A549 (American Type Culture Collection Manassas, VA, USA, catalog number: ATCC CRM-CCL-185), RAW264.7 (ATCC TIB-71), MDBK (ATCC CCL-22), and 293T (ATCC CRL-3216) cells were maintained in Dulbecco’s modified Eagle’s Medium (DMEM) supplemented with sodium pyruvate, L-glutamine, 10% heat-inactivated fetal bovine serum (FBS) and 1% penicillin–streptomycin solution in a humidified incubator with 5% CO₂. THP-1 cells (ATCC TIB-202) were grown in RPMI 1640 medium containing L-glutamine, sodium pyruvate, 1% HEPES, supplemented with non-essential amino acids, and incubated as described above.

Luria–Bertani broth was used to cultivate S. aureus (ATCC 29213) at 37°C overnight with agitation (200x rpm). They were used in the mid–late log phase at an optical density (OD) calculated at a wavelength of 600 nm of 1.0. Bacteria (500 μL) were washed 3 times with 1x PBS. Next, 20% C5-depleted (C3^+ive) or C3-depleted (C3^-ive) human serum (Complement Technology) was added to bacteria, the mixture was incubated at 37°C for 30 min for complement opsonization, every 10 min, vortexed the mixture for 30s to mix bacteria and serum thoroughly. The bacteria were rinsed three times with 1x PBS after opsonization to eliminate any unbound complement components. They were then resuspended in serum-free DMEM to be used for infection.

For constructing infection models and autophagy detection, the input multiplicity of infection (MOI) for A549 cells was 20 or 100, depending on the test. The input MOI for THP-1, MDBK, and RAW264.7 cells was 50. In intracellular-replication experiments, A549 and RAW264.7 were infected at a MOI of 20. After inoculum had been added to cells in serum-free DMEM, plates were spun at 500 × g in a centrifuge (Allegro; Beckman Coulter, Fullerton, CA, USA) for 5 min at room temperature. Cells with bacteria were incubated at 37°C for 1 h and then washed thrice with PBS. DMEM supplemented with 10% FBS and gentamycin (100 μg/mL) was added to inhibit the survival of extracellular bacteria.

In intracellular-replication experiments, before being lysed in 0.1% Triton X-100, infected cells were rinsed thrice with PBS. After being serially diluted, the lysates were applied to LB agar plates. The latter were inverted overnight in an incubator at 37°C, and bacterial colonies were counted and recorded.

Experimental procedures were carried out according to the instructions provided on the CCK-8 kit (Beyotime Institute of Biotechnology). Cells were seeded at 2 × 10³ cells/well into 96-well plates and incubated for 24 h following infection with S. aureus opsonized with C5-depleted serum (C3^+ive), C3-depleted serum (C3^-ive), human normal IgG, or PBS. At different time points (0, 2, 4, 6, and 8 h), CCK-8 reagent (10 μL) was added to the complete medium (100 μL) for further incubation. Subsequently, the OD₄₅₀ was measured by a plate reader (Multiskan FC; Thermo Scientific). Cell viability was converted from OD₄₅₀ using the formula [(As − Ab)/(Ac − Ab)] × 100%, where AS is the absorbance of experimental wells, Ac is the absorbance of control wells, and Ab is the absorbance of blank wells.

