Introduction
The gram-negative bacterium Yersinia pseudotuberculosis (Yptb) is the causative agent of a severe infection known as pseudotuberculosis.1,2 The gastrointestinal form of this infection is most common in cold temperate countries of Europe and North America.3 Serious outbreaks of a variant of the disease, referred to as Far East scarlet-like fever (FESLF), have been reported from Russia and Japan, where the clinical symptoms of the Yptb infection are more diverse and severe than those observed in Europe.3–5 In addition to damaging the gastrointestinal tract, the bacterium disseminates into internal organs rich in lymphoid elements and macrophages, such as the lungs, mesenteric lymph nodes, spleen, and liver.6,7
The bacterium has been reported to contribute to the progression of liver abscesses and sepsis in humans, especially in cases where liver disease or other immunosuppressive conditions are present, including during liver transplantation.7,8 In cases of FESLF, liver damage is detected in 50% of infected individuals.3 Liver disorders such as hemosiderosis, acute hepatitis, and chronic hepatitis have been found in infected humans as well as in many infected mammals (including primates, artiodactyls, lions, cows, and cats) and birds.7,9–11 Once in the organs, Yersinia can persist there for a long time, causing multiple waves of the disease or chronic infection.12 Particularly, FESLF is characterized as a disease that often progresses to a persistent form.13
Bacteria translocate from the intestine into the liver and other lymphoid element-rich organs via the bloodstream, after adhering to mucosal cells and subsequently penetrating adjacent Peyer’s patches.14 A subset of intestinal epithelial cells residing within the epithelium that covers gut-associated lymphoid tissue follicles (M cells) specializes in transferring bacteria from the gastrointestinal tract.15 Bacteria usually enter the liver through the portal circulation or biliary tract.16
To colonize the organs, Yptb overcome the immune response of the host organism. For this purpose, the bacterium expresses a number of cytotoxic components of the Type III secretion system (T3SS, TTSS), which suppress host phagocyte function, as has long been believed.17,18Yptb, like Yersinia pestis (the causative agent of plague) and Yersinia enterocolitica (another pathogenic species), possesses a 70-kb plasmid, pYV (known as pCD1 in Y. pestis) (42–48 MDa).5,19 This plasmid is the most common among those found in Yersinia species and ensures the pathogenicity of the strain that carries it. It encodes a unique TTSS system found only in pathogenic Yersinia, containing injectisomes that export pathogenic effector proteins with cytotoxic properties (known as Yersinia outer proteins, Yops) into the host organism.20
In addition to the Yops encoded by pYV, a number of toxins are encoded by chromosomal genes, such as Yptb-derived mitogen-superantigen (YPM), cytotoxic necrotizing factor (CNFY), and thermostable lethal toxin (TsTYp).4,20 Recent research indicates that the activity of the Yptb anti-phagocytic system depends on the expression of both plasmid and chromosomal genes.2
The geographic variability in the severity of clinical forms of the disease caused by European and Far Eastern Yptb strains is also reflected in regional differences in the virulence factors synthesized by the bacteria.4 In particular, YPM and the recently discovered Toll/interleukin-1 receptor (TIR)-domain-containing virulent protein (TcpYI) have been found almost exclusively in Far Eastern strains.3,21 Furthermore, the extreme polymorphism of the clinical manifestations of FESLF is associated with the presence of several types of toxins.22 Russian researchers have also identified that the causative agent of FESLF has a different sequence type (2ST) and a different allele of the yadA gene compared to European strains, leading to the hypothesis that Far Eastern strains form a clonal line of Yptb.4
Due to the severity of Yptb-induced liver lesions, research attention has focused on the role of Yptb protein toxins in bacterial dissemination and colonization in the liver over the last two decades. Understanding the mechanisms of action of these toxins and their contribution to fine-tuning the host immune response will aid in developing novel approaches to treating Yptb infection and related liver damage. This review aimed to summarize current data on the mechanisms and regulation of bacterial liver colonization by Yptb plasmid- and chromosome-encoded toxins. It specifically highlights the role of chromosome-encoded toxins produced by Far Eastern Yptb strains in exacerbating the infection. There is particular focus on how these toxins suppress inflammation, including through macrophage polarization via the M2 way, promoting bacterial dissemination to the liver.
Plasmid-encoded effector proteins/cytotoxins
At least six Yersinia effector proteins, including YopM, YopH, YpkA/YopO (in Y. enterocolitica), YopJ/YopP (in Y. enterocolitica), YopE, and YopT, have cytotoxic properties. With the help of some other Yops, they are translocated into host cells, primarily phagocytes.20,23 They are all important for tissue colonization, but not all have been found to be involved in liver colonization by Yersinia. During Yptb infection, Yops target professional phagocytes, predominantly neutrophils in lymphoid organs, along with dendritic cells and macrophages that are selectively targeted in different tissues.23
In addition to effector proteins and proteins that promote their translocation into the cell, the 70-kb plasmid carries genes encoding only one T3SS regulating protein. LcrF, the Yersinia T3SS master regulator, controls T3SS expression during transitions into host cells. In particular, transcriptional activation or direct binding of LcrF to the promoter regions of yopE and yopH has been reported.24 It has also been shown that antagonists of LcrF are required to control gene transcription. In turn, the transcription of lcrF is regulated by temperature through temperature-induced changes in chromosomal DNA topology.25 When Yersinia enter cells of a warm-blooded host, the temperature shift affects the structure of RNA. This sensitivity of RNA structure to changes in temperature is used by the bacterium as an RNA thermometer (RNAT) that controls translation of LcrF to activate yop gene transcription.26 LcrF is also post-transcriptionally regulated by the RNA-binding protein, carbon storage regulator system A. To increase yop gene expression, Yptb can also increase the number of pYV copies after contacting a host cell. Additionally, Yops expression in response to adverse external factors is regulated by the Rcs phosphorelay system, mediated by LcrF.25
YopM
YopM (42–54 kDa) is mostly composed of leucine rich repeats.27 YopM does not exhibit enzymatic activity but has multidirectional effects on immune cells. It promotes the colonization of the liver and other organs by Yptb, while its action is aimed at reducing inflammation and stimulating the production of the anti-inflammatory cytokine interleukin (IL)-10.28,29 Recently, it has been found that YopM stimulates IL-10 gene expression by promoting nuclear translocation of signal transducer and activator of transcription 3 (STAT3).30 Release of IL-10 upon cell exposure to YopM is important for bacterial virulence and colonization of the liver, as evidenced by experiments on the YopM mutant strain and IL-10−/− animals.29
