Brucella Phagocytosis Mediated by Pathogen-Host Interactions and Their Intracellular Survival

The Brucella species is the causative agent of brucellosis in humans and animals. So far, brucellosis has caused considerable economic losses and serious public health threats. Furthermore, Brucella is classified as a category B bioterrorism agent. Although the mortality of brucellosis is low, the pathogens are persistent in mammalian hosts and result in chronic infection. Brucella is a facultative intracellular bacterium; hence, it has to invade different professional and non-professional phagocytes through the host phagocytosis mechanism to establish its lifecycle. The phagocytosis of Brucella into the host cells undergoes several phases including Brucella detection, formation of Brucella-containing vacuoles, and Brucella survival via intracellular growth or being killed by host-specific bactericidal activities. Different host surface receptors contribute effectively to recognize Brucella including non-opsonic receptors (toll-like receptors and scavenger receptor A) or opsonic receptors (Fc receptors and complement system receptors). Brucella lacks classical virulence factors such as exotoxin, spores, cytolysins, exoenzymes, virulence plasmid, and capsules. However, once internalized, Brucella expresses various virulence factors to avoid phagolysosome fusion, bypass harsh environments, and establish a replicative niche. This review provides general and updated information regarding Brucella phagocytosis mediated by pathogen-host interactions and their intracellular survival in host cells.


Introduction
Brucellosis is a globally distributed major bacterial zoonosis characterized by abortion and reproductive failure in livestock, which seriously affects the development of animal husbandry and international trade. Although the mortality rate is low, it is harmful to human health as it can lead to a debilitating disease and serious chronic complications if left untreated [1][2][3]. The transmission of infection to humans is primarily via direct contact with animals, handling of contaminated tissues, and consumption of unpasteurized milk [1]. Furthermore, the Brucella species is classified as a category B bioterrorism agent due to its ease of transmission via aerosol [4].
No licensed human vaccine is available against brucellosis, and despite harsh therapy using a combination of two antibiotics for several weeks to months, relapses or clinical failures still occur [5,6]. Consequently, the control of human brucellosis depends on the control of brucellosis in livestock, which cannot primarily rely on vaccination alone but must be tackled in combination with proper husbandry measures [6,7]. Additionally, the widely used live attenuated Brucella vaccines for animals, such as Brucella (B.) abortus S19 and B. melitensis Rev. 1, both induce abortions, are virulent for humans, and interfere with serodiagnosis. Furthermore, Rev. 1 is resistant to streptomycin, which is a necessary As mentioned earlier, Brucella possesses the ability to cause chronic infection in many different cell types. Brucella must break the epithelial barriers to cause systemic infection. To do that, the bacteria first have to invade the host cells to initiate their intracellular infectious cycle. Therefore, adhesion to the host cells is the first and most essential step in the invasion process. Brucella expresses various adhesin molecules to mediate the host and pathogen interaction. Brucella adhesins include the sialic acid-binding proteins SP29 and SP41; BigA and BigB proteins containing the immunoglobulin-like domain; the monomeric autotransporters BmaA, BmaB, and BmaC; the trimeric autotransporters BtaE and BtaF; and collagen, vitronectin-binding protein Bp26. These molecules regulate Brucella adhesion to host cell surface molecules and extracellular matrix components. In addition to ensuring the adhesion of Brucella to the host cell surface, these adhesins are considered immunogens or vaccine candidates to activate host immunity [12][13][14][15][16][17]. A study by Castaneda-Roldan et al. [16] displayed that a Brucella surface protein SP41 can bind selectively to epithelial HeLa cells. This adhesin protein is the expressed product of the ugpB gene. Brucella ugpB knockout strain decreased bacterial internalization by 40-to 50-fold less than the wild-type strain. Besides, another surface protein with an apparent molecular mass of 29-kDa was demonstrated to target the host gangliosides structure on red blood cells. A fraction of SP29 was found as a periplasmic protein in B. melitensis [18]. A region in chromosome 1 of B. abortus contains a gene, bigA, which codes an adhesin. This adhesin mediates the adherence and invasion into Madin-Darby canine kidney (MDCK) and Caco-2 cells. By performing immunofluorescence microscopy and Western blot assay, this adhesin was proven to localize at the outer membrane and contain an immunoglobulinlike domain, respectively [19]. It is noteworthy that VirB5 is an effector of a well-known Brucella virulence factor T4SS that plays an essential role in Brucella infection and is also recognized as an adhesin to contact with the host cells. Indeed, Deng et al. [20] determined that a single domain antibody, BaV5VH4, can bind Brucella VirB5 protein, disrupting the interaction between Brucella and host macrophage cells. Though the VirB5 protein in Agrobacterium tumefaciens localizes to the T-pilus tip that contributes to host-cell recognition, its direct adhesion function in this bacterium or Brucella is still unclear and needs to be deeply clarified [21].