Prior to protein extraction, cells were rinsed with ice-cold PBS. The total protein was extracted with a cell lysis buffer for western blotting and immunoprecipitation supplemented with phenylmethylsulphonyl fluoride (1 mM; Beyotime Institute of Biotechnology). Bicinchoninic acid assay (BCA) (Thermo Scientific) was used to quantify the protein content and standardize protein loading across samples according to the manufacturer’s directions. Following dissolving lysates in 6× protein-loading buffer (TransGen Biotech) and heating at 100°C, proteins were resolved on 10%–18% polyacrylamide gels (PAG) and transferred to polyvinylidene difluoride (PVDF) membranes. The latter were blocked in blocking buffer, which is 5% non-fat milk in Tris-buffered saline (TBS) with 0.1% Tween 20 (TBST) for 1 h at room temperature and incubated with primary antibodies at 4˚C. Goat anti-C3 (Complement Technology, catalog #: A213), rabbit anti-ATG16L1 (Cell Signaling Technology, catalog #: 8089), rabbit anti-LC3A/B (Millipore Sigma, catalog #: SAB5700812-100UL), and rabbit anti- p62/SQSTM (Millipore Sigma, catalog #: P0067-200UL) were used at 1:1000 dilution in blocking buffer. Mouse anti-myc (Abcam, catalog #: ab289980), mouse anti-glyceraldehyde 3-phosphate dehydrogenase, horseradish peroxidase (HRP) conjugated (GADPH-HRP; Beyotime Institute of Biotechnology, catalog #: AF2823-50μl), and mouse anti-β-actin, HRP conjugated (Beyotime Institute of Biotechnology, catalog #: AF2815-50μl) were used at 1:2000 dilution. On the next day, PVDF membranes were washed thrice with TBST and incubated with horseradish peroxidase-conjugated secondary antibodies (Beyotime Institute of Biotechnology) diluted at 1:10000 in blocking buffer for 1 h at room temperature and detected with ECL Prime (Thermo Scientific). Gray intensity of bands was analyzed by densitometric of Image-J.

Cells were seeded on coverslips (Solarbio Life Science) at desired density. After treatment, they were fixed with 4% paraformaldehyde (PFA) for 15 min, and permeabilized with 0.1% triton X-100 for 20 min at room temperature after gently washed with iced-old PBS for 3 times. Then, 2% bovine serum albumin (BSA) in 1 × PBS was used to block the cells for 1 h at room temperature. After 3-time washes by PBS, the cells were incubated by primary antibody (1:100 dilution in 2% BSA) at 4°C overnight or at room temperature for 1 h. After 3-time washes, cells were incubated with secondary antibody conjugated with fluorophore (1:500 dilution in 2% BSA) for 1 h at room temperature. Use 1 × PBS containing 0.1% triton X-100 (PBST) to wash cells for 2 times, then add Hoechst 33342 stain (Thermo Scientific) for 5 min at room temperature. Again, use PBST to remove the unbound stains for 3 times. Cells were observed and acquired by confocal microscope Nikon AX/AX R with the microscope objectives of 10x/0.45 NA, 40x/0.95 NA, 60x/1.40 NA Oil, and 100x/1.45 NA Oil were used in this manuscript. Images were processed and analyzed by NIS-Elements Viewer software and Image-J.

Lentivirus expression vector stocks were generated by transduction, using plasmids pMDG gag-pol, pRSV–Rev, pVSV–G Env, and the pLKO.1 ATG16L1 shRNA (laboratory-retained, used as mentioned (Zhu et al., 2023)) to co-transfect 293T cells. The ATG16L1 gene was knocked down by the complementary sequence: 5’- CCGGGTCATCGACCTCCGGACAAATCTCGAGATTTGTCCGGAGGTCGATGACTTTTTG-3’ in the pLKO.1 ATG16L1 shRNA. Non-targeting control was obtained from SIGMA MISSION^® TRC-Hs 1.0 (Human). After 48 h, supernatants were collected, centrifuged, and filtered to remove cell debris. A549 cells or RAW264.7 cells were transduced with the respective lentiviruses or empty vector control lentivirus for 72 h as described previously (Xu et al., 2019). This action was followed by selection with puromycin (2 μg/mL), which was determined to kill about 95%–99% of untransduced cells. Surviving cells were amplified, and knockdown was confirmed by western blotting.

A549 cells and 293T cells were cultured to ~60% confluency and transfected with the pCMV-myc plasmid containing full-length human C3 cDNA (National Center for Biotechnology Information (NCBI) reference sequence: NM_000064.4) or human ATG16L1 cDNA (NCBI reference sequence: NM_198890.3) or empty vector using jetPRIME™ (Polyplus Transfection, Illkirch-Graffenstaden, France) according to manufacturer’s instructions. Transfection efficiency was between 70% and 80% as determined with a ZsGreen-producing or mCherry-producing control vector.