There are differences in the effect of YopM on bacterial colonization of the liver and spleen.31 Experiments with the YopM mutant strain 32777 showed that colonization of the liver, to a greater extent than that of the spleen, depended on the presence of YopM.29 Differences were also observed in the number of dendritic cells affected by YopM: in the spleen of mice infected with Y. pestis strain containing YopM, there was a decrease in only one subgroup of dendritic cells (CD8+), associated with the production of pro-inflammatory cytokines, compared to those infected by the YopM mutant; in the liver, all three major subgroups of dendritic cells decreased in numbers. However, this difference was significant only at the late stage of infection, thereby raising the question of its role in the differential effects of YopM.31 Moreover, Ye et al.31 did not find any differences in the numbers of polymorphonuclear leukocytes (PMNs) in the livers and spleens of mice infected with the wild strain of Y. pestis or the YopM mutant, indicating that YopM does not affect the recruitment of PMNs to these organs. However, experiments on the mutant ΔyopM strain of Y. pestis, in which PMNs were suppressed, showed that PMNs were selectively involved in controlling the growth of the mutant bacteria in the liver, but not in the spleen.32 These findings indicate that YopM targets PMNs in the liver to disrupt their function. Meanwhile, YopM has been shown to promote the apoptotic death of PMNs as well as inflammatory Kupffer cells (hepatic macrophages) and/or monocytes in the liver. It has also been shown that YopM-associated cell death is characteristic only of the liver, whereas cell death in the spleen is rather independent of YopM.33 Apoptosis is a programmed cell death characterized by morphological features such as cell shrinkage, plasma membrane blebbing, apoptotic body formation, chromatin condensation, and fragmentation of nuclear DNA, and is considered a non-lytic homeostatic mode of cell death.34 Caspase-3 has been identified as a cofactor or effector in YopM’s proapoptotic action on PMNs.28
After being delivered into a cell, YopM is localized in the nucleus,31 but significant amounts of the protein also reside in the cytoplasm.35 To date, the mechanisms of its action have not been fully elucidated. Thus, YopM is reported to activate the protein kinase C-related kinase 2 (PRK2) and 90-kDa ribosomal S6 kinase (RSK1).33 PRK2 is one of the members of the PRK serine/threonine kinase family that are downstream effectors of the Rho family small G proteins.36 The Rho guanosine triphosphate hydrolases (GTPases) play a role in switching important signaling pathways in the cell, followed by cytoskeleton rearrangement.37 PRKs are regulated by multiple signaling pathways in different cellular locations, and accordingly, their switching effect depends on many factors.38 One possible pathway for PRK2 activation is caspase 3-mediated proteolysis during apoptosis.39 Hentschke et al.40 found that RSK2 and PRK1 (PKN1) can also bind to YopM. However, PRK2 (PKN2) is not directly activated by YopM but indirectly by the overactivated RSK1.41 Hentschke et al.40 also suggested that activation of the extracellular signal-regulated kinase (ERK)-RSK pathway may be a common event during infection. It is worth noting that ERK has been shown to play a role in promoting cell migration and invasion.42 Additionally, YopM increased the expression of IL-10 in Y. enterocolitica-infected human macrophages through interaction with RSK.29 Supporting the hypothesis of Hentschke et al.,40 it has been shown that the interaction of PRK2 with YopM is necessary for Yptb full virulence and may lead to the inactivation of PRK-dependent pyrin (marenostrin; TRIM200) inflammasome, which is responsible for IL-1 release from macrophages in response to bacterial infection.29,43 At this level, YopM counteracts the effects of YopE and YopT, which trigger the assembly of the inflammasome.44
Inhibition of inflammasome formation by YopM is associated with downregulation of caspase-1.45 Typically, caspase-1 activation occurs at the onset of infection and is accompanied by lysosome exocytosis and pyroptosis. The latter is a form of inflammatory cell death with cell membrane rupture and release of pro-inflammatory mediators, primarily IL-1β.46 Thus, YopM may stimulate anti-inflammatory cytokine release while inhibiting pro-inflammatory cytokines, but the molecular basis for target selection has not yet been identified. Overall, the properties of YopM to induce apoptosis and abolish pyroptosis provide the toxin with the ability to kill cells while limiting inflammation.
It should also be noted that in response to various environmental signals, macrophages can transform into pro-inflammatory (M1 macrophages, “classically” activated) or anti-inflammatory (M2 macrophages, “alternatively” activated) phenotypes.47 M1 macrophages play a key role in defense against bacterial infections, while M2 macrophages are involved in tissue repair. Induction of the M1 or M2 ways occurs upon exposure to Th1 or Th2 cytokines, respectively, and certain other stimuli, including IL-10.48 In addition to the known mechanisms of cellular immunity suppression, bacteria have a relatively recently discovered and still poorly studied property of dysregulating the balance between host’s M1 and M2 macrophages. In response to various groups of bacterial pathogens, a nonspecific stress reaction of macrophages occurs, accompanied by their functional and phenotypical transformation into the M1 type, providing protection against acute infection.49 However, as shown for a number of bacteria, they cause a shift in macrophage polarization towards the M2 type, leading to decreased bactericidal activity of macrophages and contributing to chronic infection.49,50Yptb is also capable of causing the transformation of macrophages towards the M2 type. This process, according to Bi et al.,1 likely involves certain bacterial factors and plays an important role in the immunosuppressive strategy of Yptb. Due to its ability to stimulate IL-10 production,28,29 YopM is also assumed to cause a shift in macrophage polarization towards the M2 type. This assumption is supported by the anti-inflammatory effect of YopM mentioned above.29,43
YopJ
YopJ (31–32 kDa) possesses properties of a cysteine protease and deubiquitinase but is primarily known as an acetyltransferase.51,52 The protein regulates the dissemination of Yersinia from mucosal tissues but not its replication in organs.53
YopJ induces apoptotic cell death in macrophages,20 but not in neutrophils.33 YopJ-induced apoptosis of macrophages in the mucosa reduces the immune response, thereby facilitating bacterial dissemination.53,54 On the other hand, YopJ activates caspase-8, an upstream activator of caspase-3, and induces apoptosis of monocytes in lymphoid nodes,55 which, according to Bliska et al.,56 promotes the elimination of replicative cell niches, thereby limiting the rate of Yptb dissemination. To support this suggestion, Palace et al.57 have demonstrated that YopJ suppresses immune cell recruitment in mouse liver in vivo.
YopJ acetylates serine and threonine residues on mitogen-activated protein (MAP) kinase 3, transforming growth factor beta-activated kinase 1 (TAK1), and some MAP2 kinases, resulting in their inactivation. Redundantly, YopJ inhibits the c-Jun N-terminal kinases (JNK), p38 mitogen-activated protein kinase, and nuclear factor kappa B (NF-κB) signaling pathways, thereby inhibiting IL-1β transcription. Moreover, YopJ’s effects on TAK1 and other MAPKs are independent of each other, and p38 is involved only in the MAPK-dependent signaling pathway.58 Thus, YopJ predominantly induces apoptotic signals inhibiting MAPKs and NF-κB signaling.52,55
In a study by Sheppe et al.,52 prostaglandin (PG) E2, known as a lipid immunoregulator, stimulates the polarization of Y. enterocolitica-infected macrophages towards a proinflammatory M1 phenotype. Simultaneously, PGE2 reduces bacterial survival in macrophages. Both pathogenic bacteria, Y. enterocolitica and Yptb, inhibit PGE2 synthesis in macrophages, and YopJ contributes to this effect through MAPK inhibition. YopJ blocks PGE2-induced activation of IL-1β transcription by NF-κB, thereby enhancing bacterial survival in macrophages. It can be assumed that YopJ, by inhibiting PGE2 synthesis, also stimulates the polarization of macrophages towards the M2 phenotype.
