Staphylococcus aureus is an opportunistic pathogen that has adapted to long-term colonization in the human skin and nares (Jeon et al., 2020). S. aureus utilizes the adhesins to initiate the invasion process by attaching to the surface of host cell (Horn et al., 2018; Watkins and Unnikrishnan, 2020). After invasion, S. aureus induces a cytoplasmic and mitochondrial Ca²⁺ overload, which leads to both apoptotic and necrotic cell death (Stelzner et al., 2020). S. aureus infection presents as long-lasting persistent or acute diseases that are associated with significant morbidity and mortality (Turner et al., 2019). Antibiotics were most widely used to treat S. aureus infectious diseases, however, S. aureus has rapidly developed resistance to antibiotics. Approximately 90% of S. aureus strains show resistance to multiple antibiotics, resulting in decreased antibiotic application and reduced antibiotic effectiveness (Costa et al., 2018). Since methicillin-resistant S. aureus (MRSA) was identified in 1960, the infection rate with MRSA has increased globally. It leads to an increased burden on healthcare-associated expenditures (Zhen et al., 2020), and has become the main cause of bacterial infection in hospitals and communities (Lakhundi and Zhang, 2018). MRSA strains account for 5%-82% of S. aureus isolates (Köck et al., 2010; Falagas et al., 2013), leading to clinical syndromes including bacteremia (Klevens et al., 2007), one of the most severe situations of S. aureus infections with 15%-60% mortality rates (Li et al., 2021). Invasive MRSA strains possess a series of virulence factors and toxins, allowing them to spread rapidly in the community, and seriously threaten public health (Lakhundi et al., 2018). Therefore, new strategies to control S. aureus infection have gradually become the focus by manipulating and enhancing host immune defenses (Keller et al., 2020; Gauron et al., 2021).

The host immune system provides the first defense against pathogens, effectively removing intracellular pathogens in most cases. Simultaneously, autophagy also plays an essential role in resisting to pathogens (Randow et al., 2013). Autophagy is a fundamental biological process, in which pathogens are engulfed by double membrane vesicles called phagophores and eventually transported to lysosomes for subsequent degradation (Kirkegaard et al., 2004; Nakatogawa, 2020). Recent studies have demonstrated that autophagy has a crucial role in host cell defense against S. aureus (Lv et al., 2019; Gibson et al., 2020). The cell wall components of S. aureus can be detected as pathogen-associated molecular patterns (PAMPs) and then induce autophagy (Arroyo et al., 2013; Wu et al., 2016). Autophagy effectively limits S. aureus growth by fusion with the lysosome or positively regulating the phagocytosis of macrophages (Lv et al., 2019; Gibson et al., 2020). Researchers have thus tried to use the autophagy pathway to control S. aureus infection. Whereas some S. aureus strains have evolved self-defense mechanisms against autophagy degradation, and are even protected by the autophagy pathway (Schnaith et al., 2007). Once S. aureus enters the autophagosome, it transforms this “compartment” to create a hospitable environment in which it can survive and replicate (O'Keeffe et al., 2015). S. aureus being degraded by the autophagy pathway or protected by the autophagosome compartment is related to the accessory gene regulatory (agr) system which plays a crucial role in pathogenesis by coordinating virulence factors expression and bacterial density (Schnaith et al., 2007; O'Keeffe et al., 2015).

Here, we reviewed “beneficial” and “harmful” functions of autophagy in the process of S. aureus infection, as well as the mechanism by which S. aureus evades autophagy. This review is helpful to understand the interaction between hosts and S. aureus, and provides a theoretical basis for the development of new treatments for S. aureus infection.

Based on phylogenetic analyses, Queck et al. reported that S. aureus first emerged as a nonvirulent species, and only later acquired virulent functions (Queck et al., 2008). The agr quorum sensing system is the main virulence regulator of S. aureus in response to changing environmental conditions, such as adapting to low-nutrition conditions in high-cell-density populations, forming a nonpathogenic lifestyle (Queck et al., 2008). Approximately 30% of humans persistently but asymptomatically carry S. aureus in their nasopharynx (Wertheim et al., 2005). S. aureus actively adheres to promote colonization and replicates to avoid removal by nasal secretions (Foster et al., 2014).