Host Receptors
Once Brucella adheres to host cells, the binding of the pathogen to various host cell receptors activates a series of signaling pathways essential for bacterial uptake. Opsonized bacteria are internalized via complement receptors (CRs) and Fc receptors (FcRs) which promote the efficient phagocytosis of Brucella into host phagocytes. A major virulence factor of Brucella, the LPS O-chain fragment, is responsible for binding to antibodies or the C3 component of complement systems [22][23][24]. Opsonization of Brucella enhances the uptake of Brucella into human monocytes [25]. Eze et al. [22] reported that antibody and complement-mediated opsonization resulted in a profound increase in the phagocytosis of B. melitensis in murine peritoneal macrophages in individual or synergistic effects. However, another study by Rittig et al. [26] showed that the opsonization of Brucella by antibody but not complement contributed to increasing the uptake of Brucella in human monocytes. These authors proposed a reason that complement receptor 3 (CR3) has both opsonic and non-opsonic binding sites. Thus, Brucella can bind directly to this receptor without the need for the main opsonic complement fragment iC3b [26][27][28]. Although the opsonic entry plays an essential role in the early stage of Brucella infection, studies have reported that the contribution of FcRs and CRs in the phagocytosis of Brucella are still limited. Among FcRs, FCγRIIA plays a crucial role in the phagocytosis of IgG2-opsonized bacteria. Hosseini khah et al. [29] proved that FCγRIIA (CD32) and the polymorphism of this receptor are associated with susceptibility to brucellosis in humans. The most efficient phagocytic receptor among CRs is CR3 (integrin, Mac-1) [28,30].
On the other hand, host cell receptors classified as lectin and fibronectin receptors recognize non-opsonized bacteria. Scavenger receptor A (SR-A) is a well-described lectin receptor in recognition of Brucella. This receptor can effectively detect lipid A fraction of LPS [31]. Interestingly, following Brucella internalization by SR-A recognition, T4SS is considered a regulator for SR-A signal transduction that enables Brucella to establish an intracellular lifecycle in the host cells [31,32].
Among host receptors that recognize Brucella, toll-like receptors (TLRs) are most characterized for their functions in recognition and immune response against brucellosis [33]. Much experimental evidence suggests the essential role of TLRs in phagocytosis and in the host signaling pathways to protect against Brucella infection. TLR2, TLR4, and TLR6 are supposed to recognize different membrane components of Brucella, while TLR3, TLR7, and TLR9 have been found accountable for the detection of Brucella nucleic acid structure motifs [34][35][36][37][38]. Although TLRs contribute to the host protection against Brucella infection, these pathogens may interfere with TLRs-mediated immune signaling to evade recognition by the innate immune system. A TIR domain-containing protein in Brucella (TcpB) interferes with MAL-TLR4 interaction leading to the inhibition of TLR4 signaling pathway. Particularly, the DD loop of TB-8 and TB-9, which are TcpB-derived decoy peptides, is sup-posed to be able to specifically bind to the MAL TIR domain to form a dimer, then disrupt MAL TIR domain-mediated interaction leading to the inhibition of TLR4 signaling. This interference prevents DCs maturation and the cytotoxic activity of cytotoxic T lymphocytes during Brucella chronic infection [39][40][41]. TLR2 recognizes Brucella lipoprotein, but it has been revealed to have no function in controlling Brucella infection in the TLR2 knockout mice model [42,43]. TLR4 can recognize LPS, and a non-canonical lipid A of LPS at a very high concentration resulting in the promotion of the maturation phagocytic vacuoles and inducing the pro-inflammatory cytokine TNF-α [44]. Regarding receptor costimulation, Brucella can express more than one ligand for various host cell receptors. Heat-killed B. abortus simultaneously costimulates TLR2 and TLR9 in mouse DCs. This costimulation ensures bacterial phagocytosis and promotes the downstream signal transduction of these receptors [45]. In addition, TLR2 also plays a critical role in Brucella invasion into trophoblast giant cells [46]. This suggests that Brucella can infect different types of cells in the same way.