To counteract this, and in accordance with the known data that different death scenarios can simultaneously develop in a cell,59 phagocytes induce the assembly of a cytoplasmic death-inducing complex comprising receptor-interacting serine/threonine protein kinase 1 (RIPK1), Fas-associated protein with death domain (FADD), and caspase-8. This assembly is another way to limit bacterial dissemination, as RIPK1/caspase-8 induces proinflammatory cytokine release from uninfected cells.55 Recently, the role of gasdermins, pore-forming proteins, in pyroptotic cell death has been demonstrated. YopJ, mediated by RIPK1 kinase activity, triggers gasdermin (GSDM) E-dependent, but not GSDMD-dependent, neutrophil pyroptosis, indicating an exclusive role of GSDME in neutrophil pyroptosis.60 In macrophages, GSDMD has been shown to be involved in pore formation and pyroptosis.60 Mitochondrial reactive oxygen species (ROS) have been found to augment GSDMD-dependent pore formation.61 Stimulation of ROS production in mitochondria through caspase 8 and RIPK3 has been discovered as an alternative mechanism to NADPH oxidase-dependent ROS generation,58 indicating that ROS are involved in both apoptotic and pyroptotic cell death.
The Toll-like receptor (TLR) 4-dependent activation of caspase-1 by YopP can also occur in macrophages. This activation requires specific infection conditions, such as a high concentration of bacteria and prolonged contact with cells, ensuring that YopP enters macrophages at high concentrations.62 However, there are allelic variations in genes encoding YopJ and YopP proteins. Data on the activating effect of YopJ on pyroptosis were also obtained from a model of Yptb-infected mouse myeloid cells, indicating that TLR4-dependent activation of macrophages shifts cell death from apoptosis to pyroptosis.60 Thus, YopJ can be involved in regulating both apoptotic and pyroptotic cell death. Nevertheless, it remains unclear how signaling pathways switch under exposure to the effector, which determines the fate of the cell.
YpkA
YpkA (81.7 kDa) is a serine/threonine protein kinase homologous to that of eukaryotes.51 YpkA plays a significant role in Yptb virulence, as shown in experiments with mutant strains expressing catalytically inactive YpkA.63 YpkA also promotes the spread of Yptb to the brain and spleen,64 but there is no available data regarding its spread to the liver.
YpkA directly inhibits Rho GTPases, including RhoA, Rac1, and Rac2.57 To control actin remodeling necessary for macrophage phagocytosis, the protein contains a guanine nucleotide dissociation inhibitor-like domain, an N-terminal membrane localization domain, and a serine/threonine kinase domain.65
Additionally, YopO (homologous to YpkA) phosphorylates the actin-binding protein gelsolin, which is implicated in various cellular activities such as cell motility, signaling, apoptosis, and phagocytosis. Thus, YopO may attenuate host defense through its effects on gelsolin.66
YpkA phosphorylates the Gq protein alpha subunit (Gαq), a member of the family of heterotrimeric G protein alpha subunits. This leads to inhibition of the Gαq signaling cascade, which regulates a number of cell activities, including cell growth and proliferation.67 This results in inactivation of the Rho GTPases and subsequent cytoskeleton disruption,68 but there is still no full understanding of the signaling pathways that contribute to YpkA toxicity.
YpkA/YopO also induces phosphorylation of vasodilator-stimulated phosphoprotein (VASP), a regulator of actin dynamics,69 and targets other actin-regulating proteins such as the actin filament elongator Ena/VASP-like protein and the nucleation-promoting factors: Wiskott-Aldrich syndrome protein and Wiskott-Aldrich syndrome protein-interacting protein.67 Phosphorylation of these proteins presumably leads to the inhibition of actin rearrangement, as demonstrated for VASP, and further suppression of phagocytosis.70 The importance of these multiple effectors is not clear, but a role in differential tissue tropism or effector kinetics has been suggested.57
Another way to suppress host immunity, as shown for YpkA in murine macrophages, is by inducing caspase-3-mediated apoptosis,67 which may be independent of its effects on actin depolymerization.43 Additionally, co-infection with YopT and YpkA inhibited YopT-enhanced neutrophil degranulation, indicating that YpkA inhibits neutrophil degranulation, which is part of host defense.71
YopH
The protein toxin YopH (51 kDa) belongs to the class of protein tyrosine phosphatases.51 Experiments on wild-type and YopH-mutant strains have shown that YopH is crucial for the colonization of the liver and other lymphoid tissues, as well as for the virulence of pathogenic Yersinia species.71,72 YopH targets neutrophils, and one of the presumed mechanisms is as follows: the toxin causes dephosphorylation of the adapter protein Src kinase-associated phosphoprotein-2 (SKAP2).72 SKAP2 is involved in integrin-stimulated cytoskeletal changes in macrophages and neutrophils.73 In addition, SKAP2 is required for ROS production. Furthermore, YopH inhibits both ROS production in neutrophils and phagocytosis. To inhibit ROS generation, YopH blocks both SKAP2-dependent (involving ERK1/2 phosphorylation) and independent (Fc gamma receptor-stimulated phosphorylation of proximal signaling proteins such as spleen tyrosine kinase, SH2 domain-containing leukocyte protein of 76 kDa, phospholipase Cγ 2, and PML-RARA-regulated adaptor molecule-1) mechanisms.72
YopH was also shown to inhibit neutrophil degranulation independently of SKAP2.72 Another function of YopH is to prevent the uptake of Yptb by M cells and thereby reduce the dissemination of bacteria within lymph. Thus, YopH promotes bacterial colonization while reducing dissemination.15
YopE
The toxic properties of YopE (23 kDa) are associated with the modification of Rho GTPases.44,51,74 According to recent data, YopE is a GTPase-activating protein that facilitates the hydrolysis of guanosine triphosphate (GTP) into guanosine diphosphate, thereby inactivating the Rho GTPases,51 which leads to cytoskeleton disruption.44
An increase in the level of YopE is followed by restriction of bacterial dissemination to the liver.75 However, YopE partly degrades within infected phagocytes.51,76 Increased intracellular levels of YopE were shown in models of mutant strains of Y. enterocolitica serogroup O8, which avoided the degradation of YopE via the ubiquitin-proteasome pathway. The accumulation of degradation-resistant YopE was accompanied by a pronounced cytotoxic effect on infected cells and a decrease in the delivery of YopP/YopJ and YopH into cultured cells infected by Yersinia. A pathway influencing the translocation of other Yops by YopE has been suggested.75 While activated Rho GTPases stabilize pores formed by T3SS in the host cell membrane via actin cytoskeleton manipulation, YopE-induced inhibition of Rho GTPases limits pore formation and further restricts the translocation of other Yops. Thus, variations in the level of YopE and its influence on Yops translocation may depend on the activity of the proteasome, which also optimizes the infection process.75,77 This finding indicates that increased cytotoxicity limits bacterial dissemination. However, there is evidence that the direct anti-phagocytic effect of YopE released by Yptb is predominant compared to its regulatory role in Yop translocation.75
YopE was shown to inhibit caspase-1 and the release of IL-1β in macrophages.78 This effect was observed with the inhibition of Rac2, but not RhoA, and was associated with the inhibition of ROS.79 However, RhoA was the preferred target for YopE, the inhibition of which, in contrast, resulted in pyrin dephosphorylation and inflammasome activation.44 As YopM usually inactivates pyrin, YopE initiates the assembly of inflammasomes only in the absence of YopM.80 Output of pyrin dephosphorylation depends on the features of effectors, such as Rho GTPase specificity.15 However, the mechanisms targeting specific Rho GTPases remain poorly understood. Hence, despite the fact that inhibition of RhoA led to activation of the inflammasome,44 this did not restrict the survival of Yersinia in macrophages.74 Therefore, the ability of YopE to kill phagocytes through the inflammasome pathway remains unclear.
