Trastuzumab-MMAU Antibody-Auristatin Conjugates: Valine-Glucoserine Linker with Stabilized Maleimide Conjugation Improves In Vivo Efficacy and Tolerability.

PURPOSE: Antibody-drug conjugates (ADCs) have shown impressive clinical activity with approval of many agents in hematological and solid tumors. However, challenges remain with both efficacy and safety of ADCs. This study describes novel trastuzumab-auristatin conjugates with the hydrophilic MMAE prodrug MMAU, and optimization of a glycopeptide linker leading to a wider therapeutic window. EXPERIMENTAL DESIGN: Trastuzumab was conjugated with auristatin payloads via a series of linkers using a stabilized maleimide handle. The ADCs were characterized in vitro and their relative in vivo anti-tumor efficacies were assessed in HER2+ xenograft models. Relative linker stabilities and the mechanism of linker cleavage were studied using in vitro assays. Toxicity and toxicokinetics of the best performing ADC were evaluated in cynomolgus monkey (cyno). RESULTS: The trastuzumab-MMAU ADC with stabilized glycopeptide linker showed maleimide stabilization and higher resistance to cleavage by serum and lysosomal enzymes compared to a valine-citrulline conjugated trastuzumab ADC (trastuzumab-vc-MMAE). A single dose of 1 or 2 mg/kg of trastuzumab-MMAU at drug-to-antibody ratios (DAR) of 8 and 4 respectively resulted in xenograft tumor growth inhibition, with superior efficacy to trastuzumab-vc-MMAE. Trastuzumab-MMAU DAR4 was tolerated at doses up to 12 mg/kg in cyno, which represents 2- to 4-fold higher dose than that observed with vedotin ADCs, and had increased terminal half-life and exposure. CONCLUSIONS: The optimized trastuzumab-MMAU ADC showed potent antitumor activity and was well tolerated with excellent pharmacokinetics in non-human primates, leading to a superior preclinical therapeutic window. The data supports potential utility of trastuzumab-MMAU for treatment of HER2+ tumors.

A Strategy for Selective Deletion of Autoimmunity-Related T Cells by pMHC-Targeted Delivery.

Autoimmune diseases such as rheumatoid arthritis are caused by immune system recognition of self-proteins and subsequent production of effector T cells that recognize and attack healthy tissue. Therapies for these diseases typically utilize broad immune suppression, which can be effective, but which also come with an elevated risk of susceptibility to infection and cancer. T cell recognition of antigens is driven by binding of T cell receptors to peptides displayed on major histocompatibility complex proteins (MHCs) on the cell surface of antigen-presenting cells. Technology for recombinant production of the extracellular domains of MHC proteins and loading with peptides to produce pMHCs has provided reagents for detection of T cell populations, and with the potential for therapeutic intervention. However, production of pMHCs in large quantities remains a challenge and a translational path needs to be established. Here, we demonstrate a fusion protein strategy enabling large-scale production of pMHCs. A peptide corresponding to amino acids 259-273 of collagen II was fused to the N-terminus of the MHC_II beta chain, and the alpha and beta chains were each fused to human IgG4 Fc domains and co-expressed. A tag was incorporated to enable site-specific conjugation. The cytotoxic drug payload, MMAF, was conjugated to the pMHC and potent, peptide-specific killing of T cells that recognize the collagen pMHC was demonstrated with tetramerized pMHC-MMAF conjugates. Finally, these pMHCs were incorporated into MMAF-loaded 3DNA nanomaterials in order to provide a biocompatible platform. Loading and pMHC density were optimized, and peptide-specific T cell killing was demonstrated. These experiments highlight the potential of a pMHC fusion protein-targeted, drug-loaded nanomaterial approach for selective delivery of therapeutics to disease-relevant T cells and new treatment options for autoimmune disease.

Centyrin ligands for extrahepatic delivery of siRNA.

