Send to printer »

Feature Articles : Dec 1, 2009 (Vol. 29, No. 21)

China Takes Aim at RNAi

Government Agencies Are Funding Many of the Novel Research Strategies Currently Under Development
  • Vicki Glaser

China-based companies, others with facilities or collaborative ventures in China, and academic groups all committed to advancing RNA-interference (RNAi) research and therapeutic drug development were well-represented at Select Biosciences’ “RNAi Asia” meeting held recently in Kunshan, China. Most of the siRNA work going on in China is being funded by the government, with industry only playing a small role at present, according to Patrick Lu, Ph.D., president and CEO of Sirnaomics.

Some local governments—such as the Suzhou-Kunshan region where the conference took place and which promotes itself as the RNAi Valley—“want to use RNAi as an entry point into the biopharmaceutical arena and are putting a lot of effort and money into RNAi research,” Dr. Lu said.

Ling Qin, Ph.D., professor of orthopedic translational research in the department of orthopedics and traumatology at the Chinese University of Hong Kong, described his group’s work using siRNA to treat musculoskeletal disorders and, in particular, osteoporosis. The group is developing a bone anabolic siRNA therapeutic targeting severe postmenopausal osteoporosis.

Osteoporosis is an attractive target for RNAi because genes associated with the regulation of bone formation and resorption have been identified, and siRNA molecules tend to accumulate in the bone marrow after systemic administration, Ge Zhang, Ph.D., a research assistant professor in Dr. Qin’s laboratory, told GEN.

The incidence of osteoporotic fracture has doubled over the past three decades in Hong Kong. While various drugs are available to prevent and treat postmenopausal osteoporosis, except for parathyroid hormone (PTH) they all act by inhibiting bone resorption to prevent further bone loss; they do not stimulate the formation of new bone. Dr. Zhang emphasized the unmet medical need for bone anabolic therapies that do not promote bone resorption, as is the case with PTH.

“Our collaborators identified casein kinase-2 interacting protein-1 (CKIP-1) as a newly discovered intracellular negative regulator gene of bone formation without stimulating bone resorption,” he said. Dr.  Qin’s group has shown that the bone volume/total volume ratio and trabecular bone thickness and number are higher and the trabecular space is lower in female CKIP-1 knockout mice compared to wild type.

Furthermore, silencing of CKIP-1 in mice caused no abnormalities in major organs such as the liver or kidney. The group has screened a series of siRNA sequences using rat osteoblast cells and monkey and human osteoblasts in culture and has identified a cross-species siRNA with high knockdown efficiency, according to Tang Tao, Ph.D., a postdoctoral research fellow in Dr. Qin’s group.

Dr. Zhang identified the main challenge in developing a systemic siRNA drug as the need to deliver relatively large doses, increasing the cost of treatment, and the risk for harmful side effects. The group is developing a bone-targeted delivery system to enhance the tissue selectivity and knockdown efficiency of systemically administered CKIP-1 siRNA. They have also demonstrated that linking the siRNA to polyethylenimine increases its accumulation in the bone and bone marrow compared to delivery of free CKIP-1 siRNA. The next step is to evaluate the drug’s pharmacokinetics and dose-effect following systemic delivery in an aged, osteoporotic rat model.

Screening

Guangzhou, China-based RiboBio is using an EdU assay for siRNA screening. RiboBio offers functional genomics and drug target validation services and produces natural and unnatural oligonucleotides for RNAi applications. The company collaborated with the Guangzhou Institute of Biomedicine and Health to develop high-throughput siRNA synthesis and RNAi-screening platforms.

Bill Zhang, Ph.D., head of science and technology at RiboBio, provided an example of how the company screens its library of 3,000 human gene siRNAs for their effects on cell viability/cell proliferation using the Click-iT® EdU (5-ethynyl-2´-deoxyuridine) assay technology, an alternative to the BrdU assay for detecting and quantifying newly synthesized DNA that does not require DNA denaturation.

The EdU becomes incorporated into the growing DNA chain, and when exposed to a fluorescent azide (Cell-Light™ Apollo-azide) emits a fluorescent signal that indicates active cell division. The EdU staining protocol takes less than 2.5 hours, according to Dr. Zhang. Combining EdU with other dyes in the same experiment allows for multiplexing and simultaneous evaluation for cell-cycle proliferation, apoptosis, and phenotype, as well as cytotoxicity and cell-signaling pathway activation.

RiboBio’s screening campaign identified groups of siRNAs that either enhance or inhibit cell proliferation, leading to the description of 13 genes required for cell cycle progression.

