Key points

Introduction

Hemolytic uremic syndrome (HUS) is characterized by the clinical triad of microangiopathic hemolytic anemia, thrombocytopenia, and acute renal injury. It is the leading single cause of pediatric acute kidney injury. HUS is a type of thrombotic microangiopathy (TMA). Typically, it occurs after a gastrointestinal infection with a Shiga toxin (Stx)-producing pathogen. This diarrhea- or Stx-associated HUS (D+HUS or Stx-HUS) accounts for 90% of cases. Currently, there are no direct treatments and limited prevention strategies. Most patients recover from the acute illness, but there is a 1% to 4% mortality and one-third of patients are left with long-term medical problems. The major challenge facing clinicians and researchers is better understanding the disease pathogenesis in an attempt to develop new therapeutic strategies. With this in mind, attention has turned to advances in the understanding of a related condition, complement-mediated HUS. This atypical HUS subtype is now successfully treated with the C5 monoclonal antibody, eculizumab. Its success prompted clinicians to ask if it could also be used to treat Stx HUS and for researchers to examine the role of complement in the pathogenesis of this disease.

Introduction

Hemolytic uremic syndrome (HUS) is characterized by the clinical triad of microangiopathic hemolytic anemia, thrombocytopenia, and acute renal injury. It is the leading single cause of pediatric acute kidney injury. HUS is a type of thrombotic microangiopathy (TMA). Typically, it occurs after a gastrointestinal infection with a Shiga toxin (Stx)-producing pathogen. This diarrhea- or Stx-associated HUS (D+HUS or Stx-HUS) accounts for 90% of cases. Currently, there are no direct treatments and limited prevention strategies. Most patients recover from the acute illness, but there is a 1% to 4% mortality and one-third of patients are left with long-term medical problems. The major challenge facing clinicians and researchers is better understanding the disease pathogenesis in an attempt to develop new therapeutic strategies. With this in mind, attention has turned to advances in the understanding of a related condition, complement-mediated HUS. This atypical HUS subtype is now successfully treated with the C5 monoclonal antibody, eculizumab. Its success prompted clinicians to ask if it could also be used to treat Stx HUS and for researchers to examine the role of complement in the pathogenesis of this disease.

Epidemiology

Worldwide, Stx-producing Escherichia coli causes 2.8 million illnesses annually. In the United States, it causes more than 265,000 illnesses per year with 3600 hospitalizations and 30 deaths. The global incidence of Stx HUS is 0.2 to 4.28 people per 100,000 population. Seasonal variation occurs with more cases in summer. Stx HUS is more common in children, particularly those under the age of 5. E coli O157, the most common Stx-producing E coli , is responsible for 724 hospitalizations and 11 deaths per year in this age group in the United States.

Clinical course

Patients infected with a Stx-producing pathogen usually develop a profuse diarrheal illness 2 to 5 days later ( Fig. 1 ). Signs and symptoms of HUS occur after the diarrhea, but only in 10% to 15% of people infected with E coli O157. The risk of developing Stx HUS is influenced by the organisms serotype, as illustrated in the E coli O104 outbreak, where 22% of infected patients developed HUS. Host factors also play a role ( Box 1 ).

The timeline of Stx HUS. Stx E coli infection is acquired by the fecal oral route from contaminated food or water, ruminant animals (the natural reservoir), or another infected individual. After a 2- to 5-day incubation period, profuse diarrhea develops that can be bloody. Vomiting and abdominal pain can also occur. Three to 14 days later, features of HUS arise. Symptoms can be vague with pallor, lethargy, reduced urine output, swelling and easy bruising. The clinical triad of microangiopathic hemolytic anemia (MAHA), thrombocytopenia, and acute kidney injury (AKI) characterize HUS. Disease duration can vary with some patients recovering over several weeks. Others can take longer. Most will make a full recovery from HUS, but 30% will have long-standing renal or neurologic impairment.

Pathology

The pathologic feature of HUS is TMA. This term describes characteristic endothelial cell damage ( Fig. 2 A), which leads to complete or partial vessel obstruction, resulting in fragmented red blood cells. Characteristic laboratory findings include consumptive thrombocytopenia and microangiopathic hemolytic anemia. Organ ischemia occurs with symptoms determined by the vascular bed affected. In glomerular TMA, there will be evidence of renal impairment, but if TMA affects the brain neurologic symptoms develop.

Stx pathogenesis. ( A ) The pathologic feature of HUS is TMA, which describes features of endothelial cell damage, including ( 1 ) endotheliosis, ( 2 ) increased vessel wall thickness, ( 3 ) detachment of cells from the basement membrane, with ( 4 ) accumulation of protein and cellular debris in the subendothelial space. A procoagulant, activated endothelial phenotype develops with expression of von Willebrand factor attracting platelets and inducing formation of intraluminal microthrombi ( 5 ). P-selectin expression attracts neutrophils to the site. Stx carried by blood cells causes TMA by binding to Gb3 receptors on the endothelial cells. ( B ) Stx binds to Gb3 cell surface receptors. The Stx-receptor complex is internalized by endocytosis and undergoes retrograde transport to the endoplasmic reticulum via the Golgi stack. In the endoplasmic reticulum, Stx is split into 2: the α and βsubunits. The αsubunit mimics a misfolded protein and, using the hosts own machinery, is translocated to the cytosol. Here, the enzymatically active A1 fragment (27.5 kDa) causes depurination of adenosine at a highly conserved loop of 28S ribosomal RNA of the 60S ribosomal subunit, stopping protein synthesis and activating the ribosomal stress response.

