1) Design of soluble protein constructs displaying the extracellular domains of the HIV coreceptors CCR5 and CXCR4. In FY2014, we designed and characterized soluble constructs containing the extracellular domains of CCR5 previously identified as being critical to coreceptor function, i.e. the N-terminus and extracellular loop 2 (ECL2). While structural details of the HIV-1 Env glycoprotein (gp120 and gp41 subunits, as well as trimeric gp140) have been studied extensively at the atomic level (X-ray crystallography and high resolution cryo-EM), the coreceptor molecules have resisted such analyses, in large part due to their obligate complex membrane associations (they are GPCRs with 7 TM segments). Our soluble CCR5 constructs are designed to circumvent this problem. Two variant designs expressed in animal cells and tested, including: linking these domains to soluble CD4, or to a stable protein scaffold previously described as being useful to analyzing extracellular regions of other GPCRs. In assays of activation and neutralization of Env function using vaccinia-based cell fusion assays, we have found that some constructs appear to be specifically interacting with their corresponding binding sites on gp120. Further studies will employ biochemical and biophysical approaches to test this conclusion;supportive results will pave the way for collaborative X-ray crystallographic analyses of gp120-CCR5 interactions using these soluble coreceptor constructs. 2) Structure-function studies of the HIV-1 Env trimer. In FY2014, we have expanded our collaborative studies with Dr. Paolo Lusso, NIAID, on HIV-1 Env variants engineered to be 'locked'in a novel proposed conformation for gp120 in the unliganded trimeric state. in particular, we have extended previous studies on soluble monomeric gp120 to corresponding cell-associated Env trimers derived from gp160 constructs with the same modifications. The results provide further confirmation that the proposed structure indeed reflects the preferred gp120 conformation in the native Env trimer. These findings add additional structural insights beyond those recently described by the X-ray crystallographic structure of the engineered 'SOSIP'Env trimer construct. As another approach to study the native HIV-1 Env trimer, we are developing methods for real-time binding studies of antibodies and ligands to native Env expressed on pseudovirus particles, using biolayer interferometry (ForteBio Octet system). 3) Targeted killing of HIV-infected cells. In FY2014, we extended our targeting killing strategy using both novel immutoxins (for augmenting ART to cure acute infection) and chimeric antigen receptors (CARs) to achieve a 'functional cure'of chronic infection. Our previous studies (collaboration with Dr. Ira Pastan, NCI) have demonstrated highly potent and specific killing of HIV-infected cells with recombinant immunotoxins based on Pseudomonas aeruginosa exotoxin A (PE). Early collaborative studies in a humanized mouse model (Goldstein et al. J. Inf. Dis. 2000) indicated that these agents have dramatic effects when co-administered with ART during acute infection, resulting in (nearly) complete prevention of viral resurgence upon cessation of therapy. However in chronically infected BLT humanized mouse model (Denton et al. PLoS Pathogens 2014) as well as in SIV/macaque models (North et al., and Brenchley et al., unpublished), the immunotoxins significantly augmented ART activity but failed to eradicate the virus. These results suggest that immunotoxins may be of value as supplements to ART to achieve cure in HIV acute infection;to this end, collaborative studies (Pastan group) have been extended to generate new PE-based immunotoxins using recently identified broadly neutralizing mAbs. To achieve a 'functional cure'in chronically infected individuals, a more durable targeted cell killing strategy will be required. To this end, in collaboration with Dr. Steven Rosenberg, NCI, we have designed novel CD4-based chimeric antigen receptors (CARs) for genetic modification of patient T cells and adoptive transfer to the infected person. In addition to containing '2nd generation'intracellular signaling motifs, our CARs contain unique targeting domains that confer major benefits compared to 'standard'CD4-based CARs that proved ineffective in previous clinical studies: 1) enhanced potency, and 2) absence of CD4 entry receptor activity, which otherwise could lead to infection of the CAR-expressing CD8 T cells. We have designed two versions of such CARs, one containing CD4 linked to an scFv against the CD4-induced conserved bridging sheet involved in coreceptor binding, and another containing CD4 attached to the carbohydrate binding domain (CRD) of a C-type lectin, which binds to high mannose glycans universally expressed on all gp120 variants. Both CAR modalities display the indicated enhancements based on in vitro studies, and the latter is expected to be weakly (perhaps completely?) non-immunogenic. Preliminary collaborative studies (Dr. Tae-Wook Chun, NIAID) demonstrate that the CD4-scFv CARs effectively kill activated memory CD4+ T cells from ART-suppressed patients. Collaborative studies (Dr. Mario Roederer, VRC) were initiated with the CD4-scFv CAR in macaques infected with a SHIV;however the animals controlled this recombinant virus without the need for the genetically modified T cells. Studies are underway with SIV-infected macaques, using the CD4-CRD CAR (all sequences derived from corresponding rhesus macaque genes). Also planned are collaborative studies (Dr. Victor Garcia-Martinez, UNC) to test the CARs in the BLT humanized mouse model.
Showing the most recent 10 out of 14 publications
