Streptococcus lutetiensis Induces Autophagy via Oxidative Stress in Bovine Mammary Epithelial Cells

Streptococcus lutetiensis, an emerging pathogen causing bovine mastitis, has not been well characterized. We reported that S. lutetiensis was pathogenic both in vivo and in vitro and caused inflammatory reactions in the mammary gland. However, roles of autophagy and oxidative stress in the pathogenesis of S. lutetiensis-induced mastitis are unclear. In this study, an autophagy model of S. lutetiensis-infected bovine mammary epithelial cells (bMECs) was used to assess oxidative stress and autophagy flux. Expressions of Beclin1, light chain 3II, and Sequestosome 1/p62 were elevated in bMECs after S. lutetiensis infection. In addition, autophagosome and lysosome formation confirmed autophagy occurred. Based on LysoTracker Red and acridine orange, lysosome degradation was blocked, and lower expressions of lysosomal-associated membrane protein 2, cathepsins D, and cathepsins L confirmed lysosomal damage. Concurrently, the nuclear factor erythroid 2-related factor 2 (Nrf2), kelch-like ECH-associated protein 1 (Keap1), heme oxygenase 1 (HO1), and NAD (P)H: quinone oxidoreductase 1 (NQO1), and basilic proteins associated with the Nrf2/Keap1 signaling pathway, were detected. Decreased keap1 and increased Nrf2, HO1, NQO1, and reactive oxygen species (ROS) indicated increased oxidative stress. Treatment with N-Acetyl-L-cysteine (NAC), an ROS inhibitor, decreased both oxidative stress and autophagy. Therefore, we concluded that S. lutetiensis caused intracellular oxidative stress and autophagy in bMECs. In addition, crosstalk between autophagy and oxidative stress affected the autophagic flux and blocked downstream autophagy. The Nrf2-keap1-p62 pathway participated in this process, with ROS acting upstream of these effects, interfering with normal cell functions.


Introduction
Mastitis is highly prevalent in dairy cows [1], causing huge losses. Bacterial infections, often Streptococci spp., are the most common cause of mastitis [2]. S. lutetiensis, increasingly isolated from milk (Chen et al., unpublished), caused mammary gland edema and hyperemia, and we reported that bovine-derived S. lutetiensis was often resistant to multiple antibiotics [3].
Autophagy is a catabolic pathway in eukaryotic cells to remove intracellular organelles under extreme environments, to degrade damaged organelles, and control intracellular bacterial infections [4]. There are three types: macroautophagy (also called autophagy), microautophagy, and chaperone-mediated autophagy [5]. Autophagy typically involves formation of autophagosomes, fusion between autophagosomes and lysosomes, and lysosomal degradation [6]. The term "autophagic flux" includes autophagosome synthesis, delivery of autophagic substrates into the lysosome, and their subsequent degradation inside the lysosome [7]. Groups A and B Streptococci induced autophagy [8,9]; however, the ability of group D Streptococci to induce autophagy has apparently not been reported.
Reactive oxygen species (ROS) are highly reactive oxygen-containing substances. Although ROS concentrations are well regulated under physiologic conditions, if control mechanisms are disrupted, ROS concentrations can increase and may cause damage or disease. ROS is an early inducer of autophagy during nutrient deficiency [19] and can promote degradation of ubiquitinated materials [20].
The antioxidant transcription factor, Nrf2 (nuclear factor erythroid 2-related factor 2), is activated by p62 in a "noncanonical" pathway [21]. Nrf2 is an important transcription factor regulating cellular redox homeostasis. Under physiologic conditions, Nrf2 concentration is low, whereas under oxidative stress, it is activated and translocated from the cytoplasm into the nucleus [22]; this can trigger downstream target genes responsible for detoxification and elimination of harmful substances [23]. The protein expression level of Kelch-like ECH-associated protein 1 (keap1) is a negative regulator of Nrf2 [24]. There are indications of interrelations between Nrf2 and p62 and that Nrf2 has a role in dysregulation of autophagy [25].
We reported that S. lutetiensis adhered to, invaded, and destroyed bovine mammary epithelial cells (bMECs) and damaged murine mammary tissue [3]. However, roles of autophagy and oxidative stress in S. lutetiensis-induced mastitis are unclear. In this study, we used an autophagy model of S. lutetiensis-infected bMECs to assess oxidative stress and autophagy flux.
2.3. CCK-8 Assay. MAC-T cells (~4,000 cells/well) were inoculated into 96-well plates. After being infected with S. lutetiensis (MOI = 100), a CCK-8 kit was used to evaluate cell viability, in accordance with label protocols. After infection, 10 μl of CCK-8 solution was added in each well and incubated at 37°C for 2 h; finally, the absorbance was read at 490 nm (680 Multipurpose Microplate Reader, Bio-Rad Laboratories).
2.4. Infection of bMECs with S. lutetiensis. Autophagy caused by intracellular S. lutetiensis in bMECs was explored. Our intracellular infection model was modified from a previous study [24]. In the present study, bMECs seeded at 2 × 10 5 cells/well in six-well plates were infected with fresh S. lutetiensis grown with BHI broth, at multiplicity of infection (MOI) of 100. Cells were incubated for 2 h at 37°C in 5% CO 2 to allow bacterial uptake and invasion, then washed three times with phosphate-buffered saline (PBS) and DMEM/10% FBS with 100 μg/ml gentamycin (bacterial inhibiting buffer) added into each well to kill extracellular bacteria. After changing the medium to DMEM/10% FBS with 100 μg/ml gentamicin, cells were incubated for the indicated intervals at 37°C in 5% CO 2 . If not stated otherwise, inhibitors were added to incubation buffer (Figure 1(a)). To confirm the role of oxidative stress, NAC was used to pretreat cells for 3 h, followed by exposure to S. lutetiensis (MOI = 100) for 2 h and, finally, incubation in medium with 100 μg/ml gentamycin. The bMECs were incubated in 1× NAC from the start of NAC pretreatment to cell harvest (Figure 1(a)).