The cell wall-anchored proteins of S. aureus, Fnbps and IsdB, promote internalization and subsequent invasion (Zapotoczna et al., 2013; Schlesier et al., 2020). The pattern recognition receptors (PRRs) expressed on the surface of phagocytes recognize pathogens and mediate their uptake into phagosomes for later elimination (Flannagan et al., 2009). Nonprofessional phagocytes utilize endocytosis to take up S. aureus (Moldovan and Fraunholz, 2019). Once internalized by host cells, the agr system of S. aureus increases virulence factors to damage phagosomes and promote intracellular survival (Novick et al., 1993). The phagosome or endosome can fuse directly with a lysosome to acidify to low pH for degrading microorganisms (Flannagan et al., 2009; Lâm et al., 2010). However, S. aureus tolerates acidic environments, which contributes to its survival within phagolysosomes (Weinrick et al., 2004). Exposure to an acidic environment increased expression of agr system (Tranchemontagne et al., 2016). Phagosomal acidification even appears to be essential for survival of some S. aureus strains (Tranchemontagne et al., 2016). Agr positively regulates cytotoxic phenol-soluble modulins (PSMs), which mediate escape from the phagosome into the cytoplasm to avoid lysosomal killing (Grosz et al., 2014; Münzenmayer et al., 2016). The cytoplasmically located S. aureus or leaky phagosomes could be captured by autophagosomal membranes and eventually fuse with lysosomes for autophagic degradation (Fraunholz and Sinha, 2012). S. aureus is also capable of escaping or even manipulating the autophagy pathway for replication and dissemination (Vozza et al., 2021). S. aureus further evolved regulatory functions to attenuate the expression of virulence genes to reduce innate immune defenses (Boisset et al., 2007; Cheung et al., 2014). This decreases the pro-inflammatory potential of S. aureus, which is associated with chronic infection. Surprisingly, S. aureus is very responsive to external stimuli, and rapidly reverts back to the original virulent state in rich bacterial growth conditions (Tuchscherr et al., 2020).

Autophagy is considered a crucial intracellular degradation system for removing dangerous pathogens (Levine, 2005). The dynamic membrane processes of autophagy occur through regulators comprised of autophagy-related genes (ATGs) and additional factors based on the following sequential steps: autophagy initiation; phagophore formation; double-membrane nucleation and phagophore elongation; cytoplasmic microorganism engulfment; autophagosome fusion with lysosome; and cargo degradation (Kuo et al., 2018) ( Figure 1 ).

Agr system is a major gene regulator that governs the toxin production of S. aureus at the appropriate time, regulating the adhesins expression during attachment and virulence factors during infection. Agr can upregulate α-toxin to cause tissue destruction by perturbing to epithelial cell junctions (von Hoven and Husmann, 2019). α-toxin also increased S. aureus internalization within mast cells by up-regulation of β1 integrin (Goldmann et al., 2016). After internalization, the high-level expression of agr led to strong expression of toxins and exoenzymes, as well as increased expression of methicillin resistance genes, mediating the pathogenesis (Cheung et al., 2011). At the meantime, the agr locus controlled phenol-soluble modulins alpha (PSMα) has also been shown to be crucial for phagosomal escape in both professional and non-professional phagocytes (Grosz et al., 2014). When agr is absent, phagosomal escape and autophagosomal accumulation are significantly reduced as well as intracellular bacterial burden is reduced (O'Keeffe et al., 2015; Blättner et al., 2016). Additionally, agr has been shown to have the alternating function, which can reduce cytotoxicity to survive persistently within host cells and avoid the host immune system activation (Tuchscherr et al., 2011).

