- About MBL
- For Contributors
Molecular and Cellular Microbiology (MCM) | Host-Microbe Interaction and Pathogenesis
Microbiol. Biotechnol. Lett. 2022; 50(4): 441-456
Kyung-Soo Lee1,2†, Seo Young Park1†, and Moo-Seung Lee1,2*
1Environmental Diseases Research Center, Korea Research Institute of Bioscience and Biotechnology, Daejeon 34141, Republic of Korea
2Department of Biomolecular Science, KRIBB School of Bioscience, Korea University of Science and Technology (UST), Daejeon 34113, Republic of Korea
Correspondence to :
Moo-Seung Lee, firstname.lastname@example.org
Shiga toxins (Stxs) are major virulence factors from the enterohemorrhagic Escherichia coli (EHEC), a subset of Stx-producing Escherichia coli. Stxs are multi-functional, ribosome-inactivating proteins that underpin the development of hemolytic uremic syndrome (HUS) and central nervous system (CNS) damage. Currently, therapeutic options for the treatment of diseases caused by Stxs are limited and unsatisfactory. Furthermore, the pathophysiological mechanisms underpinning toxin-induced inflammation remain unclear. Numerous works have demonstrated that the various host ribotoxic stress-induced targets including p38 mitogen-activated protein kinase, its downstream substrate Mitogen-activated protein kinase-activated protein kinase 2, and apoptotic signaling via ER-stress sensors are activated in many different susceptible cell types following the regular retrograde transportation of the Stxs, eventually leading to disturbing intercellular communication. Therapeutic options targeting host cellular pathways induced by Stxs may represent a promising strategy for intervention in Stx-mediated acute renal dysfunction, retinal damage, and CNS damage. This review aims at fostering an in-depth understanding of EHEC Stxs-mediated pathogenesis through the toxin-host interactions.
Keywords: STEC, shiga toxins, hemolytic uremic syndrome, host responses, inflammation
Shiga toxins (Stxs; also called verotoxins) are produced by certain strains of Stx-producing bacteria including
STEC is a human-adapted pathogen transmitted via contaminated food and/or water. Contaminated beef was responsible for the highest proportion of STEC outbreaks (15%), followed by fresh produce (14%). In addition, pork, dairy products, alfalfa sprouts, and grains and beans have also been associated with STEC contamination [12, 13]. Reservoirs of STEC include mammals, invertebrates, fish, birds, and amphibians [14−16]. Ruminants in particular appear to be a major reservoir for STEC. As in humans, ruminants are infected with STEC via contaminated feed and drinking water, or by exposure to the feces of STEC-infected animals [16, 17]. Cattle are frequent reservoirs of STEC but display no signs of disease after infection [18−20]. Cattle lack expression of the Stxs receptor (globotriaosylceramide; Gb3) in the intestine, leading to asymptomatic spread of STEC within the cattle herd, promoting STEC transmission through the food chain . In addition to the cattle, other ruminants such as alpacas, antelopes, deer, elk, moose, water buffalo, and yaks are considered to be STEC reservoirs [22−26]. Smaller ruminants such as sheep and goats are also frequent asymptomatic carriers of STEC .
EHEC harbors a type 3 secretion system (T3SS) produced by the locus of enterocyte effacement (LEE). LEE is a 35.6 kb pathogenicity island encoded by bacteria including EHEC and EPEC. LEE encodes a functional T3SS, in addition to other secreted proteins, including the bacterial outer membrane protein intimin, and translocated intimin receptor (Tir) protein . Intimin is required for intimate bacterial adhesion to human intestinal epithelial cells. Upon contact, LEE-encoded intimin and Tir are secreted into human intestinal epithelial cells via the T3SS [29, 30]. The surface expression of fimbrial adhesions induces adhesion of STEC to host epithelial cells. Once STEC adhere, characteristic attaching and effacing lesions (A/E) are formed leading to destroy the microvilli . This attachment of
Shiga toxins (Stxs) are major virulence factors produced by
To cause systemic diseases and inflammation, Stxs require access to the bloodstream.
