Specific Inhibition of Tumor Growth by T Cell Receptor-Drug Conjugates Targeting Intracellular Cancer-Testis Antigen NY-ESO-1/LAGE-1.

Despite the significant therapeutic advances in T-cell immunotherapy, many malignancies remain unresponsive, which might be because of the negative regulation of T cells by the tumor microenvironment (TME). T cells discriminate tumor cells and normal cells through T-cell receptors (TCRs); therefore, we generated a novel type of TCR-drug conjugates (TDCs) by referring antibody-drug conjugations (ADCs) to overcome the effects of the TME on T cells while preserving the specificity of TCR for tumor recognition. We selected HLA-A2/NY-ESO-1157-165 (peptide NY-ESO-1157-165 in complex with human leukocyte antigen serotype HLA-A*02:01) as the antigen and the antigen-specific TCR (1G4113) as the carrier. By sortase A-mediated ligation, we obtained three NY-TCR-vcMMAEs and further studied their properties, antitumor activity, and toxicity in vitro and in vivo. We found that all the NY-TCR-vcMMAEs had high endocytosis efficiency and specifically killed HLA-A2/NY-ESO-1157-165 positive tumor cells. In xenograft models, one of the TDCs, NY-TCR-2M, was effectively and specifically distributed into tumor tissues and inhibited tumor growth without inducing obvious toxicity. Our results demonstrated that TCRs can be carriers of toxic payloads and that the TDCs thus formed can specifically inhibit tumor growth, neglecting the immune microenvironment.

Preclinical studies of targeted therapies for CD20-positive B lymphoid malignancies by Ofatumumab conjugated with auristatin.

Utilization of antibodies to deliver highly potent cytotoxic agents to corresponding antigen-overexpressed tumor cells is a clinically validated therapeutic strategy. Ofatumumab (OFA, trade name Arzerra) is a fully human CD20-specific antibody that is active against CD20-positive B-cell lymphoma/chronic lymphocytic leukemia cells. In order to further enhance the anticancer effect of OFA, anti-CD20 OFA has been conjugated with highly cytotoxic monomethyl auristatin E (MMAE) through a cathepsin-B-cleavable valine-citrulline (vc) dipeptide linkage to form OFA-vcMMAE and the anti-tumor activity of OFA-vcMMAE against CD20-positive B lymphoma cells are then evaluated in vitro and in vivo. As a result, conjugation of OFA with MMAE has kept the initial effector functional activities of OFA such as binding affinity, complement-dependent cytotoxicity (CDC) as well as antibody-dependent cell-mediated cytotoxicity (ADCC). In addition, the conjugation of MMAE significantly improved the cytotoxic activity of OFA against CD20-positive cells (i.e., Raji, Daudi and WIL2-S cells) but not against CD20-negative K562 cells. On the other hand, OFA-vcMMAE was modulated from the CD20-positive cell surface and then entered the lysosomes by receptor-mediated endocytosis, underwent proteolytic degradation and released active drug MMAE to induce apoptotic cell death through a caspase-3-like protease-dependent pathway. Surprisingly, OFA-vcMMAE completely inhibited the growth of CD20-positive Daudi and Ramos lymphoma xenografts in vivo, and exhibited greater anti-tumor activity than unconjugated OFA, suggesting that the anti-tumor activity of anti-CD20 antibody can be enhanced by conjugation with MMAE. In the near future, this new approach might be used as a clinical treatment of CD20-positive B lymphoid malignancies.

Engineering and refolding of a novel trimeric fusion protein TRAIL-collagen XVIII NC1.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is considered to be a promising anticancer agent because its active form TRAIL trimer is able to induce apoptosis in different tumor cell lines while sparing normal cells. However, TRAIL trimer possesses a short half-life and low stability, which turns out to be a major obstacle for the development of clinical trials. In our present study, we constructed a recombined TRAIL trimer by genetic fusion of non-collagenous domain (NC1) of human collagen XVIII or its trimerization domain (TD) to C-terminus of TRAIL via a flexible linker, and then refolded the fusion proteins using a two-step refolding approach, namely a combination of dilution and gel filtration chromatography. As a result, both recombinant proteins, TRAIL-NC1 and TRAIL-TD, were expressed in Escherichia coli as inclusion bodies, and they exhibited difficultly to refold efficiently by conventional methods. Thereby, we applied a modified two-step refolding approach to refold fusion proteins. More than 55 % of TRAIL-NC1 and 90 % of TRAIL-TD protein activity was recovered during the two-step refolding approach, and their stability was also increased significantly. Also, size exclusion chromatography showed refolded TRAIL-NC1 was a trimer while TRAIL-TD, hexamer. However, both of them exerted good apoptosis activity on NCI-H460 cells.

The endoplasmic reticulum is a target organelle for trivalent dimethylarsinic acid (DMAIII)-induced cytotoxicity.

