transmembrane Potential ( M) Assay
The bMECs were infected with N. cyriacigeorgica at a MOI of 5:1 for 1, 3, 6, 12, and 18 h at 37 • C with 5% CO 2 . Cells were collected for mitochondrial damage detection according the changes of m. The m was measured using a mitochondrial membrane potential assay kit with JC-1 (Beyotime, China) by flow cytometry. JC-1 was a dual-emission potential-sensitive probe and formed red-fluorescent aggregates in the mitochondria of the cells with higher potentials. But membrane potential collapse could result in the failure to red JC-1 in the mitochondria and the dye return to green-fluorescent monomer.

Western Blot Analysis
The bMECs were infected with N. cyriacigeorgica at a MOI of 5:1 for 1, 3, 6, 12, and 18 h at 37 • C with 5% CO 2 . Total proteins and cytoplasmic proteins without mitochondrion were respectively extracted from nocardial infected cells with RIPA lysis buffer (Beyotime, China) and Cell Mitochondria Isolation Kit (Beyotime, China). The protein concentrations were detected by the BCA method. Equivalent proteins from each sample were separated by SDS-PAGE gel, and then transferred onto PVDF membranes. Subsequently, the membranes were blocked in 5% BSA and incubated with the primary antibody for Cytochrome C (1:200, Santa, USA), caspase-9 (1:500, Santa, USA), caspase-3 (1:100, Santa, USA), and tubulin (1:1,000, Cell Signaling Technology, USA) overnight at 4 • C followed by incubation with HRP-conjugated secondary antibody (1:5,000) for 1 h at room temperature. Finally, the bands were visualized using a BeyoECL Plus ECL Kit (Beyotime, China). Densitometric analysis of the bands was quantified using Image J and these results were normalized using β-actin.

Apoptosis Analysis in Parallel with the Growth of Intracellular N. cyriacigeorgica
This experiment was carried out to evaluate the apoptosis and intracellular growth of Nocardia by killing the extracellular bacteria. The bMECs were infected with N. cyriacigeorgica at a MOI of 5:1 in DMEM/F12 for 2 h. Following co-cultures, cells were washed with PBS and treated with amikacin (50 µg/mL in DMEM/F12) for 2 h to kill extracellular Nocardia. Then, cells were washed three times with PBS and cultured in DMEM/F12 medium with 4% FBS for 0 h (control group), 1, 3, 6, 12, and 18 h at 37 • C with 5% CO 2 for the following steps: (i) for cell apoptotic analysis, cells were collected, after incubation, for apoptotic analysis by the FITC Annexin V Apoptosis Detection Kit I (BD, USA) as previously described; (ii) for total bacterial counting, the medium (1 mL) with infected bMECs was treated with 1 mL 1% Triton X-100 (v/v) for the CFU enumeration of total Nocardia. Simultaneously, the parallel infected bMECs were washed three times with PBS, further lysed with Triton X-100 and 10-folds serially diluted for CFU determination of the intracellular Nocardia. The number of Nocardia was expressed as Log 10 CFU/Well. The bacterial counting was repeated four times and each experiment was performed in quadruplicate; (iii) for observation of bacteria Gram staining was carried out. After washing with PBS, cells adhered on coverslips were fixed with 4% paraformaldehyde for 20 min and then permeabilized with 0.2% Triton X-100 (v/v) for 10 min at room temperature. Finally, cells were washed, stained by Gram staining and observed under light microscope (Olympus, Japan).

Data Interpretation
Each experiment was repeated at least three times and the data were expressed as mean ± standard deviation (SD). Statistical differences between groups were analyzed by One-way ANOVA followed by the Duncan and LSD multiple tests using SPSS 20.0 (SPSS, Inc., Chicago, IL, USA). P-value < 0.05 was regarded as statistically significant.

Adhesion and Invasion Assay
The preliminary quantitative study is shown in Table 1. The adhesion rate of total N. cyriacigeorgica increased from 4.5 to 42.8% within 3 h and this increase was time-dependent. Whereas, the invasion rate of total Nocardia also showed a time-dependent increase from 3.3 to 33.35% within 3 h. On the other hand, the invasion rate of adhered bacteria was persistent at a range of 64.4 to 77.9% within 3 h, indicating that more than 60% of the adhered Nocardia can invade into cells. Remarkably, these data showed the rapid adhesive and invasive capabilities of N. cyriacigeorgica with bMECs (Table 1).