Bacterial Engulfment
Upon translocation across the mucosal epithelial cell layer, professional phagocytes such as macrophages, DCs, and macrophage-like cells engulf the bacteria in which <10% of these phagocytosed Brucella survive and escape killing during the initial phase of infection. Following the binding of Brucella, phagocytic and non-phagocytic receptors mediate a variety of signaling pathways, leading to the remodeling of the cell membrane structure to engulf bacteria [6,8]. Generally, Brucella induces a zipper-like mechanism for internalization [47]. Up to 8 min after contact, Brucella moves in a swimming motion on the cell surface with a generalized membrane ruffling. At the same time, the cell surface is rearranged and actin polymerization is activated around the side of bacterial adherence. Notably, these events depend on the Brucella virulence factor VirB [48,49]. Moreover, the cellular prion protein (PrP c ) is a glycoprotein anchored on the outer leaflet of the plasma membrane. It is associated with phagocytosis via cytoskeletal rearrangement and the host inflammatory response. The PrP c is described as a major internalization receptor on M cells during Brucella via the oral route [50][51][52]. However, the role of PrP c in Brucella internalization is still controversial. Indeed, upon B. melitensis infection, silencing of PrP c in microglia cells did not affect bacterial phagocytosis and intracellular killing [53].
Actin polymerization is required in the penetration of Brucella into host cells. Bacterial adhesion to cell surface involves other proteins acting as second messengers including cGMP, PIP3-kinase, MAP-kinase, and tyrosine kinase that leads to activation of GTPases of Rho subfamilies such as Rho, Rac, and Cdc42, which play a critical role in the regulation of the cytoskeleton [8,54]. Additionally, a genome-wide small interfering RNA perturbation screen was performed to identify novel host factors crucial for Brucella intracellular trafficking. The results displayed that the prominently affected clusters are related to actinremodeling and phagocytosis. These clusters include ARP2/3, WASP regulatory complex, and the small GTPases Rac1 and Cdc42 [55]. Consistently, adhesion to the macrophage surface involves the activation of GTPases and F-actin polymerization where, at the early stage of infection, annexin I is also implicated in membrane fusion [8,49,54]. Besides, adhesin protein BigA was proven to induce actin-cytoskeleton rearrangement in MDCK and Caco-2 cell lines [19]. The TLR4/PI3-kinase signaling pathway is essential for smooth Brucella strain internalization but not in the rough strain. The smooth strain infection quickly elicits F-actin polymerization at the early stage of infection and fuses the early endosome faster than the rough strain [56].

