Gram-positive and gram-negative bacterial toxins in sepsis: a brief review.

Bacterial sepsis is a major cause of fatality worldwide. Sepsis is a multi-step process that involves an uncontrolled inflammatory response by the host cells that may result in multi organ failure and death. Both gram-negative and gram-positive bacteria play a major role in causing sepsis. These bacteria produce a range of virulence factors that enable them to escape the immune defenses and disseminate to remote organs, and toxins that interact with host cells via specific receptors on the cell surface and trigger a dysregulated immune response. Over the past decade, our understanding of toxins has markedly improved, allowing for new therapeutic strategies to be developed. This review summarizes some of these toxins and their role in sepsis.

Introduction

Sepsis is defined as a systemic inflammatory response syndrome (SIRS) in the presence of suspected or proven infection. It is the second most common cause of death in non-coronary intensive care units (ICU) and the tenth overall cause of death in high income countries., The incidence of sepsis in the past two decades has annually increased by 9%, to reach 240 per 100 000 people in the USA by 2013.,

Initially it was thought that the major organisms that caused bacterial sepsis were gram-negative bacteria. However, over the past 25 y it has been shown that gram-positive bacteria are the most common cause of sepsis. Some of the most frequently isolated bacteria in sepsis are Staphylococcus aureus (S. aureus), Streptococcus pyogenes (S. pyogenes), Klebsiella spp., Escherichia coli (E. coli), and Pseudomonas aeruginosa (P. aureginosa).

In order to cause disease, pathogens have to employ an array of factors known as virulence factors that protect them from the host innate immune system and enable them to cross mucosal barriers, disseminate, and replicate in distant organs., Importantly, each stage of infection involves the expression of different virulence factors depending on the stage of infection. Some of the most important bacterial virulence factors are toxins. These toxins include endotoxin or lipopolysaccharide (LPS) that is present in the outer membrane of the gram-negative bacterium and several other secreted exotoxins and enterotoxins in other bacteria. Bacterial toxins are mainly divided into three types based on their mode of action. Type I toxins disrupt host cells without the need to enter the cells. These include superantigens (SAgs) produced by S. aureus and S. pyogenes. Type II toxins, such as hemolysins and phospholipases destroy host cell membranes to invade and interrupt host defense processes within the cell. Type III toxins, also known as A/B toxins due to their binary structure; disrupt host cell defenses to allow dissemination to remote organs. The B component of these toxins binds to the host cell surface, while the A component possess the enzymatic activity to damage the cell. Several lethal toxins including Shiga toxin, cholera toxin, and anthrax lethal toxin belong to the Type III toxin family.

The host innate immune cells recognize several of the bacterial virulence factors via unique receptors called pattern-recognition receptors (PRRs). PRRs recognize conserved motifs on the pathogen surface to initiate an innate immune response. Over the last decade with major research in the field of toxins and their interaction with host cells and PRRs, there has been a wealth of knowledge in understanding sepsis. This review aims to briefly focus on our current knowledge of some important toxins and their functions.

Endotoxins

Endotoxins are the glycolipid, LPS macromolecules that make up about 75% of the outer membrane of gram-negative bacteria that are capable of causing lethal shock., The structure of LPS generally consists of a hydrophobic lipid A domain, an oligosaccharide core, and the outermost O-antigen polysaccharide. Lipid A is a di-glucosamine-based lipid that serves as a hydrophobic anchor of LPS to the microbial membrane. E. coli is known to harbor approximately 106 lipid A residues on the surface., Lipid A is a highly diverse molecule and the diversity is manifested in part in the number of fatty-acid side chains and the presence of terminal phosphate residues. Lipid A of E. coli that is hexa-acylated with side chains of 12–14 carbons has enhanced stimulatory effect of human cells compared with lipid A where the length of the side chains or the charge has been altered.- The lipid A of some human pathogens like Francisella spp., Yersinia pestis, and Helicobacter pylori contain typically only 4 or 5 acyl chains of 16–18 carbons in length and are poorly recognized by human LPS receptor known as Toll-like receptor 4 (TLR4).-

