Identification of Anti-gp41 Monoclonal Antibodies That Effectively Target Cytotoxic Immunoconjugates to Cells Infected with Human Immunodeficiency Virus, Type 1

Abstract

We are developing cytotoxic immunoconjugates (CICs) targeting the envelope protein (Env) of the Human Immunodeficiency Virus, type 1 (HIV) to purge the persistent reservoirs of viral infection. We have previously studied the ability of multiple monoclonal antibodies (mAbs) to deliver CICs to an HIV-infected cell. We have found that CICs targeted to the membrane-spanning gp41 domain of Env are most efficacious, in part because their killing is enhanced in the presence of soluble CD4. The ability of a mAb to deliver a CIC does not correlate with its ability to neutralize nor mediate Ab-dependent cellular cytotoxicity. In the current study, we seek to define the most effective anti-gp41 mAbs for delivering CICs to HIV-infected cells. To do this, we have evaluated a panel of human anti-gp41 mAbs for their ability to bind and kill two different Env-expressing cell lines: persistently infected H9/NL4-3 and constitutively transfected HEK293/92UG. We measured the binding and cytotoxicity of each mAb in the presence and absence of soluble CD4. We found that mAbs to the immunodominant helix-loop-helix region (ID-loop) of gp41 are most effective, whereas neutralizing mAbs to the fusion peptide, gp120/gp41 interface, and the membrane proximal external region (MPER) are relatively ineffective at delivering CICs. There was only a weak correlation between antigen exposure and killing activity. The results show that the ability to deliver an effective IC and neutralization are distinct functions of mAbs.

Keywords: CD4; HIV envelope; cytotoxicity; gp41; immunoconjugate; immunotoxin; persistent reservoir.

Figure 1

Comparison of binding and cytotoxicity of anti-gp41 mAbs assayed on 92UG cells. MAbs are shown ordered based on epitope location from N-terminus (top) to C-terminus (bottom). MAbs to MPER and immunodominant loop region are marked by bars. Data in red indicates presence of sCD4183 (500 ng/mL), blue the absence of sCD4183. (A) Indirect fluorescence. Mabs were tested at 3 µg/mL and detected with goat anti-human IgG-FITC as measured by flow cytometry. Binding is reported as median fluorescence with SEM. The vertical black line represents non-specific fluorescnce observed in the presence of anti-human FITC only. (B) Cytotoxicity. Anti-gp41 mAbs were tested at 125 ng/mL, and cytotoxicity was detected with ricin A chain conjugated anti-human Ig. Percent cytotoxicity was calculated as described in the text and shown as mean and SEM. (C) Correlation of cytotoxicity and binding with (top) and without (bottom) sCD4. P and R² from Pearson correlation analysis.

Figure 2

Titration of cytotoxicity of selected anti-gp41 mAbs assayed on 92UG cells. Cytotoxicity of anti-gp41 mAbs with concentrations from 0.2 to 125 ng/mL was evaluated with ricin A chain conjugated anti-human Ig in the presence of sCD4 (500 ng/mL). Percent cytotoxicity was calculated as described in the text.

Figure 3

Comparison of binding and cytotoxicity of anti-gp41 mAbs assayed on H9/NL4-3 cells. H9/NL4-3 cells are H9 cells persistently infected with the lab isolate NL4-3. The layout of figures and reagents are as in the caption of Figure 1. (A) Indirect fluorescence, (B) Cytotoxicity, (C) Correlation of cytotoxicity and binding.

Figure 4

Titration of cytotoxicity with anti-gp41 mAbs assayed on H9/NL4-3 cells.