LDH Release Assay
As shown in Figure 1, there was release of LDH at 1, 3, and 6 h post-infection, however at 12 and 18 h post-infection there was a Invasion rate of total Nocardia = Adhesion rate of total Nocardia × Invasion rate of adhered Nocardia.
FIGURE 1 | Lactate dehydrogenase (LDH) release assay in the medium. The LDH release assay was repeated three times and each experiment was performed in triplicate. Results were presented as Mean ± SD. **P < 0.01 as compared to the control group.
significant increase (p < 0.01) of LDH as compared with control group. This indicated that N. cyriacigeorgica seriously damage the cell membranes of bMECs with the passage of time.

DNA Ladder Assessment
DNA fragment is shown in Figure 2. There was no appearance of DNA ladder band in the bMECs of control group, 1, 3, and 6 h groups and also in N. cyriacigeorgica control group. But in the infected groups at 12 and 18 h, distinct typical DNA ladder bands were observed. These results indicated that N. cyriacigeorgica could induce apoptosis/necrosis in bMECs and the apoptotic DNA fragment was observed after 12 h post-infection.

Apoptosis/Necrosis Analysis
Annexin V/PI double staining was used to analyze the early apoptotic cells and late apoptotic/necrotic cells. control group. Furthermore, both at 12 h and 18 h, the late apoptosis/necrosis rate was much higher (p < 0.01) than early apoptosis rate. Additionally, the total death rate at 18 h (52.26%) was significantly higher (p < 0.01) than that at 12 h (46.99%). These data suggested that N. cyriacigeorgica could cause both apoptosis and necrosis, which was seriously intensified in time dependent manner.

Scanning Electron Microscopy (SEM)
To further confirm the adhesion, invasion, LDH release and cell death, SEM was used to test the interaction of bMECs infected with N. cyriacigeorgica and the cell damages. There were no morphological changes in control group (Figure 4A). At 6 h post-infection, morphological changes were observed as mild disruption of the some bMECs and short Nocardia filaments adhered on the surface of cells (Figures 4B,C). At 12 and 18 h post-infection, bMECs were obvious shrinkage, cytomorphosis, desquamation, and cell membrane breakage, with spherical protrusion appearance on some cell surface, abundant growing mycelium covering the cells and filaments penetrating into cell membrane both from the extracellular and intracellular (Figures 4D-I).
Particularly, at 6, 12, or 18 h, SEM results demonstrated that microvilli on the cell surface were totally wrapped by Nocardia filaments (Figures 4E,H,I). Thus, these changes to the cellular morphology suggested that the penetration and viability of N. cyriacigeorgica may be responsible for aforesaid cell damages. Finally, serious disruption of cells, loss of organelles, and intramitochondrial dense granules accumulation were observed, at 18 h post-infection (Figures 5J-L). Furthermore, Nocardial cells were found both in cytoplasm and nucleus. These observations showed a process of progressive cell necrosis induced by clinical N. cyriacigeorgica infection.

Mitochondrial Transmembrane Potential ( m) Assay
Following a relatively stable state of m in the early stage (1-6 h) of nocardial infection, there was a remarkable decrease (p < 0.01) of m in the late stage of infection viz. 12-18 h in comparison with the control group (as shown in Figure 6).

Western Blot Analysis
The western blotting results of cytochrome c, caspase-9 and caspase-3 are depicted in Figure 7. A gradual increase in mitochondrial cytochrome c release was observed after 1 h postinfection (p < 0.05), with significantly higher caspase-9 activation (p < 0.01) from 1 h and markedly enhanced caspase-3 (p < 0.05) activation from 3 h of infection as compared to the control group.