Lipid A is the single region of LPS that is recognized by the innate immune system. Picomolar concentrations of lipid A are sufficient to trigger a macrophage to produce proinflammatory cytokines like TNF-α and IL1β.- To trigger an innate immune response, the lipid A portion of LPS alone is sufficient, yet the adaptive immune response during infection is usually directed toward the O-antigen. The key pattern recognition receptor for LPS recognition is Toll-like receptor 4 (TLR4). LPS in circulation is solubilized by LPS-binding protein (LBP) in the serum. The endotoxin is then transferred to an extrinsic glycosylphosphatidylinositol-anchored membrane protein on leukocytes called CD14. CD14 can also be present in the soluble form. CD14 transfers LPS to MD2, which then binds to TLR4 to form the TLR4-MD2 receptor complex and triggers LPS recognition. Soluble MD2 non-covalently associates with TLR4, however it binds to LPS directly even in the absence of TLR4.- Once the LPS-MD2-TLR4 complex forms, the entire complex dimerizes and recruits cytoplasmic adaptor molecules, through the interaction with Toll-interleukin-1 receptor (TIR) domains.

When TLR4 is activated upon its recognition of LPS, it signals through either a MyD88 (myeloid differentiation primary response gene 88)-dependent or a MyD88-independent pathway. The MyD88-dependent pathway induces the activation of NFκB and mitogen-activated protein kinase genes leading to the release of proinflammatory cytokines, whereas the MyD88 independent pathway (also known as the TRIF pathway-Toll-interleukin-1 receptor domain-containing, adaptor-inducing interferon β) activates the Type-1 interferon-inducible genes followed by NFκB production.

The lipid A component of LPS is sufficient to cause endothelial cell injury by promoting the expression of tissue factor and proinflammatory cytokines, leading to apoptosis of these cells.- In a blood stream infection, presence of lipid A can lead to endotoxin shock. In murine TLR4, an 82-amino acids long hypervariable region is responsible for recognition of lipid A. The structure-length and the number of acyl chains are critically important in human TLR4 signaling.

Several gram-negative bacteria have developed ways to modify lipid A structure depending on the environment and host cells leading to greater resistance to host cationic antimicrobial peptides (CAMPs) and altering TLR4 recognition. CAMPs are a group of peptides produced by eukaryotes that are an important component of the innate immune responses against pathogens. Due to their cationic nature, CAMPs disrupt bacterial surface by inserting into the anionic cell wall and phospholipid membrane, thereby killing the pathogen. Studies report that an extremely low concentration of CAMPs is sufficient to modify lipid A. Modifications of lipid A are regulated by a two component system that is an environmental sensor-kinase regulator called PhoP-PhoQ in several gram-negative bacteria including S. Typhimurium. This two component system promotes the resistance of S. Typhimurium to CAMPs and also enables the pathogen to survive within human and murine macrophages. PhoP–PhoQ regulated lipid A modifications involves the deacylation of several fatty acids and also the addition of palmitate, aminoarabinose, and phosphoethanolamine to the lipid A structure. Compared with non-regulated lipid A, PhoP–PhoQ regulated lipid A modifications leads them to be less recognized and stimulatory to the TLR4 complex, a phenomenon that could lead to the persistence of infection., Acylation of lipid A is regulated by three enzymes, PagP, PagL, and LpxO in Salmonella, which catalyze the acylation, deacylation, and hydroxylation of lipid A respectively.- PagP enables the addition of C16:0 fatty acid by transferring the fatty acids from the inner membrane portion of lipid A to the outer membrane region of the molecule. PagL causes deacylation of the lipid A structure and decreases the recognition of lipid A when the pathogen colonizes host cells. Both PagP and PagL modify the recognition of lipid A by the TLR4 complex. Addition of aminoarabinose decreases the negative charge of lipid A, making it more resistant to CAMPs. Similarly, clinical isolates of P. aeruginosa that colonize the airways of cystic fibrosis patients synthesize unique lipid A molecules with an highly modified aminoarabinose and fatty-acid chains has been identified.