Globotriaosylceramide (Gb3) is the predominant mammalian cell surface receptor for most Shiga toxin families. Globotetraosylceramide (Gb4) is also reported as a minor receptor for Shiga toxin, especially the Stx2e variant. Gb3 is synthesized in the Golgi before being transported to the plasma membrane, exposing its trisaccharide moiety, and non-covalently linking its hydrocarbon ceramide (C-16 to C-24) with the plasma membrane. The B subunits of Stxs recognize the terminal alpha-1,4 di-galactose of the trisaccharide moiety . Interestingly, a single B subunit of Stx has three Gb3 binding sites . Several studies have indicated that targeting Gb3 can reduce Stxs-induced cytotoxicity. For example, addition of Gb3 to the tissue culture medium dramatically reduces the biological activities of Stxs. Furthermore, digestion of membrane glycolipids by galactosidase blocks glycolipid synthesis and subsequently reduces Stxs-mediated cytotoxicity . While numerous studies have focused on targeting Gb3 to reduce Stxs-induced cytotoxicity, it is important to note that Gb3 expression and processing are regulated by many factors. For instance, bacterial lipopolysaccharides (LPS), lymphotoxin (TNF-β), TNF-α, and interleukin-1 (IL-1) have been reported to increase the synthesis of Gb3 . Furthermore, butyrate produced by components of the microbiota were also reported to increase the expression of Gb3 .
i) Endocytosis of Shiga toxins
After binding of the Shiga toxin B subunit to Gb3, clathrin-coated pits promote the internalization of Shiga toxins into target cells . Shiga toxins themselves have been reported to mediate clathrin phosphorylation, promoting their own uptake via clathrin-dependent endocytosis . Furthermore, A subunits, bound with the host plasma membrane, stimulate clathrin-dependent uptake of the toxin by inducing toxin-receptor complex recruitment . Butyric acid also increases the internalization of Shiga toxins via clathrin-dependent mechanisms leading to cytotoxicity . Inhibition of clathrin function reduces uptake of Shiga toxins by up to 35%, but does not completely abolish it, suggesting that clathrinindependent pathways for toxin internalization also exist [61, 62]. For example, Shiga toxins stimulate activity of the soluble N-ethylmaleimide-sensitive-factor attachment protein receptor (SNARE) proteins VAMP2, VAMP3, and VAMP8 localized on the plasma membrane leading to clathrin-independent endocytosis . Interestingly, human intestinal epithelial cells, which express no Gb3, uptake Shiga toxins by micropinocytosis, confirming the presence of various routes for toxin internalization suggesting uptake mechanisms of Shiga toxins are various . After endocytosis, Shiga toxins undergo retrograde transport from the endosome to the Golgi apparatus and the ER in order to bypass degradation in late endosomes and lysosomes.
ii) Transport of Shiga toxins from the endosome to the Golgi apparatus
Shiga toxins must translocate from endosomes to the trans-Golgi network (TGN) to induce cytotoxicity. The Rab9-dependent transport pathway, used by the mannose-6-phosphate receptor (M6PR), is a major route for translocation of proteins from endosomes to the Golgi [65-68]. By contrast, Shiga toxins translocate to the TGN directly from endosomes in a Rab9-independent and calciumdependent manner [69−71]. It has also been reported that endosomal Shiga toxins are transported to the TGN in a Rab11- and clathrin-dependent manner [72, 73]. Interestingly, mutations in dynamin (dynK44A) reduce transport of Shiga toxin B subunits from endosomes to the TGN . An additional host protein that contributes to the transportation of Shiga toxins from endosomes to TGN is vesicle SNARE (v-SNARE), a transmembrane protein essential for membrane fusion . Recently, phosphoinositide binding proteins Sorting nexin 1 (SNX1) and Sorting nexin 2 (SNX2) were implicated in the transport of Shiga toxins to the TGN. Vero cells, which are sensitive to Shiga toxin stimulation, show a 40% decrease in toxin translocation to the TGN after SNX1,2 depletion by siRNA . In addition to the role of host proteins, bacterial products have also been reported to increase Shiga toxin transportation to the TGN. Butyric acid, a byproduct of human commensal bacteria, is reported to increase the transportation of Shiga toxins to the TGN, leading to cytotoxicity of target cells .
