KineTACs are fully recombinant bispecific antibodies built of human scaffolds that utilize cytokine-mediated internalization of its cognate receptor to enable highly selective lysosomal delivery of both cell surface and extracellular proteins. To demonstrate the utility of this platform, chemokine CXCL12 was chosen as it specifically binds the atypical chemokine receptor CXCR7, a decoy receptor that constitutively internalizes and recycles.We show that KineTACs bearing CXCL12 can efficiently utilize CXCR7 internalization for lysosomal degradation applications and are generalizable against various therapeutically relevant proteins .To demonstrate proof-of-concept that CXCL12 bearing KineTACs can degrade cell surface proteins, we first targeted programmed death ligand. Overexpression of PD-L1 on cancer cells leads to inhibition of checkpoint protein programmed death protein 1 and suppression of cytotoxic T cell activity.First, we generated knob-in-hole bispecifics14 in which the human CXCL12 chemokine was N-terminally fused to the knob Fc domain and the antibody sequence for atezolizumab , an FDA approved inhibitor of PD-L1, was fused to the hole Fc . CXCL12 bearing bispecifics are not limited by the light chain mispairing problem, which is common to bispecific IgGs with Fabs on both arms, enabling full assembly of KineTACs during mammalian expression.A His tag was introduced on the knob arm to allow purification of the formed bispecific from unwanted hole-hole homodimers that may form. Next, we confirmed that the PD-L1 targeting KineTAC retains binding to PD-L1 using biolayer interferometry. Furthermore, an isotype control of the CXCL12 KineTAC, which incorporates a Fab arm to the SARS-CoV-2 spike protein,vertical planters for vegetables retained binding to endogenous CXCR7 expressed on triple negative breast cancer cell line MDA-MB-231 . This data suggested that both anti-PD-L1 and CXCL12 arms of the KineTAC were functional in the bispecific context.
To determine whether CXCL12-Atz could degrade PD-L1, MDA-MB-231 cells were treated with varying concentrations of the KineTAC. After 24 hr treatment, levels of PD-L1 were quantified using western blotting, demonstrating that both glycosylated forms of PD-L1 were substantially degraded, with a maximal percent degradation of roughly 70% . Control antibodies, such as atezolizumab Fab or CXCL12 isotype, do not induce the degradation of PD-L1 either alone or in combination, indicating that PD-L1 degradation is dependent on the bispecific KineTAC scaffold . Finally, flow cytometry and western blotting was used to verify that the PD-L1 degradation observed is due to depletion of cell surface PD-L1. We next sought to determine whether the KineTAC platform could be generalized to degrade other therapeutically relevant cell surface proteins. First, we targeted human epidermal growth factor receptor 2 , which is frequently upregulated in cancer and linked to breast cancer invasiveness and tumor progression.To develop a KineTAC targeting HER2, we incorporated the antibody sequence for trastuzumab , an FDA approved HER2 inhibitor, into the KineTAC scaffold . Various breast cancer cell lines endogenously expressing HER2 were incubated for 24 hr with CXCL12-Tras. Substantial degradation of HER2 was observed in MCF7 and MDA-MB-175VII cells, with Dmax of 51 and 62%, respectively . Unsurprisingly, lower percent degradation was observed in SK-BR-3 cells, which over expresses HER2 relative to CXCR7, suggesting that the maximal percent degradation mediated by KineTACs could be dependent on the expression of the target protein relative to CXCR7. Next, we targeted epidermal growth factor receptor for degradation. EGFR is implicated as a driver of cancer progression, and EGFR inhibitors are approved for use in non-small cell lung, colorectal, and gastric cancers. We developed KineTACs targeting EGFR by incorporating cetuximab , an FDA approved EGFR inhibitor, into the KineTAC scaffold .
