Faculty Bibliography

Human parvovirus B19, which infects and lyses erythroid precursors, can cause severe anemia in patients with immunodeficiency. The incidence of parvovirus infection in adult acquired immunodeficiency syndrome (AIDS) patients is unknown. Eighty-one archival formalin-fixed, paraffin-embedded (FFPE) bone marrow biopsies from 73 AIDS adults were immunostained with monoclonal R92F6 against B19 VP1 and VP2 capsid proteins using streptavidin peroxidase and streptavidin alkaline phosphatase techniques. In addition, the same tissues were hybridized in situ with a digoxigenin-labeled parvovirus B19 DNA probe. Five FFPE bone marrows, from 3 HIV-negative patients with positive immunoglobulin M (IgM) serology for parvovirus B19, and 1 parvovirus B19-infected fetal liver were positive controls. By immunoperoxidase, all tissues were negative with R92F6 except the fetal liver, which exhibited strong positivity predominantly in viral inclusions. With immunoalkaline phosphatase, all positive controls were immunoreactive with R92F6; however, the AIDS marrows were negative. With in situ hybridization (ISH), all positive controls and 7 of 81 (8.6%) of AIDS marrows were positive for B19 parvovirus DNA. We conclude that ISH is more sensitive than R92F6 immunohistochemistry in parvovirus B19 detection. A small but significant number of bone marrows from AIDS adults shows evidence of human parvovirus B19 infection

Thrombin-treated tumour cells enhance their adhesion to platelets, fibronectin and von Willebrand factor in vitro, and enhanced their pulmonary metastasis in mice in vivo. A unique seven transmembrane spanning thrombin receptor has recently been cloned which is activated following thrombin proteolysis of the N-terminal end of the receptor with exposure of a tethered ligand. An N-terminal 14-mer (SFLLRNPNKYEPF) or 6-mer (SFLLRN) of the tethered ligand can serve as a thrombin receptor activation peptide (TRAP) by mimicking the action of thrombin on platelets, endothelial cells and smooth muscle cells. We have examined six human tumour cell lines for their response to TRAP, for the presence of this thrombin receptor mRNA by RT-PCR, protein by immunoblot and for their in vitro and in vivo response to TRAP. All six cell lines contain the receptor mRNA, and when treated with 100 microM 6-mer TRAP or 1 u/ml thrombin increase their adhesion to platelets 2-3-fold. Four of the six cell lines undergo tyrosine phosphorylation within 30 s to 1 min after exposure to 6-mer TRAP or thrombin. Thus tumour cells respond to thrombin via activation of their seven transmembrane spanning thrombin receptor

Six monoclonal IgG1-k antibodies (LK2, LK3r, LK4-55, LK5, LK6-55, LK7r) were raised against platelet membrane GPIIIa in order to study the structure-function relationship of this molecule. Antibodies were selected on their ability to react with GPIIIa by ELISA on adherent platelets, by immunoblot on platelet lysates and by fluorescence flow cytometry on intact platelets. Fluorescence reactivity varied from 3- to 202-fold greater than isotype control fluorescence. Two MoAbs reacted on immunoblot under reduced conditions (LK7r and LK3r). Two reacted with a 55 kD chymotrypsin/subtilisin digest of GPIIIa which is likely to exclude amino acids 121-348 (LK4-55 and LK6-55). Four of the MoAbs (LK5, LK3r, LK2 and LK4-55) inhibited tyrosine phosphorylation of one to four distinct bands on immunoblot. LK4-55 reacted with an N-terminal 66 amino acid fusion protein of GPIIIa near the PLA epitope (Leu 33). LK7r reacted with a 212-222 peptide reported to be an RGD fibrinogen binding site. LK2 reacted near a disintegrin-RGD binding site. Except for LK5, all inhibited ADP, collagen and thrombin-induced platelet aggregation in a heterogeneous fashion. Percentage inhibition of 125I-fibrinogen binding to platelets varied from 18% to 98%. No correlation was noted between inhibition of fibrinogen binding, location of MoAb binding on GPIIIa, reactivity of MoAb binding with GPIIIa, inhibition of thrombin-induced tyrosine phosphorylation or inhibition of platelet aggregation induced by ADP, collagen or thrombin. Thus MoAbs, binding to platelet GPIIIa at different sites, inhibit platelet aggregation in a heterogeneous manner

The role of platelets and thrombin was examined in tumor cell adhesion in vitro and metastasis in vivo. Adhesion of tumor cells to platelets was inhibited by agents inhibiting platelet integrin IIb-IIIa receptor occupancy (MoAb 10E5 and tetrapeptide RGDS) as well as IIb-IIIa ligands (polyclonal antibodies against fibronectin and von Willebrand factor (vWF)). In vivo murine experimental pulmonary metastasis (tail vein injection) could be inhibited by antibody-induced induction of thrombocytopenia and reconstituted by simultaneous injection of human platelets. Preincubation of human platelets with MoAb 10E5 (vs IIb-IIIa) inhibited reconstitution of metastasis. Thrombin activates tumor cell adhesion to platelets by activating platelets as well as tumor cells. Thrombin-activated tumor cells also enhance their adhesion to endothelial cells as well as adhesive ligands fibronectin and vWF. Experimental pulmonary metastasis is enhanced 4-400 fold by preinfusion of thrombin into mice or 10-160 fold by prior treatment of tumor cells with thrombin. The in vitro and in vivo effects of thrombin were mimicked by thrombin receptor activation peptides SFLLRNPNDKYEPF and SFLLRN. Nine of nine tumor cell lines have the seven transmembrane spanning thrombin receptor detected by the polymerase chain reaction. Thus, both platelets and thrombin contribute to experimental tumor metastasis by fostering and enhancing IIb-IIIa integrin platelet interaction with tumor cells. Since many tumor cells generate thrombin, it is proposed that tumor cells may autoactivate a metastatic phenotype