iii) Shiga toxin transport from the trans-Golgi network to the endoplasmic reticulum
Upon accessing the TGN, Shiga toxins translocate to the ER using the host retrograde transport system. Retrograde transport occurs when proteins are recycled or require localization to specific cellular compartments. For instance, the cellular adhesion protein P-selectin is transported to the TGN for repacking and recycling. The copper transporter protein Menkes is also transported by the retrograde transport system for functional localization to cellular compartments . Furthermore, Lys-Asp-Glu-Leu (KDEL) motif-bearing proteins are retro-translocated to the ER by KDEL receptors in the Golgi. Interestingly, certain pathogens hijack the host KDEL retrograde transport system to induce cytotoxicity.
iv) Release of Shiga toxins from ER to cytosol
To inhibit protein synthesis, the A1 fragment of the Shiga toxin must be released from the holotoxin. As described above, the A subunit consists of an A1 fragment with
Apoptosis is a ‘programmed cell death’ that plays a key role in the development of pathogenesis induced by Shiga toxins. Apoptosis is mediated by the sequential activation of caspases, leading to membrane blebbing, chromatin condensation, DNA fragmentation, and the subsequent formation of apoptotic bodies . Numerous studies have shown that Shiga toxins induce apoptosis in epithelial cells, endothelial cells, astrocytoma cells, monocytes, and amniotic cells by inducing ER stress leading to caspase activation. For example, Shiga toxin treatment activates caspases 6 and 9 and cleaves Bid and poly ADP-ribose polymerase (PARP), leading to apoptosis in the HEp-2 epithelial cell line . Furthermore, human proximal tubular cells (HK-2) and human leukemia monocytic cells (THP-1) also activate caspases 3 and 7 after Shiga toxin stimulation leading to apoptosis . Interestingly, the pan-caspase inhibitor carbobenzoxy-valyl-alanyl-aspartyl-[O-methyl] fluoromethylketone (Z-VAD-FMK) inhibits Shiga toxin-induced cytotoxicity, suggesting that caspases are the main regulator of Shiga toxin-induced cell death . Although the precise mechanisms of Shiga toxin-induced apoptosis remain unclear, it is well known that mitogen-activated protein kinase (MAPK) signaling contributes to the induction of apoptosis. MAPK is a serine/threoninespecific protein kinase that is activated by extracellular growth factors, mitogens, and cellular stress . MAPK signaling is composed of three major constituents including c-Jun N-terminal kinases (JNK), extracellular signal-regulated kinases (ERK), and p38 MAPK. JNK, ERK, and p38 MAPK activation are regulated by phosphorylation, leading to the activation of transcription factors that promote apoptosis and inflammation [88, 89]. Like other ribotoxins such as α-sarcin, Shiga toxins activate JNK, ERK, and p38 MAPK in epithelial and myeloid cells resulting in apoptosis and inflammation [3, 90, 91]. However, inhibition of zipper sterile-α-motif kinase (ZAK), a transduction protein in the MAPK pathway, did not block apoptotic DNA fragmentation caused by Shiga toxins, suggesting that Shiga toxins can also induce apoptosis independent of MAPK pathways .
In addition to apoptosis, Shiga toxins have been shown to induce the programmed cell death pathways of pyroptosis and necrosis. Like apoptosis, pyroptosis is regulated by caspases. Pyroptosis is activated by caspases 1, 4, 5, and 11 rather than caspases 3 and 7 in apoptosis. Pyroptosis is characterized by the formation of membrane pores by gasdermin D (GSDMD) . GSDMD pores induce cell lysis and release pro-inflammatory intracellular contents, leading to inflammation of the surrounding cells. The NLR family pyrin domaincontaining protein 3 (NLRP3) is activated in pyroptosis, which accelerates the activity of caspase 1, IL-1β secretion, and cleavage of GSDMD . Interestingly, Shiga toxins were shown to induce NLRP3 oligomerization and caspase 1 and 4 activity, leading to the cleavage of GSDMD to its active form (p30) and subsequently the induction of pyroptosis . GSDMD also localizes to the mitochondrial membrane to form pores leading to the production of reactive oxygen species (ROS) and the release of cytochrome C, which subsequently induces apoptosis and pyroptosis . Necrosis, unlike apoptosis and pyroptosis, is a caspase- and ATP-independent form of cell death. Necrosis differs from apoptosis in that there is no cell swelling or chromatin condensation, and the process induces robust inflammation. Shiga toxins were reported to induce necrosis in the breast cancer cell line T47D, indicating that multiple types of cell death can be induced by Shiga toxin exposure and that this process is likely cell type-specific .
