Mark N. Shahan

Comparison of Cell-Based and Cell-Free Protocols for Producing Target Proteins from the Arabidopsis thaliana Genome for Structural Studies

We describe a comparative study of protein production from 96 Arabidopsis thaliana open reading frames (ORFs) by cell‐based and cell‐free protocols. Each target was carried through four pipeline protocols used by the Center for Eukaryotic Structural Genomics (CESG), one for the production of unlabeled protein to be used in crystallization trials and three for the production of 15N‐labeled proteins to be analyzed by 1H‐15N NMR correlation spectroscopy. Two of the protocols involved Escherichia coli cell‐based and two involved wheat germ cell‐free technology. The progress of each target through each of the protocols was followed with all failures and successes noted. Failures were of the following types: ORF not cloned, protein not expressed, low protein yield, no cleavage of fusion protein, insoluble protein, protein not purified, NMR sample too dilute. Those targets that reached the goal of analysis by 1H‐15N NMR correlation spectroscopy were scored as HSQC+ (protein folded and suitable for NMR structural analysis), HSQC± (protein partially disordered or not in a single stable conformational state), HSQC− (protein unfolded, misfolded, or aggregated and thus unsuitable for NMR structural analysis). Targets were also scored as X− for failing to crystallize and X+ for successful crystallization. The results constitute a rich database for understanding differences between targets and protocols. In general, the wheat germ cell‐free platform offers the advantage of greater genome coverage for NMR‐based structural proteomics whereas the E. coli platform when successful yields more protein, as currently needed for crystallization trials for X‐ray structure determination. Proteins 2005.

Solution structure of thioredoxin h1 from Arabidopsis thaliana

Present in virtually every species, thioredoxins catalyze disulfide/dithiol exchange with various substrate proteins. While the human genome contains a single thioredoxin gene, plant thioredoxins are a complex protein family. A total of 19 different thioredoxin genes in six subfamilies has emerged from analysis of the Arabidopsis thaliana genome. Some function specifically in mitochondrial and chloroplast redox signaling processes, but target substrates for a group of eight thioredoxin proteins comprising the h subfamily are largely uncharacterized. In the course of a structural genomics effort directed at the recently completed A. thaliana genome, we determined the structure of thioredoxin h1 (At3g51030.1) in the oxidized state. The structure, defined by 1637 NMR‐derived distance and torsion angle constraints, displays the conserved thioredoxin fold, consisting of a five‐stranded β‐sheet flanked by four helices. Redox‐dependent chemical shift perturbations mapped primarily to the conserved WCGPC active‐site sequence and other nearby residues, but distant regions of the C‐terminal helix were also affected by reduction of the active‐site disulfide. Comparisons of the oxidized A. thaliana thioredoxin h1 structure with an h‐type thioredoxin from poplar in the reduced state revealed structural differences in the C‐terminal helix but no major changes in the active site conformation.

Wheat Germ Cell‐Free Expression System for Protein Production

The Center for Eukaryotic Structural Genomics, in cooperation with Ehime University and CellFree Sciences, has developed a novel wheat germ cell‐free technology for the production of eukaryotic proteins. Protein production and purification are robust and scalable for high‐throughput applications. The protocols have been used to express and purify proteins from Arabidopsis thaliana, human, mouse, rat and zebra fish. This unit describes expression and purification protocols for both small‐scale testing (microgram) and large‐scale production (milligram) of N‐His6‐ and N‐GST‐tagged proteins. The methods described in this unit can be used to produce both unlabeled and labeled proteins required for structure‐based determinations by NMR spectroscopy or X‐ray crystallography.

