Conceived and designed the experiments: SSFC ACYL SKC. Performed the experiments: SSFC JWCL AKML ACYL. Analyzed the data: SSFC AKML KSLL SSMC ACYL SKC. Contributed reagents/materials/analysis tools: KSLL SSMC ACYL SKC. Wrote the paper: SSFC KSLL SSMC ACYL SKC.
Current address: Eye Institute, The University of Hong Kong, Hong Kong, China
The authors have declared that no competing interests exist.
The level of endothelin-1 (ET-1), a potent vasoconstrictor, was associated with retinopathy under ischemia. The effects of endothelial endothelin-1 (ET-1) over-expression in a transgenic mouse model using Tie-1 promoter (TET-1 mice) on pathophysiological changes of retinal ischemia were investigated by intraluminal insertion of a microfilament up to middle cerebral artery (MCA) to transiently block the ophthalmic artery. Two-hour occlusion and twenty-two-hour reperfusion were performed in homozygous (Hm) TET-1 mice and their non-transgenic (NTg) littermates. Presence of pyknotic nuclei in ganglion cell layer (GCL) was investigated in paraffin sections of ipsilateral (ischemic) and contralateral (non-ischemic) retinae, followed by measurement of the thickness of inner retinal layer. Moreover, immunocytochemistry of glial fibrillary acidic protein (GFAP), glutamine synthetase (GS) and aquaporin-4 (AQP4) peptides on retinal sections were performed to study glial cell reactivity, glutamate metabolism and water accumulation, respectively after retinal ischemia. Similar morphology was observed in the contralateral retinae of NTg and Hm TET-1 mice, whereas ipsilateral retina of NTg mice showed slight structural and cellular changes compared with the corresponding contralateral retina. Ipsilateral retinae of Hm TET-1 mice showed more significant changes when compared with ipsilateral retina of NTg mice, including more prominent cell death in GCL characterized by the presence of pyknotic nuclei, elevated GS immunoreactivity in Müller cell bodies and processes, increased AQP-4 immunoreactivity in Müller cell processes, and increased inner retinal thickness. Thus, over-expression of endothelial ET-1 in TET-1 mice may contribute to increased glutamate-induced neurotoxicity on neuronal cells and water accumulation in inner retina leading to edema.
Diabetic retinopathy (DR) and other ocular diseases in diabetes, such as central retinal artery occlusion (CRAO) and glaucoma, is thought to be the consequence of retinal ischemia, leading to visual impairment and blindness
Endothelin-1 (ET-1), a 21-amino acid secretory protein synthesized in vascular endothelial cells, is a potent vasoconstrictor. A possible link between elevated plasma ET-1 level and retinopathy under ischemia has been established. Administration of ET-1 into the posterior vitreous body or the optic nerve of animal models led to physiological and cellular damages of ischemic insult, including obstruction of retinal blood flow, elevated scotopic b-wave in electroretinogram and apoptosis of cells in GCL
In order to study the pathogenic changes of ET-1 to ischemic stress in central nervous system, a hypertensive transgenic mouse model with over-expression of ET-1 in vascular endothelial cells using Tie-1 promoter (TET-1 mice) has been generated
The use of animals in this study was conducted according to the requirements of the Cap. 340 Animals (Control of Experiments) Ordinance and Regulations, and all relevant legislation and Codes of Practice in Hong Kong. All the experimental and animal handling procedures were approved by the Faculty Committee on the Use of Live Animals in Teaching and Research in The University of Hong Kong (Permit number #634-01).