Western
Blotting. For Western blot analysis, after being infected with S. lutetiensis for defined intervals, MAC-T cells were collected and lysed with 200 μl RIPA buffer (with 1% phenylmethylsulfonyl fluoride) for 10 min on an ice-plate, then centrifuged at 12,000 × g for 15 min and protein concentrations quantified with a BCA protein quantity reagent kit. Equal volumes of lysates were separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane by Wet blotting. Blots were blocked in Tris-buffered saline (TBS) containing 5% skim milk powder. Membranes were incubated with primary antibodies (Beclin 1, LC3 II/I, p62, CTSD are referred to GAPDH, and LAMP2, CTSL, Nrf2, Keap1, HO-1, and NQO-1 are referred to α-tubulin) at 4°C overnight, washed with TBS buffer with Tween 20, then incubated with secondary antibodies for 1 h at RT. Blotting signals were detected with an enhanced chemiluminescence Western blot detection system, images collected, and relative band density analyzed with Image J software (NIH, Bethesda, MD, USA).   3 Oxidative Medicine and Cellular Longevity 2.11. Immunofluorescence Staining. Cells were seeded on sterile coverslips placed in 6-well plates. After cells were infected with S. lutetiensis for 2 h, they were fixed in 4% paraformaldehyde for 10 min and then permeabilized with 0.2% Triton X-114 in PBS for 15 min. After being washed with PBS, cells were blocked with 3% BSA for 1 h at RT. Slides were incubated with anti-α-tubulin antibody (1 : 200 diluted in PBS) overnight at 4°C and then incubated with peroxidase-conjugated AffiniPure (diluted 1 : 100 in PBS) second antibody for 1 h at 37°C, before nuclei and intracellular bacteria were stained with DAPI. Between steps, cells were thrice-washed with PBS. Ultimately, all slides were mounted with antifade mounting medium and examined on a Nikon A1HD25 confocal microscope with a 100x oil immersion objective. Imaging used laser wavelengths of 488 and 561 nm.