Formation and Maturation of Brucella-Containing Vacuoles
Internalization occurs via lipid raft microdomains at the macrophage cell membrane that are rich in cholesterol, glycosylphosphatidylinositol, and gangliosides, and contribute to the intracellular trafficking of Brucella [8,52]. Following interaction with the cell mem- brane of professional or non-professional phagocytes and around 5 min after entry, BCVs enter the endocytic pathway, initiated by the fusion with early endosome, characterized by Rab5, early endosomal antigen 1, and the transferrin receptor (TfR) markers [57,58]. At the next stage of internalization, BCVs continue to fuse with the late endosomes that undergo transient acidification, leading to changes in bacterial gene expression and intracellular survival associated with the late endosomal markers Rab7, RILP, and Lamp1 during 4 h post-infection (pi) [55,59,60]. The presence or absence of these lysosomal markers in the BCVs maturation can be used to evaluate the effect of Brucella virulence factors on its intracellular survival. Brucella expresses various effector molecules to modulate its intracellular trafficking. For example, Brucella can secrete a lysozyme-like protein SagA identified as a muramidase. This enzyme participates in the early stage of Brucella intracellular trafficking by avoiding BCVs and lysosome fusion. In particular, it regulates the recruitment of Lamp1 in the BCVs maturation by an unknown mechanism [61]. In addition, Brucella membrane fusogenic protein interacts with phospholipid vesicles to favor membrane fusion and is involved in the recruitment of lysosomal markers, Lamp1 [62]. However, the specific mechanism in this event is still unclear [63]. A study by Naroeni et al. [64] revealed that live B. suis prevented the fusion of B. suis-containing phagosomes with the lysosome compartment while this association was clearly observed using killed B. suis.
The interaction between BCVs and endoplasmic reticulum (ER) from 4 to 8 h pi results in the formation of replicative BCVs (rBCVs) with the presence of several ER markers such as calnexin, calreticulin, and Sec61β [60]. It is noteworthy that interaction between lipid rafts and O-chain of smooth Brucella LPS plays a vital role in the mediation of the bacterial entry that leads to the development of specialized membrane-bound compartments known as replicative phagosomes or replicative vacuoles [65]. The continuity of rBCV membranes with the ER was observed unmistakably by applying the 3D-CLEM approach. This study proposed a model where rBCVs are integrated into the ER meshwork [66]. On the other hand, CβG is well-known as a virulence factor of Brucella sp. and can modulate lipid raft organization. CβG probably acts on lipid raft to favor the recruitment of lipid raft marker flotillin-1 from 0.5 to 8 h pi. In addition, CβG regulates the lysosome fusion and enables BCVs to associate with ER [67,68].

Host Bactericidal Effect against Brucella
Following bacterial engulfment by host cells, most ingested Brucella that are localized in the endosomal BCVs (eBCVs) are killed within the phagolysosome compartment in the last stage of phagosome maturation during 4 to 8 h pi. Using a fluorescent probe specifically for proteolytic activity, during the first 6 h pi, the BCVs clearly showed fluorescent intensity emission, and this remarkable fluorescence was not detected at 12 h pi [69]. The phagolysosome compartment displays many effectors that limit Brucella growth and survival such as reactive oxygen species (ROS), nitric oxide (NO), and antimicrobial peptides [70,71]. Nitric oxide and lysozyme activities are employed to monitor RB51 vaccine efficiency in cows [72]. This is because NO can react with structural elements, metabolic enzymes, nucleic acid, and it inhibits bacterial secretion system expression. A study by Hu et al. [73] indicated that thioredoxin-interacting protein (TXNIP), a multifunctional protein in metabolic diseases, is involved in the production of NO and ROS, leading to the control of bacyccterial intracellular survival. These authors proposed a novel host pathway, iNOS/NO-TXNIP-iNOS/NO, related to Brucella T4SS expression. Aside from NO, although ROS causes DNA, protein, and lipid damage, Brucella possesses xthA-1 gene that encodes a protein that contributes to oxidative stress resistance [74]. On the other hand, rough Brucella induces monocytes to secrete higher CXC, CC chemokines (GRO-α, IL-8, MCP-1α, MCP-1β, MCP-1, and RANTES), pro-inflammatory (IL-6, TNF-α), and anti-inflammatory (IL-10) cytokines than the smooth strain [75]. The lacking of the O-chain of LPS interferes with the fusion between the BCVs and lysosomes [76]. Therefore, the decrease in bacterial viability is more dramatic with strains that are lacking in the O-chain of LPS.

