The treatment of bacterial infections has changed dramatically with the development of antibiotics, and many lives have been saved; in the era of the COVID-19 pandemic, tuberculosis, a widespread and deadly disease, has not been conquered, but it has been set aside as an infectious disease. On the other hand, bacterial infections are still strictly a major threat to people’s lives and a major treatment target, one of which is sepsis caused by a rapid and massive bacterial load (1) and another is the emergence of antimicrobial resistance bacteria spurred by antibiotic overuse (2). Infectious diseases other than sepsis, such as tuberculosis, infectious gastroenteritis, bacterial pneumonia, and food poisoning, are limited to localized areas, and dysfunction of the tissues of the infected foci comes to the fore. Sepsis is defined as organ damage due to an inadequate host response to infection (3). In addition to the classic treatment of infusion, removal of infected lesions, and respiratory circulatory support, treatment is aimed at normalizing coagulation abnormalities to maintain organ microcirculation (4). Nevertheless, more than 11 million lives are lost annually to sepsis, making it the cause of nearly 20% of deaths worldwide (5). One of the reasons that current medical treatment for sepsis has been so hampered is that the host’s immune system forms a troublesome response to sepsis. That response is the coexistence of an excessive inflammatory response and a prolonged state of immunosuppression (6). The former, also called a cytokine storm, is characterized by an overproduction of inflammatory cytokines as the predominant phenotype (7). Much of the pathology of sepsis is associated with this unhelpfully exuberant reaction of the host, which is thought to be a common end pathway that occurs with viral infections as well as bacterial infections, and suppression of excess cytokines and regulation of their receptors is thought to reduce the disease state (8). However, the major proinflammatory cytokines IL-1 (9) and TNF-α (10) which are major inflammatory cytokines in sepsis, have failed to improve the survival of sepsis. In addition, inhibitors of Toll-like receptor (TLR) 4, which detects many bacteria and transduces intracellular signals that trigger inflammation, have also failed to improve sepsis survival (8). Despite such disappointment, transcriptome analysis of leukocytes from patient blood in sepsis revealed that up to 80% of the pathways of cellular function are altered and that inflammatory and regulatory mechanisms are simultaneously driven in the first few hours after onset (11). The setbacks in these clinical trials and the genetic approach to pathophysiology have led to a major shift in our current understanding of the pathogenesis of sepsis, in which host immunity to sepsis is a conflict between attack and suppression, far from its original goal of eliminating pathogens (12). This understanding of the pathogenesis has led to a search for therapeutic strategies that achieve homeostasis of host immune function.

For antibiotics, the invention of new drugs in the nearly 30 years since the 1940s, a golden age, has been a wonderful scientific breakthrough that has led to an overly optimistic fantasy that bacterial infections will cease to be a threat to humanity (2). However, in the half-century since the 1970s, only a few new classes of antibiotics have been invented, and in addition, we have been handicapped by the disastrous situation with multidrug-resistant bacteria. Despite advances in understanding the life cycles of bacteria and long-awaited advances in molecular biology and genetics, biology and medicine today are far behind the good old days of the past 30 years in terms of progress in the field of infectious disease treatment. In the past half century, mankind has not only used invented antibiotics in large quantities in medicine but also abused them in livestock in search of economic rewards (13). The result has been a situation in which bacteria that have acquired multidrug resistance, also known as superbugs, have become rampant. Tuberculosis remains an uncontrolled infectious disease worldwide, and is the leading cause of mortality among mono infectious diseases, with 1.4 million deaths per year (14). Currently, it is estimated that 1/4 of the earth’s population is infected with Mycobacterium tuberculosis, the majority of which is considered to be in the latent stage, but reactivation is a common occurrence (15). Although the current standard of care is to continue the four-drug combination for at least six months (16), reinfection cannon be completely prevented, and 18% of these infections are caused by multidrug-resistant organisms (14, 17). Modern medicine has devoted many resources, and with the struggles of the field, has managed to prevent the global spread of antibiotic-resistant bacteria from becoming a pandemic threat like COVID-19. Given that the doubling time of bacteria is in hours, the speed of their molecular evolution is tremendous. Considering that even if a new drug is invented, it takes a certain amount of time for its clinical development, this weasel-word is extremely at disadvantage for humans. Nevertheless, new antibiotics are being developed to win the battle, and in addition, methods to attenuate bacterial toxins and phage-based methods (18, 19).

