Carboxypeptidase B2 Is Protective in a Mouse Model of Shiga Toxin-Induced Hemolytic Uremic Syndrome

Hemolytic uremic syndrome (HUS) is characterized by a triad of microangiopathic hemolytic anemia, thrombocytopenia, and acute renal failure. The most common cause of HUS is Shiga toxin (STX)-producing E. coli, and eculizumab, a monoclonal antibody against complement C5, has shown clinical efficacy in some patients. Carboxypeptidase B2 (CPB2) is a metalloprotease activated by the thrombin/thrombomodulin complex that inactivates a number of inflammatory mediators, including complement C3a, and C5a by removing their C-terminal arginine. We hypothesized that in a murine model of STX-induced HUS, Cpb2^-/- mice would have exacerbated disease compared to wild type (WT) mice due to excessive C3a and/or C5a in the absence of CPB2. A mouse model of STX-induced HUS was established by giving STX and LPS toxins intraperitoneally. Cpb2^-/- mice had worse survival than WT (37% survival vs. 87% at 48h, p=0.0156). At 48h, severe thrombocytopenia developed in both WT and Cpb2^-/- mice (WT: 0.096x10⁶/μL; Cpb2^-/-: 0.054x10⁶/μL) compared to controls (1.2x10⁶/μL; p>0.0001 vs. either WT or Cpb2^-/-), with Cpb2^-/- mice showing worse thrombocytopenia. Renal insufficiency was worse in Cpb2^-/- mice than WT mice (BUN at 48h: 85 mg/dL vs. 37 mg/dL, p=0.0074; creatinine: 1.33 mg/dL vs. 0.23 mg/dL; p=0.0112, for Cpb2^-/- and WT mice respectively, compared with normal baseline BUN and creatinine of 19 mg/dL and 0.1 mg/dL). Cpb2^-/- mice developed worse anemia than WT (hemoglobin 9.8 g/dL vs. 12.4 g/dL, p=0.001 in Cpb2^-/- vs. WT mice respectively). At 48h, liver function was worse in Cpb2^-/- mice than WT mice, while plasma LDH was increased in Cpb2^-/- mice more than WT mice. Using a standardized health score, the Cpb2^-/- mice were worse than WT mice at all time points. Thus this model recapitulates STX-induced HUS with the Cpb2^-/- mice having worse disease than WT. If the animals were treated with STX alone, there were no deaths in either genotype at 48h and only 37.5% mortality in Cpb2^-/- mice by 60h compared with no deaths in WT mice. BUN, creatinine, liver enzymes and LDH were increased in both genotypes treated with STX alone compared to untreated mice, but there was no significant difference between the genotypes. Treatment with LPS alone caused thrombocytopenia in both WT and Cpb2^-/- mice and LDH, BUN and creatinine levels were higher in Cpb2^-/- mice than in WT mice, but there was no death at 48h and no drop in hemoglobin. Thus while either STX alone or LPS alone caused pathological conditions in the mice, the typical triad of HUS was only present when STX and LPS were given in combination. The Cpb2^-/- mice had worse disease than WT mice consistent with our hypothesis on the role of CPB2 in inactivating C3a and/or C5a in STX-induced HUS. The potential efficacy of C3a and/or C5a blockade and anti-thrombotic agents will be tested in this model.