Results
3.1. Successful Cell Infection Model. S. lutetiensis was observed in MAC-T cells of the treated group but was absent in the control group (Figure 1(b)). Based on the CCK-8 assay, continuous infection with S. lutetiensis at MOI = 100 for 2 h did not decrease cell activity (Figure 1(c)). However, in ultrastructure images, S. lutetiensis was present in MAC-T cells, and there was an autophagic vesicle membrane structure around the bacteria (Figure 1(d)).

Intracellular S. lutetiensis-Induced Autophagy.
Infection with S. lutetiensis substantially increased expression of Beclin 1 (Figure 2(a)). Furthermore, in bMECs treated with S. lutetiensis, conversion of LC3I to LC3II increased significantly ( Figure 2(b)), starting from the 1 st hour postinfection (hpi) and remaining elevated for the following 3 h, when a green fluorescent protein (GFP)-LC3 labeled autophagosome was detected by confocal imagine (Figure 2(d)). Autophagosomes (fluorescent spots) increased with interval after infection. The protein expression level of SQSTM1/p62 tended to decrease at 0.5 hpi, and then increased from 1 to 3 hpi (Figure 2(c)).

Autophagy Flux Disorder.
To determine whether the autophagy flux was blocked after infection with S. lutetiensis, mCherry-GFP-LC3, which can label lysosomes, was transfected into MAC-T cells ( Figure 3). Yellow fluorescence (green fluorescence merged with red fluorescence) became increasingly prominent during the postinfection period, with red mottled fluorescence peaking at 3 hpi (Figure 3(a)), and yellow fluorescence spots quantified (Figure 3(b)).

S. lutetiensis Reduced Lysosomal pH.
Infection with S. lutetiensis increased LTR staining, consistent with lysosomal acidification (Figures 4(a) and 4(b)). Red spots increased over time and accumulated around the nucleus, from 0 to 3 h, with a modest decline at 4 h. After staining with AO, the color changed over time. They changed from 1 h, peaking at 3 h, and a slight decrease at 4 h.
3.5. Degradation of Impaired Lysosomes. During the final stages of autophagy, autophagosomes fuse with lysosomes and are subsequently degraded. Thus, LAMP2 protein, detected via Western blotting, decreased over time (Figure 4(c)). Lysosome-related proteins CTSD and CTSL were also detected (Figures 4(d) and 4(e)). All three kinds of lysosome-associated proteins decreased over 3 hpi, indicating lysosomal degradation was impaired after S. lutetiensis infection of MAC-T.  (Figures 7(a)-7(c)). In addition, treatment for 2 h with S. lutetiensis (MOI = 100) induced a time-dependent increase in ROS production, although the effect was abolished at 4 h ( Figure 6(b)). In addition, NAC abolished ROS production (Figures 7(a)-7(c)) and consequently oxidative stress (Figures 7(d)-7(f)), with reductions in extent of changes in GSH, SOD, and MDA. Furthermore, Nrf2, HO-1, and NQO-1 were decreased significantly compared to the treated group, whereas Keap1 protein was recovered (Figures 7(g)-7(j)). We next assessed whether ROS was involved in S. lutetiensis-induced autophagy. S. lutetiensis increased Beclin 1 protein level in bMECs, as well as the LC3 II/I ratio and p62 degradation, whereas NAC prevented these effects (Figures 8(a)-8(c)). The lysosomal protein LAMP2 was decreased after infection, but after NAC, was increased compared to the control group (Figure 8(d)). However, neither CTSD nor CTSL expression was significantly different between the treatment group and NAC-supplemented group (Figures 8(e) and 8(f)). In addition, LTR fluorescence was used to examine autophagosome-lysosome fusion (Figure 8(g)). These results demonstrated that increased ROS production was an upstream event contributing to S. lutetiensis activation of autophagy.