As described above, the A subunit of Shiga toxin inhibits protein synthesis by depurinating eukaryotic ribosomes, leading to the induction of the unfolded protein response (UPR) and ER stress. The UPR is activated when misfolded and/or unfolded proteins accumulate in the ER . The UPR is an adaptive response against ER stress that reduces unfolded protein load to maintain cellular homeostasis. The UPR signal is activated by three major UPR stress sensors, inositol-requiring protein 1 (IRE1), protein kinase RNA-like ER kinase (PERK), and activating transcription factor 6 (ATF6). When UPR is initiated, cellular membrane resident IRE1 forms a homodimer promoting auto-phosphorylation and removal of introns of the X box-binding protein 1 (XBP1u) to make spliced XBP1 (XBP1s). XBP1s is a transcription factor for UPR target genes that regulate autophagy, apoptosis, quality control, and folding [98, 99]. PERK is an additional UPR sensor located on the cellular membrane, which is released from the BiP complex by UPR before homo-dimerizing. Dimerized phosphorylated PERK subsequently activates eukaryotic initiation factor 2α (eIF2α) by phosphorylation. Phosphorylated eIF2α directly inhibits global translation to recover from ER stress . In addition, phosphorylated eIF2α induces activating transcription factor 4 (ATF4), which promotes UPR gene expression including
Calcium (Ca2+) influx is a crucial regulator of cell survival and death mediated by ER stress . Intracellular Ca2+ is mainly stored in the ER lumen to re-fold proteins using Ca2+-binding chaperones . Ca2+ is also involved in the induction of apoptosis . Bcl-2-associated X protein (Bax), which is upregulated by ER stress, leads to the release of Ca2+ from the ER lumen and subsequently increases Ca2+ level in the mitochondria. Increased Ca2+ levels in the mitochondria subsequently promote release of cytochrome C, inducing apoptosis. Shiga toxins, as an ER stress inducer, are also reported to induce calcium influx in HeLa cells and platelets . Interestingly, an antagonist of the P2X1 receptor – a known calcium channel – reduces Ca2+ influx in the ER lumen, leading to increased cell survival following Shiga toxin stimulation of HeLa cells . Furthermore, the P2X1 antagonist reduced the release of Shiga toxin-induced pathogenic microvesicles, suggesting that Ca2+ influx is also associated with the pathogenicity of Shiga toxins .
Membrane-derived extracellular vesicles (EVs) are essential for maintaining cellular homeostasis and communicating with other cells. Initially, EVs were shown to discard excess cell membrane and unnecessary membrane receptors such as transferrin . Over time, additional biological functions of EVs have been identified. For instance, transferrin receptor families carried by EVs are transferred to recipient cells, leading to phenotypic changes in the target cell [109−111]. EVs derived from dendritic cells containing MHC complexes were also shown to amplify T cell activation and immune responses, suggesting that EVs play a key role in immune regulation . EVs are subdivided into two major groups, exosomes and microvesicles. Exosomes are ~30−200 nm in size and contain cellular components including proteins, receptors, and/or RNAs [113, 114]. Microvesicles are ~200−1000 nm in size and contain cellular components [115−117]. The biogenesis of microvesicles and exosomes display marked differences in that microvesicles are generated from cellular membrane budding while exosomes are derived from endosomes. More recently, a pathogenic role of microvesicles in Shiga toxin-induced diseases was reported. The quantity of circulating microvesicles in the blood is increased in late-phase HUS patients. Furthermore, Shiga toxins were detected in blood-borne microvesicles originating from neutrophils, platelets, and red blood cells, suggesting that Shiga toxins transfer to target organs from blood via microvesicles [47, 118−120]. Furthermore, Shiga toxins were reported to induce the release of complement-coated microvesicles derived from red blood cells leading to acute disease progression, suggesting that complement activation contributes to the Shiga toxin-mediated pathogenesis via microvesicles . Similarly, it is reported that exosomes derived from Stx2a-intoxicated human macrophages also contain Stx2a. Not only Stx2a but also pro-inflammatory exosomal mRNAs were detected in the exosomes suggesting the pathogenic role of exosomes in HUS development. Interestingly, blockade of exosomal biogenesis by GW4869 ameliorates cytotoxicity in target cells suggesting new therapy for HUS development . The role of exosomes in HUS development however, remains unclear. Thus, the regulated release of exosomes and the function of exosomes in HUS development should be considered in major Stx target cells including human macrophage and various renal cells.
