Severe Mucha–Habermann‐Like Ulceronecrotic Skin Disease in T‐Cell Acute Lymphoblastic Leukemia Responsive to Basiliximab and Stem Cell Transplant

A 5‐year‐old girl with T‐cell acute lymphoblastic leukemia (T‐ALL) developed a progressive eruption of crusted papules and ulcerative plaques involving 80% of her body surface area with histopathology consistent with febrile ulceronecrotic Mucha–Habermann disease (FUMHD), although multiple specimens also contained clonal leukemic cells. Her skin disease was refractory to many classic treatments for FUMHD, including methotrexate, and became so severe that concern about superinfection prevented intensification of chemotherapy for her malignancy. The addition of basiliximab promoted gradual improvement of the skin, allowing for chemotherapy intensification and subsequent bone marrow transplantation, after which the eruption resolved completely. This report describes a severe case of FUMHD‐like eruption associated with clonal leukemic cells that improved with basiliximab, suggesting anti‐CD25 therapy as a novel treatment for ulceronecrotic skin disease in the setting of high interleukin‐2 levels.     

Platelets inhibit fibrinolysis in vitro by both plasminogen activator inhibitor-1-dependent and -independent mechanisms

Platelet-rich thrombi are resistant to lysis by tissue-type plasminogen activator (t-PA). Although platelet alpha-granules contain plasminogen activator inhibitor-1 (PAI-1), a fast-acting inhibitor of t-PA, the contribution of PAI-1 to the antifibrinolytic effect of platelets has remained a subject of controversy. We recently reported a patient with a homozygous mutation within the PAI-1 gene that results in complete loss of PAI-1 expression. Platelets from this individual constitute a unique reagent with which to probe the role of platelet PAI-1 in the regulation of fibrinolysis. The effects of PAI-1-deficient platelets were compared with those of normal platelets in an in vitro clot lysis assay. Although the incorporation of PAI-1-deficient platelets into clots resulted in a moderate inhibition of t-PA-mediated fibrinolysis, normal platelets markedly inhibited clot lysis under the same conditions. However, no difference between PAI-1-deficient platelets and platelets with normal PAI-1 content was observed when streptokinase or a PAI-1-resistant t-PA mutant were used to initiate fibrinolysis. In addition, PAI-1-resistant t-PA was significantly more efficient in lysing clots containing normal platelets than wild-type t-PA. We conclude that platelets inhibit t-PA-mediated fibrinolysis by both PAI- 1-dependent and PAI-1-independent mechanisms. These results have important implications for the role of PAI-1 in the resistance of platelet-rich thrombi to lysis in vivo.     

Kringle glycosylation in a modified human tissue plasminogen activator improves functional properties

Native tissue plasminogen activator (ntPA) has a variable glycosylation site on its kringle-2 domain. We have examined the effects of kringle glycosylation on functional properties by studying the simplified tPA molecule, tPA-6. tPA-6 is composed of kringle-2 and the serine protease domains and, like ntPA, cells expressing tPA-6 process it into two glycoforms: the monoglycosylated tPA-6-primary (tPA-6P, type II) with N- linked glycosylation at Asn-448 in the serine protease domain and diglycosylated tPA-6-variant (tPA-6V, type I) with glycosylation at Asn- 448 and at Asn-184 in kringle-2. When the two glycoforms were separated, we found that purified tPA-6V had reduced fibrin-stimulated plasminogenolytic activity toward Glu-plasminogen when compared to purified tPA-6P. However, in the presence of fibrin, tPA-6V unexpectedly exhibited a sixfold increase in selectivity toward Lys- plasminogen. In addition, tPA-6V was less susceptible than tPA-6P to plasmin-mediated conversion to the two-chain form. By site-directed mutagenesis of tPA-6, we eliminated variable glycosylation at Asn-184 and engineered a new glycosylation signal at a remnant site in the kringle. This derivative, designated tPA-6D, was secreted with complete kringle glycosylation. Like the naturally occurring tPA-6V, tPA-6D had lower rates of fibrin-stimulated Glu-plasminogen activation, increased specificity toward Lys-plasminogen, and greater resistance to plasmin digestion. Although the activity of tPA-6D could be stimulated by fibrin, its activity was not stimulated significantly by fibrinogen, and in human plasma the rate of fibrinogen depletion was reduced threefold. Although fibrin binding to kringle-2 of tPA-6D was slightly improved, there was a substantial increase in the dissociation constant (kd) for lysine binding, demonstrating a lack of correlation between these ligand-binding sites. Overall, our data demonstrate the marked effect of kringle glycosylation on functional properties. In addition, we have generated a derivative with properties that potentially improve clot specificity and single-chain half-life and reduce the potential for plasminogen activation in the plasma.