The 293T cells in 10-cm dishes were grown to 60% confluency and transfected with pCMV-myc-C3-ZsGreens and/or pCMV-myc-ATG16L1-mCherry by jetPRIME™ (Polyplus Transfection). After 24 hours, cells were washed with ice-cold PBS and lysed with cell lysis buffer for Western blotting and immunoprecipitation (BIB) analysis. Cell lysates were centrifuged at 14,000 × g for 5 min at 4˚C and precleared with normal rabbit IgG and protein A+G agarose (BIB) at 4˚C for 1 h with rotation to remove nonspecific binding. Concerning the mixture of cell lysates and C3^+ive-S. aureus, the collected supernatants after lysis were incubated with C3^+ive-S. aureus directly for 2 h to remove nonspecific binding. Protein samples were centrifuged at 2,500 rpm for 5 min at room temperature. Then, the supernatant was collected and incubated with rabbit anti-C3 primary antibody or rabbit anti-ATG16L1 primary antibody and protein A+G agarose (BIB) at 4˚C overnight with rotation. Then, the A+G agarose was washed six times with lysis buffer. Bound proteins were eluted by boiling with 6× protein-loading buffer (TransGen Biotech) for 10 min. Proteins were resolved on 8%–15% polyacrylamide gels (BIB) and analyzed by Western blotting analysis with antibodies for C3 and ATG16L1.

A549 cells were infected with S. aureus or C3^+ive S. aureus at 20 CFU/cell. After 4 h, cells were washed with PBS, fixed with 2.5% glutaraldehyde, dehydrated, and embedded in plastic resin for viewing with a transmission electron microscope (HT7800; Hitachi, Tokyo, Japan).

Statistical analyses were carried out using GraphPad Prism 9 software. The number of biological replicates used herein was estimated based on a power analysis with the following assumptions: the standard deviation would be ~20% of the mean; P < 0.05 when the null hypothesis was false; the effect size (Cohen’s d) was between 1.0 and 2.0. Percent LC3-positive phagocytes or CFU counts were determined for significance with the unpaired parametric t-test for two groups, with ANOVA for multiple groups, and corrected for Student’s t test and post hoc Tukey’s multiple comparisons. Randomization or blinding was not undertaken in this study.

To test the function of C3 in autophagy regulation, we used human C5-depleted (C3^+ive) serum, which allowed C3 deposition but avoided MAC formation, to incubate with S. aureus. The S. aureus strain we used was modified to stably express green fluorescent protein (GFP), and the level of GFP was used as a marker for S. aureus infection (Li et al., 2021). To ascertain if S. aureus with C3 deposition (C3^+ive-S. aureus) could be internalized in cells, A549, THP-1, MDBK, and RAW264.7 cells were infected at 100 CFU/cell. GFP and F-actin (stained by phalloidin-rhodamine) were observed within the cells at 4 hours post-infection (hpi), indicating C3^+ive -S. aureus was internalized regardless of the origin of species of the cells used, or whether the cells were of myeloid lineage ( Figure 1A ). Next, we chose a non-phagocytic cell type (A549) and a phagocytic cell type (RAW264.7) to examine if C3 remained attached to the surface of S. aureus within host cells. A549 and RAW264.7 cells were infected with C3^+ive -S. aureus at 100 CFU/cell, and three-dimensional (3D) images were taken by a confocal microscope. The results showed C3 colocalized with S. aureus within the cells, suggesting C3 was still bound with internalized S. aureus ( Figures 1B, C ). Internalized C3^+ive-S. aureus was hypothesized to induce autophagy as the microtubule-associated protein light chain 3 (LC3), a marker of autophagosomes, was found colocalizing with C3^+ive-S. aureus ( Figure 1D ). LC3 signals increased in a time-dependent manner, and nearly all bacterial particles were surrounded by autophagosomes at 4 hpi. These results confirmed that C3 deposition was internalized with bacteria and colocalized with autophagic machinery.