Pathogenesis

Stx-producing bacteria infect the large intestine causing colitis. The toxin produced crosses the gastrointestinal epithelium and enters the circulation possibly via globotetraosylceramide (Gb4) cellular receptors, but monocytes and neutrophils may also be involved. Stx targets cells that express globotriosylceramide (Gb3) receptors. Free Stx has never been detected in patient blood. Erythrocytes, neutrophils, and platelets have been implicated in Stx carriage, but this may not be via Gb3 receptors. Recently, it was reported that neutrophils bind Stx via toll-like receptor 4 (TLR4). It is hypothesized that TLR4 has a lower affinity for Stx than Gb3, so when Stx arrives at a Gb3 expressing organ, such as the kidney, it preferentially detaches from the circulating cells and binds to the Gb3-expressing cells.

The complement system in Shiga toxin-associated hemolytic uremic syndrome pathogenesis

In Vitro Evidence

Shiga toxin inhibits protective complement regulators

Stx can bind to and inhibit the complement regulatory function of complement factor H (CFH) and CFH-related protein 1, which may make cells vulnerable to complement-mediated damage ( Fig. 3 ). Stx targets the membrane-binding portions of CFH, so although it can still regulate complement in fluid phase, it cannot on the cell surface. This phenomena is the same as that produced by the CFH loss-of-function mutations associated with atypical, complement-mediated HUS. Furthermore, glomerular endothelial cell expression of the membrane-bound complement regulator CD59 and the complement and coagulation regulator thrombomodulin (TM) were reduced by Stx. TM loss may be due to increased cellular shedding, which could explain the high serum TM levels found in Stx HUS patients.

The role of complement in Stx HUS pathogenesis. Stx can bind to and inhibit protective CFH ( 1 ) and reduces expression of CD59 and TM ( 2 ). It can also directly activate complement via the alternative pathway via interactions with C3 ( 3 ). Endothelial cell P-selectin also hydrolyses C3 to C3a and C3b ( 4 ). The anaphylatoxin C3a enhances P-selectin expression and reduces TM ( 5 ). C3 deposits occur on the endothelium leading to activation of the alternative pathway before thrombin is seen ( 6 ). The alternative pathway is illustrated ( 7 ). C3b combines with the Bb fragment of factor B to form the C3 convertase, C3bBb. This is mediated by factor D. The C3 convertase causes an amplification loop hydrolyzing C3 to produce more C3b, combining with the C3 convertase to form a C5 convertase (C3b2Bb); this hydrolyses C5 to C5a (anaphylatoxin) and C5b, which combines with C6-C9 to form the MAC. Cell lysis occurs, exposing the subendothelial matrix. Inflammation caused by the anaphylatoxins activates endothelial cells, leading to a procoagulant phenotype with von Willebrand factor expression. Activation of the complement system is known to active the coagulation system also, leading to the development of TMA. RBC, red blood cell.

Shiga toxin activates complement

Stx can directly activate complement predominately via the alternative pathway (see Fig. 3 ). Endothelial cells pretreated with Stx and exposed to human serum expressed P-selectin can bind and activate C3 (see Fig. 3 ). The formation of C3a further enhances P-selectin expression, decreases TM expression, and thrombosis ensues. In vitro, C3 complement deposits occurred before thrombin deposition, suggesting that complement deposition could trigger thrombosis. Inhibiting complement activation abolished these effects.

Mouse models

Murine models of Stx HUS are used to study the disease but do not precisely replicate human pathology because mice lack glomerular Gb3 receptors. Instead, they develop renal tubular damage. Heterozygous CFH-deficient mice and C5-deficient mice developed the same tubular damage as wild-type mice, suggesting complement does not play a role in this murine model. Mice treated with higher doses of Stx and lipopolysaccharide (LPS) do develop some intraglomerular platelet clumps and endothelial swelling; this is the best published Stx HUS model to date. In this model, mice have reduced glomerular TM expression with C3 and C9 glomerular deposits, suggesting complement activation. Mice lacking the TM complement regulatory (lectin) domain had a more severe clinical phenotype when challenged with Stx and LPS. Similarly challenged factor B knockout mice showed no C3 deposits and less thrombocytopenia, and renal function was maintained. In this murine model of Stx HUS, alternative pathway blockade through genetic knockout was protective. Further studies suggested a role for complement activation specifically in the specialized glomerular epithelial cell, the podocyte, highlighting that Stx has effects on other cell types too. Further work is needed to confirm these findings in patients.