Apoptosis Analysis in Parallel with the Growth of Intracellular N. cyriacigeorgica
The growth of intracellular Nocardia showed that the apoptosis rate was increased significantly (p < 0.01) from 12 h compared with control group (Figures 8A,B). At 12 and 18 h, most of the dead cells were at the early apoptotic stage (p < 0.01). In addition, the total death rate of bMECs at 18 h (19.70%) was significantly higher (p < 0.01) than at 12 h (15.80%). Strikingly, Gram staining showed that the Nocardia were short rod-shaped with no considerable growth during 1-6 h; nevertheless, they germinated into unique long mycelial form at 12 and 18 h, leading to pathomorphological changes of infected cells ( Figure 8C). Bacterial counting presented an interesting results that there were no significant difference of the number of total Nocardia at different incubation time compared with control group. Similar results for the enumeration of intracellular Nocardia were also found within 12 h. This was in accordance with the apoptosis analysis, which presented that the apoptosis significantly increased at 12 and 18 h. The intracellular bacteria significantly decreased (p < 0.01) at 18 h compared with control group (Figure 8D). The unchanging number of Nocardia and the results of Gram staining revealed that both intracellular and extracellular N. cyriacigeorgica grew from small rods to mycelia rather than to reproduce during 18 h incubation (Figures 8C,D). During the robust growth phase of Nocardia at 12-18 h, cell apoptosis was aggravated remarkably (Figures 8A-C).  SEM and TEM showed that N. cyriacigeorgica possess the adhesion and penetration abilities. Significantly, we observed that N. cyriacigeorgica could promptly adhere to and penetrated into the bMECs within 10 min. These results were in accordance with a previous study which reported 68% adherence rate and 70.8% penetration rate of N. asteroides GUH-2 (N. cyriacigeorgica GUH-2) to endothelial cells in brain within 15-25 min exposure period (Beaman and Ogata, 1993). The findings of this study and some previous studies suggested that Nocardia has the capability of adhesion, invasion and viability within various host cells both in vitro and in vivo (Beaman and Ogata, 1993;Beaman, 1994, 1998).

