Pathogenesis of sepsis: new concepts and implications for future treatment

Gram negative sepsis

Most cases of Gram negative sepsis are caused by Enterobacteriaceae such as E coli and Klebsiella species. Pseudomonas aeruginosa is the third commonest cause. Gram negative infections usually occur in the lung, abdomen, bloodstream, or urinary tract.

Lipopolysaccharide is an important component of the outer membrane of Gram negative bacteria and has a pivotal role in inducing Gram negative sepsis.6 Lipopolysaccharide binding protein in host cells binds to lipopolysaccharide in the bacteria and transfers it to CD14.7 CD14 is a protein anchored in the outer leaflet of the plasma membrane, although it also exists as a soluble plasma protein that attaches lipopolysaccharide to CD14-negative cells, such as endothelial cells. CD14 is located in the extracellular space and therefore cannot induce cellular activation without a transmembrane signal transducing coreceptor.

A series of remarkable investigations have recently led to the identification of Toll-like receptor 4 (TLR4) as the coreceptor for lipopolysaccharide. Toll receptors were first discovered in Drosophila, where they were found to have a role in the defence of flies against fungi and Gram positive bacteria.8 Toll-like receptors were then identified in other species. Human Toll-like receptors, like their homologues in insects and other mammalian species, are type I transmembrane proteins with an extracellular leucine rich repeat domain and an intracellular domain homologous to the interleukin 1 receptor. Genetic studies in mice showed that mutations in the Tlr4 gene were linked to resistance to lipopolysaccharide, providing evidence that TLR4 was an essential component of the lipopolysaccharide receptor complex.9 MD-2, a secreted protein associated with the extracellular domain of TLR4, has also recently been shown to have an important role in responsiveness to lipopolysaccharide.10

Gram positive sepsis

Staphylococci (mainly Staph aureus and coagulase-negative staphylococci) and streptococci (Strep pyogenes, viridans streptococci, Strep pneumoniae) are the commonest causes of Gram positive sepsis. They are usually responsible for infections of skin and soft tissue, infections associated with intravascular devices, primary bloodstream infections, or respiratory infections. Gram positive organisms can cause sepsis by at least two mechanisms: by producing exotoxins that act as superantigens (see definition below) and by components of their cell walls stimulating immune cells.11

Superantigens are molecules that bind to MHC class II molecules of antigen presenting cells and to Vβ chains of T cell receptors. In doing so, they activate large numbers of T cells to produce massive amounts of proinflammatory cytokines. Staphylococcal enterotoxins, toxic shock syndrome toxin-1, and streptococcal pyrogenic exotoxins are examples of bacterial superantigens.

Gram positive bacteria without exotoxins can also induce shock, probably by stimulating innate immune responses through similar mechanisms to those in Gram negative sepsis. Indeed, Toll-like receptor 2 (TLR2) has been shown to mediate cellular responses to heat killed Gram positive bacteria and their cell wall structures (peptidoglycan, lipoproteins, lipoteichoic acid, and phenol soluble modulin).12

Toll-like receptors

Of the 10 human Toll-like receptors identified so far, seven interact with microbial motifs (table 1).4 For example, TLR2 binds components of the cell wall of Gram positive bacteria as well as ligands derived from other pathogens, TLR5 is the receptor for bacterial flagellin,14 and TLR9 is required for cellular activation by unmethylated CpG motifs of bacterial DNA.15 Cooperation between Toll-like receptors is necessary to respond to certain pathogens, such as Gram positive bacteria and yeast (zymosan).16 Several signal transducing pathways are activated after microbial ligands bind to Toll-like receptors (figure).17

The fact that different microbial products bind to different Toll-like receptors, the existence of receptor specific signalling pathways, and the idea of differential expression of Toll-like receptors by tissues and organs strongly suggest that the innate immune system is tailored in a pathogen and tissue specific manner. Expression of immune genes and host responses to infections will vary depending on the structural and biochemical composition of the invading pathogen. If confirmed, these hypotheses point to the need to develop pathogen specific approaches to treatment.

Other soluble and membrane associated proteins have recently been shown to be involved in recognising bacteria or microbial products. These include peptidoglycan recognition proteins and triggering receptor expressed on myeloid cells (TREM-1). Additional information on these proteins is available on bmj.com

Microbial drugs and pattern recognition molecules

Designing new drugs to neutralise microbial products or block their interaction with specific receptor on immune cells is an attractive concept. Potential targets include lipolysaccharide binding protein, CD14, TLR4, and MD-2 for Gram negative sepsis, and CD14, TLR2, and TLR6 for Gram positive sepsis. Monoclonal antibodies against CD14 are being evaluated in phase II studies. Several intracellular signalling molecules, such as MyD88 and the mitogen-activated protein kinase are other possible therapeutic targets. However, inactivating molecules that are pivotal to innate immunity can be harmful, as shown by the increased sensitivity to bacterial sepsis in mice with mutations of the Tlr4 gene.23 Careful selection of patients with severe infections associated with a high probability of death will therefore be essential.

Macrophage migration inhibitory factor

Macrophage migration inhibitory factor is a cytokine that has recently been shown to be important in innate immunity and sepsis.24 It is constitutively expressed in large amounts by immune, endocrine, and epithelial cells and is rapidly released after exposure to microbial products and pro-inflammatory cytokines. Macrophage migration inhibitory factor regulates innate immune responses to endotoxin and Gram negative bacteria by modulating the expression of TLR4, enabling macrophages and other cells at the front line of defences to respond quickly.25 High levels of macrophage migration inhibitory factor have been detected in patients with inflammatory and infectious diseases, including severe sepsis and septic shock.26

Immunoneutralisation of macrophage migration inhibitory factor or deletion of the Mif gene protects mice against lethal endotoxaemia, Gram positive toxic shock syndromes, and experimental bacterial peritonitis. Conversely, mice injected with macrophage migration inhibitory factor together with live bacteria or microbial toxins have increased death rates.26^–28 This factor thus has the potential to endanger life when expressed in excess during sepsis. Development of drugs to block the production of macrophage migration inhibitory factor or inhibit its function may help treat severe sepsis and other inflammatory diseases.

High mobility group protein 1

High mobility group protein 1, a protein previously known as DNA binding protein regulating gene transcription and stabilising nucleosome formation, has recently been described as a “late” mediator of inflammation and sepsis.29 Patients with septic or haemorrhagic shock have raised serum concentrations of high mobility group protein 1, and concentrations are associated with patient's outcome. Use of polyclonal antibodies to block high mobility group protein 1 in mice protects them against lipopolysaccharide induced acute lung injury and lethal endotoxaemia.30

Genetic studies of susceptibility to sepsis

Several gene polymorphisms have been associated with increased susceptibility to sepsis (see bmj.com for more information). Testing for polymorphisms of important genes may help to identify people who are at increased risk of sepsis when exposed to virulent bacteria and who may benefit from targeted immunomodulatory therapies.