Following 24 hrtreatment with CXCL12-Ctx, EGFR levels were dramatically reduced in HeLa cells, with a Dmax of 84% observed . This result was recapitulated in various breast and lung cancer cell lines, including MDA-MB-231, A431, and NCI-H292 .Using a previously described antibody against CUB domain-containing protein 1, we observed near complete degradation of CDCP1 after 24 hr treatment of HeLa cells, with a Dmax of 93% . KineTACs also enabled the degradation of tumor-associated calcium signal transducer 2 , the over expression of which has been linked to tumor progression in a variety of tumors. In MCF7 cells, we observed a Dmax of 51% after treatment with TROP2 targeting KineTAC . We then tested whether KineTACs are active to degrade the checkpoint protein PD-1 in CD8+ T cells isolated from primary human peripheral blood mononuclear cells. T cells were then activated, causing over expression of PD-1 on the cell surface along with other activation markers . Activated T cells were then treated for 24 hr with a PD-1 targeting KineTAC, which incorporates the antibody sequence for nivolumab , an FDA approved PD-1 inhibitor . Following treatment with CXCL12-Nivo, cell surface PD-1 levels were dramatically reduced, with a Dmax of 82%, compared to nivolumab isotype control, which is known to induce slight internalization of PD-1 .Overall, these results demonstrate the generality of the KineTAC platform for degrading a variety of cell surface proteins for degradation .Next, we sought to determine which properties, such as binding and receptor signaling, of the KineTAC are critical for efficient degradation. Alongside CXCR7, CXCL12 binds the signaling receptor CXCR4, which upon agonism will cause downstream signaling followed by receptor internalization and degradation. Signaling through CXCR4 could be counter-productive if using KineTACs to target cancer drivers, as CXCR4 over expression and agonism is linked to tumor metastasis.Thus, avoiding CXCR4 signaling could be an important consideration in developing KineTACs for therapeutic purposes.
To test this, KineTACs bearing previously described antagonistic variants of CXCL12 , that retain binding to both CXCR7 and CXCR4 but prevent CXCR4 signaling, were incorporated into the KineTAC scaffold with atezolizumab.Following 24 hr treatment in MDA-MB-231 cells, all three antagonistic variants retained the ability to degrade PD-L1 to a similar degree compared to CXCL12WT . This data suggests that CXCL12 signaling through CXCR4 is not critical for degradation, allowing flexibility to include or eliminate it depending on the biology we wish to affect. Next, the dependence on binding affinity to target receptor was probed by introducing alanine mutations into key interacting residues of atezolizumab’s complementary determining regions known to interact with PD-L1 based on the known structure of the complex. We generated a library of alanine scanned mutants of atezolizumab with a range of binding affinities and corresponding kinetic parameters to PD-L1, as measured with BLI . The atezolizumab mutants were then introduced into the KineTAC scaffold with CXCL12WT and tested for their ability to degrade PD-L1 . Correlating the PD-L1 levels post treatment to the different kinetic parameters of these binders,vertical farming technology we find that degradation is correlated to the KD and the dissociation rate , but not to the association rate. Of the mutants tested, wild-type atezolizumab had the highest binding affinity and induced the greatest level of PD-L1 degradation. Therefore, over this affinity range,the levels of degradation are dependent on the binding affinity of the antibody arm to the target protein. To determine whether a pH-dependent antibody binder against the target protein would affect degradation, BMS936559, an anti-PD-L1 antibody reported to release PD-L1 in acidic conditions29, was introduced into the KineTAC scaffold. Treatment with CXCL12- BMS936559 compared to CXCL12-Atz showed that pH-dependent release of PD-L1 slightly decreases the maximal level of degradation observed . This result is not due to differences in KD as atezolizumab and BMS936559 are reported to have similar binding affinities to PD-L1.30 To investigate whether the binding epitope on the protein of interest could impact degradation, we introduced additional HER2 and EGFR targeting antibodies into the KineTAC scaffold that have been described to bind different epitopes. For HER2, pertuzumab , which is known to bind a distinct epitope from trastuzumab on HER231, was introduced into the KineTAC scaffold . Following 24 hr treatment of MCF7 cells, we find that CXCL12-Tras is superior to CXCL12-Ptz at lower concentrations, indicating that epitope can alter the dose response to KineTACs . For EGFR, we introduced five different antiEGFR binders , necitumumab , and matuzumab 32 into the KineTAC scaffold. Following 24 hr treatment of HeLa cells, we observe that some epitope binders, such as necitumumab and matuzumab, retain similar levels of EGFR degradation compared to CXCL12-Ctx, while other epitope binders, such as depatuxizumab, nimotuzumab, and panitumumab, abrogate or impair the ability to degrade EGFR .
