Cirrhosis is associated with low CD4(+) T cell counts in the absence of HIV infection. Discordance between low absolute CD4(+) T cell counts and normal CD4(+) T cell percentages may be attributable to portal hypertension and splenic sequestration. Our findings have significant implications for the use and interpretation of absolute CD4(+) T cell counts in HIV-infected patients with advanced liver disease.

Many bacteria that spread in the skin produce enzymes that digest extracellular matrix components. Borrelia burgdorferi spreads from a skin inoculation site to form the characteristic erythema migrans skin lesion. It was determined that B. burgdorferi does not produce collagenase, elastase, hyaluronidase, or other enzymes that digest extracellular matrix components. However, B. burgdorferi bound human plasmin, plasminogen (Pgn), and urokinase-type plasminogen activator (uPA). When spirochetes were sequentially incubated with Pgn and uPA, bioactive plasmin was generated on the surface of B. burgdorferi. B. burgdorferi did not produce an endogenous Pgn activator. Fluorochrome-conjugated uPA and Pgn colocalized to the terminus of the spirochete. In a mouse model, uPA-treated B. burgdorferi were more infectious than control spirochetes. Binding of host uPA and Pgn to form a bioactive extracellular matrix protease on B. burgdorferi represents a mechanism that could facilitate dissemination and localization of spirochetes to sites of vascular injury.

There is no effective medical treatment for primary sclerosing cholangitis (PSC), a chronic cholestatic liver disease that usually progresses to cirrhosis and liver failure. The aim of this study was to determine the safety and efficacy of etanercept, an inhibitor of tumor necrosis factor, in the treatment of PSC. Ten patients with clinically active PSC were studied. All had elevated serum alkaline phosphatase levels, cholangiograms that were diagnostic of PSC, and liver histology consistent with PSC. Five patients had elevated serum bilirubin levels, five had pruritus, eight had failed to respond to ursodiol and/or methotrexate, and six had rapidly recurring dominant bile duct strictures. Patients were to receive etanercept, 25 mg subcutaneously twice weekly, for 6 months if there were no side effects and for 1 year if there was evidence of efficacy after 6 months. Biochemical tests of liver function did not improve in any patient. Mean serum bilirubin levels increased significantly, from 2.0 to 3.6 mg/dl (P = 0.026). Two of the five patients with pruritus had resolution of pruritus during treatment with etanercept, recurrence when etanercept was stopped, and resolution when it was restarted, although there was little change in liver enzymes or bilirubin levels. There was no decrease in the rate of stricture formation and there were no side effects. Etanercept, at the dosage used, was well tolerated but not effective in the treatment of PSC. It may be helpful in treating pruritus due to cholestasis.

Borrelia burgdorferi binds human urokinase plasminogen activator (uPA), which cleaves plasminogen (Pgn) to plasmin. The ability of other Borrelia species to bind uPA, Pgn, or both was investigated. Borrelia coriacae, Borrelia garinii, Borrelia parkeri, Borrelia anserina, and Borrelia turicatae were compared with infectious and noninfectious B. burgdorferi isolates. All Borrelia species lacked endogenous proteases capable of digesting casein, but all species bound human uPA and Pgn, generating Pgn-dependent proteolytic activity. There were no significant differences in the amount of plasmin, Pgn, or uPA bound per spirochete of the different species. On unfixed borreliae, fluorochrome-conjugated uPA bound to all species. Early binding was at the terminus of B. burgdorferi, whereas diffuse binding was observed on B. coriacae, B. garinii, B. parkeri, and B. turicatae. These studies demonstrate that binding of human uPA and Pgn to borreliae occurs on multiple species with apparent differences in surface distribution.