Understanding the pathogenesis of sepsis is directly linked to drug development, which is moving toward therapies that can eliminate pathogens while balancing the active and regulatory systems of the immune system (6). One of the molecular basis of sepsis is the transformation of the energy supply system of immune cells, and various compounds related to PGC1α, which activates mitochondrial biosynthesis, are being investigated for their efficacy in the treatment of sepsis (20). In the treatment of antibiotic-resistant bacteria, the use of immune checkpoint inhibitors that block inhibitory signals in T cells, TGF-β to activate T cells, M1-like macrophage adaptive transfer, and strategies such as the administration of gelsolin, an endogenous protein, to enhance the pathogen clearance of macrophages are beginning to be explored (5). Methodologies to intervene in host immunity and promote pathogen elimination are beginning to emerge in the form of specific methods and compounds.

In this review, we focus on therapeutic strategies for infectious diseases through intervention in the host rather than approaches to the pathogens themselves. In the field of cancer therapy, the development of drugs aimed at killing the cancer cells themselves and the intervention of host immunity have made remarkable progress (21). Although immunity plays a major role in infectious diseases, the host approach has been neglected to date. Defense against microorganisms is mediated by the effector mechanisms of innate and adaptive immunity. Innate immunity is mainly responsible for defense in the early stages of infection, whereas adaptive immunity, together with innate immunity, provides a stronger and more specific response, and establishes a sustained defense posture with immune memory (22). The balance between these host immune responses and the acquisition of microbial resistance determines whether infection is established (23). The initial response of the host to bacterial infection is recognition of the bacteria by cells possessing pattern recognition receptors, release of inflammatory cytokines such as IL-1β, TNF-α, and IL-6, vasodilation, and increased vascular permeability (24). This leads to the accumulation of leukocytes, mainly neutrophils, which are non-specialized phagocytes, and more phagocytes in the infected foci. These cells phagocytose extracellular bacteria and infected cells and serve as the first line of defense in bacterial clearance (25). Inflammatory cytokines activate adaptive immunity, leading to enhanced antibody production by B cells, and opsonized bacteria are subject to phagocytosis by phagocytes, while T cells produce a variety of cytokines, including IFN-γ, to enhance bacterial killing by phagocytosis (26). Antibodies, together with activated complement, cause bacterial neutralization and lysis and play a role in the host defense system. In the early stages of infection, phagocytosis plays a central role in bacterial killing. On the other hand, bacteria that cause intracellular infection ensure their survival and replication by disabling the phagolysosomal system, which is the executor of intracellular disinfection.

Mycobacterium tuberculosis, which causes intracellular infection, can cause delayed-type hypersensitivity and tissue damage. Slow-growing Mycobacterium tuberculosis evades the killing of the phagolysosomal system and survives intracellularly, resulting in persistent stimulation of T cells and macrophages and the formation of granulomas (27). This granuloma and the solid tumor microenvironment share common features of immunosuppressive conditions such as lymphocyte exhaustion/elimination, macrophage polarization to M2-like phenotype, hypoxia, immunomodulatory cytokines such as TGF-β/IL-10, and infiltration of myeloid-derived suppressor cells (5). This similarity reminds us that host-directed therapy, which has been successful in anticancer therapy, could bear great fruit in infectious diseases. Among host immune mechanisms, the phagolysosomal system is considered to be at the center of pathogen control and an appropriate target for infection control. In order to examine the possibility of intervention in the phagolysosome system in host-directed therapy, the molecular mechanism of the pathway from phagocytosis to phagosomes reaching lysosomes is discussed from the perspective of host-pathogen interaction. Finally, the current status and future potential of drug discovery targeting the phagolysosome pathway will be discussed.

