Chao-Yie Yang, MD, PhD

Discovery of small molecule complement inhibitors as the treatment for PNH
Pubmed Author Name: 
Yang Chao-Yi
Original Research Center: 
University of Michigan

Uncontrolled complement activation caused by a gene (PIG-A) mutation in hematopoietic stem cells has been characterized in PNH. Patients suffer from hemolysis, thrombosis and bone marrow failure. Current FDA-approved only treatment for PNH is eculizumab which is expensive, unable to eradicate PNH clone, not orally available, requires clinic-bound infusions via iv access and lifelong therapy. In this proposed work, we will discover and make rational designs to small molecules guided by protein structures to develop complement inhibitors. The promising small molecule inhibitors identified from this work will be evaluated and used to develop orally-available therapeutics to treat PNH patients.

First Year Report: 

Uncontrolled complement activation caused by a gene (PGI-A) mutation has been characterized in PNH. Current only FDA-approved treatment for PNH is eculizumab which is expensive, unable to eradicate PNH clone, not oral available and requires lifelong administration in clinics. Another molecule (AMY-101) based on the discovery of compstatin has received orphan status in Europe and US for treatment of PNH in 2014. AMY-101 inhibits C3, a key component of the complement activation system responsible for the cause of PNH, and the use of AMY-101 is a new therapeutic strategy to treat PNH. First in-human trial of AMY-101 will start in 2015. In preclinical study, AMY-101 was administered in monkey via intravenous injection. Currently, no oral drug is available to treat PNH. In this proposed work, we will discover and develop small molecules to inhibit complement activation with the guide of protein structures information. Promising small molecules identified from this work will be used to develop oral-available drugs for clinical development to treat PNH patients.

For the first part of the project, we first studied the structure of compstatin and generated a structural model recapitulating the important functional groups in compstatin. Using the structural model, we screened against 514957 compounds available from a commercial compound library to select a smaller set of 76800 compounds using a computer program. To determine which of the 76800 compounds fit well to C3c and capture the important interaction similarly to compstatin, we performed a second computational docking simulation which finds the best way each molecule fits into the C3c. The computer program also provides theoretical estimates of the potency of each compound to C3c. Finally, we selected 24 compounds for biochemical evaluations. Currently, we are waiting for the collection of patients’ blood samples to validate these compounds. We will use the same workflow to continue screening against millions of other small molecules libraries. This will be a cost-effective approach to discovery new candidate compounds. Once we confirmed that the compounds bind to C3 with good potencies, we will select the best candidates for additional inhibitor optimization.

A second therapeutic strategy proposed in this work is to target another protein, Factor D, that controls the complement activation via the alternative pathway. We first experimentally screened against the Factor D using 2670 fragment compounds to identify novel compounds that bind with Factor D. These fragment compounds are generally small in sizes but have great potentials to be chemically modified into potent molecules that can be made orally bioavailable. This screening allowed us to identified 11 promising candidates including several new classes of compounds that were not reported previously. Using the information of the new molecules identified from our screening experiment and the templates from Novartis’ inhibitors, we have designed new classes of Factor D inhibitors based on computer-aided drug design method. The computer-aided drug design method allows us to evaluate if our proposed chemical modifications of the small molecules will be compatible with the pocket in Factor D and set priority of which modifications should be tried first. With the assistance of the computer-aided design, we can make the drug design process more efficient as opposed to the traditional trial-and-error medicinal chemistry approach.

In the first year of this project, we have constructed critical computational models, established experimental biochemical tools to evaluate the selected compounds, discovered novel chemical molecules used for drug design, and devised a computer-aided drug design workflow. New compounds have been synthesized and will be evaluated using PNH patients’ blood samples. We anticipated that we will obtain encouraging data in the second year and publish the results to the research community to benefit the therapeutic development for treating this rare blood disorder.  

Final Report: 

Uncontrolled complement activation caused by a gene (PGI-A) mutation found in Paroxysmal nocturnal hemoglobinuria (PNH) is a main cause for intravascular hemolysis. Though effectively reducing hemolysis in the majority of PNH patients, the treatment cost of eculizumab (an antibody to C5 which is a downstream component of the complement system) remains a significant financial burden. Another molecule (AMY-101) based on the discovery of compstatin has been approved in Europe and US to treat complement related orphan diseases. Specifically, AMY-101 inhibits C3, a key upstream component of the complement activation, and is approved to treat C3 glomerulopathy in 2015. Its clinical use to treat PNH is still under investigation. Emerging and vibrant drug development activities in complement-targeted therapy in the past few years are encouraging and welcoming news for PNH patients awaiting affordable and effective treatments. Administration of eculizumab and AMY-101 to patients are via intravenous injection performed by healthcare professionals at the hospitals. No oral drug is currently available to treat PNH patients who require a long-term therapy in clinics. Our goal in this project is to discover small molecules that inhibit two upstream component proteins regulating the complement activation. Unlike eculizumab and AMY-101, these molecules can potentially be advanced into oral drug development to treat PNH.

In the first aim of the project, we have searched 514957 compounds available from a commercial compound library to select a smaller set of 76800 compounds that match with some chemical groups of compstatin using a computer program. The compounds were then subject to computational docking calculations which find the best way each molecule fits into the C3c protein. The computer program also estimates how strong each compound interacts with C3c in theory. Using compstatin (a known C3c inhibitor) as a reference compound, we have selected 24 compounds for biochemical evaluations. This is a cost-effective approach to discovery new candidate compounds for biological evaluation from large of amount of molecules available but difficult to be tested in a reasonable time. We plan to adopt this approach to continue searching for best candidates for follow-up inhibitor development.

In the second aim, we developed compounds to inhibit Factor D that controls an important pathway to affect the rate of complement activation. Similar to Aim 1, we screened commercially available compounds using computer programs and discovered two new molecules confirmed by our experimental validation. We have also successfully obtained crystal structures that provide atomic details of how three other compounds inhibit Factor D. The structural information provides critical guidance when we designed and modeled four novel different series of compounds. We are currently evaluating their activities that will provide feedback to our drug design pipeline. In addition, we have determined that our compounds have favorable stability profiles in mouse liver enzymes. This indicates that good amount of the compounds can enter the blood stream after intestine absorption through oral administration in mouse. Follow-up evaluation in human liver enzymes will be performed. We have also validated that these compounds inhibit complement activation in the rabbit blood cells induced by human serum. Our data showed that our compounds effectively inhibit Factor D and its associated activation of the complement system. Evaluation of these inhibitors in the PNH patients’ blood sample is planned and will be conducted in the near future.

We have established a drug design platform involving a strong team of scientists through the support from AA&MDSIF. We have made a significant progress especially in the Factor D inhibitor development and have presented our findings in the 2016 Bone Marrow Failure Disease Scientific Symposium hosted by AA&MDSIF. We received positive feedbacks from physician scientists working to treat PNH. We also submitted a manuscript documenting our results to benefit the research community. Additional research supports from our institution have been applied based on these data. We will continue to work on these projects initiated by the support from AA&MDSIF support and contribute to the research community dedicated to the therapeutic development for treating this rare blood disorder.