Letter to the Editor: Hypothetical protein At2g24940.1 from Arabidopsis thaliana has a cytochrome b5 like fold ∗

Progesterone is believed to exert rapid non-genomic actions through its interaction with membrane associated progesterone receptors (MAPRs) (Bramley, 2003; Li and O’Malley, 2003). BLAST sequence searches (Altieri et al., 1995) for mammalian MAPRs and putative MAPRs from plants have identified that these proteins all contain a cytochrome b5-like ligandbinding domain (Mifsud and Bateman, 2002). Interestingly, unlike cytochrome b5 itself, these MAPRs domains appear not to bind heme and not to be involved in redox reactions. Their distinct biological functions suggest that these steroid receptors adopt the cytochrome b5 domain as a template in order to build their own ligand-binding pockets (Mifsud and Bateman, 2002). The Center for Eukaryotic Structural Genomics is engaged in determining the three-dimensional structures of novel proteins from eukaryotic gene families. Its target selection algorithm selected Arabidopsis thaliana putative protein At2g24940.1 for structure determination. The biochemical function of At2g24940.1 currently is unknown. Its ∼40% sequence identity with mammalian MAPR suggests that At2g24940.1 may act as a steroid binding protein. In addition, its sequence is distantly similar to that of cytochrome b5 (Figure 1). Here we describe the threedimensional structure of At2g24940.1 as determined by NMR spectroscopy. At present, no structure of a MAPR is available from the Protein Data Bank. Thus, the structure of At2g24940.1 may provide clues to the function of a class of steroid binding proteins in plants.

Abstract 1838: Efficacy of ribonuclease QBI-139 in combination with standard of care therapies

RNA has been recognized as a drug target for cancer therapy, as evidenced by the ongoing clinical trials of RNAi and antisense therapies. An alternative approach that circumvents the delivery and stability issues of those drugs is to harness the activity of naturally occurring RNA degradation enzymes. Human ribonuclease (RNase) variants have been generated with diminished binding to their natural inhibitor inside cells, which allows the new proteins to kill cancer cells. One of these RNase variants, referred to as QBI-139, demonstrated efficacy against multiple tumor types in xenograft models of human cancer and was selected for an ongoing first in human Phase I clinical trial. Non-small cell lung and ovarian cancer are two of the indications where QBI-139 has shown significant efficacy as a single agent in xenograft models. In planning for the next stage of clinical trials, the impact of combinations of QBI-139 plus standard of care agents in these diseases was determined both in vitro and in vivo. QBI-139 showed additive and synergistic activity in combination with cisplatin and docetaxel, which are standard of care for non-small cell lung and ovarian cancer, respectively. Fixed ratios of QBI-139 and the combination drug based on their EC50 values were tested. The Combination Index (CI) was determined using median effect analysis. Synergy was demonstrated with QBI-139 in combination with cisplatin against non-small cell lung cancer (A549 and H1975). An additive effect of QBI-139 and cisplatin was seen in ovarian (Ovcar-3) cancer. In ovarian cancer (SK-OV-3), a combination of docetaxel and QBI-139 was synergistic with QBI-139. The benefit of the combinations were repeated in xenograft models of both cancers and were signifcat relative to each agent alone. Non-small cell lung and ovarian cancer represent critical unmet needs and new, differentiated treatment approaches such as QBI-139 combination therapy are critical. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1838. doi:1538-7445.AM2012-1838

Abstract 4405: RNases: Approaching RNA as a target for cancer therapy.

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Ribonucleases (RNases) are a unique class of anti-cancer drugs that target the RNA in cancer cells without the complications of delivering RNA. Efficacious RNases have diminished binding to the naturally occurring RNase inhibitor protein in the cytosol of cells. Since human RNase I binds tightly to the inhibitor, changes to the amino acid residues in the contact region were made and resulted in the lead candidate referred to as QBI-139. This variant maintains 95% sequence identity with the wild type human RNase 1. Broad efficacy and a strong therapeutic window were seen across multiple tumor types in xenograft models and QBI-139 was advanced to an ongoing Phase I clinical trial. For the Investigational New Drug application package, rat and dog toxicology studies were used to aid in selection of the starting dose for humans. Exposure levels (pharmacokinetics, PK) are an important factor in evaluating whether toxicity and efficacy will be comparable across the species. A complicating issue in the evaluation of the results is the range of routes of administration across species, specifically intraperitoneal injection (mice), intravenous injection (rat and dog) and intravenous infusion (human). One parameter that is consistent is rapid clearance, which is likely due to the small size of QBI-139 (15 kiloDaltons). With preliminary results from the early stage clinical trial, the exposure levels across the experiments will be compared. The results demonstrate that the exposure profiles in the clinic where the route of administration is intravenous infusion, is most similar to that in the mouse efficacy models, which is intraperitoneal injection. A human RNase variant, called QBI-139 provides an innovative drug development approach to targeting RNA. In model systems, QBI-139 has shown broad in vivo anti-tumor activity alone and in combination with standard of care therapies and a strong tolerability profile. The knowledge gained about the impact of PK on therapeutic effect will be valuable as the method of administration is selected and confirmed in the next stage of clinical trials. Citation Format: Laura E. Strong, John Kink, Baigen Mei, Mark Shahan, Ronald T. Raines. RNases: Approaching RNA as a target for cancer therapy. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4405. doi:10.1158/1538-7445.AM2013-4405

