How Do the Virulence Factors of Shigella Work Together to Cause Disease?

Abstract

Shigella is the major cause of bacillary dysentery world-wide. It is divided into four species, named S. flexneri, S. sonnei, S. dysenteriae, and S. boydii, which are distinct genomically and in their ability to cause disease. Shigellosis, the clinical presentation of Shigella infection, is characterized by watery diarrhea, abdominal cramps, and fever. Shigella's ability to cause disease has been attributed to virulence factors, which are encoded on chromosomal pathogenicity islands and the virulence plasmid. However, information on these virulence factors is not often brought together to create a detailed picture of infection, and how this translates into shigellosis symptoms. Firstly, Shigella secretes virulence factors that induce severe inflammation and mediate enterotoxic effects on the colon, producing the classic watery diarrhea seen early in infection. Secondly, Shigella injects virulence effectors into epithelial cells via its Type III Secretion System to subvert the host cell structure and function. This allows invasion of epithelial cells, establishing a replicative niche, and causes erratic destruction of the colonic epithelium. Thirdly, Shigella produces effectors to down-regulate inflammation and the innate immune response. This promotes infection and limits the adaptive immune response, causing the host to remain partially susceptible to re-infection. Combinations of these virulence factors may contribute to the different symptoms and infection capabilities of the diverse Shigella species, in addition to distinct transmission patterns. Further investigation of the dominant species causing disease, using whole-genome sequencing and genotyping, will allow comparison and identification of crucial virulence factors and may contribute to the production of a pan-Shigella vaccine.

Keywords: Shigella; Shigellosis; bacterial pathogenesis; type III secretion system; virulence effectors.

Figure 1

Infectious cycle of Shigella (Roerich-Doenitz, 2013) modified from Schroeder and Hilbi (2008). Entry into the colonic epithelium is mediated in two ways: M-cell membrane ruffling, and epithelial barrier destabilization. Entry via M-cells is achieved through membrane ruffling (1), and the bacillus is then transported to the M-cell pocket, where it is endocytosed by resident macrophages (2). Epithelial barrier destruction is mediated by pro-inflammatory (IL-1) and chemotaxic cytokines (IL-8). IL-8 produced by neighboring epithelial cells recruits PMN leukocytes (12), which travel from the basolateral to the apical colonic epithelium, destabilizing the junctions between the epithelial cells and allowing further invasion of Shigella (5). Induction of pyroptotic macrophage death occurs after Shigella escape from the phagocytic vacuole (3 and 6). Caspase-1, when activated, cleaves and activates IL-1β and IL-18, leading to the release of these pro-inflammatory cytokines (4) (Jennison and Verma, 2004). Uptake of Shigella is a macropinocytic process at the basolateral membrane of epithelial cells (7). Stimulation of Rho-family GTPases triggers actin polymerisation and then depolymerization, forming filopodial and lamellipodial extensions of the epithelial membrane, leading to engulfment of the bacilli. Lysis of the macropinocytic vacuole allows Shigella to gain access to the epithelial cytoplasm, where it rapidly multiplies, escapes autophagy and fragments the Golgi (8 and 9). Exploitation of the epithelial actin assembly machinery allows Shigella to move both intra- and intercellularly (10). Protrusions mediated by bacilli are actively endocytosed by the clathrin-mediated endocytic pathways at intercellular junctions, and the double membrane vacuole is lysed to give Shigella access to the neighboring cells cytoplasm (11). PMN leukocytes eventually eliminate Shigella infection from the colonic epithelium (13).

Figure 2

Genomic organization of the entry region on plasmid pWR100 (Roerich-Doenitz, 2013). Genes are clustered in two operons, the ipa/ipg and the mxi/spa operon. They are colored in the legend according to their protein class, some of which, such as T3SS effectors, are detailed in the text. Secretion machinery refers to the components that build the T3SS. Translocators are components of the translocon, a pore inserted into the host membrane to allow effector translocation and chaperones are components that stabilize individual effectors prior to secretion from the bacterium. Regulators modulate T3SS expression and function but they are largely beyond the scope of this review. This figure was modified from the virulence plasmid map by Buchrieser et al. (2000).

Figure 3

Schematic drawing of the Shigella Type III Secretion System (Roerich-Doenitz, 2013). The architecture of Shigella T3SS is based on the virulence plasmid mxi-spa locus (Figure 2), which encodes proteins that produce the T3SS apparatus, a structure composed of a 60 nm hollow extracellular needle, a transmembrane domain, and a cytoplasmic bulb (Blocker et al., 1999).

Figure 4

Role of IpaD, IpaB, and IpaC in regulation of the T3SS (Martinez-Argudo and Blocker, 2010). Four hydrophilic IpaD molecules and one hydrophobic IpaB molecule are localized at the tip of the T3SS needle in the “off” state. IpaB senses host cells upon contact of the T3SS needle tip, and inserts into the membrane to signal “translocators on.” Secretion of effectors is signaled by conformational changes in IpaD via a signal transduction pathway to the base of the secreton. Four IpaC may travel up the secreton and associate with the needle tip, one atop each IpaD, and insert around the single IpaB to form a translocon in the epithelial cell membrane. This signals “effectors on” and early effectors are injected into the cytoplasm of the host epithelial cell (Veenendaal et al., 2007).