Therapeutics

“RNAi as a tool for genome-wide screening is in its early days,” said Hakim Djaballah, Ph.D., director of the high-throughput screening core facility at Memorial Sloan Kettering Cancer Center (MSKCC), and the challenges it presents largely reflect the overall poor quality of the early findings and misleading conclusions published in the scientific literature.

“Off-target effects were too often ignored,” and the belief still persists that siRNA sequences are specific and will only hit the intended target, he added. Chemical modifications to the passenger strand of the siRNA hairpin are helping to minimize these off-target effects and remedy this problem.

The data generated from genome-wide siRNA screens and, in particular, the high hit rates, make it difficult to prioritize hits for further analysis. “We have struggled with different ways to analyze the data,” Dr. Djaballah said. “Most of the fancy methods described don’t really make sense,” he asserted, adding, “I gave up on the up-front algorithms. If something comes out of a screen we want to make sure it is not an artifact before going after the biology.” Recently, his group began using back-end knowledge-based strategies for RNAi hit prioritization, including relying on database searching to rule out unlikely candidates. “We have also built some database tools around 3´-UTR regions.”

Researchers at MSKCC are using chemically modified siRNA duplexes to perform genome-wide RNAi screens to identify genes involved in cell-cycle regulation and cell viability and proliferation. At “RNAi Asia,” Dr. Djaballah presented data from RNAi screens intended to probe the biogenesis of microRNAs (miRNAs) and to help define the role of miRNAs in the transformation of cancer cells. The miRNA signatures described to date have shown that in cancer cells some miRNAs are greatly overexpressed and others may be deleted.

In addition to studying the rewiring of cancer cells, the group is employing drug modifier screens to understand why some drug candidates fail in late-stage clinical testing. MSKCC is using high-content assays to screen drug-sensitive and drug-resistant cells against high and low concentrations of a drug to identify genes that affect drug sensitivity in a defined patient population. This information could then be used to predict patient response and drive patient recruitment and selection for clinical trials.

Dr. Lu spoke about the work under way at Sirnaomics to develop a cancer therapeutic targeting breast carcinoma and non-small-cell lung carcinoma using a multitargeted siRNA cocktail designed to silence multiple disease-related genes. At the core of this program is Sirnaomics’ algorithm for siRNA design and its Tri-Blocker™ platform for producing the active pharmaceutical ingredient (API).

Using a polymer-based carrier for systemic delivery of the API, the company has validated its siRNA cocktails in mouse xenograft and orthotopic models. Sirnaomics is exploring other delivery vehicles—including ligand-directed cell targeting, infrared-activated nanoparticles, and nanomicrospheres for oral delivery—as well as combination therapy with monoclonal antibodies and other drugs to enhance the antitumor effects of the siRNA cocktails.

Design of the cocktail begins with a predictive algorithm that uses pathway analysis to select sequences most likely to inhibit disease gene-related targets. The algorithm aims to optimize the thermodynamic properties and RNA-induced silencing complex (RISC) binding activity of the siRNAs, avoid immune-stimulating motifs, minimize off-target effects, protect against potential intellectual property conflicts, and ensure homology between human and mouse to facilitate testing and development. The selection process identifies up to three genes that can be targeted by one API that contains a mixture of siRNA duplexes.

In his presentation, Dr. Lu demonstrated the synergistic effects evident both in vitro and in vivo of the multitarget platform, which can take advantage of the cross-talk between disease pathways.

In addition to the STP-502 therapeutic against solid tumors, Sirnaomics has four other siRNA products in development: STP-601 for ocular neovascularization diseases, STP-702 for pandemic flu (H5N1 and H1N1), and STP-705 for scarless wound healing.

Jinkang Wang, Ph.D., vp and CSO at Biomics Biotechnologies, introduced his company’s two main technology platforms and its therapeutic siRNA development programs. The first platform is a multiplex siRNA technique in which three siRNA sequences—each of which can target a different gene—are combined on a single strand. The strand is digested inside a cell to release the three distinct siRNAs.

“The technology can increase efficacy, reduce the development cost, and bypass the 21-mer patent protection,” said Dr. Wang. Biomics has also demonstrated a greatly reduced interferon response with the multiplex siRNA compared to a control siRNA.

The company’s other technology platform presents a new method for gene-specific Entire siRNA Target (EsT) library construction that uses the EcoP 151 enzyme to generate siRNAs that cover all possible 19–23 base-pair target binding sites.