RNA interference (RNAi) offers the potential to treat disease at the earliest onset by selectively turning off the expression of target genes, such as intracellular oncogenes that drive cancer growth. However, the development of RNAi therapeutics as anti-cancer drugs has been limited by both a lack of efficient and target cell-specific delivery systems and the necessity to overcome numerous intracellular barriers, including serum/lysosomal instability, cell membrane impermeability, and limited endosomal escape. Here, we combine two technologies to achieve posttranscriptional gene silencing in tumor cells: Centyrins, alternative scaffold proteins binding plasma membrane receptors for targeted delivery, and small interfering RNAs (siRNAs), chemically modified for high metabolic stability and potency. An EGFR Centyrin known to internalize in EGFR-positive tumor cells was site-specifically conjugated to a beta-catenin (CTNNb1) siRNA and found to drive potent and specific target knockdown by free uptake in cell culture and in mice inoculated with A431 tumor xenografts (EGFR amplified). The generalizability of this approach was further demonstrated with Centyrins targeting multiple receptors (e.g., BCMA, PSMA, and EpCAM) and siRNAs targeting multiple genes (e.g., CD68, KLKb1, and SSB1). Moreover, by installing multiple conjugation handles, two different siRNAs were fused to a single Centyrin, and the conjugate was shown to simultaneously silence two different targets. Finally, by specifically pairing EpCAM-binding Centyrins that exhibited optimized internalization profiles, we present data showing that an EpCAM Centyrin CTNNb1 siRNA conjugate suppressed tumor cell growth of a colorectal cancer cell line containing an APC mutation but not cells with normal CTNNb1 signaling. Overall, these data demonstrate the potential of Centyrin-siRNA conjugates to target cancer cells and silence oncogenes, paving the way to a new class of anticancer drugs.

Bioanalytical workflow for novel scaffold protein-drug conjugates: quantitation of total Centyrin protein, conjugated Centyrin and free payload for Centyrin-drug conjugate in plasma and tissue samples using liquid chromatography-tandem mass spectrometry.

AIM: Alternative scaffold proteins have emerged as novel platforms for development of therapeutic applications. One such application is in protein-drug conjugates (PDCs), which are analogous to antibody-drug conjugates. METHODOLOGY: Liquid chromatography-mass spectrometry methods for quantitation of total protein, conjugate and free payload for a PDC based on Centyrin scaffold were developed. Tryptic peptides generated from a region of the Centyrin that does not contain a conjugation site, and another that has the conjugation site with the linker-payload attached were used as surrogates of the total and conjugated Centyrin, respectively. CONCLUSION: The methods were successfully applied to analysis of samples from mice to quantify the plasma and tissue concentrations. This same workflow can potentially be applied to other PDCs and site-specific antibody-drug conjugates.

LC/MS/MS Bioanalysis of Protein-Drug Conjugates-The Importance of Incorporating Succinimide Hydrolysis Products.

Bioanalysis of antibody-drug conjugates (ADCs) is challenging due to the complex, heterogeneous nature of their structures and their complicated catabolism. To fully describe the pharmacokinetics (PK) of an ADC, several analytes are commonly quantified, including total antibody, conjugate, and payload. Among them, conjugate is the most challenging to measure, because it requires detection of both small and large molecules as one entity. Existing approaches to quantify the conjugated species of ADCs involve a ligand binding assay (LBA) for conjugated antibody or hybrid LBA/liquid chromatography/tandem mass spectrometry (LC/MS/MS) for quantitation of conjugated drug. In our current work for a protein-drug conjugate (PDC) using the Centyrin scaffold, a similar concept to ADCs but with smaller protein size, an alternative method to quantify the conjugate by using a surrogate peptide approach, was utilized. The His-tagged proteins were isolated from biological samples using immobilized metal affinity chromatography (IMAC), followed by trypsin digestion. The tryptic peptide containing the linker attached to the payload was used as a surrogate of the conjugate and monitored by LC/MS/MS analysis. During method development and its application, we found that hydrolysis of the succinimide ring of the linker was ubiquitous, taking place at many stages during the lifetime of the PDC including in the initial drug product, in vivo in circulation in the animals, and ex vivo during the trypsin digestion step of the sample preparation. We have shown that hydrolysis during trypsin digestion is concentration-independent and consistent during the work flow-therefore, having no impact on assay performance. However, for samples that have undergone extensive hydrolysis prior to trypsin digestion, significant bias could be introduced if only the non-hydrolyzed form is considered in the quantitation. Therefore, it is important to incorporate succinimide hydrolysis products in the quantitation method in order to provide an accurate estimation of the total conjugate level. More importantly, the LC/MS/MS-based method described here provides a useful tool to quantitatively evaluate succinimide hydrolysis of ADCs in vivo, which has been previously reported to have significant impact on their stability, exposure, and efficacy.