When considering host-directed therapy for bacterial infections, it is important to divide bacteria into those that cause extracellular infections and those that cause intracellular infections. Bacteria that cause extracellular infections include Staphylococcus aureus, Streptococcus pyogenes, Streptococcus pneumoniae, Escherichia coli, Vibrio cholerae, Clostridium tetani, Neisseria meningitidis, and Corynebacterium diphtheriae. On the other hand, bacteria that cause intracellular infections include Mycobacterium, Listeria monocytogenes, and Legionella pneumophila (22). Microorganisms can be classified into three types: 1) obligate extracellular growth parasites, which cannot grow inside the cell but only outside, 2) facultative intracellular growth parasites, which can grow both inside and outside the cell, and 3) obligate intracellular growth parasites, which can grow only inside the cell. Obligate extracellular growth parasites are eliminated by phagocytes and have developed resistance mechanisms against phagocytosis (22). S. aureus is classically recognized as a bacterium that causes only extracellular infections such as furuncles, carbuncles, impetigo, abscesses, septicemia, necrotizing pneumonia, and biofilm formation (28). Recent studies have shown that S. aureus can survive and proliferate intracellularly, which is a major factor in pathogenesis, making it a second category of bacteria (29) (Horn). Bacteria belonging to this category have evolved the ability to neutralize the phagolysosome. The third category of bacteria includes rickettsia and chlamydia, which are dependent on the host in terms of membrane structure and metabolism, respectively, but the immune mechanisms against them are beyond the scope of this review and should be referred to the cited review (30).

Innate immunity to extracellular infections is centered on complement activation, phagocyte activation, and inflammatory responses, and the final execution mechanism of bacterial elimination depends on the phagolysosomal system. Phagocytes directly recognize bacteria via mannose and scavenger receptors and enhance phagocytosis (25). In addition, both peptidoglycan, a major membrane component of Gram-positive bacteria, and LPS, an endotoxin of Gram-negative bacteria, activate the alternative complement pathway to opsonize bacteria. Like complement, bacteria opsonized by antibodies enhance phagocytosis (26). Extracellular bacterial protein antigens cause activation of CD4+ T cells, also assisting phagocytosis. Although neutralization and lysis of bacteria by antibodies are important defense systems, phagolysosomes as the final executor of bacterial elimination are central to bactericidal activity. Bacteria that produce intracellular infections have found a microenvironment (niche) within the phagocyte that is isolated from strong adaptive immunity and have acquired mechanisms that allow them to survive and replicate there. These bacteria have evolved mechanisms to disable the phagolysosomal system within the phagocyte and hijack the phagosome to survive (27). Adaptive immunity attempts to execute bacterial clearance through activation of the phagolysosomal system by recruiting phagocytes with the CD40 ligand signal and INFγ by CD4+ T cells. In the process of escape from the phagosome, the host can trigger a mechanism by which CD8+ T cells, upon receiving the signal, eliminate the infected cell itself (22). This section on host-pathogen interactions describes the general effector function of the host’s phagolysosomal system on pathogens.

The immune system quickly detects invading bacteria in the body and timely initiates phagocytosis as the appropriate response to eliminate the threat ( Figure 1 ) (25). Phagocytes are estimated to make up less than 1% of all cells in the body (31). The ability of these cells to adequately patrol and scavenge throughout the body is critical for defense against foreign enemies (32). Although phagocytes form a constant protrusion (33), and signals from the calcium-sensing receptor (CaSR), a G protein coupled receptor, regulate phosphatidylinositide phosphorylation plasma membrane remodeling (34), and polymerization of the branched actin network just below the plasma membrane (35). Pathogen-sensing receptors include the pattern recognition receptors (PRRs) such as TLR4 that directly bind to pathogen surface structures (24), Fcγ receptors, and complement receptors interact with antibodies or complement that opsonize pathogens, and their interactions play a role in signaling to phagosome formation (25). The pattern recognition receptors (PRRs) involved in bacterial infections are Toll-like receptors (TLRs), nucleotide oligomerization domain (NOD)-like receptors (NLRs), C- and C- type lectin receptors (CLRs), and absent in melanoma-2 (AIM2)-like receptors (ALRs). Ten TLRs have been identified in humans and are present in dimeric form at the plasma membrane or phagosomal membrane (36). In the plasma membrane, TLRs exist as homodimers or heterodimers and recognize lipids, proteins, lipoproteins, and other components of microorganisms. On the other hand, in phagosome membranes, they exist as homodimers and recognize microbial nucleic acids (37). TLR1- TLR2 and TLR2-TLR6 are expressed in monocytes, dendritic cells, and are involved in the recognition of triacyl lipopeptide, lipoprotein, lipopeptide, lipoteichoic acid, arabinomannan, peptidoglycan. TLR4 homodimer is expressed in macrophages and dendritic cells and binds to lipopolysaccharide. TLR4 homodimer is expressed in macrophage and dendritic cells and recognizes lipopolysaccharide. TLR5 homodimer is expressed in intestinal epithelial cells and senses flagellin (24). NOD1 in the cytoplasm of intestinal epithelial cells and macrophages recognizes γ-D-glu-meso-diaminopimelic acid in the cell wall of gram-negative bacteria, while NOD2 recognizes muramyl depeptide in the cell wall of all bacteria (38). CLRs are expressed on macrophages and dendritic cells and play a critical role in anti-fungal immunity. They include the mannose receptor, which recognizes mannose units repeated on the surface of bacteria such as mycobacterium and induces phagocytosis, and the Asian glycoprotein receptor family, which includes Dectin-2, which recognizes mannose-capped lipoarabinomannan (39). ALRs are PRRs that recognize intracellular double strand DNA and do not participate in innate immunity but are involved in apoptosis (40).