YopE, like other Yops, is a multifunctional protein. The toxin can be recognized by phagocytes, with YopE being an antigen for T and B cell responses. Activation of phagocytes may play a role in the progression of chronic lymph adenopathy.79
Similar to YopH, the toxin prevents the uptake of Yptb by M cells. This fact may explain the more efficient penetration of Yptb into M cells during early infection stages when YopE concentration is still low.15
YopT
The cytotoxin YopT (35.5 kDa), like YopE and YpkA, is a Rho-modifying toxin.44,51,81 However, as a protease, it inactivates Rho GTPases, such as RhoA, Rac1, RhoG, and CDC42 in vitro, by cleaving the C-terminus, leading to their release from the membrane.44,81 Meanwhile, in vivo, YopT induces the shift of RhoA, but not Rac1 or CDC42, from the cell membrane to the cytosol, indicating that RhoA is the primary target for the bacterially translocated YopT.81 However, YopT has been shown to selectively target different Rho GTPases depending on the cell type.67 YopT is highly cytotoxic, and it is the only Yersinia effector protein that can destroy actin filaments in HeLa cells.82
Similar to YopE, in relation to inhibiting RhoA GTPase, YopT also induces pyrin dephosphorylation in the absence of YopM, leading to the activation of inflammasome assembly, further activation of caspase-1, and eventually cell death by pyroptosis. As is known, pyroptosis limits bacterial growth in cells and their spread to deep tissues.44 However, there is no experimental confirmation of the effect of YopT on bacterial dissemination.
YopT acts slower on pyrin dephosphorylation compared to YopE. For this reason, YopE is considered to play a dominant role in triggering this process.44 However, YopT is able to decrease the translocation rate of YopE possibly by competing for the same Rho GTPases.71 The exact mechanisms of this action remain poorly studied.
Chromosome-encoded toxins
Some chromosome-encoded genes in Yptb contribute to the coordinated activation of the expression of the 70-Kb plasmid (pCD1). For example, the ctgA (chromosomal T3SS secretion gene) mutant strain released fewer Yops than the parental strain.83 Moreover, recently, some gene products (YlrA, YlrB, and YlrC) representing chromosome-encoded T3SS effector proteins of Y.pestis have been discovered for the first time. These proteins promote the optimal survival of Y. pestis in the presence of macrophages.84 However, the putative role of these proteins in the virulence of Yersinia has only begun to be studied.
The chromosome-encoded toxins of Yersinia, whose action does not depend on the presence of the 70-Kb plasmid, have been previously described.3,4,20 Over the past 20 years, evidence has emerged that toxins such as YPM, CNFY, TsTYp, and TcpYI may also contribute to the modulation of immune response and bacterial spread.
CNFY
CNFY catalyzes the deamidation of glutamine in molecules of small Rho GTPases, blocking their ability to hydrolyze GTP. In contrast to YopE, CNFY constitutively activates small Rho GTPases.85,86 The molecular weight of CNF family proteins is usually estimated at approximately 114 kDa, but this applies to cytotoxic necrotizing factor 1 (CNF1) of Escherichia coli,87 which shares only about 65% identity with CNFY.86
The CNFY toxin is an important virulence determinant expressed predominantly during the early stage of infection.87 Temperature control is crucial for its activation upon entry into warm-blooded hosts; increased protein levels are recorded only after a shift to 37°C. However, the secondary structure of the protein in Yptb YPIII is impaired at temperatures up to 42°C, suggesting that cnfY RNAT confers temperature-dependent translational control to its downstream gene. The RNAT-controlled synthesis of CNFY is crucial for Yptb virulence and its spread to the liver, but the cnfY gene is not controlled by LcrF.88
RhoA GTPase plays a crucial role in CNF-promoted bacterial spread, as it is involved in cell membrane destabilization.89 Specifically, CNFY activates Rho GTPases, enhancing the translocation of certain Yops.85,86,90
CNF-induced activation of Rho GTPases is associated with its ability to induce the formation of giant multinucleated cells. Studies using CNF1 have shown that toxin-induced stimulation of Rho protein activity leads to cellular DNA replication without concomitant cell division, resulting in multinucleation.4 CNFY also induces dermonecrotic reactions in vivo.85 Studies with CNF1 indicate that RhoA inactivation and Rac1 presence are critical for CNFs’ ability to prevent cell apoptosis by stimulating the phosphoinositide 3-kinase (PI3K)/protein kinase B (Akt)/NF-κB signaling pathway, while CNF-mediated RhoA activation opposes pyrin inflammasome activation.90
Additionally, CNFY-induced activation of Rho GTPases leads to STAT3 activation in human embryonic kidney cells, suggesting this mechanism operates in various cell types.90 STAT3 activation may lead to increased IL-10 release,30 but its role in cell death during CNFY exposure remains unclear.
In the absence of CNFY, transcripts of enzymes involved in lymphatic tissue damage repair and regeneration (e.g., Adamts4, Adamtsl2, Lox, Wisp2/CCN5, Pappa2, and Retnlg) are found in lower numbers compared to wild-type-infected tissue,91 suggesting CNFY’s association with inflammatory response and tissue injury. The pro-inflammatory actions of CNFs are evidenced by the activation of cytokines IL-6 and IL-8, as well as mediators like monocyte chemotactic protein 1 and macrophage inflammatory protein-3 in CNF1-intoxicated endothelial cells.92
Heine et al.91 demonstrated that CNFY played a role in the virulence of Yptb, promoting dissemination in lymphatic tissues and organs, including the liver, of infected mice, and inducing inflammatory reactions that lead to extensive tissue destruction. These authors suggested that CNFY accelerates bacterial dissemination through Rho GTPase-activated phagocyte motility and facilitates bacterial penetration into damaged tissues. However, this requires further study.
Meanwhile, CNFY-induced caspase-3/7-mediated apoptosis was observed in the human prostate cancer cell line LNCaP,93 indicating that the cellular response to CNFY exposure depends on the cell type.