Various autophagic components or processes, such as LC3-associated phagocytosis (LAP), contributes to defense against bacterial infection (Munoz-Sanchez et al., 2020). Studies have identified that phagocytosed S. aureus in neutrophils triggers LAP, and activates the formation of LC3-associated phagosomes (“LAPosomes”) with a single-membrane structure (Prajsnar et al., 2021). To gain a deeper understanding of the role of C3 on autophagy, transmission electron microscopy (TEM) was carried out to visualize the formation of autophagosomes in A549 cells with internalized S. aureus and C3^+ive-S. aureus at 20 CFU/cell. If internalized bacteria mainly induced LAP, then LAPosomes with a single membrane would be formed in the cytoplasm, but autophagosomes with double a membrane were observed which confirmed that both S. aureus and C3^+ive-S. aureus stimulated xenophagy. The typical submicroscopic structure of A549 cells was shown in Figure 2A . Cells treated with rapamycin induced autophagy and formed autophagosomes (marked with ▴) ( Figure 2B ). Consistent with the autophagosomes stimulated by rapamycin, S. aureus and C3^+ive-S. aureus infection also drove the formation of autophagosomes in A549 cells. When internalized in the cytoplasm, S. aureus ( Figure 2C ) and C3^+ive-S. aureus ( Figure 2D ) were positioned with double membranes (marked with ⇒), and were engulfed as the membrane elongated to form autophagosomes. The bacterial cells in autophagosomes were later degraded as the fusion of autophagosome and lysosome began to occur. These results suggest that S. aureus and C3^+ive -S. aureus induced autophagy as an antimicrobial strategy within A549 cells, whereas C3^+ive -S. aureus resulted in more autophagic vesicles ( Figure 2E ).

The composition of serum is very complicated. To exclude the effect observed was caused by other non-C3 serum components, we used a C3-depleted serum (C3-negative, C3^-ive) to incubate S. aureus (C3^-ive-S. aureus) as a parallel control. S. aureus was grown to the exponential phase and re-suspended in phosphate-buffered saline (PBS) with 20% C5-depleted (C3^+ive) or C3-depleted (C3^-ive) serum. Incubated S. aureus was then used to infect A549 cells at 100 CFU/cell. Intracellular C3 was detected by immunofluorescence assay (IFA). We observed that C3 was localized to S. aureus following bacterial infection, most S. aureus were C3 positive or with partial C3 deposition at 2, 4, and 6 hpi ( Figure 3A ).

Endogenous LC3 was monitored in A549 cells by IFA at 2 and 4 hpi. There was a significant increase in the colocalization of C3^+ive-S. aureus with LC3 compared to S. aureus, indicating that C3 contributed to LC3 targeting. Meanwhile, the C3^-ive-S. aureus also caused more LC3 targeting than S. aureus alone, suggesting other serum ingredients may contribute to the process, although playing a less significant role than C3 ( Figure 3B ).

LC3-I to LC3-II conversion is a well-accepted marker for autophagic activation (Tanida et al., 2008). We monitored the levels of LC3-I and LC3-II and observed a greater conversion of LC3-I to LC3-II in C3^+ive-S. aureus-infected cells than C3^-ive-S. aureus-infected ones, suggesting the involvement of C3 in autophagic activation ( Figure 3C ). Also, compared with incubation with PBS, S. aureus incubated with 20% C3-depleted (C3^-ive) serum facilitated LC3-I to LC3-II conversion, which was in accordance with our previous results.

SQSTM1/p62 protein (p62) is a cellular protein selectively wrapped into the autophagosomes and subsequently degraded by proteases in the autolysosomes (Yoshii and Mizushima, 2017). The expression of p62 is negatively correlated with autophagic flux and was also used as an autophagic marker (Liu et al., 2014). To further demonstrate the effect of C3 on autophagic flux, p62 was examined by Western blotting analysis. The level of p62 protein expression decreased upon S. aureus invasion, and C3 deposition exacerbated this decrease, indicating C3 upregulated autophagic flux ( Figure 3D ). To further confirm the role of C3 deposition on autophagic flux, p62 was measured in A549 cells after S. aureus and C3^+ive-S. aureus infection at different time points after infection (2, 4, 6, 8 hpi) without Leupeptin inhibition. As shown in Figure 3E , p62 gradually reduced with the duration of infection, and C3^+ive-S. aureus caused a more rapid reduction than S. aureus. Overall, these data demonstrated that C3 deposition of S. aureus promoted autophagic activation and accelerated the autophagic flux.