DISCUSSION
In the log-phase N. cyriacigeorgica depicted strong penetrative and adhesion ability than the stationary-phase and only live log-phase organisms can penetrate into cells (Beaman and Beaman, 1998). Nocardia can adhere to cells through surface adhesion molecules, live, and heat-killed Nocardia both can bind to the cell surface. Most of studies demonstrated that a filamentous tip associated 43-kDa protein of N. asteroides GUH-2 played a dominant role in attachment and invasion to pulmonary epithelial cells and HeLa cells (Beaman and Beaman, 1998). The antiserum against the 43-kDa antigen inhibited apical adhesion and penetration to pulmonary epithelial cells, and prevented spread to the brain. Through gene sequence analysis of N. cyriacigeorgica, a series of putative mammalian cell entry proteins (mce) of N. cyriacigeorgica (Vera-Cabrera et al., 2013;Zoropogui et al., 2013) were found and these proteins were essential for pathogens to attach, enter and survive in the host cells (Zhang and Xie, 2011). Additionally, adhesion of Nocardia was associated with the cell surface structures, such as microvilli. Cell surface provides a direct connection with the bacteria. As evident from the SEM results, this presented that the microvilli of bMECs can capture Nocardia. This phenomenon was also observed previously on other cells Beaman, 1994, 1998).
After exposure to bMECs, N. cyriacigeorgica infection led to perforation in cell membranes, release of LDH, fragmentation of nuclear membrane, chromatin condensation and mitochondrial degeneration, which can caused untimely collapse of the entire cell. Previous studies suggested that the pathogenicity of Nocardia is dependent on the virulence factors and toxins. The putative virulence factors of N. cyriacigeorgica included catalase, superoxide dismutase, hemolysin, invasion, protease, mammalian cell proteins (mec), mycolic acids, nitrate reductase, and PE/PPE/PGRS family proteins through gene analysis (Vera-Cabrera et al., 2013;Zoropogui et al., 2013), among most of them had been proved. Catalase and superoxide dismutase of N. cyriacigeorgica probably acted important roles in defense against deleterious superoxide and reactive oxygen species (ROS) during intracellular killing by phagocytes (Wu et al., 2006). The presence of nitrate reductase for  The western blot assays of cytochrome c, cleaved caspase-9 and cleaved caspase-3 expression at 1, 3, 6, 12, and 18 h after N. cyriacigeorgica infection in bMECs. Results were presented as Mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 as compared with the control group.
N. cyriacigeorgica suggested an ability to grow under lowoxygen conditions in stimulated macrophages (Zoropogui et al., 2013). Significantly, mycolic acids, the major and specific lipid components of cell envelope, have important implications in the pathogenesis of Mycobacterium, Nocardia and Rhodococcus, and Corynebacterium (Elamin et al., 2012;Verschoor et al., 2012). The cell wall-associated lipids of Nocardia induced the production of the proinflammatory cytokines and inhibited important macrophage microbicidal effects (Trevino-Villarreal et al., 2012).
Bacteria often elicited cell apoptosis as a survival strategy. Pathogens have evolved a series of toxins and virulence factors to modulate host cell death (Lamkanfi and Dixit, 2010). During the first 6 h of Nocardia infection, there were only several From 1-6 h, only few short rods of Nocardia can be observed (black arrows); while from 12 to 18 h, short rods grew into long mycelia (black arrows). (D) The number of total Nocardia and intracellular Nocardia. Data were presented as Mean ± SD of at least three independent experiments. *P < 0.05, **P < 0.01 as compared with the control group. ## P < 0.01 as compared between the rate of early and late apoptosis/necrosis at the same time point. § § P < 0.01 represents the significant differences in total cells death rate between 12 and 18 h.
short nocardial filaments on the cell surface imparting mild morphological changes in bMECs, and there were also no obvious cytotoxic effects and m, following no or low number of dead cells. Nevertheless, during the late stage of infection, short nocardial mycelia grew into long and strong mycelia, which can invade and penetrate into the bMECs, and even grew within cells; simultaneously, N. cyriacigeorgica was found to significantly elicit cell apoptosis, and exacerbated changes in cell morphology and ultrastructure, with a typical fragmentation of DNA and a collapse of m from 12 h. Besides, there was robust release of LDH from 12 h, which was associated with the serious cell membrane breakage and the large number of late apoptotic/necrotic cell in this period. A previous study demonstrated that live Nocardia can invade into cells and significantly increase in apoptosis rate in infected cells compared to heat-killed Nocardia (Barry and Beaman, 2007). Considering these above results, we can assume that the penetration and aggressive growth of N. cyriacigeorgica mycelia might be one of the factors to induce death of bMECs. Additionally, on PC12 cells, Nocardia culture filtrate was found to induce apoptotic morphology after 24 h-treatment and increased release of LDH after 48 h (Loeffler et al., 2004), indicated that the metabolites of Nocardia are toxic to cells.