CNFY has been intensively studied worldwide since isolation from YPIII strain in 2002.85 It’s important to note that all the described experiments were conducted with CNFY isolated from YPIII bacteria.94 The full-length cnfY gene encoding CNFY has only been identified in Yptb strains YPIII and IP2666 of serogroup III. Many other Yptb and all Y. pestis strains contain various deletions within cnfY.89 The presence of cnfY increases the virulence of Yptb, presumably enhancing the translocation of certain Yops. However, considering that Yops such as YopH, Yop M, and YpkA reduce inflammatory responses,91 while CNFs predominantly stimulate inflammation,92 these proteins are unlikely to be associated with increased virulence of wild-type bacteria. It has been suggested that the increased virulence induced by CNFY in Yptb is rather a consequence of decreased translocation of YopJ, the most potent inhibitor of the innate immune response,53,54 as mRNA levels of YopJ were found to increase in ΔcnfY mutants.91 Thus, the specific roles of certain Yops in CNFY effects remain to be elucidated.
Earlier, in 1996, a heat-labile protein toxin (HLT, 200 kDa) was isolated and purified in Russia from the YPIII strain 2517 obtained from France.22,94 It is noteworthy that the first information about the Yptb heat-labile toxin (at that time the bacterium was considered to belong to the genus Pasteurella) appeared in 1969.17 According to Somova et al.,95 HLT causes a local dermonecrotic reaction and death in laboratory animals when administered parenterally.
Severe dystrophic and necrotic changes were found in organs, mainly in the liver and kidneys, already in the initial stages (3–6 h) after the administration of HLT. At 48 h post-administration, necrobiotic and necrotic changes in hepatocytes with a perifocal inflammatory reaction were observed in the liver. Subsequently, mass spectrometric analysis and a comparative study of the virulent and toxic properties (dermonecrotic reaction, heat lability, and similar immune response) of CNFY and HLT on macrophages identified HLT as CNFY.94 However, these findings need further confirmation. In particular, the molecular targets of HLT remain unexplored.
It has also been reported that in the Far East of Russia, there are strains having the cnfY gene, which differs from that of the YPIII strain by the presence of three nonsynonymous mutations, resulting in amino acid substitutions Asn192Ser, Glu304Gly, and Ala394Thr (registered as “allele 2 of the cnfY gene (KR028010)”). Nevertheless, a comparison of the activity of the Russian and American variants of the toxin has revealed their complete similarity.4 However, most isolates in Russia and all isolates obtained from patients with FESLF (0:1b serotype) contain an inactivated variant of CNFY. This gene (registered in the GenBank database as “Yptb cnf gene allele 1 (KR028011)”) contains two deletions and a number of nucleotide substitutions, one of which, T355G, causes the formation of a stop codon.96
Previously, a study of HLT activity showed that the toxin stimulated lactate and succinate dehydrogenase activities in resident macrophages after the first 2–6 h of incubation, while simultaneously decreasing the activity of the inducible nitric oxide synthase system.97 Nitric oxide (NO) is a marker of M1 macrophages, whereas M2 macrophages demonstrate an oxidative metabolic profile.48 These findings are in favor of the hypothesis that this toxin can cause macrophage polarization through the M2 pathway in the initial period of incubation. Apparently, the reparative changes in the transcriptome in wild-type-infected tissues compared to those of ΔcnfY mutant infected cells may also be associated with the shift from M1 to M2 macrophage phenotypes in response to CNFY action,91 which can reduce inflammation induced by Yops translocation.
A significant accumulation of NO by macrophages in response to HLT was recorded at 24 h.97 Thus, the toxin exhibits biphasic action, inducing a presumed pro-inflammatory effect at a later stage of infection, likely related to polarization towards the M1 macrophage type.
However, there is a question regarding the role of CNFY inactivation in worsening infection during FESLF. It is likely that the characteristics of the bacterial strain serotype 0:1b, prevalent in patients with FESLF, including the presence of an additional plasmid pYV82 (82 MDa),94 may influence the ratio and activity of effector proteins translocated under CNFY exposure. The presence of this plasmid, which has only been recorded in Russian strains in addition to pYV, leads to a more severe clinical course of the disease compared to strains that have only pYV.4
Obviously, a balance between the effects of CNFY and other Yops is crucial for the infection outcome. However, the mechanisms regulating this balance in FESLF remain unclear.
YPM
YPM (14 kDa) is encoded by the chromosomal gene ypm.3 This superantigen binds directly to MHC II molecules, rapidly activating both innate immune cells and up to 20% of T cells.2 YPM also activates hepatotoxic CD4(+) T cells.5 In contrast to Gram-positive bacterial superantigens, which increase inflammatory cytokine production by phagocytes, YPM stimulates IL-4 release and strengthens the Th2 immune profile.3,5,98
YPM enhances the toxicity of Yptb in systemic infections but not in gastroenteric infections in mice. Injection of YPM into mice induces massive hepatocyte necrosis. Further, transaminase activity in plasma of YPM-containing Yptb-infected mice increases.5,98 It has been suggested that granzymes (cytotoxic serine proteases) released from YPM-stimulated CD4+ T cells damage the liver via anoikis, a variant of apoptosis.5 Data from a YPM mutant indicates that the superantigen promotes bacterial colonization in the mouse liver.99
YPM is produced by all clinical strains from Far Eastern regions causing FESLF but is absent in almost all European non-FESLF-causing strains. Superantigen-producing strains can be categorized into three clusters containing YPMa, YPMb, or YPMc encoded by the ypmA, ypmB, and ypmC genes.3 Each of these variants activates a specific clone of T-lymphocytes.100 YPMa is detected in 98% of Far Eastern clinical strains, YPMc in five strains of serotype O:3 from Western countries and Japan, while YPMb has not been found in clinical strains at all.101 This suggests that YPMa is implicated in the mechanisms underlying the severe form of the disease.
Furthermore, higher titers of anti-YPM were observed in patients presenting systemic symptoms such as lymphadenopathy, transient renal dysfunction, and arthritis, but not in patients with liver diseases.102 This raises questions about the role of the toxin in Yptb colonization in the liver.
TsTYp
TsTYp (45 kDa) is produced by bacteria of I-VI serovariants identified in various regions of Russia and other countries.22,103,104 It is produced by more than 80% of Yptb strains isolated from patients with FESLF and from the environment. TsTYp causes a general vascular reaction in the liver and other organs that results in dystrophic and necrotic changes. Diffuse and focal hyperplasia of Kupffer cells has been recorded, indicating their involvement in the inflammatory response.22 TsTYp has been used as an antigen in a test system for the diagnosis of pseudotuberculosis.105,106 This toxin causes death in mice when administered parenterally. TsTYp toxicity depends on the temperature during its isolation and is lower at 6–8°C compared to that at 37°C. Such a wide temperature spectrum of TsTYp activity suggests toxin involvement in Yptb pathogenicity in various organisms, including invertebrates.22
In the last two decades, evidence has emerged that TsTYp is able to affect the mechanisms of phagocytic activity in neutrophils and monocytes.22,106In vitro studies have shown that TsTYp has an apoptosis-modulating effect on rat neutrophils, dependent on its concentration and incubation time.107
Apoptosis is closely associated with ROS levels, and antioxidant enzymes limit the damaging effect of ROS on cells.108,109 TsTYp (in the concentration range of 0.2–2.5 µg/mL) had a significant impact on mononuclear leukocytes after 30 min of incubation in vitro, causing suppression of ROS generation at a low but not higher concentration. In neutrophils, the toxin had the opposite effect, corresponding to changes in phagocytic activity.22 The response of mononuclear cells even to a lower concentration of TsTYp indicates that these cells are more sensitive to the action of TsTYp than neutrophils, despite neutrophils being the first line of defense against bacteria.50,110 Such early suppression of the functional activity of mononuclear leukocytes at low concentrations of TsTYp can be considered one of the strategic mechanisms of bacterial suppression of host immunity.