Abstract 2573: Combinations of QBI-139, a clinical stage ribonuclease drug

Proceedings: AACR 102nd Annual Meeting 2011‐‐ Apr 2‐6, 2011; Orlando, FL RNA has been recognized as a drug target for cancer therapy, as evidenced by the ongoing clinical trials of RNAi and antisense therapies. An alternative approach that circumvents the delivery and stability issues of RNAi and antisense is to harness the activity of naturally occurring enzymes that degrade RNA. The ribonuclease approach is partially validated by objective clinical responses seen in clinical trials of ribonuclease (RNase) discovered in frog eggs. Our research has focused on creating variants of human RNases with the ability of frog RNase to evade the human RNase inhibitor. Diminished binding of mammalian RNases to the inhibitor have been shown to endow the RNase with the ability to kill cancer cells. One of the RNase variants, called QBI-139, demonstrated particularly strong anti-cancer activity against a variety of tumor types in in vivo models. QBI-139 is differentiated from the frog RNase because QBI-139 is 95 % identical to a native human RNase, while the frog RNase is only 20% similar. This difference may be what has led to a significantly improved tolerability profile for QBI-139 across species. QBI-139 was selected for clinical development and is now in a Phase I trial of solid tumors. In charting the clinical development path for QBI-139, the ability of the RNase to work in combination with a variety of standard of care agents had to be evaluated. Based in the single agent activity of QBI-139 in colon, non-small cell lung, ovarian and pancreatic cancers, the potency of QBI-139 in combination with the standard of care agents for these indications was explored. In in vitro screening, QBI-139 has demonstrated additive and synergistic activity in combination with cisplatin, 5-FU and docetaxel. The treatments were evaluated using fixed ratios of QBI-139 and the combination drug based on their EC50 values. The Combination Index (CI) was determined using median effect analysis. A CI of 1 indicates additive effects, whereas a CI < 1 indicates synergy. Synergy was demonstrated with QBI-139 in combination with cisplatin against solid tumors of non-small cell lung cancer (A549 and H1975) and an additive effect in ovarian (Ovcar-3) cancer. 5-FU and docetaxel also were synergistic with QBI-139 against solid tumors of colon (HCT-116) and ovarian (SK-OV-3) cancer, respectively.The goal of this project is development of an innovative, differentiated therapy, QBI-139, to provide clinical benefit to patients with colon, non-small cell lung and ovarian cancer, which are among the seven deadliest forms of cancer in the United States. In addition to being serious diseases, effective treatment of these cancers also represent unmet medical needs. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 2573. doi:10.1158/1538-7445.AM2011-2573

Abstract 5390: QBI-139, a human RNase variant in a phase I trial, works in combination