Male heterozygous (He) TET-1 mice were generated by microinjection of the ET-1 construct, which contains the mouse ET-1 cDNA with SV40 polyA driven by the Tie-1 promoter
Mice at the age of 6 to 8 weeks were sacrificed, their eyes were cut underneath the optic nerve and the retinae were dissected under dissecting microscope (Stemi SV6; Carl Zeiss, Thornwood, NY), frozen in liquid nitrogen and stored in a −80°C. Retinal tissues were homogenized in 1 ml ice-cold TRI REAGENT® (Molecular Research Center, Cleveland, Ohio) and the total RNA from retina was isolated according to manufacturer's protocol. 1.5 µg of total RNA was treated with DNase I (Boehringer, Ingelheim, Germany) and reverse transcription was performed with Oligo(dT)18, Superscript™ II RNase H- reverse transcriptase (Invitrogen, Carlsbad, CA), DTT and dNTP mixture. The primer sequences and condition for rt-PCR of ET-1 has been previously mentioned
Age-matched Hm TET-1 and non-transgenic (NTg) mice at the age of 8 to 10 weeks were subjected to retinal ischemia as previously described
Serial paraffin sections of 7 µm thick from non-ischemic and ischemic eyeballs were prepared using the microtome (Microm HM315, Germany). The sections were deparaffinized in xylene, rehydrated with a graded series of ethanol, and stained briefly with hematoxylin and eosin (Sigma-Aldrich, St. Louis, MO). To perform immunocytochemistry (ICC) of GFAP, glutamine synthetase (GS) and AQP-4, sections of ischemic and non-ischemic retinae were blocked with 1.5% normal goat serum (Vector Laboratories, San Francisco, CA) for 1 hour, followed by incubation in diluted primary antibodies overnight at 4°C. Incubation of blocking serum served as control. The primary antibodies and their concentrations were rabbit anti-GFAP (1∶2000; Dako, Carpinteria, CA), rabbit anti-AQP-4 (1∶ 800; Chemicon International, Temecula, CA) and rabbit anti-GS (1∶ 300; Santa Cruz Biotechnology, Santa Cruz, CA). The sections were subsequently incubated with biotinylated goat anti-rabbit secondary antibody (Vector Laboratories), and immunoreactive signals were visualized by incubation with the avidin-biotin-peroxidase complex (Vector Laboratories) for 30 min and diaminobenzidine tetrahydrochloride (Zymed Laboratory, San Francisco, CA) for 2 min. Finally, the sections were counterstained with hematoxylin, dehydrated, cover-slipped and mounted with mounting medium.
Retinal sections containing the optic nerve that were stained with hematoxylin and eosin were chosen for viewing. Two images, one on either the nasal or temporal side, were taken at a region about 300 µm away from the optic nerve using the 20× objective (Olympus IX71, Olympus, Japan) as previously described
Digital photographs were taken using 20× objectives from the nasal and temporal sides (about 300 µm from the optic nerve) in each of the contralateral and ipsilateral retinae. In each photo, the thickness of three regions in the retina was measured: inner nuclear layer (INL), from inner plexiform layer (IPL) to inner limiting membrane (ILM), and from INL to ILM representing the thickness of the whole retina
All data were expressed as means ± SEM. Statistical analyses were performed using one-way ANOVA with Bonferroni's post hoc tests, Fisher's exact test, Mann-Whitney Test, and Kruskal-Wallis Test by Statistical Package for the Social Sciences software (SPSS 12.0, SPSS). Difference with
Total ET-1 mRNA expression in 6- to 8-week-old in Hm TET-1 mouse retina was about 2.5 times higher than that of NTg mice by semi-quantitative rt-PCR (
A: Ethidium bromide-stained agarose gels of rt-PCR products of ET-1, showing increased ET-1 mRNA expression in transgenic mouse retina. The molecular size marker (M) was 1 kb plus marker. B: Histograms showing the levels of ET-1 mRNA expression normalized to that of GAPDH. n = 4 for each experimental group. *:
Immunoreactivity of ET-1 was up-regulated in the endothelial cells (arrows) of Hm TET-1 mice compared with the corresponding NTg mice. Scale bar = 25 µm.
Cellular morphology in retinal sections of ischemic and non-ischemic retinae of NTg and Hm TET-1 mice were examined. Similar to our previous results, there was presence of cells with pyknotic nuclei in the GCL
|
|
|
|
|
|
|
|
12 | 0 | 0 | 0 | 12 | 12 |
|
0 | 7 | 4 | 1 | 11 | 12 |
|
14 | 0 | 0 | 0 | 14 | 14 |
|
0 | 4 | 2 | 8 | 6 | 14 |
Glial cell reactivity in the ipsilateral and contralateral retinae of NTg and Hm TET-1 mice was studied by immunocytochemical staining of GFAP (
Compared with the contralateral retinae of NTg and Hm TET-1 mice, GFAP signal was up-regulated in the ipsilateral retinae in the astrocytes around capillaries in the inner limiting membrane (arrowheads) and the Müller cell processes in IPL (arrows). n = 5 for each experimental group. Scale bar = 50 µm.
