Affinity hemodialysis for antiviral therapy. II. Removal of HIV-1 viral proteins from cell culture supernatants and whole blood.

BACKGROUND: HIV-1 gp120 may play a role in the progression from HIV infection to AIDS. AIMS: We investigated affinity hemodialysis for removing gp120 from cell culture and whole blood. METHODS: Anti-gp120 antibodies covalently coupled to agarose beads were packed into columns or hollow-fiber hemodialysis cartridges. Supernatants from HIV-infected HL2/3 cells or gp120 containing whole blood were pumped over the columns and gp120 measured by ELISA. RESULTS: Anti-gp120 agarose removed approximately 90% of HIV-1 gp120 from HL2/3 cultures in 30-60 min. Capture was antibody-dependent (F105 > IDX 1121 > ABI 13-108). Affinity hemodialysis also efficiently captured gp120 from buffer in a first-order, flow-rate-dependent fashion (t((1/2)) = 13 min at 0.9 ml/min). Clearance was faster than calculated diffusion (t((1/2)) approximately 2.5 h) suggesting significant convective transport. gp120 removal from blood was slower (t((1/2)) = 1.4 h). CONCLUSION: Affinity hemodialysis efficiently clears gp120 from cell culture fluids and blood and may be useful in slowing the progression to AIDS.

Affinity hemodialysis for antiviral therapy. I. Removal of HIV-1 from cell culture supernatants, plasma, and blood.

We tested an affinity hemodialysis technique designed to efficiently remove HIV and toxic viral proteins from blood. Miniature polyethersulfone hollow-fiber dialysis cartridges (200-500 nm pore) were packed with anti-HIV antibodies covalently coupled to agarose beads and sealed inside the cartridge. Cell culture fluids, plasma, or infected blood (7-15 ml) containing HIV-1 were circulated over the cartridge at 0.7-10 ml/min and the rate of removal of HIV measured by PCR and p24 ELISA. The technique removed up to 98% of HIV-1 particles from cell culture supernatants. Affinity hemodialysis also efficiently captured cultured HIV from human blood plasma (90%) and native HIV from infected blood (83% to 100%). Viral capture followed first-order kinetics (t(1/2) = 2.8 h). Variations in antibody type, matrix linkage (protein G versus direct coupling), bead pore size, and temperature of operation (25-37 degrees C) had only small effects. Although some binding was nonspecific, direct binding to the immobilized antibodies appeared to be the predominant mechanism.