Evaluation of a Centyrin-Based Near-Infrared Probe for Fluorescence-Guided Surgery of Epidermal Growth Factor Receptor Positive Tumors.

Tumor-targeted near-infrared fluorescent dyes have the potential to improve cancer surgery by enabling surgeons to locate and resect more malignant lesions where good visualization tools are required to ensure complete removal of malignant tissue. Although the tumor-targeted fluorescent dyes used in humans to date have been either small organic molecules or high molecular weight antibodies, low molecular weight protein scaffolds have attracted significant attention because they penetrate solid tumors almost as efficiently as small molecules, but can be infinitely mutated to bind almost any antigen. Here we describe the use of a 10 kDa protein scaffold, a Centyrin, to target a near-infrared fluorescent dye to tumors that overexpress the epidermal growth factor receptor (EGFR) for fluorescence-guided surgery (FGS). We have developed and optimized the dose and time required for imaging small tumor burdens with minimal background fluorescence in real-time fluorescence-guided surgery of EGFR-expressing tumor xenografts in murine models. We demonstrate that the Centyrin-near-infrared dye conjugate (CNDC) binds selectively to human EGFR+ cancer cells with an EC50 of 2 nM, localizes to EGFR+ tumor xenografts in athymic nude mice and that uptake of the dye in xenografts is significantly reduced when EGFR are blocked by preinjection of excess unlabeled Centyrin. Taken together, these data suggest that CNDCs can be used for intraoperative identification and surgical removal of EGFR-expressing lesions and that Centyrins targeted to other tumor-specific antigens should prove similarly useful in fluorescence guided surgery of cancer. In addition, we demonstrate that the CNDC is detected in the NIR region of the spectrum and can be utilized for fluorescence-guided surgery (FGS). In addition, we propose that with its eventual complete clearance from EGFR-negative tissues and its quantitative retention in the tumor mass for >24 h, a Centyrin-targeted NIR dye should provide excellent tumor contrast when injected at least 6-8 h before initiation of cancer surgery in human patients.

Engineering a targeted delivery platform using Centyrins.

Targeted delivery of therapeutic payloads to specific tissues and cell types is an important component of modern pharmaceutical development. Antibodies or other scaffold proteins can provide the cellular address for delivering a covalently linked therapeutic via specific binding to cell-surface receptors. Optimization of the conjugation site on the targeting protein, linker chemistry and intracellular trafficking pathways can all influence the efficiency of delivery and potency of the drug candidate. In this study, we describe a comprehensive engineering experiment for an EGFR binding Centyrin, a highly stable fibronectin type III (FN3) domain, wherein all possible single-cysteine replacements were evaluated for expression, purification, conjugation efficiency, retention of target binding, biophysical properties and delivery of a cytotoxic small molecule payload. Overall, 26 of the 94 positions were identified as ideal for cysteine modification, conjugation and drug delivery. Conjugation-tolerant positions were mapped onto a crystal structure of the Centyrin, providing a structural context for interpretation of the mutagenesis experiment and providing a foundation for a Centyrin-targeted delivery platform.

Site-specific protein modification using immobilized sortase in batch and continuous-flow systems.