An extra experiment was performed to know the cytopathic and apoptotic effects of only intracellular Nocardia, which was coupled with study regarding the growth of intracellular bacteria. For this purpose the extracellular bacteria were killed with amikacin and apopotosis and growth of the intracellular bacteria was evaluated. Interesting, the results showed that apoptosis and bacterial growth have a parallel relationship. These effects were significantly exhibited during 12 and 18 h. Importantly, there was no significant changes in the total number of Nocardia but the bacteria grew into mycelial form, imparting cell damage and apoptosis. At 12 and 18 h of infection, a rapid FIGURE 9 | Schematic illustration of mitochondrial-caspase induced apoptosis. Depolarization of mitochondrial transmembrane ( m) causes release of cytochrome c (Cyt C), which may initiate caspase cascade. Cyt C bonds with apoptotic protease-activating factor 1 (Apaf-1) (not shown in the Figure) and activates caspase-9, this cleaves and activates caspase-3, which triggers the apoptosis. Translocation of endonuclease G (Endo G) from the mitochondria to the nucleus and nuclear activation of DNA fragmentation factor (DFF) (not shown in the Figure) caused by caspase activation induced nucleosomal DNA fragmentation during apoptosis (Kitazumi and Tsukahara, 2011). growth of Nocardia was noted without reproduction, which indicated that the aggressive growth of intracellular Nocardia may act as an important role in the cell injury leading to cell death.
These data suggested that N. cyriacigeorgica infection and invasion can induce both apoptotic and necrotic changes in bMECs after 6 h, corroborating with the previous findings in which Nocardia induced apoptosis in PC12 cells and HeLa cells (Tam et al., 2002;Loeffler et al., 2004;Barry and Beaman, 2007). In Hela and PC12 cells, live Nocardia promoted a significant DNA fragmentation during 6 h-exposure period (Tam et al., 2002;Barry and Beaman, 2007). Previous studies also showed that Nocardia localized in substantia nigra and then underwent a rapid growth, inducing apoptosis of dopaminergic cells and failure of inflammatory response in 24 h (Kohbata and Beaman, 1991;Tam et al., 2002). Cell apoptosis is a double-edged sword both for host and bacteria during infection. For the host, cell apoptosis is a defense mechanism against pathogenic bacteria to prevent the release of intracellular bacteria and the spread of bacteria. Whereas, for bacteria, they induce cell apoptosis to block the spilling of cellular contents, to suppress the motivation of inflammatory reaction and then to evade host defenses. Several bacterial pathogens can cause cell apoptosis; whereas, many intracellular pathogens, such as Mycobacterium tuberculosis, can protect infected cells from apoptosis for its survival (Faherty and Maurelli, 2008;Butler et al., 2012). However, in the process of infection, N. cyriacigeorgica, being an intracellular pathogen, caused apoptosis and necrosis in bMECs.
The ultrastructural pathology of TEM showed the mitochondrial damage. To further validate our results, the collapse of m and release of mitochondrial cytochrome c were determined, presenting a decrease of m and an increase of cytochrome c release. On the other hand, western blot analysis suggested the promotion of casepase-9 and casepase-3 activation. Likewise, similar effects on mitochondria and caspase-3 activity have been proved in the Nocardia-induced Hela apoptosis at a MOI of 5:1co-cultured for 5 h (Barry and Beaman, 2007). Altogether, from these findings we can draw the conclusion that N. cyriacigeorgica can induce apoptosis of bMECs mediated by a mitochondria-caspase dependent pathway. Mitochondrial dysfunction to release proteins from the intermembrane space into the cytosol is the pivotal focus in the process of apoptosis (Wang and Youle, 2009). As for apoptotic signaling, cytochrome c was a critical apoptogenic factor and was capable of initiating the caspase cascade (Kagan et al., 2009). Normally, cytochrome c is localized in the mitochondrial intermembrane space. During the infection of N. cyriacigeorgica, the opening of the mitochondrial permeability transition pore induced depolarization of mitochondrial transmembrane m, release of apoptogenic factors and loss of oxidative phosphorylation (Tait and Green, 2010). Cytochrome c transposed from the mitochondrial membrane space to the cytosol; then bond to apoptotic protease-activating factor 1 (Apaf-1), triggered the formation of the apoptosome, inducing recruitment and activation of caspase-9. Caspase-9 cleaves and activates executioner caspase-3, which triggers the apoptosis (Ow et al., 2008;Tait and Green, 2010). Simultaneously, translocation of endonuclease G (Endo G) from the mitochondria to the nucleus and nuclear activation of DNA fragmentation factor (DFF) caused by caspase activation induced nucleosomal DNA fragmentation during apoptosis (Kitazumi and Tsukahara, 2011). DNA fragmentation is a hallmark of apoptosis and is one of the last consequences of apoptosis (Kitazumi and Tsukahara, 2011). The major schematic representation of mitochondrial-caspase induced apoptotic pathway studied in this work is elucidated in Figure 9.
In summary, these results of the present study demonstrated that clinical N. cyriacigeorgica was able to adhere and invade the bMECs, causing disruption of cell membrane and mitochondrial degeneration. Our data supports the hypothesis that N. cyriacigeorgica induced host cell death via an apoptotic mechanism and the later stage of N. cyriacigeorgica infection followed mitochondrial-dependent apoptotic pathway in bMECs. These findings provide that the apoptosis of bMECs was induced via triggering the caspase cascade and might present important insights into the mechanisms of N. cyriacigeorgica infections in bovine mammary tissues.

ETHICS STATEMENT
The present study was conducted in accordance with the ethical guide lines of China Agricultural University (CAU), Beijing.
Furthermore, prior to the initiation of this work, proper approval was granted by the departmental committee of College of Veterinary Medicine, CAU.

AUTHOR CONTRIBUTIONS
WC planned, performed the experiments, and then wrote manuscript. XG and YL performed the experiments, LZ and TA helped in analysis of flow cytometry experiments, GL collaborated in writing manuscript, MS and TA revised and corrected the manuscript, JG analyzed the data and BH designed and evaluated the research. All authors have read and approved the final manuscript.

FUNDING
This research was supported by Ministry of Education in China major project (No. 313054) and the National Natural Science Foundation of China (No. 3151101034 and NO. 31572587).