Studies have shown that the autophagy-related protein, ATG16L1, interacted with the complement C3d fragment (Sorbara et al., 2018). As C3 deposition on the bacterial surface continues to undergo hydrolysis by Factor H and produce multiple fragments (C3b, iC3b, C3dg, C3d), we chose to further confirm whether C3 fragments could interact with ATG16L1 by immunoprecipitation (Gordon et al., 1988; Zarantonello et al., 2023). Total lysates of 293T cells overexpressing ATG16L1 were mixed with C3^+ive-S. aureus. Following the immunoprecipitation of C3, ATG16L1 was found co-precipitated with C3 fragments ( Figure 4A ). To rule out artificial effect caused by protein overexpression, we examined the interaction of endogenous ATG16L1 with the overexpressed full-length C3 within cells by co-transfecting 293T cells and A549 cells with ATG16L1 and C3 overexpressed vectors. In 293T cells, C3β was found to pulldown ATG16L1 and the reciprocal IP showed the same results ( Figure 4B ). In addition, the full-length C3 colocalized with ATG16L1 according to confocal microscopy ( Figure 4C ). Taken together, these data demonstrated that ATG16L1 interacted and colocalized with full-length C3 and its cleaved products.

The higher level of autophagy induced by C3^+ive-S. aureus was not accompanied by a change in ATG16L1 expression ( Figure 3C ). We hypothesized that ATG16L1 was recruited from dispersion by C3, which then controlled the location of the autophagosome and drove the conversion of LC3-I to LC3-II. A549 cells were infected with C3^+ive-S. aureus and ATG16L1 were traced by immunofluorescence staining. Compared with S. aureus and C3^-ive-S. aureus, C3^+ive-S. aureus caused more aggregation of ATG16L1 ( Figure 4D ). To further confirm the regulatory role of ATG16L1 and C3 in autophagy, ATG16L1-knockdown cell lines were constructed using lentivirus-mediated short hairpin RNA (shRNA). C3 increased autophagy in wild-type A549 cells, but this increase disappeared in ATG16L1-knockdown A549 cells ( Figure 4E ). Correspondingly, compared with wild-type cells, LC3 expression was reduced in ATG16L1-knockdown cells, indicating that ATG16L1 affected the efficiency of autophagic targeting and the total cellular autophagy level.

Transfected full-length C3 could bind to ATG16L1 ( Figure 4B ). Hence, we explored if full-length C3 without deposition could also promote autophagy. C3 was transfected into A549 cells before S. aureus infection. Compared with C3 deposited on S. aureus, endogenous C3 lost the ability to regulate autophagy ( Figure 4F ), which suggested that extracellular cleavage of C3 by complement factors was essential for the regulation of autophagy. Taken together, these findings suggested that although both C3 deposition and full-length C3 interacted with ATG16L1, the C3 deposition on the surface of S. aureus rather than full-length C3 could upregulate autophagy by recruiting and interacting with ATG16L1.

Autophagy contributes to intracellular bacterial restriction by targeting invading agents to lysosomes for degradation. We aimed to determine if the C3 deposition facilitated autophagy-dependent restriction of internalized S. aureus. We normalized the amount of S. aureus, C3^+ive-S. aureus, C3^-ive-S. aureus and IgG-S. aureus before infecting mammalian cells, and measured the replication of bacteria after infection. The levels of internalized S. aureus in A549 cells ( Figure 5A ) and RAW264.7 cells ( Figure 5B ) decreased dramatically when bacterial cells were coated with C3. Meanwhile, the degree of proliferative restriction mediated by C3 was further increased by addition of an autophagy inducer, rapamycin, and was suppressed by autophagy inhibitors (3-Methyladenine, bafilomycin A1 and chloroquine) ( Figures 5C, D ). Importantly, the reduction of bacterial proliferation caused by C3 deposition no longer occur when ATG16L1 was depleted ( Figure 5E ).

Next, the viability of A549 and RAW264.7 cells after infection was examined by the CCK-8 assay. Remarkably, host cells displaying highest viability were those infected by C3^+ive-S. aureus, suggesting the protective role of C3 coating on host cells ( Figures 5F, G ). Meanwhile, cells infected with C3^-ive-S. aureus and IgG-S. aureus exhibited similar viability, indicating the dominant role of C3 in preventing bacterial proliferation and protection host cells. Taken together, these results demonstrated that the C3 deposition facilitated autophagy-dependent limitation of internalized S. aureus by ATG16L1 and had a protective effect on cells ( Figure 6 ).