The EVade™ Ribonucleases (RNases) are a new class of anti-cancer drugs that target the RNA in cancer cells. These EVade™ RNases are variants of native mammalian RNases that have been converted to anti-cancer agents by diminishing their binding to their natural inhibitor protein. One of these EVade™ RNases, called QBI-139, has advanced to human clinical trials. QBI-139 shares 95% sequence identity with the wild type human RNase 1. QBI-139 has shown tumor growth inhibition against a variety of cancers, including: colon, non-small cell lung, ovarian, pancreatic and prostate cancers. QBI-139 is currently in a Phase I clinical trial in patients with solid tumors. A frog RNase called Ranpirnase (Alfacell) that shares only 20% sequence identity with human RNase 1 has been in clinical trials and responses have been seen in esophageal, non-small cell lung and breast cancers as well as unresectable mesothelioma. The next stage of clinical development of QBI-139 will require selection of a specific indication as well as an existing treatment regimen to be used in combination with QBI-139. The Phase I trial may provide insights about sensitive tumor types. However, mouse models of cancer will be used to select the combinations for development. QBI-139 has shown efficacy against non-small cell lung cancer (A549 cell line) in xenograft models. Cisplatin is a standard of care treatment for non-small cell lung cancer in patients. The ability of QBI-139 and cisplatin to work together to cause tumor growth inhibition was determined in a non-small cell lung cancer xenograft model (A549 cell line). Cisplatin (3 mg/kg) and QBI-139 (15 mg/kg) were used as controls. The combination of cisplatin and QBI-139 (3 and 15 mg/kg, respectively) was also tested. Cisplatin treatment resulted in tumor growth inhibition (TGI) of 60% while QBI-139 caused 76% TGI. The combination of the two drugs resulted in a TGI of 96%, strongly supporting the use of QBI-139 in combination with cisplatin for the treatment of lung cancer. Even the combination of the two drugs exhibited minimal impact on the health of the animals. EVade™ Ribonucleases are novel agents that provide a new way to target the RNA in cancer cells and are an innovative new tool for oncologists. QBI-139 has a variety of benefits, including: a novel mechanism of action with clinical support, broad in vivo anti-tumor activity alone and in combination with standard of care therapies and a strong tolerability profile. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 5390.

Abstract C42: QBI‐139, a human RNase variant in a phase I trial, has broad in vivo efficacy

The RNA in cancer cells is the target of a new class of anti‐cancer drugs: the EVade™ Ribonucleases (RNases). The EVade™ RNases are variants of native RNases that have been modified to diminish their binding to their natural inhibitor, allowing the EVade™ RNases to kill cancer cells. These RNase variants are broadly efficacious, competitive with approved drugs. A lead candidate, called QBI‐139, has been selected for clinical devleopment and is currently in a Phase I clinical trial in patients with solid tumors. The next stage of development requires selection of a specific indication for a Phase II trial of QBI‐139. In addition to the Phase I trial, mouse models of cancer will be used to select the tumor type for future development. The efficacy of QBI‐139 in in vivo models of human cancer (e.g. lung, colon and prostate) has been examined relative to FDA approved cancer therapies. QBI‐139 was tested at different doses in nude mouse xenograft models using a variety of tumor types and multiple cell lines within the different tumor types (e.g. prostate cancer cell lines DU145 and PC3). Each model compared the efficacy of QBI‐139 with control groups which included: vehicle, and an effective commercial chemotherapeutic specific for each tumor type (e.g. docetaxel, 5‐fluorouracil and cisplatin). The human cancer cells were grown in culture and approximately 5 ×106 cells were implanted subcutaneously into the flanks of (nu/nu) nude mice. The tumors are allowed to grow to ∼75 mm3 prior to treatment and all drugs were administered intraperitoneally (ip) once weekly. Calculated tumor volumes in mm3 and animal body weights (g) were recorded twice weekly. The percent tumor growth inhibition (TGI) used as a measure of efficacy was determined by comparing final and starting tumor volumes of groups treated with drug relative to the vehicle. A favorable response was seen for QBI‐139 against prostate cancer (DU145) compared to docetaxel. The TGI of QBI‐139 was 83% compared to 84% for docetaxel. Treatment with QBI‐139 did not lead to observable toxicity and the mice all gained body during the course of treatment. In contrast, docetaxel treatment resulted in death for one mouse after 4 treatments (day 42) and another after 6 treatments (day 52). In colon cancer (HT‐29), the efficacy of QBI‐139 was 58%, similar to the results for 5‐fluorouracil (54%). Additional results will be provided, including for multiple cell lines of human non‐small cell lung cancer. QBI‐139 is a broadly efficacious agent that provides a new way to attack the RNA in cancer cells and an innovative new tool for oncologists. The results above will be considered with the results of the Phase I trial to select an indication for a Phase II clinical trial. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):C42.