Transpeptidation catalyzed by sortase A allows the preparation of proteins that are site-specifically and homogeneously modified with a wide variety of functional groups, such as fluorophores, PEG moieties, lipids, glycans, bio-orthogonal reactive groups and affinity handles. This protocol describes immobilization of sortase A on a solid support (Sepharose beads). Immobilization of sortase A simplifies downstream purification of a protein of interest after labeling of its N or C terminus. Smaller batch and larger-scale continuous-flow reactions require only a limited amount of enzyme. The immobilized enzyme can be reused for multiple cycles of protein modification reactions. The described protocol also works with a Ca(2+)-independent variant of sortase A with increased catalytic activity. This heptamutant variant of sortase A (7M) was generated by combining previously published mutations, and this immobilized enzyme can be used for the modification of calcium-senstive substrates or in instances in which low temperatures are needed. Preparation of immobilized sortase A takes 1-2 d. Batch reactions take 3-12 h and flow reactions proceed at 0.5 ml h(-1), depending on the geometry of the reactor used.

Cys-scanning disulfide crosslinking and bayesian modeling probe the transmembrane signaling mechanism of the histidine kinase, PhoQ.

Bacteria transduce signals across the membrane using two-component systems (TCSs), consisting of a membrane-spanning sensor histidine kinase and a cytoplasmic response regulator. In gram-negative bacteria, the PhoPQ TCS senses cations and antimicrobial peptides, yet little is known about the structural changes involved in transmembrane signaling. We construct a model of PhoQ signal transduction using Bayesian inference, based on disulfide crosslinking data and homologous crystal structures. The data are incompatible with a single conformation but are instead consistent with two interconverting structures. These states differ in membrane depth of the periplasmic acidic patch and the reciprocal displacement of diagonal helices along the dimer interface. Studies of multiple histidine kinases suggest this repacking might be a common mode of signal transduction in sensor His-kinase receptors. Because a similar scissors model has been ruled out in CheA-linked chemoreceptors, the evidence suggests that sensor His-kinase and CheA-linked receptors possess different signaling mechanisms.

Transmembrane polar interactions are required for signaling in the Escherichia coli sensor kinase PhoQ.

PhoQ is the transmembrane sensor histidine kinase of the bacterial phoPQ two-component system, which detects and responds to divalent cations and to antimicrobial peptides, and can trigger virulence. Despite their ubiquitous importance in bacterial signaling, the structure and mechanism of the sensor kinases are not fully understood. In particular, the mechanism by which the signal is propagated through the transmembrane (TM) region remains unclear. We have identified a critical asparagine residue in the second TM helix of PhoQ. Replacement of this Asn202 with a variety of hydrophobic amino acids results in a protein that is blind to signal, fails to activate transcription of PhoQ-dependent genes, and abrogates transcription when coexpressed with wild-type PhoQ. Analysis of other two-component kinase sequences indicated that many such proteins contain similarly conserved polar residues, and the structure of one such domain shows a polar residue proximal to an extended cavity near the center of the TM bundle. We therefore examined the role of Asn202 in PhoQ. Our analysis indicated that its kinase function is dependent on the polarity of Asn202, rather than its precise structure or position in the TM region; it can be displaced up or down one turn of TM helix 2, or even moved to the adjacent TM helix 1. The presence of polar TM amino acids among many diverse sensor kinases suggest a widespread mechanism of two-component signal transduction; we speculate that they might stabilize underpacked water-containing cavities that can accommodate conformational changes required for switching from phosphatase to kinase-competent conformations.

Engineered single- and multi-cell chemotaxis pathways in E. coli.

We have engineered the chemotaxis system of Escherichia coli to respond to molecules that are not attractants for wild-type cells. The system depends on an artificially introduced enzymatic activity that converts the target molecule into a ligand for an E. coli chemoreceptor, thereby enabling the cells to respond to the new attractant. Two systems were designed, and both showed robust chemotactic responses in semisolid and liquid media. The first incorporates an asparaginase enzyme and the native E. coli aspartate receptor to produce a response to asparagine; the second uses penicillin acylase and an engineered chemoreceptor for phenylacetic acid to produce a response to phenylacetyl glycine. In addition, by taking advantage of a 'hitchhiker' effect in which cells producing the ligand can induce chemotaxis of neighboring cells lacking enzymatic activity, we were able to design a more complex system that functions as a simple microbial consortium. The result effectively introduces a logical 'AND' into the system so that the population only swims towards the combined gradients of two attractants.