Ribotoxic stress was first discovered by site-specific modification of ribosomal peptidyl transferase. The plant toxin ricin A and fungal ribotoxin α-sacrin act on eukaryotic 28S ribosomal RNA, targeting the peptidyl transferase in the center of ribosomes, leading to ribosome inactivation and activation of JNK . Like other ribosome-inactivating proteins (RIPs), Shiga toxins inactivate eukaryotic ribosomes by cleaving the N-glycosidic bond at A-4324 in 28S ribosomal RNA, leading to the induction of ribotoxic stress . For instance, Shiga toxin-susceptible cells including human macrophagelike differentiated THP-1 cells, kidney tubular cells, and retinal pigment epithelial cells induce acute activation of ribotoxic stress markers including PERK, p38 MAPK, JNK, ERK, CHOP, and DR5α after toxin stimulation [2, 3, 91, 122]. Ribotoxic stress markers such as p38 MAPK, JNK, and ERK are also involved in pro-inflammatory signaling pathways. For instance, cellular stress such as ROS induces phosphorylation of p38 MAPK, JNK, and ERK to increase the production of pro-inflammatory cytokines including TNF-α and IL-1β . Similarly, phosphorylation of p38 MAPK, JNK, and ERK by Shiga toxin exposure enhances the production of pro-inflammatory cytokines. Inhibition of p38 MAPK, JNK, and ERK extensively reduce the production of pro-inflammatory cytokines including IL-1β and IL-8 . Furthermore, inhibition of p38 MAPK signaling by PKR inhibitors also reduces cytokine production including TNF-α and GM-CSF, suggesting a key role of ribotoxic stress markers in Shiga toxin-induced inflammation .
Antibiotics are not recommended for the treatment of Shiga toxin-mediated diseases. Bactericidal antibiotics activate the lytic cycle of STEC. This enhances the production of Shiga toxins through the activation of the lysogenic cycle promoter. For instance, the quinolone antibiotic ciprofloxacin induces robust expression of the
Antibodies against Shiga toxins are produced in STEC-infected patients, suggesting that the adaptive immune system could play a role in HUS treatment [128, 129]. Natural antibody-production in humans is typically longer than the disease course of HUS. There is therefore a need to generate humanized antibodies for the treatment of HUS
Since the major receptor of the Shiga toxin is Gb3, there are several trials aimed at the treatment of HUS with Gb3 analogs to inhibit the internalization of toxin by target cells expressing Gb3. SYNSORB Pk, which consists of silicon dioxide particles with covalently linked trisaccharides, was the first Gb3 analog used to treat HUS. SYNSORB Pk was tested in 145 HUSdiagnosed children, resulting in no significant difference in the number of deaths . The reasons for treatment failure are likely that (i) SYNSORB Pk was administrated after HUS development, (ii) Shiga toxins are not free floating but more likely cell- and/or exosomes/microvesicles associated, and (iii) the binding affinity of Shiga toxins with Gb3 is higher than SYNSORB Pk. Thus, SUPER TWIG Gb3 analogs that show higher affinity with Gb3 than SYNSORB Pk were developed [136, 137]. Interestingly, SUPER TWIG successfully bound to Shiga toxin with a higher affinity and prevented the uptake of Shiga toxins by target cells. In addition, phagocytosis of Shiga toxin by macrophages is increased, leading to the survival of mice after toxin exposure. Acrylamide polymers of Gb3, which have a higher affinity to Gb3 than SUPER TWIG, also reduced Shiga toxin-mediated lethality in a mouse model . Interestingly, orally administrated polymers reduced lethality even after colonization of the bacteria in the gut, suggesting a new therapeutic agent to treat HUS.