The protrusion-captured target induce clustering of phagocytic receptors, and the immunoreceptor tyrosine-based activation motif (ITAM) in their intracellular domain (41) is productively phosphorylated by Src-family tyrosine kinases (SFKs), spleen tyrosine kinases (Syk) (42). As soon as the phagocytosis signal begins to amplify and a transient increase in PI (4, 5)P₂ occurs, conversion occurs from PI (4, 5)P₂ to PI (3–5)P₃ by PI3K recruited to adaptor proteins (43). PI (3–5)P₃ surges recruits phospholipase Cγ which breaks down PI (3–5)P₃ to produce diacylglycerol (DAG) and inositol (1, 4, 5)-triphosphate (IP₃) (44). DAG acts as a second messenger for signaling between phagocytic receptors (45), while IP₃ provides calcium spike from the endoplasmic reticulum (ER) into the cytoplasm (46). These two 2nd messengers cooperate to activate small G protein Rap1, which mediates the “inside-out” response of integrin (47). PI3K activates Rho family GTPases that facilitate cytoskeletal remodeling directly and through the GEF. This activation dynamically alters the cytoskeleton to form phagocytic cups. NADPH oxidase 2 (NOX2), which is responsible for the generation of reactive oxygen species (ROS) that cause oxidative bursts, engages in the newly formed phagocytic cups (48).

On the other hand, there are systems that prevent phagocytosis, which phagocytoses pathogens and apoptotic cells, from running amok and eliminating normal cells. In cancer research, immune checkpoints have been identified as entities of T cell regulatory mechanisms (49). Immune checkpoint inhibitors such as anti-programmed cell death protein 1 (PD-1; pembrolizumab and nivolumab) (50), anti-cytotoxic T lymphocyte-associated protein 4 (CTLA-4; ipilimumab and tremelimumab) (51), anti-PD-1 ligand 1 (PD-L1: atezolizumab, avelumab and durvalumab) (52) have developed, and demonstrated to significantly improve outcome in patients suffered from devastating cancers ( Figure 2A ). In innate immunity, phagocytosis checkpoints recognize “Don’t eat me” signals during the phagocytosis process, and are beginning to be recognized as important new targets for cancer immunotherapy (53). The discovery was signal-regulatory protein α (SIRPα) expressed in the myeloid lineage (54). Upon the interaction of SIRPα and CD47, the intracellular domain of SIRPα, an immunoreceptor tyrosine-based inhibitory motif (ITIM) recruits SH2-containing protein tyrosine phosphatase 1 (SHP1) or SHP2, preventing myosin IIA dephosphorylation, subsequent rearrangement of the cytoskeleton, and the phagocytosis they form (55) ( Figure 2B ). Although PD-1 expression is an important marker of T cell exhaustion (56), it also is expressed on various immune cells, including macrophages. The interaction with PD-1 and PD-L1 provides a suppressive signal for the phagocytosis of tumor-associated macrophages (TAMs) (57). Cancer cells can escape macrophage-induced phagocytosis by expressing PD-L1. Sialic acid-binding immunoglobulin-like lectin (SIGLEC), which contains inhibitory receptor motifs (ITIMs) in its intracellular domain, is induced at the surface of macrophages and its expression confers a poor prognosis in cancer patients (58). The ligand for SIGLEC is CD24, and this interaction serves as an entity for anti-phagocytic action (59).