LPS induces inflammatory cells to express a number of proinflammatory cytokines including IL-8, IL-6, IL-1β, IL-1, IL-12, and IFNγ;, however, TNFα seems to be of critical importance during endotoxic shock- and causes tissue damage. In some clinical studies and animal models of sepsis, anti-TNF antibodies have shown to help in the treatment of septic shock. Mice lacking the TNF receptor have an attenuated response to endotoxins., During LPS-induced shock, TNFα, in addition to inducing anti-inflammatory cytokines such as IL-10 and IL-4, also triggers the expression of proinflammatory cytokines, IL-1, IL-6, and IL-8 among others. Apart from cytokine induction, TNFα also induces nitric oxide synthase (iNOS) and cyclooxygenase (COX-2) that catalyze the production of nitric oxide (NO) and prostaglandin E2 (PGE2)., Both NO and PGE2 are vasodilators that may cause the reduction in the migration of neutrophils to the site of infection by inhibiting the endothelium–leukocyte binding.- LPS in combination with TNFα induces apoptosis of the endothelium layer in several tissues including intestine, lungs, and thymus. Several strategies to ameliorate endotoxin shock have been tested in both preclinical and in clinical trials. Despite the compelling evidence that LPS is a major factor in the pathophysiology of septic shock, recent trial targeting lipid-A portion of LPS with a drug called eritoran was unable to improve outcome in a large phase 3 clinical trial.

Superantigens

Superantigens (SAg) are one of the most potent toxins produced by bacteria, namely, S. aureus and S. pyogenes. They are non-glycosylated proteins that have a relatively low molecular weight. SAgs produced by S. aureus include staphylococcal enterotoxins SE (A–E) and toxic shock syndrome toxin-1 (TSST-1), while the toxins produced by S. pyogenes include streptococcus pyrogenic exotoxin A and C (SPEA and SPEC), and the streptococcal mitogenic exotoxin Z (SMEZ). These toxins are capable of producing a massive cellular immune response that could lead to a fatal toxic shock. Unlike conventional antigens that are processed by antigen presenting cells and presented to T cells through the MHC-II molecules, SAgs bind directly to the outer leaflet of MHC-II molecules- specific domains of the variable portion of β-chain (Vβ) of the T-cell receptor.,- This allows for bypassing the processing by antigen presenting cells and stimulates most T cells. In addition to binding to MHC-II and the Vβ-chain, it has been recently shown that SAgs also engage a third receptor, CD28, which is a costimulatory molecule on T cells.- SAg bind directly at the homodimer interface of CD28 to cause toxicity by inducting a cytokine storm. Unlike conventional antigens that normally activate <0.01% of T cells, SAgs activate >20% of T cells by binding to the MHC-II and T-cell receptor directly.- This leads to a massive induction of proinflammatory T-helper 1 (Th1) cytokines including tumor necrosis factor (TNF), interferon γ (IFN γ), and interleukin-2 (IL-2).,,, Further details on superantigens can be found in the paper by Reglinski and Sriskandan in this issue of Virulence.

P. aeruginosa and Exotoxin A

In ICUs P. aeruginosa, a gram-negative bacterium is among the top five organisms causing pulmonary, urinary tract, soft-tissue, and bloodstream infections. P. aeruginosa express several virulence factors such as flagella, pili, and LPS that play an important role in their pathogenesis. However, the toxins of P. aeruginosa are some of the most potent factors these organisms express and secrete. Apart from the toxins they secrete, P. aeruginosa also inject one set of toxins directly into host cells through a macromolecular syringe called Type III secretion system.