TsTYp also modulated the level of cAMP in phagocytes in a concentration-dependent manner, and the changes were biphasic in the studied concentration range of 1–5 µg/mL. A similar biphasic effect was shown for the adenylate cyclase toxin of Bordetella pertussis when incubated for 2 h with rat fat cells. The cholera toxin stimulated a linear accumulation of GTP, closely associated with that of cAMP, in a concentration range of 1–3 µg/mL, and at 3–5 µg/mL, the toxin began to decrease the GTP level.111 Moreover, the biphasic concentration mechanism of effect (also called hormetic) is basic and common to biological systems, representing stimulatory or overcompensation responses to relatively low damaging doses of agents of various origins.112 A biphasic dose-dependent response to TsTYp has also been demonstrated in mouse leukocytes in vivo. At a dose of 0.01 mg/mouse, TsTYp promoted leukopenia; at a dose of 0.1 mg/mouse, TsTYp stimulated leukocytosis, in addition to the release of pro-inflammatory cytokines and the synthesis of anti-inflammatory cytokine IL-4.18
However, the concentration-dependent impact of TsTYp on cAMP accumulation in human peripheral blood monocytes was opposite to that in neutrophils, which was consistent with the opposite effects of the toxin on ROS levels in these two types of leukocytes.22 These results are consistent with reports indicating an inverse relationship between increased levels of cAMP and stimulation of the immune response.113 The involvement of cAMP in the mechanisms of the apoptosis-modulating effect of TsTYp on rat neutrophils was demonstrated using caffeine, an inhibitor of cAMP phosphodiesterase activity. The combined addition of caffeine and the toxin reduced the level of apoptosis compared to the effect of TsTYp alone, indicating that an increase in toxin-induced apoptosis occurs against the background of a reduced cAMP level.107 This agrees with available data on the cAMP-dependent suppression of apoptosis in neutrophils.114
For Yptb, it has been shown that the inhibition of ROS production in blood phagocytes promotes bacterial colonization in the host’s organs.58,115 However, the effect of TsTYp on bacterial colonization in the liver has not been studied yet.
The toxin does not have ADP-ribosyltransferase activity, which is a feature of some other enterotoxins that enhance the accumulation of cAMP.22 The mechanisms of TsTYp’s effect on cAMP accumulation remain unclear. As known, the signaling cascades following the increase in cAMP levels generally include the activation of protein kinase A (PKA), which catalyzes the phosphorylation of a number of proteins, including the cyclic AMP response element-binding protein transcription factor (CREB), and subsequent expression of target genes.116,117 Activation of the cAMP/PKA signaling pathway may lead to a decrease in ROS generation,118 as observed in mononuclear cells exposed to low concentrations of the toxin.22
Cyclic AMP can also induce activation of a PKA-independent signaling pathway, which includes the binding of cAMP by proteins of the exchange proteins activated by the cAMP family, involved in the regulation of cell migration.116 Apparently, through the modulation of cAMP accumulation, TsTYp can counteract or, vice versa, promote the effects of Yops and CNFY depending on the concentration and stage of infection. Further study on TsTYp-induced cAMP activation downstream effects will help to understand their interaction.
Of particular interest is the role of cAMP in the activation of ERK 1/2 through ROS-dependent activation of Ras GTPases.119 The ERK 1/2 - MAPK signaling pathway, in turn, can contribute to the polarization of macrophages towards the M2 phenotype.120 In addition, Ras stimulates the activity of PI3K.116 The PI3K/Akt signaling pathway may contribute to an increase in the number of M2 macrophages.121 Apparently, these mechanisms induce a cAMP-dependent shift towards a decrease in the content of Th1 lymphocytes and, respectively, M1 macrophages when exposed to bacterial toxins that increase the level of cAMP in host cells.122 Based on these data, it can be assumed that the cAMP-elevating effect of TsTYp in macrophages may also induce a shift in macrophage polarization towards the M2 phenotype, facilitating Yptb survival in cells. This assumption was confirmed through model experiments on marine invertebrates. Yptb, like other Yersinia species, successfully adapts to the external environment.123Yptb is able to survive for a long time in seawater and in the bodies of some marine invertebrates, including sea cucumbers.22 The two types of phagocytes in these animals differ phenotypically and functionally and display markers (NO level or arginase activity), suggesting that these two types of phagocytes are analogs of M1 and M2 macrophages.124 TsTYp dysregulates the functional activity of the two types of phagocytes and causes a shift towards a dominance of the activity of cells with a phenotype similar to that of M2 macrophages.125
TcpYI
The recently discovered TcpYI (24.6 kDa) was isolated from FESLF serotype I strains. TcpYI contains a C-terminal hydrophobic 115 amino acid domain similar to that of the TIRI domain family. The TcpYI gene is located on a genomic island known as the Yersinia adhesion pathogenicity island. TcpYI protein causes virulence in the absence of the pYV virulence plasmid. The toxin promotes intracellular survival of Yptb within macrophages and in mice spleens in vivo.21 As for the effects of TcpYI in the liver, there is no available data to date.
As shown previously by Rana et al.,126 a common strategy for Tcp-containing bacteria to suppress host defenses involves targeting pyroptosis-inducing TLR4 signaling via specific protein-protein interactions. Another mechanism through which some Tcp-containing bacteria demonstrate virulence is their nicotinamide adenine dinucleotide (NAD+)-hydrolase activity, leading to NAD+ depletion and cell death.127 TcpYI blocks TIR signaling by binding to TIR,21 which may protect host cells from NAD+ depletion.128 However, the downstream effects of TcpYI remain unstudied.
