The role of serum proteins in sepsis

Keywords: :

• host-pathogen interactions
• LPS
• LTA
• sepsis
• Serum Protein

Sepsis, defined as infection with resulting systemic inflammation, afflicts millions of patients worldwide and may be increasing in incidence.¹ Severe sepsis leads to hypotension, impaired systemic perfusion, organ dysfunction and has a mortality rate of approximately 25%. One of the fundamental tenets of our current understanding of sepsis is that an overly exuberant host inflammatory response to infection, rather than infection itself, is largely responsible for the organ dysfunction and death that results. Clinical outcomes in sepsis have been steadily improved by focusing on the early administration of organized supportive care and appropriate antibiotic therapy.¹ The clinical strides that have been made by adopting these supportive strategies are in marked contrast to both clinical and basic science work that has focused on modulating the host response itself. We have no effective therapies targeting the systemic inflammation induced in sepsis, a situation made all the more apparent by the results of two recent trials involving drotrecogin α [recombinant activated protein C, (Eli Lilly, Indianapolis, IN USA)] and eritoran tetrasodium [an anti-Toll-like receptor (TLR)-4 compound, (Eisai, Tokyo, Japan)]. The former was the only drug specifically indicated for severe sepsis and has now been withdrawn from the market, the latter represented a promising new class of drugs targeting a cellular pathway believed to be crucial to the systemic inflammatory response induced by sepsis. Neither showed any benefit in large clinical trials.²

Why have we failed to identify sepsis-specific therapies with clinical benefit? Leaving aside the problems associated with the large-scale clinical trials encompassing the wide spectrum of disease processes that can cause sepsis (e.g., bowel perforation, pneumonia, meningitis, bacteremia) it seems likely that our understanding of the process of systemic inflammation in sepsis is still rudimentary.³ For one, there is a tacit assumption in many studies that inflammation in sepsis is a “final common pathway” regardless of the inciting stimulus, and that results gleaned from a model in which sepsis is induced, for example, by lipopolysaccaride (LPS), a component of Gram-negative bacterial cell walls, will be applicable to sepsis induced by other agents. This presumed common pathway of inflammation can be influenced at many levels, but a large body of work has focused on membrane-bound receptors and their downstream signaling pathways as the key modulators of systemic inflammation in sepsis. Work on TLR4 resulted in the 2011 Nobel Prize in Physiology or Medicine and yet, as mentioned above, TLR4 and other membrane-bound receptors have proven elusive as therapeutic targets in sepsis.⁴ The description and manipulation of these receptor pathways and their downstream cellular effects remain important areas of inquiry. Still, before infectious agents can activate these pathways, they must interact with other inflammation modulating systems that can alter the systemic inflammatory response, such as circulating serum proteins.

Triantafilou et al.’s study in this issue of Virulence⁵ addresses two of the implicit assumptions of much sepsis research: namely, that intracellular signaling is the dominant regulator of the inflammatory response, and that the inflammatory response has a similar mechanism regardless of the inciting stimulus. The investigators sought to examine the role of plasma proteins in modulating sepsis-induced inflammation. This is an earlier phase of the host response to infection than those initiated by cell surface receptors, and may be a crucial step in regulating the activation of such inflammatory pathways. Triantafilou et al. are by no means the first to investigate the role of serum proteins in sepsis-related inflammation, but this stage of the host response has received relatively little attention. The investigators cast a wide net seeking to identify plasma proteins that bind bacterial moieties, to measure the downstream effects of this plasma protein binding, and to correlate the levels of these proteins with survival in the serum of septic patients. They also distinguish between the effect of plasma proteins on the inflammatory responses initiated by LPS and lipoteichoic acid (LTA), which are cell wall components of Gram-negative and Gram-positive bacteria respectively.

There is no clear-cut hypothesis in this study regarding the role of serum proteins in regulating sepsis-related inflammation, and in that sense the work represents a bit of a fishing expedition. Nonetheless, there are a plethora of interesting findings that generate questions for future study. First, the authors identify a number of serum proteins that bind bacterial cell wall products, including hemoglobin, apolipoprotein, LDL, albumin, transferrin and holotransferrin. Many of these have been previously shown to bind LPS, but their LTA binding has not been previously described. Although these proteins selectively bound LPS and LTA, their effects on the subsequent LPS/LTA induced inflammatory response was not uniform, suggesting protein-specific immunomodulation. The authors show that another serum protein, lipopolysaccharide-binding protein (LBP), has essentially opposite effects on tumor necrosis factor-α (TNF-α) production induced by LPS (in which LBP binding enhances TNF-α production) and by LTA (in which LBP binding inhibits TNF-α production), further demonstrating the specificity of the response. Finally, survivors of sepsis were shown to have higher levels of LPS/LTA-binding plasma proteins than matched septic patients who went on to die.

This study opens the doors onto a complex interaction between pathogens, serum proteins, and the downstream inflammatory cascade that intuitively seems important in the understanding and regulation of the septic response. It raises more questions than it answers—how do the various proteins interact to fine-tune the inflammatory response; what are the biochemical mechanisms whereby serum proteins bind and present pathogenic moieties; is the association between survival and increased levels of plasma proteins in septic patients causal, or merely an association with better pre-morbid health? Only with a more complete understanding of the nuances of the septic response can we understand and ultimately favorably manipulate it. The observations made and questions raised in this paper push us a step closer to developing that understanding.