The purpose of present study was to characterize the endoplasmic reticulum stress and generation of ROS in rat liver RLC-16 cells by exposing to trivalent dimethylarsinous acid (DMAIII) and compared with that of trivalent arsenite (iAsIII) and monomethylarsonous acid (MMAIII). Protein kinase-like endoplasmic reticulum kinase (PERK) phosphorylation was significantly induced in cells exposed to DMAIII, while there was no change in phosphorylated PERK (P-PERK) detected in cells after exposure to iAsIII or MMAIII. The generation of reactive oxygen species (ROS) after DMAIII exposure was found to take place specifically in the endoplasmic reticulum (ER), while previous reports showed that ROS was generated in mitochondria following exposure to MMAIII. Meanwhile, cycloheximide (CHX) which is an inhibitor of protein biosynthesis strongly inhibited the DMAIII-induced intracellular ROS generation in the ER and the phosphorylation of PERK, suggesting the induction of ER stress probably occurs through the inhibition of the protein folding process. Activating transcription factor 4 (ATF4) and C/EBP homologous protein (CHOP) mRNA were induced by all three arsenic species, however, evidence suggested that they might be induced by different pathways in the case of iAsIII and MMAIII. In addition, ER resident molecular chaperone glucose-regulated protein78 (GRP78) was not affected by trivalent arsenicals, while it was induced in positive control only at high concentration (Thapsigargin;Tg), suggesting the GRP78 is less sensitive to low levels of ER stress. In summary, our findings demonstrate that the endoplasmic reticulum is a target organelle for DMAIII-induced cytotoxicity.

Soluble Expression of Fc-Fused T Cell Receptors Allows Yielding Novel Bispecific T Cell Engagers.

The specific recognition of T cell receptors (TCR) and peptides presented by human leukocyte antigens (pHLAs) is the core step for T cell triggering to execute anti-tumor activity. However, TCR assembly and soluble expression are challenging, which precludes the broad use of TCR in tumor therapy. Herein, we used heterodimeric Fc to assist in the correct assembly of TCRs to achieve the stable and soluble expression of several TCRs in mammalian cells, and the soluble TCRs enable us to yield novel bispecific T cell engagers (TCR/aCD3) through pairing them with an anti-CD3 antibody. The NY-ESO-1/LAGE-1 targeted TCR/aCD3 (NY-TCR/aCD3) that we generated can redirect naïve T cells to specific lysis antigen-positive tumor cells, but the potency of the NY-TCR/aCD3 was disappointing. Furthermore, we found that the activation of T cells by NY-TCR/aCD3 was mild and unabiding, and the activity of NY-TCR/aCD3 could be significantly improved when we replaced naïve T cells with pre-activated T cells. Therefore, we employed the robust T cell activation ability of staphylococcal enterotoxin C2 (SEC2) to optimize the activity of NY-TCR/aCD3. Moreover, we found that the secretions of SEC2-activated T cells can promote HLA-I expression and thus increase target levels, which may further contribute to improving the activity of NY-TCR/aCD3. Our study described novel strategies for soluble TCR expression, and the optimization of the generation and potency of TCR/aCD3 provided a representative for us to fully exploit TCRs for the precision targeting of cancers.

Elimination of melanoma by sortase A-generated TCR-like antibody-drug conjugates (TL-ADCs) targeting intracellular melanoma antigen MART-1.

Most tumor-associated proteins are located inside tumor cells and thus are not accessible to current marketed therapeutic monoclonal antibodies or their cytotoxic conjugates. Human leukocyte antigen (HLA) class I can present peptides derived from intracellular tumor-associated proteins and somatically mutated proteins on the cell's surface, forming an HLA/peptide complex as tumor-specific antigens for T cell receptor (TCR) recognition. Therefore, HLA-mediated presentation of intracellular tumor antigen peptides provides a viable way to distinguish tumor cells from normal cells, which is important for broadening antigen selection, especially for antibody-drug conjugates (ADCs) regarding their highly cytotoxic payload. We applied sortase A-mediated conjugation to develop TCR-like ADCs (i.e., EA1 HL-vcMMAE) targeting intracellular MART-1 protein, a melanocyte-differentiating antigen specific for metastatic melanomas, via the cell surface HLA-A2/MART-126-35 peptide complex. Homogenous EA1 HL-vcMMAE (drug to antibody ratio of 4) efficiently eliminated melanoma cells in xenograft mouse models with no obvious toxicity at the therapeutic dosage. Trametinib, an MEK inhibitor serving as an HLA expression enhancing agent, augmented the TL-ADCs' efficacy both in vitro and in vivo by upregulating MART-126-35 peptide presentation, thus providing a strategy for overcoming the limitation of antigen presentation level for TL-ADCs. Hence, our findings validate the strategy of using sortase A-generated TL-ADCs to target tumor-specific intracellular proteins, with or without agents present, to increase presenting TCR epitope peptides.

