Debra Cheung

A model for familial exudative vitreoretinopathy caused by LPR5 mutations

We have identified a mouse recessive mutation that leads to attenuated and hyperpermeable retinal vessels, recapitulating some pathological features of familial exudative vitreoretinopathy (FEVR) in human patients. DNA sequencing reveals a single nucleotide insertion in the gene encoding the low-density lipoprotein receptor-related protein 5 (LRP5), causing a frame shift and resulting in the replacement of the C-terminal 39 amino acid residues by 20 new amino acids. This change eliminates the last three PPP(S/T)P repeats in the LRP5 cytoplasmic domain that are important for mediating Wnt/beta-catenin signaling. Thus, mutant LRP5 protein is probably unable to mediate its downstream signaling. Immunostaining and three-dimensional reconstructions of retinal vasculature confirm attenuated retinal vessels. Ultrastructural data further reveal that some capillaries lack lumen structure in the mutant retina. We have also verified that LRP5 null mice develop similar alterations in the retinal vasculature. This study provides direct evidence that LRP5 is essential for the development of retinal vasculature, and suggests a novel role played by LRP5 in capillary maturation. LRP5 mutant mice can be a useful model to explore the clinical manifestations of FEVR.

Diverse gap junctions modulate distinct mechanisms for fiber cell formation during lens development and cataractogenesis.

Different mutations of α3 connexin (Cx46 or Gja8) andα 8 connexin (Cx50 or Gja8), subunits of lens gap junction channels, cause a variety of cataracts via unknown mechanisms. We identified a dominant cataractous mouse line (L1), caused by a missense α8 connexin mutation that resulted in the expression of α8-S50P mutant proteins. Histology studies showed that primary lens fiber cells failed to fully elongate in heterozygous α8S50P/+ embryonic lenses, but not in homozygous α8S50P/S50P, α8-/- andα 3-/- α8-/- mutant embryonic lenses. We hypothesized that α8-S50P mutant subunits interacted with wild-typeα 3 or α8, or with both subunits to affect fiber cell formation. We found that the combination of mutant α8-S50P and wild-type α8 subunits specifically inhibited the elongation of primary fiber cells, while the combination of α8-S50P and wild-type α3 subunits disrupted the formation of secondary fiber cells. Thus, this work provides the first in vivo evidence that distinct mechanisms, modulated by diverse gap junctions, control the formation of primary and secondary fiber cells during lens development. This explains why and how different connexin mutations lead to a variety of cataracts. The principle of this explanation can also be applied to mutations of other connexin isoforms that cause different diseases in other organs.

Knock-in of α3 connexin prevents severe cataracts caused by an α8 point mutation

A G22R point mutation in α8 connexin (Cx50) has been previously shown to cause a severe cataract by interacting with endogenous wild-type α3 connexin (Cx46) in mouse lenses. Here, we tested whether a knocked-in α3 connexin expressed on the locus of the endogenous α8 connexin could modulate the severe cataract caused by the α8-G22R mutation. We found that the α3(-/-) α8(G22R/-) mice developed severe cataracts with disrupted inner fibers and posterior rupture while the α3(-/-) α8(G22R/KIα3) lens contained relatively normal inner fibers without lens posterior rupture. The α8-G22R mutant proteins produced typical punctate staining of gap junctions between fiber cells of α3(-/-) α8(G22R/KIα3) lenses, but not in those of α3(-/-) α8(G22R/-) lenses. Thus, we hypothesize that the knocked-in α3 connexin subunits interact with the α8-G22R connexin subunits to form functional gap junction channels and rescue the lens phenotype. Using an electrical coupling assay consisting of paired Xenopus oocytes, we demonstrated that only co-expression of mutant α8-G22R and wild-type α3 connexin subunits forms functional gap junction channels with reduced conductance and altered voltage sensitivity compared with the channels formed by α3 connexin subunits alone. Thus, knocked-in α3 connexin and mutant α8-G22R connexin probably form heteromeric gap junction channels that influence lens homeostasis and lens transparency.

NHE8 Is Essential for RPE Cell Polarity and Photoreceptor Survival

A new N-ethyl-N-nitrosourea (ENU)-induced mouse recessive mutation, identified by fundus examination of the eye, develops depigmented patches, indicating retinal disorder. Histology data show aberrant retinal pigment epithelium (RPE) and late-onset photoreceptor cell loss in the mutant retina. Chromosomal mapping and DNA sequencing reveal a point mutation (T to A) of the Slc9a8 gene, resulting in mutant sodium/proton exchanger 8 (NHE8)-M120K protein. The lysine substitution decreases the probability of forming the 3rd transmembrane helix, which impairs the pore structure of the Na+/H+ exchanger. Various RPE defects, including mislocalization of the apical marker ezrin, and disrupted apical microvilli and basal infoldings are observed in mutant mice. We have further generated NHE8 knockout mice and confirmed similar phenotypes, including abnormal RPE cells and late-onset photoreceptor cell loss. Both in vivo and in vitro data indicate that NHE8 co-localizes with ER, Golgi and intracellular vesicles in RPE cells. Thus, NHE8 function is necessary for the survival of photoreceptor cells and NHE8 is important for RPE cell polarity and function. Dysfunctional RPE may ultimately lead to photoreceptor cell death in the NHE8 mutants. Further studies will be needed to elucidate whether or not NHE8 regulates pH homeostasis in the protein secretory pathways of RPE.

Characterization of Mouse Mutants with Abnormal RPE Cells

Retinal pigment epithelium (RPE) is essential for the function and survival of photoreceptor cells by playing supporting roles including shedding the outer segments of the photoreceptor cells, removing metabolic wastes, transporting nutrients and maintaining visual cycle. RPE defects have been found in various human retinal disorders, such as age-related macular degeneration (Zarbin, 1998), Best disease (Petrukhin et al., 1998; Marmorstein et al., 2000), Sorsby fundus dystrophy (Weber et al., 1994; Ruiz et al., 1996; Della et al., 1996), and childhood-onset severe retinal dystrophy (Gu et al., 1997). Animal models with RPE defects have been used to study the molecular basis for the function of the RPE cells. The Royal College of Surgeons (RCS) rat, a model for recessive inherited retinal degeneration, is characterized by the dysfunction of RPE due to a null mutation of the receptor tyrosine kinase Mertk gene (D’Cruz et al., 2000). RPE cells fail to shed the outer segments of the photoreceptor cells in the RCS rat (Mullen et al., 1996). Recently, mice with a targeted disruption of the Mertk gene manifest retinal dystrophy similar to RCS rats (Duncan et al., 2003). In addition, mutated Mertk gene has been identified in patients with retinitis pigmentosa (Gal et al., 2002).