Brucella Intracellular Survival
In the early stage of phagocytosis, a small proportion of BCVs allows Brucella to start proliferating. Some evidence for this event include the inhibition of Rab7 recruitment, which is essential for phagolysosome fusion, and the acidification of this compartment, which triggers the expression of T4SS. Besides, Brucella chromosomal replication was observed at 8 h pi within Lamp1-positive BCVs [77]. Up-to-date studies revealed that effectors of T4SS have crucial roles in inhibiting host immune responses and promoting intracellular trafficking and growth during Brucella infection. In particular, RicA can interact with Rab2 GTPase, which is critical for Brucella intracellular trafficking [60]. VceC and VecA are related to host autophagy and apoptosis. BtpA and BtpB are required to modulate host immunity and energy metabolism. In addition, other T4SS effectors also play different roles in the host-Brucella interaction that have been extensively studied, clarifying their specific functions [78][79][80]. A study by Martinez-Nunez et al. [81] demonstrated that BvrS/R transcriptionally regulates T4SS VirB. A bvrS and bvrR mutant resulted in low levels of the VirB1, VirB5, VirB8, and VirB9. In addition, this study clarified that the regulator protein BvrR binds directly to the VirB promoter. The BvrR/BvrS system is one of the virulence factors that interacts with other diverse virulence factors simultaneously to ensure the intracellular growth fate of Brucella.
Brucella can start multiplying in macrophages at 12 to 24 h pi [69]. At this time, two events are in progress. First, the pathogen readily reaches its safe haven to replicate. Lastly, the pathogen has set up its specific virulence genes. The conversion from eBCVs to rBCVs is a hallmark of Brucella intracellular growth. In addition to the beneficial functions of T4SS in promoting Brucella intracellular replication, a study by Casanova et al. [55] discovered two components of the trimeric vacuolar protein sorting complex, VPS35 and VPS26A, which are related to the evasion of the lysosomal degradative pathway. In this infection stage, Brucella is in a multi-membranous compartment, supplemented with several autophagosomal bodies and ER markers. The association of Brucella with the host ER establishes the replicative niche. By reaching this organelle, Brucella can take advantage of the biosynthetic enzymes, connect to the local nutrient supply, and avoid phagolysosome fusion [76,77,82]. On the other hand, Kohler et al. [76] proposed the term "brucellosome" for the replicative niche of Brucella, which is poor in nutrients, low in oxygen tension, and with a neutral pH microenvironment. However, Brucella still furtively exploits these harsh environments as stimuli to induce virulence genes involved in their intracellular trafficking and growth [83]. Once adapted to the intramacrophage environment, Brucella extends its intracellular persistence indefinitely, contributing to systemic proliferation and dissemination to other targeted cells or tissues. Furthermore, at this late stage of infection, the switch from rBCVs to the autophagic BCVs (aBCVs) mediates bacterial release from infected cells. Autophagy-initiation proteins ULK1, Beclin 1, ATG14L and PI3-kinase activity is required for the aBCVs formation. This event is exploited by Brucella for cell-to-cell spreading [84][85][86]. However, once again, Brucella continues displaying an effector of the T4SS system, BspL, to delay the formation of aBCVs that benefit the optimal intracellular replication before disseminating to other cells. At the same time, BspJ, a nucleomodulin, directly or indirectly regulates host cell apoptosis to complete its intracellular cycle [87,88]. As mentioned, the microenvironment inside these vacuoles has limited nutrients. Interestingly, Brucella behaves as a stealthy pathogen. It has adapted to these harsh intracellular environments in the host cells by setting up various virulence factors such as the heat shock proteins DnaK and ClpB. These virulent factors play an important role in Brucella resistance against some stresses but do not favor intracellular growth [89].