Schnaith et al. reported agr-regulated factor(s) activated autophagy could prevent the maturation of S. aureus-containing phagosomes in human epithelial cells (Schnaith et al., 2007). Subsequently, the agr regulated α-toxin was shown to be necessary for eliciting autophagy, but the autophagic response was dysfunctional and the induced autophagosomes were not acidic. Additionally, α-toxin-deficient S. aureus strains were unable to activate the autophagy pathway (Mestre et al., 2010) ( Figure 2 ). In addition, an agr-specific factor was discovered that manipulates the autophagy network to provide an intracellular niche for S. aureus in human PMNs, but whether it is α-toxin has yet to be determined. The normal autophagic flux, expression of LC3II and p62, was disrupted in PMNs containing S. aureus (Mulcahy et al., 2020). Agr-positive S. aureus leads to the accumulation of autophagy inducer p53 in PMNs, driving transcriptional activation of pro-autophagic membrane protein damage-regulated autophagy monitor (DRAM). DRAM can directly mediate p53-induced autophagy and enhance the accumulation of autophagosomes in cells in order to maintain a survival niche for S. aureus. Within these induced autophagosomes, S. aureus are protected and ultimately facilitates dissemination. S. aureus survival rate is significantly reduced using an agr-deficient mutant, suggesting that the agr locus is crucial for autophagy-mediated intracellular survival (Mulcahy et al., 2020). Similarly, the agr mutant showed a significantly reduced intracellular survival rate in mouse phagocytes because they fail to accumulate LC3-II⁺ autophagosomes and are delivered efficiently to lysosomes (O'Keeffe et al., 2015). These results indicate that agr-regulated factors determined the ability of S. aureus for autophagy targeting and avoidance of lysosomal degradation in host cells ( Figure 2 ). However, in human osteosarcoma cells, agr-positive S. aureus strains were more efficiently entrapped in autophagosomes than agr-negative S. aureus (Mauthe et al., 2012). Additionally, a recently study showed the absence of agr regulated PSMs increased S. aureus long-term survival in human endothelial cells (Siegmund et al., 2021). Thus, a comprehensive analysis of different S. aureus strains as well as various cell types is required to elucidate the interplay between agr and autophagy.

The emergence of antibiotic-resistant strains of bacteria requires the continuous development of new antibiotics. However, drug development is a very long and expensive process. The exploration of new drugs for effective treatment of S. aureus infection is a difficult task. Many years of research have produced a few drugs, such as penicillin, vancomycin, and cephalosporin. However, once the pathogen becomes intracellular, antibiotics will not achieve the expected effect, and new antibiotics may have an impact on cell function. In this case, development of new molecules must be explored in order to defeat S. aureus. Autophagy, as an innate immune response mechanism, degrades S. aureus in cells. Strains with high agr activity are usually able to escape and replicate intracellularly using autophagy, while strains lacking agr systems are usually unable to escape the autophagosomes and are eventually degraded.

Appropriate doses of the autophagy modulators could be an effective strategy for controlling infection. A recent study has reported the natural coumarin derivative daphnetin (DAPH) effectively enhances autophagic pathway to exert an anti-bacterial effect against S. aureus (Zhang et al., 2019). Moreover, selenium has been shown to inhibit the proliferation of S. aureus by promoting autophagy pathway in S. aureus infected mouse macrophages (Zang et al., 2020). Regardless of S. aureus strain identity, their methods of escaping autophagy pathway usually involve blocking autophagy flux. The recently developed pH-responsive polymersome (Xu et al., 2020) loaded with LC3 and p62, disintegrates after encountering lysosomes with low pH, releasing the loaded proteins to supplement autophagy flux, which could be a new strategy. However, the situation is more complicated than expected, the fact that autophagy inducers seem to be beneficial for treating S. aureus infections, but in turn might facilitate other bacterial infections (Escoll et al., 2016). Therefore, the use of autophagy modulators should be highly cautious. Additionally, the ability of S. aureus to escape and survive in the cytosol are dependent on both the strain and cell type. Treatment with autophagy inhibitors was shown to reduce S. aureus load, and the autophagy induction by rapamycin restored replication of S. aureus (Schnaith et al., 2007; Bravo-Santano et al., 2018). It is difficult to perform corresponding treatment of S. aureus infections with different genetic backgrounds and different targeted specific cell types.

We need novel approaches to suppress intracellular S. aureus load with minimal side effect to the host. And obviously, the significance of eliminating intracellular bacteria for effective treatment of persistent S. aureus infections has received more attention. The vancomycin encapsulated within liposomes was shown to be taken up efficiently by Kupffer cells and killed intracellular S. aureus, which reduced the mortality of mice, whereas free vancomycin could not (Surewaard et al., 2016). Combining autophagy modulators with the liposomes may be a promising strategy. The recent focus on developing strategies for intracellular S. aureus is encouraging and may lead to more effective treatments in the near future.

MW performed the literature survey and wrote the draft. ZF critically reviewed and improved the manuscript. HH contributed to critical evaluation and finalizing of the review. All authors contributed to the article and approved the submitted version.

This study was supported by the National Key R&D Program of Intergovernmental Key Projects in China (2018YFE0101700).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.