On the basis of weight, exotoxin A of this organism is the most toxic compound it produces. Exotoxin A is part of an enzyme family called mono-ADP-ribosyltransferase. The toxin affects the protein synthesis in host cells by catalyzing the ADP ribosylation of eukaryotic elongation factor 2, much like the mechanism of diphtheria toxin. It is released by P. aeruginosa as a proenzyme that is toxic to animals and cultured cells but has very low enzymatic activity. This toxin undergoes partial proteolysis by the serine endoprotease called furin, and then enters host cells through receptor mediated endocytosis. Exotoxin A is internalized into clatherin coated vesicles and moves into the endosomes. The LD50 of exotoxin A was shown to be 0.2 µg in a 20 g mouse by the intraperitoneal route of administration. Between 80% and 90% of all clinical isolates of P. aeruginosa have demonstrated exotoxin A production in vitro., It is presumed to escape into the cytosol through a translocation event. Studies have demonstrated that domain la of exotoxin A is the primary region of the toxin involved in cellular binding. In vivo studies with mice injected with purified exotoxin A lacking the la domain showed attenuation of toxicity compared with mice injected with native exotoxin A. Administration of IVIG that are enriched in neutralizing antibodies to exotoxin A, however, led to no clinical improvement in patients with established Pseudomonas bacteremia.

Bacillus anthracis and Toxins

Bacillus anthracis (B. anthracis), the causative agent of the disease anthrax is a gram-positive bacteria that is able to survive in the environment in the spore form. The disease is generally contracted mainly through three routes, namely, cutaneous, gastrointestinal, and the inhalation routes.- In spite of appropriate therapy, all the three routes of infection can lead to fatal disease as a result of sepsis and shock-like symptoms. The inhalation route generally leads to the highest fatality and is a serious bioterrorism threat today. The toxins of B. anthracis play a vital role in the pathogenesis of the disease. The toxins are made up of three secreted proteins working in binary combinations, namely protective antigen (PA), lethal factor (LF), and edema factor (EF)., The PA combines with EF to form the edema toxin (ET) and with LF to form the lethal toxin (LT).

LF, a 90-kD zinc protease consisting of 4-folding domains, is known to recognize six out of the seven mitogen-activated protein kinases, 1–4, 6, and 7. These are bound by domains II and III and cleaved at the N-terminus by domain IV.- The cleavage, results in the possible disruption of downstream signaling, mainly the inactivation of ERK1/2 (extracellular-signal-regulated-kinases), p38, and SAPK (stress-activated protein kinases)/JNK (Jun N-terminal kinases) pathways that are important for the activation of immune responses. LT induces apoptosis in different cell types including Human umbilical vein endothelial cells by disrupting the ERK, p38, and JNK/SAPK pathways, with the ERK pathway being of upmost importance. LT affects the translocation of tight junction proteins and alters the cytoskeleton reorganization by reducing levels of F-actin and blocking localization of vascular endothelial cadherin. In human endothelial cells that are TNF-induced, LT amplifies expression of vascular cell adhesion molecule-1 that results in vasculitis and barrier disruption of cells.- Lymphocytic processes like T-cells activation, proliferation, and cytokine production are shown to be suppressed by both LT and ET.- The mechanism of T-cell suppression is the direct effect of LT cleaving MAPKKs, whereas ET suppresses T-cell processes by elevating the level of cAMP activity. Both LT and ET prevent chemotaxis of T cells and macrophages by reducing the activation of MAPK to different chemokines. While EF plays a greater role in disrupting the neutrophil migration and cytokine production, LF is directly lethal to macrophages and prevents dendritic cell maturation., These deleterious effects of LF and EF on the immune cells impair phagocytosis of inhaled spores and vegetative forms of B. anthracis, allowing them to be transported to lymph nodes. A hemorrhagic and septic medistinitis develops accompanied by high-grade bacteremia, septic shock, and death. The toxic nature of both LT and ET results in bacterial dissemination to remote organs resulting in widespread tissue necrosis and death. A number of therapeutic strategies have been developed in an attempt to block the effects of anthrax toxins and improve outcomes.

Conclusion

Treatment of sepsis still remains a serious concern and challenge in hospitals. Bacterial toxins from both gram-positive and gram-negative bacteria allow the pathogen to modulate host defenses through their interaction with cells enabling the bacteria to escape the innate immune system to remote organs. The type of toxin plays a major role in the outcome of disease. Over the last decade our understanding of the mechanisms by which these toxins modulate host defense has tremendously improved. This could enable a more efficient way of targeting the toxins and better clinical outcomes.