The RNAi therapeutics initiatives under way at Biomics are targeting age-related macular degeneration (AMD), liver cancer, and hepatitis B virus (HBV). Dr. Wang described the company’s nonvirus liver delivery system, which, he says, has demonstrated its ability to deliver an siRNA to the liver and knock-down the ApoB gene, resulting in reduced blood cholesterol levels.

AMD, a degenerative disease affecting the eye, lends itself to localized siRNA delivery via direct injection into the eye. The goal of treatment is to inhibit the abnormal angiogenesis characteristic of AMD that can lead to blindness. Biomics has designed a small ligand interfering RNA (sliRNA) that does not silence a target gene via a RISC mechanism, but instead targets a toll-like receptor (TLR-3) on the cell surface, activating signaling pathways that block angiogenesis.

In animal studies, the sliRNA “performed with comparable or even better efficacy compared to Avastin (bevacizumab, a monoclonal antibody-based angiogenesis inhibitor), and much better than VEGF-targeted siRNA, stated Dr. Wang.

Biomics is collaborating with Benitec and using the Australian company’s viral delivery system together with its EsT library technology as the basis for siRNA drug development to treat HBV.

Knowledge-Based Design

Joerg Dennig, Ph.D., global product manager for siRNA at Qiagen, offered practical recommendations on how to set up an RNAi screen to increase the likelihood of getting true positive hits. “The screen has to be efficient and specific,” he said, emphasizing the importance of optimizing protocols across the entire workflow, including siRNA design, transfection, and data analysis.

Design parameters should focus on potency, specificity, and avoidance of off-target effects. Qiagen applies an siRNA design algorithm called BioPredsi, which it licensed from Novartis, combined with in-house bioinformatics modules. As more information on RNAi pathways becomes available, the ability to target a unique gene region will become more reliable.

Similarly, identification of interferon motif sequences allows design algorithms to bias siRNA sequence selection against sequences known to induce an interferon response. Knowledge-based design approaches also try to avoid sequences that might mimic miRNA effects. “It is not possible yet to predict the target of all endogenous miRNAs,” said Dr. Dennig, “but using this approach we can minimize the risk that one of our siRNA sequences could have miRNA effects.”

He also talked about the need to select a transfection reagent that can achieve efficient and selective siRNA transfection in specific cell lines. He highlighted Qiagen’s HiPerFect transfection reagent, which contains a combination of cationic and neutral lipids.

The most time-consuming aspect of the process, besides design of the siRNAs, is setting up the screen. Analysis of the data is also time-intensive and may take several months. A typical screen, for example, might include four different siRNAs for 20,000 genes. Using more than one siRNA in a screen provides a control factor to help determine whether a particular phenotype is a true positive and the result of targeted knock-down of gene expression, or represents instead a nonspecific siRNA effect.

To validate hits from siRNA screens, Qiagen pursues several approaches: performing smaller, focused screens using additional siRNAs; carrying out functional, high-content screens to explore the biological effects of the siRNA; and designing rescue experiments using a plasmid-expression construct that replaces the knocked-down gene to restore the wild-type phenotype. Another strategy to rule out nonspecific RNAi effects involves siRNAs that target untranslated gene regions, as the siRNAs will not target recombinant constructs.

Qiagen maintains an inventory of at least four siRNAs for every human and mouse gene at scales of 20 nmol to 0.1 nmol, and offers a range of expression plasmids carrying human and mouse genes for rescue experiments. It also provides custom synthesis of 1–10 siRNAs for a target gene and large-scale siRNA synthesis (10 mg to 10 g) for in vivo animal studies and preclinical applications. The company’s GeneGlobe Web portal is a knowledge resource containing pathway information and resources on products and protocols for RNAi experiments.

Alexander Vlassov, Ph.D., staff scientist, molecular biology reagents division at Life Technologies, described the company’s dual strategy for producing Silencer® Select siRNAs by selecting for “hyperfunctional siRNAs” that have enhanced potency and specificity and chemically modifying the duplexes with locked nucleic acid, which enhances strand bias and specificity of the siRNA molecules.

The design aspect of the strategy relies on a set of sequence features the company has identified as potency selection criteria and has built into a design algorithm. Additional bioinformatic filters minimize the potential for off-target hits and microRNA-like effects to optimize siRNA specificity.

Dr. Vlassov described two recent advances to the Silencer Select siRNA platform, which was introduced about 18 months ago. The siRNAs are now available at smaller scale, 1 nmol in 96- and 384-well plates and 2-D tubes, for rapid validation studies in vitro. Silencer Select siRNAs are also available at 250 nmol scale, in vivo grade, and new data supports their performance when delivered systemically in mice.