Data on the activity of Yptb plasmid-encoded effector proteins and chromosome-encoded toxins are summarized in Table 1.2,3,5,15,18,21,22,28–30,36,39,41,43–45,51–58,60,62,64,66–73,75,78,81,82,85,86,89–92,98,99,107,125
Table 1Anti-phagocytic and dissemination-regulating effects of protein cytotoxins/toxins of Yersinia pseudotuberculosis
| Protein | Enzyme activity | Cellular effect | Target molecule | Dissemination/liver colonization effect |
|---|
| YopM | No activity28 | Stimulation of IL-10 release28
Inhibition of IL-1 release29
Caspase-3-mediated apoptosis39
Suppression of caspase-145 | PRK236
RSK141
STAT330 | Promotes liver colonization28 |
| YopJ | Acetyltransferase, cysteine protease and deubiquitinase51,52 | Apoptosis of macrophages in the mucosa53,54
Apoptosis of monocytes in lymphoid nodes55
Prevention of inflammasome assembly52,55,58
Blocking IL-1β secretion55
Increase in release of proinflammatory cytokines55,68
Pyroptosis in neutrophils60
IL-1β release60
Pyroptosis in macrophages60,62 | Caspase-855
TAK158
NF-κB52,55,58 MAPKs58
JNK58
p3858
PGE252
Cytoplasmic death–inducing complex (RIPK1, FADD and caspase-8)55
GSDME60
GSDMD60
Caspase-162 | Promotes bacterial dissemination to lymphoid tissues53,54 Not obligatory for colonization53
Limits rates of dissemination55–57
Suppresses immune cell recruitment into the liver57 |
| YpkA | Serine threonine kinase.51 | Disruption of the actin cytoskeleton70
Suppression of phagocytosis70 and neutrophil degranulation71
Caspase-3-mediated apoptosis43,67 | RhoA, Rac1,
and Rac257
Gαq67
VASP69
EVL67
WASP67
WIP67
Gelsolin66
NFκB68
ROS68 | Promotes bacterial dissemination to brain and spleen64
No data for the liver |
| YopH | Tyrosine phosphatase71 | Inhibition of ROS production and phagocytosis in neutrophils72
Inhibition of neutrophil degranulation, decrease in immune response72
Preventing the uptake of Yptb by M cells15 | SKAP2
Syk, SLP-76, PLCγ2, and PRAM-172,73 | Promotes colonization of lymphoid tissues/liver71,72
Reduces bacterial dissemination15 |
| YopE | GTPase-activating protein51,75 | Promotion of cytoskeleton disruption44
Inhibition/activation of caspase-1 activity and release of IL-1β in macrophages78
Prevents the uptake of Yptb by M cells15
Limits the translocation of other Yops75 | Rho GTPases44
Rac 2/ Rho A78 | Restricting of bacterial dissemination to the liver75
Inhibits bacterial dissemination71 |
| YopT | Protease51 | Disruption of cytoskeleton82
Activation of inflammasome assembly44 | Rho GTPases44,51,81 | No data available |
| CNFY | Catalyzes the deamidation of glutamine in Rho GTPases, constitutively activating them85,86 | Cell membrane destabilization89
Reduction of phagocytosis89
Pyrin inflammasome assembly inhibition89
Preventing apoptosis89
IL-6, IL-8, MCP1, and MIP-3α release91 | Rho GTPases85,86
Rho A89
Rac, PI3K/ Akt/NF-κB89
STAT389 | Promotes bacterial dissemination and liver colonization90,92
Promotes bacterial dissemination89 |
| YPM | No activity3 | IL-4 release, strengthening the Th2 immune profile3,5,98 | MHC II molecules2 | Promotion of bacterial growth rate99 |
| TsTYp | No activity22 | A biphasic changes in pro-and anti-inflammatory cytokine release in mice18
Inhibition of the phagocytic activity of neutrophils and macrophages22
Apoptosis in neutrophils107
Modulation of ROS generation in monocytes and neutrophils22
Dysregulation of the functional activity of the two types of phagocytes of the sea cucumber125 | cAMP22 | No data available |
| TcpYI | No activity21 | Decrease in macrophage activity21 | TLR signaling21 | Promotes bacterial dissemination with lymph21 |
Multidirectional regulations of survival and dissemination of Yptb by protein toxins
Both Yops and chromosome-encoded toxins provide bacteria with protection from the host’s immune system.3,4 All of them have a toxic effect on the cells of the phagocytic system, causing their death or decrease in activity (Fig. 1). Simultaneously, most proteins exert multidirectional effects on the dissemination of Yptb and its further colonization in the liver (Table 1).
Yersinia usually replicates as an extracellular pathogen that evades phagocytosis by neutrophils and monocytic cells in lymphoid tissues. Facultatively, bacteria are able to survive and replicate in phagocytes, leading to increased pathogenicity.129 Once inside a cell, bacteria avoid the host’s immune defenses and can also enter lymphoid organs, including the liver. In this case, the immune cell dies, which may be accompanied by the release of antibacterial substances. Moreover, apoptotic dendritic cells release antimicrobial agents that can prime CD8+ T cells.130 Hence, the death of immune cells has been considered a strategy for Yersinia to eliminate host phagocytes. On the other hand, macrophage death, in the context of Yersinia infection, is protective for the host animal.50,131 Thus, at least two scenarios of phagocyte death may unfold simultaneously.57,61,69 The choice of the most beneficial variant for the bacterium is ensured by cytotoxic Yops and other toxins.
Typically, the activation of caspase-1 occurs at the beginning of infection, immediately after entry into the cell, and is accompanied by lysosome exocytosis and pyroptosis.41,132 The latter is a form of inflammatory cell death, followed by the release of proinflammatory mediators that have antimicrobial effects.76 Activation of caspase-1 was shown only for YopT and YopE,44,77 and under certain conditions, for YopJ.59 YopE could also inhibit the enzyme, but only in the absence of YopM,77 while YopM, vice versa, inhibited the activation of caspase-1 by YopT,44 indicating that YopM plays an important role in maintaining the balance of cytotoxicity between these two Yop proteins.
The action of most cytotoxic effector proteins is generally aimed at preventing pyroptosis-associated inflammasome assembly and caspase-1 activation (YopM, YpkA, and YopE),45,62 and causing cell death through apoptotic mechanisms (YopJ, YopH, and TsTYp),53,54,107,123 while keeping the phagocyte cell membrane intact. One might expect that TcpYI and CNFY are also capable of inducing apoptosis since they block TIR signaling associated with Toll receptor activation,21 and NO production,95 respecrively, involved in apoptosis regulation mechanisms.61
It should also be noted that the early response of macrophages to different infections is associated with polarization towards the M1 phenotype, while in chronic infection, polarization occurs via the M2 way.133,134 As a result, several toxins and effector proteins of Yptb display activities that may be associated with their ability to induce anti-inflammatory M2-type polarization of macrophages. Thus, YopM and YPM induce the release of IL-10 and IL-4, respectively.5,28,98 These cytokines belong to the Th2 immune profile.5,98 The ability of TsTYp to dysregulate the activities of two types of phagocytes in sea cucumbers has been shown in model-based experiments followed by the activation shift in one of them, considered analogous to M2 macrophages.126 One result of such polarization, as demonstrated in tuberculosis infection, is that M2 macrophages contain more bacteria compared to M1 cells.134 This suggests that polarization towards the M2-type macrophages promotes the survival and growth of bacteria in macrophages.
Protection from pro-inflammatory substances may be expected to promote the replication of bacteria within a cell and their dissemination inside phagocytes to the liver and other organs. Indeed, as shown in Table 1, YopM and YpkA promoted bacterial dissemination and colonization in the liver.28,62 Speculatively, CNFY could be considered in this series.92 Regarding TcpYI, it promotes Yptb dissemination to the spleen,21 but its effect on the liver is unknown.