Determination of the physiological dimer interface of the PhoQ sensor domain.

PhoQ is the transmembrane sensor kinase of the phoPQ two-component system, which detects and responds to divalent cations and antimicrobial peptides and can trigger bacterial virulence. Despite their ubiquity and importance in bacterial signaling, the structure and molecular mechanism of the sensor kinases is not fully understood. Frequently, signals are transmitted from a periplasmic domain in these proteins to the cytoplasmic kinase domains via an extended dimeric interface, and the PhoQ protein would appear to follow this paradigm. However, the isolated truncated periplasmic domain of PhoQ dimerizes poorly, so it has been difficult to distinguish the relevant interface in crystal structures of the PhoQ periplasmic domain. Thus, to determine the arrangement of the periplasmic domains of Escherichia coli PhoQ in the physiological homodimer, disulfide-scanning mutagenesis was used. Single cysteine substitutions were introduced along the N-terminal helix of the periplasmic region, and the degree of cross-linking in each protein variant was determined by Western blotting and immunodetection. The results were subjected to periodicity analysis to generate a profile that provides information concerning the C(beta) distances between corresponding residues at the interface. This profile, together with a rigid-body search procedure, side-chain placement, and energy minimization, was used to build a model of the dimer arrangement. The final model proved to be highly compatible with one of the PhoQ crystal structures, 3BQ8, indicating that 3BQ8 is representative of the physiological arrangement. The model of the periplasmic region is also compatible with a full-length PhoQ protein in which a four-helix bundle forms in the membrane. The membrane four-helix bundle has been proposed for other sensor kinases and is thought to have a role in the mechanism of signal transduction; our model supports the idea that signaling through a membrane four-helix bundle is a widespread mechanism in the transmembrane sensor kinases.

Investigation of the mechanism of resistance to third-generation cephalosporins by class C beta-lactamases by using chemical complementation.

The widespread use of antibiotics to treat bacterial infections has led to the continuing challenge of antibiotic resistance. For beta-lactam antibiotics, the most common form of resistance is the expression of beta-lactamase enzymes, which inactivate the antibiotics by cleavage of the beta-lactam core. In this study, chemical complementation, which is a general method to link the formation or cleavage of a chemical bond to the transcription of a reporter gene in vivo, was employed in combination with combinatorial mutagenesis to study the mechanism by which the class C beta-lactamase P99 might evolve resistance to the commonly administered third-generation cephalosporin cefotaxime. The chemical complementation system was first shown to be able to distinguish between the wild-type (wt) class C beta-lactamase P99 and the clinically isolated extended-spectrum class C beta-lactamase GC1 in the presence of cefotaxime. The system was then employed to evaluate the activity of mutants of wt P99 towards cefotaxime. A number of single-point mutations at position 221 (Tyr in wt P99) were identified that conferred resistance towards inhibition by cefotaxime, with as much as a 2000-fold increase in k(cat) and a 100-fold increase in k(cat)/K(M) (k(cat)=the rate of catalysis; K(M)=the Michaelis constant), as compared to those of the wt enzyme. Finally, the chemical complementation system was employed in a high-throughput screen to identify a number of mutants of P99 that have multiple mutations around the substrate-binding pocket that increase resistance towards cefotaxime inhibition. The catalytic turnover of cefotaxime by the most active mutant identified was 5500 times higher than that of the wt P99. The resistant mutants suggest a mechanism by which a number of mutations can confer resistance by increasing the flexibility of the Omega loop and altering the positioning of residue 221. Thus, as illustrated in this study, chemical complementation has the potential to be used as a high-throughput screen to study a wide range of enzyme-drug interactions.