Discussion
S. lutetiensis invades cells to evade immune defenses and survives within those cells [3]. Herein, the ability to induce autophagy was achieved following optimization of S. lutetiensis dose in MAC-T cells, without compromising cell viability. This model provided clear evidence that S. lutetiensis 5 Oxidative Medicine and Cellular Longevity induced autophagy in MAC-T cells by increasing oxidative stress and that there was a close relationship between oxidative stress and autophagy. Furthermore, these findings provided insights into potential mechanisms underlying S. lutetiensis-induced autophagy (Figure 9).
In this study, S. lutetiensis was regarded as a repelling substance; the classical vesicles of autophagosome compartments were observed and engulfed by an autophagosome, typical characteristics indicating the host activated its defense mechanism, including initiation of autophagy [26]. Moreover, Beclin 1, which is essential for both autophagy and lysosomal enzyme transport [27], was increased after infection with S. lutetiensis. Furthermore, LC3 II, an autoph-agy marker with levels related to the intensity of autophagy [28], uses GFP-LC3 for confirming autophagy accumulation [29]. Therefore, expression of LC3 II was assessed by GFP-LC3 detected with confocal microscopy after MAC-T cells were infected with S. lutetiensis, with a time-independent tendency for green spots. Collectively, all of these outcomes indicated that S. lutetiensis induced autophagy. SQSTM1/p62 acts as a relative protein to mediate degradation of its recognition substrate [30,31]. Excessive buildup of SQSTM1/p62 sequestered keap1, an adaptor of the E3-ubiquitin ligase complex for Nrf2 [32]. In the present study, expression of SQSTM1/p62 protein began to rise from the second time point (0.5 h after infection) during S.  Oxidative Medicine and Cellular Longevity lutetiensis infection in bMECs. Meanwhile, expression of Nrf2 increased, whereas keap1 decreased from 0 to 3 hpi, consistent with previous studies [24]. The mCherry-GFP-LC3 is an adenovirus probe to detect the rate of autophagic flux [33]. In this study, accumulation of yellow fluorescence was observed by mCherry-GFP-LC3 after S. lutetiensis infection, indicating that the autophagic flux was blocked at fusion of autophagosomes and lysosomes. In addition, S. lutetiensis induced increases in the LC3 II/I ratio and p62 degradation, providing further support for autophagy  Autophagy formation is the first step of autophagy flux. In this step, autophagosomes can fuse with lysosomes or late endosomes and then fuse with lysosomes [34]. Lysosomal degradation is an essential factor for autophagy flux, and a slightly acidic environment for lysosomes promotes proteolytic enzyme degradation of organelles [35]. LTR is a reliable probe to detect changes in intercellular pH, as it accumulates in an acidic environment, with the intensity of red fluorescence corresponding to the pH [36,37]. In addition, AO is a specific probe that can be trapped in lysosomes [38]. In the present study, the intensity of LTR peaked at 3 h after

Oxidative Medicine and Cellular Longevity
S. lutetiensis infection, with AO staining having a similar outcome. These two approaches both provided evidence that lysosomes accumulated due to a blockage of the downstream autophagy flux, which differed from a report that group A Streptococcus was efficiently killed within a lysosome-fused autophagosome compartment [9]. Cathepsins are the major lysosomal proteases involved in autophagic degradation, wherein CTSD and CTSL are two abundant lysosomal proteases [39]. When activation of cathepsin proteases requires acidification, the altered pH decreased protein degradation [40]. CTSD is involved in intracellular catabolism in lysosomal compartments [41]; by activating CTSB, CTSL can hydrolyze proteins, hormones, and phagocytic bacteria [42]. In this study, expression of CTSD and CTSL decreased initially, but subsequently recovered to a basal level, indicating that intracellular invasion of S. lutetiensis had a negative effect on lysosomes, and cathepsins had self-healing ability.
As an essential mechanism for eliminating damaged organelles and exogenous foreign body, autophagy is regulated by various cellular process such as nutrient deficiencies, oxidative stress, and other influencing factors [43]. Autophagy and oxidative stress in a S. lutetiensis-infected model have apparently not been reported. Herein, we determined that oxidative stress had an essential role in S. lutetiensisinduced cyto-injury. As a transcription factor, Nrf2 regulates

