TGFBIp-associated corneal dystrophy : exploring structure-based drug development strategies for disease prevention and treatment
Date of Issue2018-02-06
School of Biological Sciences
Singapore Eye Research Institute
Corneal Dystrophies are a group of inherited disorders localized to various layers of the cornea that affect corneal transparency and visual acuity. The deposition of insoluble protein materials in the form of extracellular amyloid fibrils or intracellular cysts is pathognomonic. The gene associated with the stromal and Bowman’s related Corneal Dystrophy is TGFBI. The protein product of the gene is known as Transforming Growth Factor-induced protein (TGFBIp). This protein is a 68 KDa protein with 683 amino acids with 4 Fasciclin-like domains. In pathological conditions, the protein accumulates as insoluble deposits in various forms. The severity, clinicopathologic variations, age of the onset, and location of the deposits depend on the type of amino acid alterations in the protein. Until 2006, 38 different pathogenic mutants were reported in the TGFBI gene associated with various phenotypes of corneal dystrophy. To date, there are more than 65 reported mutations in the gene, and 84% of the mutations are found in the 4th FAS1 domain, making it a mutational hotspot. Hence our research focused mainly on the 4th FAS1 domain and mutations involved in the region. There is no effective treatment to prevent, halt, or reverse the deposition of TGFBIp. The main aim of this project is to understand the aggregation mechanism of mutant TGFBIp in the cornea. We have sought different approaches to understand the disease pathology and mechanism of protein aggregation and deposition. The first was a proteomics-based approach where we aimed to identify the composition of various proteins present in the amyloid corneal deposits of the patients with different mutations reported in TGFBI. The study aimed to understand the clearance mechanism of the mutant protein and compare it with the wild-type (WT) protein. The study also aimed to identify specific proteases and proteolytic mechanism involved in clearing the mutant protein in the cornea. The proteolytic processing of mutant TGFBIp protein compared to the WT protein may provide insight to the disease pathology. The proteomics data was vital to understand the role of specific proteases and the proteolytic clearance activity in WT and mutant protein. We identified enrichment of serine protease HTRA1 in the amyloid deposits of corneal dystrophy patients compared to the normal healthy controls. The corneal amyloid deposits are also enriched with amyloid-associated proteins, non-fibrillar amyloid proteins and TGFBIp itself. The study also showed that the patient corneal amyloid deposit samples showed enrichment of certain tryptic peptide fragments like TGFBIp 515-533, TGFBIp 571-588 and TGFBIp 611-642, which are either not present or present only in low amounts in healthy controls. The study showed the possibility of mutant protein processing by serine protease HTRA1, which may give rise to peptides that potentially act as amyloid fibril seeds and enhance the aggregation process. The second approach was to understand the mechanism of fibril formation using a peptide model. Based on the mass spectrometry analysis, we identified many short peptide residues (TGFBIp 515-533, TGFBIp 571-588 and TGFBIp 611-642) enriched in the amyloid corneal deposits of patients compared to the healthy control cornea. The main aim of the study was to understand the molecular-functional relationship of TGFBIp-derived mutant peptides where the amino acid substitutions decreased the overall net charge and their effects on the propensity of the peptides to form amyloid fibrils. We modified the 23 residue peptide (TGFBIp E611PVAEPDIMATNGVVHVITNVLQ633) with clinically relevant cationic charged amino acid substitution and investigated the amyloid-forming properties of the mutant peptides using various biophysical approaches. The peptide sequences were predicted to have a very high propensity to form amyloid fibres. The study also aimed to investigate the structure of amyloid fibrils derived from the peptides by Solid-state NMR. The study may be used to understand the fibrillation process, intermediate states of fibril formation, and will be useful for drug discovery and treatment of TGFBIassociated Corneal Dystrophy. The study provides an insight into the effects of substitution of individual amino acid on the rate of amyloid formation in the peptides, the structure, biological and biophysical properties of the amyloid fibrils formed by the peptides. We also explored how a position specific and type of amino acid substitution hugely contributes to the amyloidogenic property of the peptides. The third was a drug discovery approach to identify small fragments or compounds that can bind to the TGFBIp and modulate the protein, thereby delaying the aggregation process or changing the proteolytic processing of the mutant protein and preventing the formation of highly amyloidogenic peptide seeds. The lead compound identified will be tested to see if it can alter the thermodynamic stability of the protein and thereby stabilize the protein or delay the formation of toxic oligomers or fibrils. The compounds may possibly play a role in altering the proteolytic processing of the mutant TGFBIp. Thus the screening of 2500 compounds from the Maybridge library using initial Weak affinity chromatography and secondary 15N-HSQC-based assays have identified 3 lead compounds that bind to the mutant protein near a predicted binding pocket. The lead compounds were added to the proteins, and limited proteolysis by trypsin showed that the addition of the compounds stabilised the secondary structure of the protein and prevented the formation of highly amyloidogenic peptide seeds. The compounds, when bound to the mutant proteins, may alter the proteolytic processing for efficient clearance or to prevent the formation of smaller peptides. The lead compounds are to be taken to the next stage of the drug discovery process. A combination of computational, proteomics, structural and biophysical studies along with the clinical data provided an insight into the mechanism of cornea-specific protein aggregation and identification of lead compounds that may play a role in protein processing and proteolysis. Even the smallest progress made in each of the different approaches will provide additional information to the basic understanding of the mechanism of protein aggregation, disease progression, and development of novel therapeutic strategies to treat TGFBI-associated Corneal Dystrophies.