Figure 5

IcsA mediates actin-based motility and IcsB prevents autophagy. IcsA is localized in a unipolar fashion to promote efficient unidirectional intracellular movement parallel to the long axis of Shigella for movement intra- and intercellularly. IcsB masks the Atg5 recognition site on IcsA, recruits Toca-I, and prevents the formation of septin cages.

Figure 6

VirA and IpaJ mediate Golgi fragmentation, which may contribute to entry vacuole disruption. There is a level of semi-redundancy between VirA and IpaJ, as a double virA⁻ipaJ⁻ knockout is required to completely abolish Shigella-induced Golgi fragmentation. IpaJ is the dominant effector, and targets ARF1-GTP for N-myristoylated cleavage at the exposed di-glycine motif. ARF1 is irreversibly lost from the Golgi membrane, leading to inhibition of Golgi trafficking (Burnaevskiy et al., 2013). VirA is capable of interacting and inactivating many Rab GTPases to mediate Golgi fragmentation, including Rab1 at the ER and Rab33 at the Golgi (Dong et al., 2012). This causes disruption of ER-to-Golgi trafficking, leading to fragmentation of the Golgi. Inactivation of Rab35 at the recycling endosome may prevent recruitment of additional membrane to the Shigella entry vacuole. Replication of Shigella within the entry vacuole without additional membrane may then allow vacuole lysis. IpgD may also contribute to vacuole lysis by recruiting Rab11 to macropinosomes, which have been shown make contact with the entry vacuole to accelerate vacuolar rupture (Mellouk et al., ; Weiner et al., 2016).

Figure 7

Shigella modulates the innate immune response in epithelial cells. NFκB can be activated via many pathways during Shigella infection, including genotoxic stress during invasion and recognition of the lipopolysaccharide on the bacterial surface. This may explain the redundancy of Shigella effectors, many of which have the ability to interfere with these pathways, leading to efficient inhibition of NFκB. IpgD increases levels of PI5P, which activates PI3K and Akt and inhibits p53 and caspase 3. OspC3 sequesters the p19 pre-cursor of caspase 4, which prevents its heterodimerization with p10 and inhibits capase 4 activation. OspI deamidates Ubc13, which inhibits ubiquitination of TRAF6 and prevents activation of the TRAF6-NFκB pathway. IpaH0722 ubiquitylates TRAF2, leading to its proteasomal degradation and preventing recruitment of IKK and activation of NFκB. Polyubiquitination of NEMO/IKKγ is dependent on IpaH9.8 binding to A20 binding inhibitor of NFκB (ABIN-1), leading to proteasomal degradation of NEMO/IKKγ, activation of IκBα, and inhibition of NFκB. IpaH9.8 also localizes to the nucleus where it catalyses ubiquitination of U2AF³⁵, leading to its degradation and consequently reduced il-8 expression and decreased neutrophil recruitment. By an unknown mechanism, OspG prevents SCF^β−TrCP from ubiquitinating IκBα, preventing degradation of IκBα and activation of NFκB. OspZ blocks the nuclear translocation of p65, a subunit of NFκB, thereby inhibiting activation of NFκB. Threonine desphosphorylation of MAPKs by OspF leads to inactivation of ERK1/2 and p38 MAPK pathways and inhibition of histone H3 phosphorylation (H3pS10). This produces an inaccessible chromatin conformation at the gene promoters of inflammatory cytokines and chemokines, meaning NFκB cannot activate their transcription.

Figure 8

S. flexneri serotype composition and conversion. LPS O-antigen modifications include glucosylation (the addition of glucosyl groups), mediated by the gtr operon and the addition of O-acetyl groups, achieved by the O-acetyltransferase protein encoded by the oac gene. This is the basis of serotype conversion, as Shigella begins with serotype Y and the basic O-antigen structure, and further modification produces different serotypes (Allison and Verma, 2000).

Figure 9

IpgD inhibits T-cell migration and IpaD induces B-cell apoptosis. (A) The PIP2 concentration at the plasma membrane is responsible for the dynamic interchange between active and inactive conformations of ezrin, radixin, and moesin (ERM) proteins (Konradt et al., 2011). ERM proteins are involved in cell-cortex organization of T-cells, and their active conformations localize at the membrane in response to chemokine stimulation to allow T-cell migration toward chemoattractants (Konradt et al., 2011). In the subcapsular sinus of the lymph node (Salgado-Pabón et al., 2013) Shigella can invade T-cells or inject IpgD via the T3SS into T-cells to reduce PtdIns(4,5)P₂ concentration. This prevents cell polarisation induced by active ERM proteins and T-cell migration to sites of infection (Konradt et al., 2011). (B) Bacterial co-signals increase pro-apoptotic proteins, induce loss of mitochondrial membrane potential (MMP), and also upregulate tlr2 mRNA, leading to increased TLR2 expression on the surface of the cell. IpaD signaling via the TLR2-1 heterodimer leads to up-regulation of FAS-associated death domain (FADD) protein, which ultimately induces B-cell apoptosis (Nothelfer et al., 2014).