Conclusions
Brucella sp. is a facultative intracellular pathogen. It can infect professional and nonprofessional phagocytes. To establish its intracellular niche, it has to invade host cells, initiated by bacterial adherence to the host cell surface. The detection of Brucella by various host cell receptors is the second phase of bacterial phagocytosis (Figure 1). Once Brucella is detected by either opsonic or non-opsonic receptors, various signaling pathways are activated to engulf and uptake it into the host cells. In the typical way of phagocytosis, BCVs enter the endosomal pathway from the fusion steps with early endosomes, late endosomes, and lysosomes resulting in bacterial elimination. Nevertheless, as a stealthy pathogen, Brucella facilitates its intracellular replication by exploiting host cell resources. More than that, Brucella has various effectors that interfere with antibacterial mechanisms to ensure its intracellular survival (Figure 2). Furthermore, all Brucella functional proteins related to the interaction between Brucella and host cells is summarized in Table 1. important role in Brucella resistance against some stresses but do not favor intracellular growth [89].

Conclusions
Brucella sp. is a facultative intracellular pathogen. It can infect professional and nonprofessional phagocytes. To establish its intracellular niche, it has to invade host cells, initiated by bacterial adherence to the host cell surface. The detection of Brucella by various host cell receptors is the second phase of bacterial phagocytosis (Figure 1). Once Brucella is detected by either opsonic or non-opsonic receptors, various signaling pathways are activated to engulf and uptake it into the host cells. In the typical way of phagocytosis, BCVs enter the endosomal pathway from the fusion steps with early endosomes, late endosomes, and lysosomes resulting in bacterial elimination. Nevertheless, as a stealthy pathogen, Brucella facilitates its intracellular replication by exploiting host cell resources. More than that, Brucella has various effectors that interfere with antibacterial mechanisms to ensure its intracellular survival (Figure 2). Furthermore, all Brucella functional proteins related to the interaction between Brucella and host cells is summarized in Table 1.

Brucella Proteins
Functions Cyclic β-1,2-glucan Important for circumventing host cell defenses, and modulate lipid raft organization VirB T4SS Mediating intracellular survival and circumventing host immune responses SP29, SP41 Sialic acid-binding proteins, in bacterial adherence BigA, BigB Proteins containing the immunoglobulin-like domain, in bacterial adherence BmaA, BmaB, BmaC The monomeric autotransporters, in bacterial adherence Brucellosis is a worldwide disease causing severe economic and public health problems. Therefore, understanding the characteristics of the causative agent, its interaction with host cells and specific virulence factors is vital in eradicating brucellosis. Several recent studies focus on understanding the interaction between Brucella and its host, and the mechanism of how it works, but still the available data are limited and the complete mechanisms involved are unclear. In fact, in comparison to other bacteria such as Mycobacterium species, Salmonella species, and Helicobacterium species, more research has been recently invested in Brucella species. However, there are still many gaps in understanding Brucella phagocytosis, and many questions remain unanswered particularly regarding the interaction between molecules related to F-actin polymerization including cofilin, profilin, ARP2/3, and WASP that renders importance in the context of Brucella infection. In addition, searching for more virulence factors implicated during Brucella phagocytosis also deserves more attention. Therefore, more extensive efforts are necessary to study this stealthy pathogen deeper in order to discover alternative therapies or even more effective vaccines to eliminate brucellosis completely. Brucellosis is a worldwide disease causing severe economic and public health problems. Therefore, understanding the characteristics of the causative agent, its interaction Figure 2. Intracellular trafficking cycle of Brucella in host cells. After recognizing Brucella by different receptors, host cells activate the signaling pathway resulting in F-actin polymerization at the binding site with bacteria. Then the BCVs undergo fusion with early, late endosome, and finally lysosome. At the last phase of BCVs' maturation, Brucella is killed by nitric oxide, ROS, and digestive enzymes that localize in the phagolysosome compartment. However, Brucella has the ability to avoid the phagolysosome fusion to reach its intracellular niche at ER. Brucella SagA protein is secreted to interfere with the interaction between BCVs and lysosomes. Once localizing in ER, Brucella can proliferate and mature into aBCVs. This aBCVs formation depends on distinct autophagy proteins including ULK1 and Beclin 1, which contribute to bacterial egress and the formation of infection foci resulting in dissemination to other cells and tissues. Besides, Brucella activates a T4SS-BspL effector to delay the formation of aBCVs, and ultimately reach maximum proliferation.

Conflicts of Interest:
The authors declare no conflict of interest.