Hetero-modification of TRAIL trimer for improved drug delivery and in vivo antitumor activities.

Poor pharmacokinetics and resistance within some tumor cell lines have been the major obstacles during the preclinical or clinical application of TRAIL (tumor-necrosis-factor (TNF)-related apoptosis-inducing ligand). The half-life of TRAIL114-281 (114 to 281 amino acids) was revealed to be no more than 30 minutes across species. Therefore maleimido activated PEG (polyethylene glycol) and MMAE (Monomethyl Auristatin E) were applied to site-specifically conjugate with the mutated cysteines from different monomers of TRAIL successively, taking advantage of steric effects involved within TRAIL mutant conjugations. As a result, TRAIL trimer was hetero-modified for different purposes. And the resulting PEG-TRAIL-vcMMAE conjugate exhibited dramatically improved half-life (11.54 h), favourable in vivo targeting capability and antitumor activities while no sign of toxicity in xenograft models, suggesting it's a viable therapeutic and drug delivery strategy.

Site-specific PEGylation of a mutated-cysteine residue and its effect on tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL).

Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a promising antitumor agent that specifically induces apoptosis in broad-spectrum tumor cell lines, meanwhile leaving normal cells unaffected. Unfortunately, the clinical development of TRAIL was hampered, and could be attributed to its instability, bioavailability or poor delivery. Although N-terminal specific PEGylation provides a means to improve the pharmacokinetic and stability of TRAIL, it took a bit longer time to accomplish the PEGylation process than expected. We therefore designed another PEGylation approach, mutated Cys-SH site-specific PEGylation, to conjugate methoxypoly(ethylene glycol) maleimide (mPEG-MAL) with TRAIL (95-281) mutant N109C. Asn-109 was chosen as the PEGylated site for it is a potential N-linked glycosylation site. It was shown that ~90% TRAIL mutant N109C could be PEGylated by mPEG-MAL within 40 min. And mPEG(MAL)-N109C was revealed to possess superior in vitro stability and antitumor activity than N-terminal specifically PEGylated TRAIL (114-281) (mPEG(ALD)-TRAIL(114-281)). What's more, mPEG(MAL)-N109C exhibited more therapeutic potentials than mPEG(ALD)-TRAIL(114-281) in tumor xenograft model, benefitting from better drug delivery and bioavailability. These results have demonstrated mutated Cys-SH specific PEGylation is an alternative to site-specifically PEGylate TRAIL efficiently and effectively other than N-terminal specific PEGylation.

Novel conjugation of tumor-necrosis-factor-related apoptosis-inducing ligand (TRAIL) with monomethyl auristatin E for efficient antitumor drug delivery.

Monomethyl auristatin E (MMAE) is conjugated with TNF-related apoptosis-inducing ligand (TRAIL) via a linker that is stable in extracellular fluid, while it is cleaved by cathepsin once the conjugate has entered a tumor cell, thus activating the antimitotic mechanism of MMAE. The TRAIL-MMAE conjugate is a conceptually viable therapeutic strategy with improved in vitro antitumor activity, cell circle arrest and specific accumulation in tumor to treat TRAIL-resistant tumors.

Influence of various polymorphic variants of cytochrome P450 oxidoreductase (POR) on drug metabolic activity of CYP3A4 and CYP2B6.

Cytochrome P450 oxidoreductase (POR) is known as the sole electron donor in the metabolism of drugs by cytochrome P450 (CYP) enzymes in human. However, little is known about the effect of polymorphic variants of POR on drug metabolic activities of CYP3A4 and CYP2B6. In order to better understand the mechanism of the activity of CYPs affected by polymorphic variants of POR, six full-length mutants of POR (e.g., Y181D, A287P, K49N, A115V, S244C and G413S) were designed and then co-expressed with CYP3A4 and CYP2B6 in the baculovirus-Sf9 insect cells to determine their kinetic parameters. Surprisingly, both mutants, Y181D and A287P in POR completely inhibited the CYP3A4 activity with testosterone, while the catalytic activity of CYP2B6 with bupropion was reduced to approximately ~70% of wild-type activity by Y181D and A287P mutations. In addition, the mutant K49N of POR increased the CLint (Vmax/Km) of CYP3A4 up to more than 31% of wild-type, while it reduced the catalytic efficiency of CYP2B6 to 74% of wild-type. Moreover, CLint values of CYP3A4-POR (A115V, G413S) were increased up to 36% and 65% of wild-type respectively. However, there were no appreciable effects observed by the remaining two mutants of POR (i.e., A115V and G413S) on activities of CYP2B6. In conclusion, the extent to which the catalytic activities of CYP were altered did not only depend on the specific POR mutations but also on the isoforms of different CYP redox partners. Thereby, we proposed that the POR-mutant patients should be carefully monitored for the activity of CYP3A4 and CYP2B6 on the prescribed medication.