Non-specialized phagocytes involved in bacterial infections include epithelial cells, endothelial cells, osteoblasts, and fibroblasts. Epithelial cells, which play another major role among non-specialized phagocytes, cover the outer and luminal surfaces of the body and organs. Depending on where they are present, epithelial cell functions include absorption in the lungs and intestinal tract, secretion in the kidneys and stomach, and material transport in the trachea and oral-nasal cavity. Regardless of their location, the basic function of epithelial cells is to interact closely with the external environment and, in particular, to serve as the first line of defense in the immune system (60). Epithelial cells possess pattern recognition receptors to detect external hazards, but do not express receptors to capture opsonized pathogens as do specialized phagocytes. Therefore, phagocytosis of pathogens by epithelial cells is initiated by two following main methods. One is a trigger mechanism in which the cytoskeleton is restructured by effector molecules secreted by the bacteria, forming ruffles on the plasma membrane, and the other is a zipper mechanism in which the bacteria attach to proteins involved in cell adhesion, such as integrins and cadherins (61). After internalization, the phagolysosome system takes over for clearance.

Bacterial degradation in first line innate immunity is carried out through three main pathways ( Figure 1 ): phagolysosome, xenophagy, and LAP (Microtubule-associated proteins 1A/1B light chain 3B (LC3)-associated phagocytoisis), each of which is followed by a vesicle: phagosome, autophagosomes, and LAPosomes, respectively, through fusion with lysosomes (60). In case of the first pathway, it begins with internalization by phagosomes after pathogen recognition, during which signaling occurs in the cell, leading to phagosome maturation (25) The phagosomes are then translocated to the lysosomes. Subsequently, the pathogen is degraded by phagolysosomes that are generated by fusion with lysosomes (60). On the other hand, autophagy is triggered when the imported bacterium attempts to escape from the phagosome, causing vesicle damage (62). Autophagy to enclose pathogens is called xenophagy, whose efficacy largely depends upon lysosomal function. In the third pathway, LC3, which plays a major role in autophagy, is embedded in the phagosome membrane in the form of LC3 modified by phosphatidylethanolamine (referred to as LC3-II), resulting in the formation of the LAPosome. The LAPosome is partitioned by a single membrane like the phogosome, unlike autophagy, which has a double membrane beginning in the phagophore (63). Its major feature is the quick recruitment of NADPH and the burst of reactive oxygen species that its enzymatic activity leads to, and the maturation of the LAPosome is faster than that of conventional phagocytosis. Review of bacterial killing in LAP and xenophagy are beyond the scope of this review; see other reviews (62, 63).

Salmonella typhimurium interferes with TLR4 recognition and signaling through deacylation and palmitoylation of lipid A present on their surfaces (137). Helicobacter pylori dephosphorylates lipid A to escape TLR4 (138). Haemophilus influenzae, H. aegyptius, Streptococcus peumonie, and Neisseria gonorrhoeae secrete proteases that selectively cleave immunoglobulin A, which is responsible for opsonization (139, 140). Complement also plays a role in opsonization, but some bacteria exploit the elaborate activation pathway of complement. Neisseria mengitidis is a bacterium that avoids phagocytosis by mimicking the host complement factor H (fH), the regulatory substance of complement, on its own surface (141). Staphylococcus aureus also recruits fH to its surface by secreting a substance called SdrE (142). YopH, a protein tyrosine phosphatase produced by Yersinia spp., dephosphorylates host phosphotyrosine proteins and prevents phagocytosis (143). Vibrio parahaemolyticus secretes an inositol polyphosphate 5-phosphatase, VPA0450, which disrupts host cell membrane integrity and causes blebbing (144). This tactic is also used by Shigella flexneri, which secretes IpgD, an inositol 4-phosphatase, as its virulence factor, causing plasma membrane blebbing by converting PI (4, 5)P2 to PI (5)P (145). Although not a bacterium, m128L encoded by Myxoma virus has a high homology to CD47, which is known to inhibit host phagocytosis as a “don’t eat me” signal (53). The “don’t eat me” signal is essential for establishing the lethal infection through inhibition of host phagocytosis (146). Although there are no reports of cases in which the bacteria themselves encode CD47 mimic, enhanced expression of CD47 in cells infected with S. typhi and Borrelia burgdorferi has been reported (147). The upregulation of CD47 expression occurs through signaling from PRRs and is also enhanced by inflammatory cytokines, suggesting that the CD47-SIRPα axis may work to suppress excessive inflammatory responses. Bacteria make good use of this host regulation mechanism to aid immune evasion. On the other hand, M. tuberculosis, a phagocytosis-dependent intracellular parasite, is unique in that it does not enhance CD47 expression, unlike other bacteria.