Mixed quantum mechanical/molecular mechanical (QM/MM) study of the deacylation reaction in a penicillin binding protein (PBP) versus in a class C beta-lactamase.

The origin of the substantial difference in deacylation rates for acyl-enzyme intermediates in penicillin-binding proteins (PBPs) and beta-lactamases has remained an unsolved puzzle whose solution is of great importance to understanding bacterial antibiotic resistance. In this work, accurate, large-scale mixed ab initio quantum mechanical/molecular mechanical (QM/MM) calculations have been used to study the hydrolysis of acyl-enzyme intermediates formed between cephalothin and the dd-peptidase of Streptomyces sp. R61, a PBP, and the Enterobacter cloacae P99 cephalosporinase, a class C beta-lactamase. Qualitative and, in the case of P99, quantitative agreement was achieved with experimental kinetics. The faster rate of deacylation in the beta-lactamase is attributed to a more favorable electrostatic environment around Tyr150 in P99 (as compared to that for Tyr159 in R61) which facilitates this residue's function as the general base. This is found to be in large part accomplished by the ability of P99 to covalently bind the ligand without concurrent elimination of hydrogen bonds to Tyr150, which proves not to be the case with Tyr159 in R61. This work provides an essential foundation for further work in this area, such as selecting mutations capable of converting the PBP into a beta-lactamase.

Correlation between catalytic efficiency and the transcription read-out in chemical complementation: a general assay for enzyme catalysis.

High-throughput assays for enzyme catalysis that can be applied to a broad range of chemical reactions are key to advances in directed evolution and proteomics. Recently, we reported such a general assay, chemical complementation, which links enzyme catalysis to reporter gene transcription in vivo using the yeast three-hybrid assay. In this proof-of-principle experiment, it was shown that a wild-type beta-lactamase enzyme could be isolated from a pool of inactive mutants using a lacZ screen. Ideally, however, such an assay should be able to distinguish enzymes based on their catalytic activity. Thus, here, we set out to determine if the catalytic efficiency of an enzyme variant does in fact correlate with its level of transcription activation in the chemical complementation assay. First, the reaction mechanism for the cleavage of the beta-lactam substrate used in the chemical complementation proof-of-principle experiment was determined. Then a series of beta-lactamase variants was designed to span several orders of magnitude in k(cat)/K(m). The activity of each variant was determined both in vitro using purified enzyme and in vivo in the chemical complementation transcription assay. Beta-lactamase variants spanning three-orders of magnitude in k(cat)/K(m) could be distinguished in the assay, and the catalytic efficiency of each variant correlated with its level of transcription activation in vivo. These results establish the suitability of chemical complementation for the directed evolution of enzymes with improvements in catalytic activity and for profiling the relative substrate specificities of a group of enzymes in proteomics applications.

Identification of residues critical for catalysis in a class C beta-lactamase by combinatorial scanning mutagenesis.

Despite their clinical importance, the mechanism of action of the class C beta-lactamases is poorly understood. In contrast to the class A and class D beta-lactamases, which contain a glutamate residue and a carbamylated lysine in their respective active sites that are thought to serve as general base catalysts for beta-lactam hydrolysis, the mechanism of activation of the serine and water nucleophiles in the class C enzymes is unclear. To probe for residues involved in catalysis, the class C beta-lactamase from Enterobacter cloacae P99 was studied by combinatorial scanning mutagenesis at 122 positions in and around the active site. Over 1000 P99 variants were screened for activity in a high-throughput in vivo antibiotic resistance assay and sequenced by 96-capillary electrophoresis to identify residues that are important for catalysis. P99 mutants showing reduced capability to convey antibiotic resistance were purified and characterized in vitro. The screen identified an active-site hydrogen-bonding network that is key to catalysis. A second cluster of residues was identified that likely plays a structural role in the enzyme. Otherwise, residues not directly contacting the substrate showed tolerance to substitution. The study lends support to the notion that the class C beta-lactamases do not have a single residue that acts as the catalytic general base. Rather, catalysis is affected by a hydrogen-bonding network in the active site, suggesting a possible charge relay system.