Since Shiga toxins bind Gb3, there are several trials aimed at delivering molecules to Gb3 expressing cells using B subunits of Shiga toxin. For instance, human cancer cells including breast cancer cells , ovarian cancer cells , prostate cancer cells , and testicular cancer cells , which express high levels of Gb3, are natural targets for Shiga toxin B subunits. Cytotoxic compounds such as topoisomerase I inhibitor SN38 and benzodiazepine RO5 linked with toxin B subunits are reported to inhibit the cancer cells described above , . Indeed, natural holotoxins have emerged as a novel cancer therapy. For instance, intratumoral or intraperitoneal injection of Shiga toxin inhibited tumor growth in a murine metastatic fibrosarcoma model , human malignant meningiomas , human bladder carcinoma , human renal carcinoma , and human astrocytoma . Interestingly, no side effects were observed in Shiga toxin-treated murine models for more than 50 days . Due to the lack of other virulence factors including LPS, adhesins, and bacterial proteases, purified Shiga toxins may not be sufficient to induce HUS. Therefore, purified toxins alone may not be able to induce side effects such as bloody diarrhea . Although murine models successfully show the potent anti-cancer effects of Shiga toxins, several problems remain. For instance, not only cancer cells express a high level of Gb3. Other cell types including human renal cells express Gb3, which may promote off-target effects of anti-cancer therapies using B subunits of Shiga toxins. However, the Shiga toxins or Bsubunit of the holotoxin might be exploited in immunotherapy or cancer patient treatment.
Finally, in the review, we discussed our current knowledge about the Shiga toxin-mediated signaling pathways in host cells and provide a concise review of therapeutic applications demonstrated by targeting identified host-directed factors to mitigate the pathological conditions in the infected human body. STEC are food-borne bacteria that are distributed among mammals, invertebrates, fish, birds, and amphibians. Recently, many more people have been at increased risk for developing life-threatening extraintestinal complications such as HUS following infection with Shiga toxin-producing bacteria. Even though many virulence factors of STEC are revealed, it is hard to reduce the pathogenesis of HUS by targeting virulence factors of STEC due to the lack of specificity and bacterial host immune evasion. As seen in the German outbreak, STEC including hybrid strain EHEC/EAEC evolves through the recombination of various STEC genes resulting in new virulence factors. Thus, it is essential to understand more precisely the interaction between the Shiga toxins and the host immune system to reduce the pathogenesis of STEC. Besides therapeutic options that target Shiga toxins themselves including neutralizing antibodies and Gb3 analogs, emerging interventions to inhibit Shiga toxins and/or STEC have been recently developed focusing on the host cellular responses. For instance, targeting exosomes induced by Shiga toxins is one of the candidates for neutralizing the toxins. Lee
Furthermore, the microbiome that modulates the host cellular responses is reported to reduce STEC-mediated pathogenesis. For instance,
This work was supported by the KRIBB Research Initiative Program (KGS1352221, KGM5322214) and a grant from the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (NRF-2018M3A9H3023077, 2021M3A9H3016046) and and by the Basic Science Research Program through the National Research Foundation of Korea (NRF) (2022R1A2C1003699) and also the "Cooperative Research Program for Agriculture Science and Technology Development (Project No. PJ015001022022)" Rural Development Administration, Republic of Korea.
The authors have no financial conflicts of interest to declare.
Ji-Won Park, Jin-Mi Park, Sangmi Eum, Jung Hee Kim, Jae Hoon Oh, Jinseon Choi, Tran The Bach, Nguyen Van Sinh, Sangho Choi, Kyung-Seop Ahn, and Jae-Won LeeMicrobiol. Biotechnol. Lett. 2022; 50(4): 574-583 https://doi.org/10.48022/mbl.2206.06004
Joong Hyun ShimMicrobiol. Biotechnol. Lett. 2022; 50(3): 354-360 https://doi.org/10.48022/mbl.2204.04003
Ho KimMicrobiol. Biotechnol. Lett. 2022; 50(1): 157-163 https://doi.org/10.48022/mbl.2108.08004