Although we are far from having a complete picture of the diverse strategies by which bacteria neutralize host immune attack within the cell, learning the molecular mechanisms is the first thing we must do to win the war against bacteria. The strategy of bacteria that acquire permissive niches intracellularly as intracellular pathogens (IPs) (148) provides a great clue for the construction of host-directed therapeutics (HDTs) (27). M. tuberculosis, a leading IP, has long been a major scourge to mankind due to its high prevalence and high mortality rate on a global scale (149), and is one of the most carefully investigated (150). M. tuberculosis is thought to have evolved in such a way to struggle with the host immune system that an exhaustive list of protein and lipid effectors produced by the bacillus has been compiled (151). In addition to M. tuberculosis, other potential IPs include Rickettsia rickettsia, Chlamydia trachomatis, Legionella peumophila, Coxiella burnetiid, Brucella abortus, and Salmonella enterica; Cryptococcus neoformans and Aspergillus fumigatus among fungi. Candida albicans uses other intracellular organelles such as mitochondria as a habitat. The mechanisms by which they evade the phagosome pathway have been intensively studied, and detailed molecular mechanisms have been elucidated (152).

The application of immune checkpoint blockade to infectious diseases has been investigated in the context of the interrelationship between innate and adaptive immunity, rather than between infected cells and innate immunity. Interventions on the PD1-PD-L1 axis are effective in animal studies against infectious diseases such as malaria, toxoplasma, leishmania, and Listeria ( 197). The results of a phase 1/2 trial of nivolumab in sepsis have overcome safety concerns, including the development of autoimmune disease (198). In cancer therapy, although macrophages have been intensively studied as targets for intervention in various aspects, phagocytosis that macrophages execute is recognized as a promising drug discovery (199). The findings of the phagocytosis checkpoint may provide clues to the treatment of infectious diseases (53). The inhibition of the PD-1-PD-L1 axis as a phagocytosis checkpoint enhances phagocytosis in liver Kupffer cells and prevent bacterial infection (200). A study using CD47 KO mice also reported that E. coli pneumonia showed better recovery compared to wild type (201).

It was reported that monocytes from imatinib-treated patients with leukemia showed an increased production of V0a3 and V0c and had more acidic lysosomes. Sera from those patients, which are added to the cell culture of macrophages, enforced more acidic lysosomes and of M. tuberculosis ( 202). One compound was reported to promote regulated assembly of V-ATPase, KU-60119, which was identified as an ATM inhibitor that inhibited assembly of V1E and V1G through phosphorylation of the latter (119). ZLN005 was originally recognized as a peroxisome proliferator-activated receptor gamma coactivator 1-α (PGC1α) activator (203). By administering ZLN005 to the cecum perforation ligation (CPL) model of sepsis and analyzing intraperitoneal cells, ZLN005 was shown to be a transcription factor EB (TFEB) activator involved in lysosome biogenesis as well as a lysosomal acidifier (204). The compound significantly improved overall survival and drastically reduced intraperitoneal bacterial load in mice at only 2 h after administration of ZLN005 in in vivo sepsis model. These results indicate that lysosomal acidification is a therapeutic target for sepsis.