In the non-ischemic retinae of NTg and Hm TET-1 mice, immunoreactivity of GS was mainly present in the astrocytes, Müller cell bodies in INL and Müller cell processes from ILM to OLM (
Up-regulation of GS signal was found in the Müller cell processes in IPL (arrows) and Müller cell bodies in INL (arrowheads) in the ipsilateral retina of Hm TET-1 mice compared with the contralateral retina of Hm TET-1 mice and the ipsilateral retina of NTg mice. n = 5 for each experimental group. Scale bar = 50 µm.
Expression of a water transport protein, aquaporin-4 (AQP4), was determined for the presence of edema and water accumulation of glial cells in ischemic retina. In the contralateral retinae of both NTg and Hm TET-1 mice, positive immunoreactivity of AQP4 was found in the Müller cell processes from ILM to OLM, astrocytes around blood vessels in GCL, Müller cell bodies in INL and astrocytic endfeet in OPL (
Up-regulation of AQP4 signal was found in the Müller cell processes in IPL (arrows) and Müller cell bodies in INL (arrowheads) in the ipsilateral retina of Hm TET-1 mice compared with the corresponding contralateral retina, while similar intensity of AQP4 signal was present in the contralateral and ipsilateral retinae of NTg mice. n = 5 for each experimental group. Scale bar = 50 µm.
According to our semi-quantitative observation, the major phenotype of ischemic injury in the ipsilateral retina of Hm TET-1 mice was “severe” while that of NTg mice was “mild”. Therefore, representative photomicrographs from each experimental animal group were selected after careful screening for retinal thickness measurement. Similar thicknesses of INL, IPL to ILM and inner retinal layer were found between the contralateral retinae of Hm TET-1 and NTg mice (
|
|
|
|
|
57.2±3.5 | 29.8±0.8 | 87.0±2.8 |
|
61.5±3.5 | 30.5±1.9 | 92.0±5.2 |
|
56.2±3.1 | 28.9±1.8 | 85.2±3.9 |
|
71.7±3.9* | 29.0±1.9 | 100.6±3.3* |
Mild inner retinal ischemia was induced by MCAO in mice with an absence
In this study, the ischemic model induced by middle cerebral artery occlusion also resulted in a mild transient retinal ischemia with neuronal cell damage in mice with F1 hybrid background, as suggested by increased GFAP immunoreactivity in the Müller cell processes and astrocytes, and the presence of pyknotic nuclei in the cells in GCL in the NTg ipsilateral retinae. Thus, slight cellular and structural changes were present in the NTg ipsilateral retinae, including pyknotic nuclei were present in less than 80% of cells in GCL, slightly increased GS expression in the Müller cell processes and no significant changes in AQP4 immunoreactivity and inner retinal thickness. On the contrary, more severe cellular and structural changes were observed in the ipsilateral retina of Hm TET-1 mice, including presence of pyknotic nuclei in more than 80% of cells in GCL, prominent up-regulated GS expression in Müller cell processes, up-regulation of AQP-4 immunoreactivity and significant increase in inner retinal thickness. Under these prominent cellular changes observed in the ischemic retina, we may deduce that Hm TET-1 mice with over-expression of ET-1 in endothelial cells would exacerbate the effects of neuronal cell death and retinal edema after OA occlusion. Previous studies already showed significant loss of cells in GCL layer following intraocular injection of ET-1 to the optic nerve
Down-regulation of GS expression has been shown in Müller cells after ischemia and reperfusion
AQP4 is the main water channel protein which regulates the bi-directional movement of water across membranes and involved in the maintenance of the ionic and osmotic balance
This study showed that the transgenic mice model with over-expression of endothelial ET-1 may enhance the effects of glutamate-induced neurotoxicity after induction of inner retinal ischemia and reperfusion with exacerbated neuronal cell death. Moreover, TET-1 mice may also contribute to increased water accumulation in the glial cells, leading to retinal edema with increased inner retinal thickness. The TET-1 mouse model after mild transient inner retinal ischemia may serve as a suitable model to study pathological changes in ischemic retinopathy simulating disease conditions like DR, glaucoma, peripheral or central ischemia which may lead to ocular ischemia or stroke, and is useful for therapeutic studies on these disorders.
The authors would like to thank Marcella Ma, Maggie Ho and Michael Koon for performing the microinjection experiment and generating the TET-1 mice as well as Janice Law and Alvin Cheung for technical advice.