Faculty Directory

Chia-Yang Liu

Associate Professor

Chia-Yang Liu

Office: Opt 511


Curriculum Vitae

  • M.S.–1985 (National Taiwan University)
  • Ph.D.–1993 (University of Cincinnati)
Brief of research interesting

The main focus of my research is to use conditional gene loss-of-function and gain-of-function mouse model to elucidate the signal transduction and transcriptional regulations of ocular anterior segment development and diseases. We have established several transgenic mouse phenotypes which resemble human ocular diseases such as ocular surface squamous neoplasia (OSSN), Keratoconjunctivitis sicca (dry eye symdrome), Blephar-ophimosis, Ptosis, and Epi-canthus inversus Syndrome (BPES), and congenital juvenile angle closure glaucoma.

1. Krt12rtTA/rtTA/tetO-FGF-7 double transgenic mouse model for OSSN.  This mouse strain displays a corneal papillomatous squamous cell carcinoma with profound neovascularization. These phenotypes resemble the human OSSN, which is the most common ocular surface cancer, but its etiology and pathological progression are unknown. We demonstrated that nuclear localization of beta-catenin in the corneal epithelium of Krt12rtTA/rtTA/tetO-FGF-7 double transgenic mice following Dox administration. This is in agreement with the notion that activation of receptor tyrosine kinases (RTK) by their cognate growth factor ligands can activate extracellular signal-regulated kinases-p90 ribosomal S6 kinase (ERK-p90RSK) as well as phosphatidylinositol 3-kinase (PI3K)-protein kinase B (PKB)/AKT (PI3K-PKB/AKT) pathways that down-regulate glycogen synthase kinase (GSK-3-beta) activity. This down-regulation leads to the stabilization and the nuclear accumulation of beta-catenin causing enhanced mitogenesis and metastasis, as evidenced in a wide variety of human cancers. We are currently using genetic approaches to link if beta-catenin activation is required for FGF-7 induced OSSN in mouse. Our preliminary results demonstrate that corneal hyperplasia by excess FGF-7 can be prevented by ablation of beta-catenin gene (Ctnnb1) in quadruple Krt12rtTA/rtTA/tetO-FGF-7/tet-O-Cre/Ctnnb1lox(Ex2-6)/lox(ex2-6) mice fed with Dox. Moreover, forced expression of Ctnnb1 gain-of-function mutant in differentiated (K12-positive) corneal epithelia disrupts normal corneal epithelial homeostasis and causes neovascularization and hyperplastic transformation. This provides the first indication that beta-catenin is downstream of FGF-7 signaling cascade. Interestingly, pannus surgically removed from human OSSN patients also exhibits excess FGF-7 and nuclear translocation of beta-catenin. These findings implicate that excess FGF-7 may associate with human OSSN.

 2. K14-rtTA/tetO-Cre/RosaLSL-dn-MAML1 triple transgenic mouse model for goblet cell morphogenesis and dry eye syndrome. Conjunctival goblet cells synthesize mainly mucins to lubricate ocular surface, essential for normal vision. Notch signaling has been associated with goblet cell differentiation in intestinal and respiratory tracts, but its function in ocular surface remains elusive. Our preliminary results show that conditional inhibition of canonical Notch signaling by expressing dominant negative mastermind-like-1 (dn-MAML1) in ocular surface epithelia impairs goblet cell differentiation during and following development. As compared to the wild-type ocular surface (OSWt), expression of dn-MAML1 in ocular surface (OSdn-MAML1) causes conjunctival epithelial hyperplasia, aberrant desquamation, failure of goblet cell differentiation, and subconjunctival inflammation. In addition, OSdn-MAML1 down-regulates Pax-6 expression, alters keratin expression pattern, and results in epidermal metaplasia, leading to keratoconjunctivitis sicca (dry eye) syndrome. We are currently studying the molecular target of Notch signaling in goblet cell differentiation.

 3. Kera-rtTA/tetO-Cre/R26floxedN1-ICD triple transgenic mouse model for eyelid development and congenital ptosis. Blephar-ophimosis, Ptosis, and Epi-canthus inversus Syndrome (BPES) is an autosomal dominant genetic disorder characterized by craniofacial defects that mainly affect the development of the eyelids. People with BPES are at an increased risk of developing vision problems such as myopia or hyperopia, strabismus and amblyopia affecting one or both eyes. Mutations in the transcription factor forkhead box L2 (FOXL2) structure gene cause 70 percent of BPES. The FOXL2 gene provides instructions for making a protein that is involved in the development of the eyelids and the ovaries before birth. Approximately 30 percent of people with BPES do not have an identified FOXL2 structure gene mutation; the cause of the condition in these people is unknown but FOXL2 gene regulation maybe altered. We demonstrated that Notch1 activation serve as the upstream control of expression of FoxL2 by periocular mesenchyma, which are destined for levator smooth muscle development of the eyelids. We are currently using this new knowledge to unravel signaling mechanisms of the Notch pathway in regulating FoxL2 gene expression governing levator smooth muscle differentiation during eyelid morphogenesis in mouse.

 4. Krt12rtTA/rtTA/tetO-TGF-alpha double transgenic mouse model for ocular development and congenital anterior segment dysgenesis (ASD). The anterior segment of the vertebrate eye is constructed by proper spatial development of ectodermal cells, which then become the corneal epithelium and lens, the neuroectoderm (posterior iris and ciliary body) and the cranial neural crest (corneal stroma, corneal endothelium and anterior iris). Although coordinated interactions between these different cell types are presumed to be essential for proper spatial positioning and differentiation, the requisite intercellular signals remain undefined. We established Krt12rtTA/tetO-TGF-alpha mouse strain where corneal epithelium-specific expression of TGF-alpha at different developmental stages can be examined and compared. Induction of excess TGF-alpha during or following development resulted in fibrosis and angiogenesis; leading to corneal opacification. Histological examinations revealed that induction of TGF-alpha initially causes corneal epithelial hyperplasia.  As the mouse develops, degeneration of the corneal epithelium occurs accompanied with profound inflammation and neovascularization. In addition, the cornea and iris, which normally form an opened angle, adhere to each other, compromising the formation of the irido-corneal angle, resulting in the loss of the anterior chamber.  Our data suggests that the levels of biologically active TGF-alpha in the aqueous humor must be under tight control to maintain proper corneal morphogenesis and homeostasis.  This proper spatial development is required to prevent adherence of the iris to the corneal stroma; therefore permitting the normal formation of the anterior segment. The Krt12rtTA/tetO-TGF-alpha transgenic mouse line can be a potential model for congenital and secondary glaucoma.