Introduction
Mechanisms for blood coagulation in the presence of bacterial infections are typically attributed to immune responses and inflammation, yet until now, the correlation has been limited to the observation that clotting factors are activated in the presence of certain bacterial components. Infections ranging from Staphylococcus aureus to Escherichia coli are well known to produce clotting factors, and disseminated intravascular coagulation is commonly associated with sepsis and the onset of multiple organ failure1. However, the molecular mechanisms linking bacterial concentration to initiation of the coagulation cascade are complex and not well understood2. In their recent work, Kastrup et al.3 used a microfluidic in vitro model system to demonstrate that clustering of bacteria can raise the local concentration of coagulation factors and thereby initiate coagulation by exceeding a local threshold concentration independent of the total bacterial load.
Blood coagulation is an important and particularly devastating consequence of bacterial infections that lead to sepsis and, in particular, septic shock. Over 33,000 deaths per year are associated with septic shock, and mortality rates approach 50%; intensive efforts for new drug targets such as interruption of the thrombin receptor pathway are underway4, but basic mechanisms are poorly understood. Less common but equally complex clinical situations have been encountered with victims of bioterrorism-related attacks with Bacillus anthracis and with food-borne infections involving Bacillus cereus in people with compromised immune systems. The challenge in understanding the relationship between bacterial infections and coagulation has been related in part to the complexity of the coagulation cascade and the large number of potential pathways and mechanisms involved in its activation. A second challenge has been related to the absence of appropriate model systems for coagulation—simple in vitro systems are needed in which coagulation can be monitored in a controlled environment in the presence of a variety of bacterial species.
The Ismagilov group has pioneered the use of microfluidic systems to study coagulation, and in particular the initiation and propagation of coagulation5. These systems, comprising transparent polydimethylsiloxane micromolded polymer constructs, have formed the basis of a diverse array of lab-on-a-chip systems and have been used for topologically complex microfluidic networks6. Kastrup et al.3 have constructed microfluidic devices that mimic vascular networks to investigate the role of shear stress in coagulation and to study the influence of the spatial distribution of the clotting activator tissue factor on coagulation. This latter approach has now been applied to probe the role of the spatial distribution of bacteria in the initiation of coagulation; B. cereus localized on the surface of a microfluidic channel initiated coagulation at much lower total bacterial levels than when dispersed evenly in solution, as shown in Figure 1. The effect is dependent on the particular species; similar experiments with E. coli exhibited no such spatially dependent enhancement. The effect, termed 'quorum acting', differs from the well-known phenomenon of quorum sensing7 in that the latter is defined by alterations in the cellular phenotype itself, whereas this new effect is characterized by alterations to the external environment induced by spatial localization of the cells.
Figure 1: Coagulation induced by the presence of bacteria is governed by spatial localization, or clustering, rather than by the total concentration of a bacterial species in the blood.
When the bacteria are uniformly distributed throughout the bloodstream, as in the left image, coagulation is not induced. However, when bacterial clustering occurs, as in the right image, local threshold concentrations of clotting factors are exceeded, resulting in coagulation.
Katie Ris-Vicari
Full size image (40 KB)Microfluidic devices enable the study of various phenomena in the presence of precisely controlled flow conditions, and Kastrup et al.3 used this capability to demonstrate that clusters of B. cereus still initiate coagulation in the presence of flowing whole blood. Experiments also confirmed that specific coagulation factors must be present for the quorum acting process, while others are not essential, and therefore, a coagulation network in which quorum acting bypasses some of the initiation points can be mapped. Numerous Bacillus species activate coagulation in vitro; one interesting finding is that B. anthracis requires a specific metalloprotease in order to trigger coagulation. Mouse models injected with B. anthracis also exhibited a strong correlation between the bacterial clustering and coagulation, but differences between mouse and human infection and pathophysiology complicate the picture considerably.
The ability of microfluidic models to simulate coagulation-related phenomena is an exciting advance with many potential applications in basic research and in the development of new therapeutic approaches. Septic shock and related conditions have long been associated with coagulation, and yet a complete picture of the mechanisms activating and controlling this correlation has not emerged. With new insights into quorum acting phenomena and the ways in which bacteria communicate to influence not only their own behavior but also that of the local environment, a deeper understanding of infection-induced coagulation is possible. Many questions remain, however, as effects such as coagulation in the presence of B. anthracis seen in vitro and in animal models may not be consistent with recent human clinical experience8. Future work along the lines of Kastrup et al.3 is likely to be very instructive in addressing key questions, such as the relationship between bacterial infections and cardiovascular disease, the evolution of cooperative effects within host responses and within bacterial populations, and approaches towards reducing the incidence of coagulation related to microbial contamination and biofilms in vascular access management.