YopE has been shown to promote Yptb colonization in lymphoid tissues, but different thresholds of the toxin are observed between different tissues. The effect also depends on the level of its anti-phagocytic activity and ROS inhibition.78 These findings support the idea that inhibition of the inflammatory profile of cells facilitates Yptb colonization in organs. However, the effect of YopE on bacterial colonization in the liver has not been studied to date.
Evidence that YopJ-induced apoptosis facilitates Yptb dissemination has supported the assumption about the relationship between the type of host cell death and Yptb dissemination and colonization in the liver, where apoptosis is the preferred option for bacterial dissemination.53,54 However, Viboud and Bliska obtained opposite results,20 showing that YopJ-induced apoptosis was followed by slowing down Yersinia dissemination. These inconsistencies may be explained by differences in the models used, highlighting the need to consider various factors when assessing the effects of toxins, including their interplay.
Many facts indicate that Yptb limits toxicity during long-term infection.4,120 To maintain a balance between the death of the host cell and saving a niche for survival and spread, Yptb uses a panel of toxic proteins, whose complex action provides fine-tuning of the bacterial survival mechanism. The proteins exert their action through different signaling pathways (Fig. 1). This provides reliability of regulation, where damage to one mechanism can be compensated by another. Moreover, the activity of the crucial pathways that determine the type of cell death is controlled simultaneously by several proteins, which can both mutually enhance each other’s effects and compete with each other. Thus, the activity of the caspase-1–inflammasome complex is regulated by YopM, YopJ, YopE, and YopT, with only YopM displaying exclusively inhibitory activity.44,45,60 Thus, considering that YopE can suppress caspase-1 activity, YopM appears to shift a balance between YopE-induced pyroptosis and apoptosis towards the latter. However, YopM also activates caspase-3,28 and YopJ supports this pro-apoptotic effect through the activation of caspase-8.1,20
In turn, YopE, YopТ, YpkA, and also CNFY are involved in the manipulation of the cytoskeleton, disturbances of which are closely associated with cell death.66,75,82,85,86 Yops modulate the activity of Rho GTPases in different ways, including by competing between certain proteins (YopE, YopТ), and presumably by binding to different Rho GTPases, mainly RhoA, or Rac1 and Rac2. These proteins have a wide range of activities, particularly in regulating the actin cytoskeleton and coordinating cell migration.74 Additionally, these GTPases stabilize the formation of pores through which Yops are translocated. Therefore, YopE, by inhibiting Rho GTPases, can reduce the entry of other Yops into the cell and thus decrease the possibility of extrusion and replication of Yptb in the cell.75 It may be assumed that YopТ and YpkA are also involved in the regulation of pore formation. CNFY, which has been shown to increase the translocation of Yops into a cell, can be expected to act in the same mode by regulating pore formation.94
YopH, like YopJ, inhibits ROS-dependent signaling by two independent mechanisms, indicating the importance of this pathway and its redundant regulation to increase signaling reliability.72,120,132 TcpYI is assumed to be involved in the suppression of inflammasome assembly by inhibiting the TIR signaling pathway, which leads to pyroptosis.135
TsTYp appears to induce apoptosis via a cAMP-dependent pathway.22 However, there is a lack of experimental data to elucidate details.
Experiments with staphylococcal enterotoxin superantigen have shown that the activation of T cells is accompanied by apoptosis of antigen-presenting macrophages.136 Based on these findings, YPM may be expected to induce apoptosis through T-cell activation mediated by MHC class II molecules. However, in view of some differences between YPM and superantigens of Gram-positive bacteria,98 this idea also needs to be tested.
The multiplicity of bacterial effector proteins and toxins is apparently associated with their different roles and importance for Yptb at different stages of infection and under different environmental conditions. In particular, temperature is a significant factor influencing the expression of plasmid- and chromosome-encoded genes.83 In addition, there are differences in the mechanisms of the effects of certain toxins on macrophages and neutrophils. Thus, YopJ is known to cause apoptosis in macrophages, whereas neutrophils do not undergo apoptosis in response to YopJ exposure. The apoptosis-inducing effect of YopM may be most strongly manifested on neutrophils in the liver.31,50,137 Monocytes are also more sensitive to the apoptotic effects of TsTYp than neutrophils.22 This difference determines the preferential target cells for toxins with different levels of activity at certain stages of infection. It also determines the selectivity of colonization of one or another organ, since they differ in cell composition.138
The increase in bacterial resistance to antibiotics observed over recent decades requires novel drugs with new mechanisms of action. A promising approach is the development of drugs targeting the regulation of bacterial survival and dissemination. One such target is the T3SS of Yptb, and the search for its inhibitors is already underway.139,140 However, it should be borne in mind that the pathogenicity of Yptb is not limited to the T3SS, and Yops proteins interact both with each other and with proteins encoded by chromosomes, and the effects they cause are specific to different cells of the immune system.
Conclusion
Recent studies have shown that the anti-phagocytic defense and bacterial dissemination in host organs are “fine-tuned” by both plasmid- and chromosome-encoded proteins. Toxins such as YPM, TcpYI, and CNFY have been shown to stimulate bacterial dissemination even in the absence of Yops. For efficient bacterial replication and colonization of host organs, phagocyte death by a non-inflammatory type, such as apoptosis, is favored by Yersinia. However, apoptosis may be accompanied by the activation of pro-inflammatory mechanisms of antibacterial defense, leading to pro-inflammatory cell death, namely pyroptosis. As a result, bacteria need to maintain a balance between killing cells via pyroptosis or apoptosis. Yops proteins regulate this balance using different host cell signaling pathways, often duplicating the effects of several proteins on crucial molecular components triggering cell death, and also by modulating the activity and/or rate of translocation of some proteins by others. The chromosome-encoded proteins can directly affect the mechanisms of cell death activation, and some of them even appear to influence the outcome of Yops activity, changing the rate of release or translocation of toxins into the cell or interfering with them. The chromosome-encoded proteins released by Far-Eastern strains play an important role in aggravating the course of Yptb infection. Some Yops and chromosome-encoded proteins, such as YopM, YPM, and TsTYp, induce metabolic changes in macrophages, indicating a shift towards an anti-inflammatory M2 phenotype. This provides new insights into the mechanisms of anti-phagocytic effects of Yptb toxins.
Meanwhile, a number of studies have shown that the effect of Yops on the colonization of different organs is specific, and the colonization of the liver differs from that of the spleen. Moreover, individual Yops may have different effects on bacterial colonization of the liver. However, the liver has often been overlooked in studies of lymphoid organ colonization. Further studies are required to elucidate the mechanisms of the effect of certain Yop proteins and chromosome-encoded toxins on Yptb colonization in the liver. This may provide novel approaches for the development of new types of drugs to target blocking the replication and dissemination of bacteria for the treatment of Yersinia-induced liver injury and FESLF.
Declarations
Acknowledgement
The author thanks Dr. N.F. Timchenko and Dr. I.Yu. Dolmatov for valuable advice in preparing the manuscript.
Funding
This work was implemented under the state assignment of the Ministry of Science and Higher Education of the Russian Federation (No. 124022100077-0).
Conflict of interest
The author declares no conflict of interests related to this publication.
Authors’ contributions
LSD is the sole author of the manuscript.