Mode of inhibition of HIV-1 Integrase by a C-terminal domain-specific monoclonal antibody

Abstract

Background: To further our understanding of the structure and function of HIV-1 integrase (IN) we developed and characterized a library of monoclonal antibodies (mAbs) directed against this protein. One of these antibodies, mAb33, which is specific for the C-terminal domain, was found to inhibit HIV-1 IN processing activity in vitro; a corresponding Fv fragment was able to inhibit HIV-1 integration in vivo. Our subsequent studies, using heteronuclear nuclear magnetic resonance spectroscopy, identified six solvent accessible residues on the surface of the C-terminal domain that were immobilized upon binding of the antibody, which were proposed to comprise the epitope. Here we test this hypothesis by measuring the affinity of mAb33 to HIV-1 proteins that contain Ala substitutions in each of these positions. To gain additional insight into the mode of inhibition we also measured the DNA binding capacity and enzymatic activities of the Ala substituted proteins.

Results: We found that Ala substitution of any one of five of the putative epitope residues, F223, R224, Y226, I267, and I268, caused a decrease in the affinity of the mAb33 for HIV-1 IN, confirming the prediction from NMR data. Although IN derivatives with Ala substitutions in or near the mAb33 epitope exhibited decreased enzymatic activity, none of the epitope substitutions compromised DNA binding to full length HIV-1 IN, as measured by surface plasmon resonance spectroscopy. Two of these derivatives, IN (I276A) and IN (I267A/I268A), exhibited both increased DNA binding affinity and uncharacteristic dissociation kinetics; these proteins also exhibited non-specific nuclease activity. Results from these investigations are discussed in the context of current models for how the C-terminal domain interacts with substrate DNA.

Conclusion: It is unlikely that inhibition of HIV-1 IN activity by mAb33 is caused by direct interaction with residues that are essential for substrate binding. Rather our findings are most consistent with a model whereby mAb33 binding distorts or constrains the structure of the C-terminal domain and/or blocks substrate binding indirectly. The DNA binding properties and non-specific nuclease activity of the I267A derivatives suggest that the C-terminal domain of IN normally plays an important role in aligning the viral DNA end for proper processing.

Figure 1

Ala substitution of residues in the NMR-determined epitope of mAb33 decrease antibody binding. (A) Model of the HIV-1 CTD dimer. Left panel: The NMR-determined SH3-fold structure of the CTD is displayed with one subunit colored in yellow and the other in blue. DNA binding residues E246, K264, K266, and R262 are shown in ball-and-stick representation and labeled with black lettering. Residues of the putative mAb33 epitope, as determined by NMR techniques [28], are displayed in green space filling representation. K244 is shown in grey, as it is unlikely to be a critical component of the mAb33 epitope (see Panel B). The long arrow points to residue W243 at the interface between subunits in this NMR dimer of the CTD which forms a "saddle" with both K264 residues extending into the cleft proposed to bind DNA [19]. Right panel: An orthogonal view rotated 90 degrees about the displayed axis. This view shows residues known to be involved in DNA binding on one face of the subunit and adjacent to the mAb33 epitope. (B) ELISA data for CTD substituted HIV-1 sIN proteins and their interactions with mAb33. A high binding microtiter plate was coated with 50 ng of antigen (sIN or one of the Ala substituted sIN proteins) and incubated overnight at 4°C. The plates were then blocked with bovine serum albumin, washed, and incubated with the mAb33 1° antibody which was serially diluted (2-fold) from a starting concentration of 250 ng per well. A standard ELISA protocol was then followed using an alkaline phosphatase conjugated 2° antibody against the kappa chain. The relative binding efficiency of mAb33 to the IN proteins was determined by measuring the absorbance at 405 nm.

Figure 2

Enzymatic Activity of HIV sIN and the CTD substituted sIN proteins. Processing (solid bars) and joining (open bars) activities of the sIN derivatives are shown relative to that of sIN, whose activity is set at 100%. Assay conditions are described in Materials and Methods.

Figure 3

Non-specific nuclease activity of I267A sIN derivatives. Reaction conditions are described in Materials and Methods. Bl represents a blank reaction in which the enzyme was omitted. (-) indicates duplicate reactions in which the Merck HIV-1 inhibitor L-708,906 was absent. Either 10 or 100 μM of the L-708,906 inhibitor was added as indicated. These concentrations of inhibitor are expected to affect the processing as well as the joining activity of wild type HIV-1 IN [33].