Further, the degradation observed for each binder is not correlated to binding affinity . This data highlights the dependence of KineTAC-mediated degradation on target binding epitope.Next, we asked whether glycosylation of the KineTAC Fc domain at the N297 position would impact degradation. The N297G mutation is commonly introduced into IgGs to produce an aglycosylated form to eliminate effector function. However, glycosylation at N297 can impart greater stability and favorable pharmacokinetic properties.The glycosylation site at N297 was re-introduced to the CXCL12-Atz scaffold and the degradation efficiency between the glycosylated and aglycosylated forms compared. We find that glycosylation at N297 does not significantly impact PD-L1 degradation levels . Thus, the improved stability and pharmacokinetic properties of KineTACs can be utilized for in vivo use without major disruption to degradation efficiency. Finally, we determined whether the bispecific antibody construct used could influence levels of degradation. To this end, a Fab fusion construct in which CXCL12 is fused to the N-terminus of the atezolizumab Fab heavy chain via a flexible Avidin tag linker was co-expressed with atezolizumab Fab light chain . The CXCL12-Atz Fab fusion retained binding to PD-L1 Fc fusion as measured by BLI . After 24 hr treatment in MDA-MB- 231, the levels of PD-L1 were measured by western blotting. While the bispecific IgG construct caused significant degradation of PD-L1, the Fab fusion was unable to induce significant degradation, with a Dmax of only 20% observed . The differences in degradation between these two constructs could be due to several factors, including construct rigidity and linker length. Overall, this data highlights that the bispecific IgG is a useful KineTAC scaffold. We next sought to evaluate the mechanism of KineTAC-mediated degradation. To determine whether KineTACs catalyze degradation via the lysosome or proteasome, MDA-MB-231 cells were pre-treated with either media alone, bafilomycin , orMG132 . After 1 hr pre-treatment, cells were treated with CXCL12-Atz for 24 hrs. We observed that bafilomycin pre-treatment inhibited degradation of PD-L1, while MG132 had no effect, demonstrating that KineTACs mediate degradation via delivery of target proteins to the lysosome . Immunofluorescent microscopy revealed complete removal of EGFR from the cell surface following 24 hr CXCL12-Ctx treatment as compared to the cetuximab isotype, further highlighting that KineTACs induce robust internalization of target proteins . Furthermore, KineTAC-mediated degradation occurs in a time-dependent fashion, beginning after 6 hrs post-treatment with CXCL12-Atz, with the levels of PD-L1 continuing to decrease over time to near complete degradation at 48 hrs . We next wished to determine which of the two receptors is more important for degrading surface proteins in the context of KineTACs. To confirm that KineTAC mechanism of action occurs via CXCR7 and not CXCR4, RNA interference was used to knockdown the levels of CXCR4 in HeLa cells. After 48 hr transfection with CXCR4-targeting or control siRNA pools, cells were treated with CXCL12-Ctx for 24 hrs . Western blotting analysis revealed that EGFR degradation levels are unchanged with CXCR4 knocked down. This data highlights that CXCR4 is not necessary for efficient degradation and suggests that KineTACs operate through CXCR7-mediated internalization . Furthermore, KineTACs bearing CXCL11, a chemokine that specifically binds CXCR7 and CXCR3 but not CXCR434, are capable of degrading both PD-L1 and EGFR . This result further highlights that CXCR7 is the receptor responsible for KineTAC-mediated degradation and demonstrates the exciting opportunity for using alternative cytokines in the KineTAC scaffold to degrade target proteins. We next used quantitative mass spectrometry to determine whether proteome-wide changes occur following KineTAC treatment. Both the surface-enriched and whole cell lysates were analyzed following 48 hr CXCL12-Atz or CXCL12-Ctx treatment compared to PBS treated control in MDA-MB-231 or HeLa cells, respectively. For PD-L1 degradation, the surface-enriched sample revealed no significant changes to the proteome, with PD-L1 being the only protein down regulated in CXCL12-Atz treatment compared to control . Whole cell proteomics also revealed that no major changes are occurring . PD-L1 was not detected in the whole cell sample, likely due to low abundance of cell surface proteins relative to cytosolic proteins.