12
Oxidative Medicine and Cellular Longevity a cluster of oxidative stress-inducible genes in cells. There is considerable evidence that the Nrf2-Keap1 antioxidant system has an important cytoprotective defense mechanism in alleviating oxidative insults [44,45]. Meanwhile, Keap1 can indirectly assess activation of Nrf2, as the Nrf2-Keap1 pathway is a regulator for the endogenous antioxidant response [46]. Therefore, Nrf2-Keap1 system integrity is maintained by an autophagy pathway [47], and activation of Nrf2 is prolonged if autophagy is disrupted [25]. In response to oxidative stress, Nrf2 transcription factor could induce expression of p62 [48,49]. Our study implied that transportation of oxidized proteins to autophagosomes for degradation can decrease oxidative injury. In response to oxidative stimuli, keap1 lost its ability to bind with Nrf2. Therefore, we speculate that Nrf2 may enter the nucleus and bind to the antioxidant response element (ARE), to further regulate downstream genes HO-1 and NQO-1, as they are key oxidative stress indicators regulated by Nrf2 [24]. Furthermore, HO-1 and NQO-1 proteins in MAC-T cells were elevated after S. lutetiensis infection. Recent research demonstrated that NAC could block ROS production and further reversed LC3 II accumulation which trig-gered by coenzyme Q 0 , and the same mechanism of NAC also occurred in chrysin-induced autophagy [50,51]. We used NAC, an efficient ROS inhibitor [52,53]. Modulation of antioxidant enzymes (SOD, GSH, and MDA) indicated inhibition of oxidative stress. Meanwhile, expression of NQO-1 and HO-1 was restored to basal levels; therefore, we inferred that Nrf2 may be a key cytokine in oxidative stress and autophagy. As arguably the most influential intracellular signaling molecule, ROS regulates cell function and promotes homeostasis [54]. There are indications that oxidative stress can be a vital stimulus, via regulation of ROS, to stimulate autophagy [21,55]. Autophagy and oxidative stress have an intricate relationship in many diseases [26]; hyperthermia can enhance both autophagy and ROS generation, implying potential associations between autophagy and oxidative stress [56]. Moreover, several studies reported the relationship between oxidative damage and autophagy [57], although this has apparently not been reported in bMECs. In this study, ROS were rapidly produced after exposure to S. lutetiensis. Notably, after 2 h, S. lutetiensis induced accumulation of LC3II and degradation of p62, suggesting  Figure 9: Putative signal pathway of S. lutetiensis inducing autophagy in bovine mammary gland epithelia though oxidative stress. Schematic diagram illustrating that S. lutetiensis can invade MAC-T cells and lead to autophagy by stimulating mitochondrial oxidative stress. In this process, the ROS-Nrf2-keap1 pathway has an essential role. ROS is upstream of these effects. Black arrows and red bars indicate stimulation and inhibition, respectively. 13 Oxidative Medicine and Cellular Longevity that ROS generation preceded initiation of autophagy. Furthermore, NAC effectively blocked autophagy activation by S. lutetiensis. As a main component of oxidative stress, ROS could be the link between autophagy and oxidative stress in bMECs.
The current study, apparently the first investigation of the role between autophagy and oxidative stress induced by S. lutetiensis, identified crosstalk between autophagy and oxidative stress. The limitation of this study was that an in vitro bMEC infection model was used to evaluate autophagy and oxidative stress. In future studies, we plan to investigate whether similar interactions occur in vivo in mammary tissue.

Conclusions
We characterized autophagy induced by oxidative stress in bMECs infected with S. lutetiensis isolated from bovine mastitis. We concluded that S. lutetiensis induced autophagy of bMECs by upregulating oxidative stress, although lysosomes accumulated due to a blockage of the downstream autophagy flux. In this process, there was crosstalk between autophagy and oxidative stress; the latter affected autophagy by intervening in the Nrf2-keap1-p62 pathway, with ROS acting upstream of these effects.

Data Availability
All data generated or analyzed during this study are included in this published article.

Ethical Approval
This work had received approval for research ethics from the Ethical Committee of the College of Veterinary Medicine, China Agricultural University (CAU), Beijing. Furthermore, this study was conducted according to standard ethical guidelines implemented at CAU (SYXK, 2016-0008), and a certificate of approval is available upon request.

Conflicts of Interest
The authors declare they have no actual or potential competing financial interests.