Site-specific conjugation technology frequently relies on antibody engineering to incorporate rare or non-natural amino acids into the primary sequence of the protein. However, when the primary sequence is unknown or when antibody engineering is not feasible, there are very limited options for site-specific protein modification. We have developed a transglutaminase-mediated conjugation that incorporates a thiol at a “privileged” location on deglycosylated antibodies (Q295). Perhaps surprisingly, this conjugation employs a reported transglutaminase inhibitor, cystamine, as the key enzyme substrate. The chemical incorporation of a thiol at the Q295 site allows for the site-specific attachment of a plethora of commonly used and commercially available payloads via maleimide chemistry. Herein, we demonstrate the utility of this method by comparing the conjugatability, plasma stability, and in vitro potency of these site-specific antibody–drug conjugates (ADCs) with analogous endogenous cysteine conjugates. Cytotoxic ADCs prepared using this methodology are shown to exhibit comparable in vitro efficacy to stochastic cysteine conjugates while displaying dramatically improved plasma stability and conjugatability. In particular, we note that this technique appears to be useful for the incorporation of highly hydrophobic linker payloads without the addition of PEG modifiers. We postulate a possible mechanism for this feature by probing the local environment of the Q295 site with two fluorescent probes that are known to be sensitive to the local hydrophobic environment. In summary, we describe a highly practical method for the site-specific conjugation of genetically nonengineered antibodies, which results in plasma-stable ADCs with low intrinsic hydrophobicity. We believe that this technology will find broad utility in the ADC community.

Over the past two decades, antibody drug conjugates (ADCs) and small molecule drug conjugates (SMDCs) have widely employed valine-citruline and related cathepsin-cleavable linkers due to their stability in plasma and their rapid cleavage by lysosomal cathepsins. However, a number of recent studies have illustrated that these linkers are subject to cleavage by exogenous enzymes such as Ces1C and neutrophil elastase, thus resulting in off-target release of drug. As such, there is a need to diversify the portfolio of ADC linkers in order to overcome nonspecific drug release. Rather than targeting cathepsins, we began with an “enzyme agnostic” screen in which a panel of 75 peptide FRET pairs were screened for cleavage in lysosomal extracts and in plasma. Unexpectedly, a series of Asn-containing peptides emerged from this screen as being cleaved far more quickly than traditional ValCit-type linkers while retaining excellent stability in plasma. Catabolism studies demonstrated that these linkers were cleaved by legumain, an asparaginyl endopeptidase that is overexpressed in a variety of cancers and is known to be present in the lysosome. MMAE-containing ADCs that incorporated these new linkers were shown to exhibit highly potent and selective cytotoxicity, comparable to analogous ValCit ADCs. Importantly, the Asn-containing linkers were shown to be completely stable to human neutrophil elastase, an enzyme thought to be responsible for the neutropenia and thrombocytopenia associated with ValCitPABC-MMAE ADCs. The legumain-cleavable ADCs were shown to have excellent stability in both mouse and human serum, retaining >85% of the drug after 1 week of incubation. Moreover, the corresponding small molecule FRET pairs exhibited <10% cleavage after 18 h in mouse and human serum. On the basis of these results, we believe that these new linkers (AsnAsn in particular) have significant potential in both ADC and SMDC drug delivery applications.

No abstract

Purpose: To evaluate the chemical and physical stability of an admixture containing cefepime and vancomycin in a single volume of lactated Ringer solution at refrigerated temperatures. Methods: Cefepime 2000 mg and vancomycin 1000 mg were, respectively, reconstituted with 10 and 20 mL of sterile water for injection (SWFI) per manufacturer instructions. This resulted in cefepime and vancomycin concentrations of 200 and 50 mg/mL, respectively. The resulting cefepime and vancomycin solutions at 10 and 20 mL, respectively, were drawn up and injected into 1000 mL lactated Ringer solution. Aliquot samples were obtained on days 0 to 9, visually inspected for gross incompatibility, and then stored at −80°C. Samples were thawed on the day of the analysis and run through ultraperformance liquid chromatography. Area under the concentration-time curve (AUC) on each day was compared with baseline AUC values. Chemical stability was defined as an AUC more than 93% of the baseline value. Results: No evidence of gross physical incompatibility was observed by visual inspection. Cefepime and vancomycin replicants were more than 94.5% and 98% of baseline AUC values. Therefore, all sample replicants were found to be more than 93% of their baseline AUC value. Conclusion: An admixture containing cefepime 2000 mg and vancomycin 1000 mg in 1000 mL lactated Ringer solution appears to be chemically and physically stable at refrigerated temperatures for up to 9 days.

No abstract