Retinal Gene Expression and Visually Evoked Behavior in Diabetic Long Evans Rats
Total Page:16
File Type:pdf, Size:1020Kb
Retina Retinal Gene Expression and Visually Evoked Behavior in Diabetic Long Evans Rats Stefanie J. Kirwin,1,2 Suzanne T. Kanaly,3 Candice R. Hansen,1 Belinda J. Cairns,3,4 Meina Ren,3 and Jeffrey L. Edelman1 PURPOSE. Patients with diabetic retinopathy may experience abnormally high glucose levels. Diabetic patients without vas- severe vision loss due to macular edema and neovascularization cular changes in the retina experience abnormal color vision,2 secondary to vascular abnormalities. However, before these dark adaptation,3 and contrast sensitivity.4 The oscillatory po- abnormalities become apparent, there are functional deficits in tential (OP) of the electroretinogram (ERG) is anomalous in contrast sensitivity, color perception, and dark adaptation. The diabetic patients4–8 and rodents,9–13 which is consistent with goals of this study are to evaluate early changes (up to 3 altered function of supporting cells in the retina, such as months) in retinal gene expression, selected visual cycle pro- impaired glutamate metabolism14–16 and increased glial fibril- teins, and optokinetic tracking (OKT) in streptozotocin (STZ)- lary acidic protein (GFAP) expression in Mu¨ller cells.13,15,17–19 induced diabetic rats. Other glial cells, such as astrocytes and microglia, are also 17 METHODS. Retinal gene expression in diabetic Long Evans rats affected. There is evidence of inflammatory changes in the 20–24 25–27 was measured by whole genome microarray 7 days, 4 weeks, eyes of diabetic patients and animals and cell death of 28–33 and 3 months after the onset of hyperglycemia. Select gene and various retinal neurons. protein changes were probed by polymerase chain reaction In addition to inner retinal pathology, photoreceptor viabil- 5,31,34 (PCR) and immunohistochemistry, respectively, and OKT ity is impacted by hyperglycemia. Before photoreceptor thresholds were measured using a virtual optokinetics system. cell death, the visual cycle is altered in diabetic rodents as evidenced by impaired rhodopsin regeneration35 and reduced RESULTS. Microarray analysis showed that the most consistently 11-cis-retinal concentration in the retina.36 The visual cycle is affected molecular and cellular functions were cell-to-cell sig- initiated in photoreceptors but also involves other cell types naling and interaction, cell death, cellular growth and prolifer- intimately associated with photoreceptors, such as retinal pig- ation, molecular transport, and cellular movement. Further ment epithelium (RPE) and Mu¨ller cells. On light stimulation, analysis revealed reduced expression of several genes encod- 11-cis-retinal is converted to all-trans-retinal in the first step of ing visual cycle proteins including lecithin/retinol acyltrans- phototransduction (see Ref. 37 for review), followed by recy- ferase (LRAT), retinal pigment epithelium (RPE)-specific pro- cling of all-trans-retinal to 11-cis-retinal by the visual cycle, tein 65 kDa (RPE65), and RPE retinal G protein–coupled which takes place outside the photoreceptor.37,38 It involves receptor (RGR). These molecular changes occurred simultane- several binding proteins as well as enzymes such as lecithin/ ously with a decrease in OKT thresholds by 4 weeks of diabe- retinol acyltransferase (LRAT),39 RPE65,40 11-cis-retinol dehy- tes. Immunohistochemistry revealed a decrease in RPE65 in the drogenase (RDH5),41 and RPE retinal G protein–coupled re- RPE layer of diabetic rats after 3 months of hyperglycemia. ceptor (RGR).42 Patients with mutations in LRAT, RPE65, and CONCLUSIONS. The data presented here are further evidence that RGR exhibit severe early-onset retinal dystrophies.37,38 Rod inner retinal cells are affected by hyperglycemia simultane- opsin regeneration relies on visual cycle reactions occurring in ously with blood retinal barrier breakdown, suggesting that the RPE, whereas an alternative visual cycle occurring in Mu¨ller glial and neuronal dysfunction may underlie some of the early cells is associated with cone opsin regeneration (see Refs. 43, visual deficits in persons with diabetes. (Invest Ophthalmol Vis 44 for reviews). Sci. 2011;52:7654–7663) DOI:10.1167/iovs.10-6609 One previous study using gene microarray in the diabetic rat model assessed 5147 genes and expressed sequence tags45 iabetic retinopathy is a major sight-threatening disease in 1 as opposed to current whole genome technology, which Dworking age people in the developed world, and its probes more than 41,000 rat genes and transcripts. Further- severity is clinically classified by vascular changes that can lead more the analysis was limited to individual genes grouped into to blindness. There is ample evidence, however, that early broad functional classes45 rather than specific pathways or functional abnormalities are caused by nonvascular effects of biological networks. Others explored changes in a single cell type, specifically Mu¨ller cells.25 Finally, Brucklacher et al.26 performed a whole genome microarray with pathway analysis From the Departments of 1Biological Sciences and 3Pathology, in the retinas of diabetic Sprague-Dawley rats that show inflam- 27 Allergan, Inc., Irvine, California. matory changes in the retina not detected in other strains. Present affiliation: 2Eye.Q.Consulting, Corp., Brea, California; and Evaluating a rat strain that lacks a potentially overshadowing 4Takeda San Francisco, Inc., San Francisco, California. inflammatory response may reveal more subtle, but equally Supported by Allergan, Inc. important, changes in metabolic, signaling, and other path- Submitted for publication September 20, 2010; revised April 28, ways. Changes other than inflammation are likely present and June 6, and July 26, 2011; accepted August 8, 2011. partially responsible for functional deficits obvious in changes Disclosure: , Allergan, Inc. (E, C); , Aller- S.J. Kirwin S.T. Kanaly in ERG,10,46 defective ␥-aminobutyric acid signaling,46,47 and gan, Inc. (E); C.R. Hansen, Allergan, Inc. (E); B.J. Cairns, Allergan, 25,45 Inc. (E); M. Ren, Allergan, Inc. (E); J.L. Edelman, Allergan, Inc. (E) abnormal gene expression reported in earlier microarrays. Corresponding author: Stefanie J. Kirwin, Department of Biologi- The data presented here show evidence of abnormal cell cal Sciences, Allergan, Inc., 2525 Dupont Drive, Irvine, CA 92612; metabolism, cellular interaction and proliferation, and de- [email protected]. creased expression of retinal genes involved in the visual cycle Investigative Ophthalmology & Visual Science, September 2011, Vol. 52, No. 10 7654 Copyright 2011 The Association for Research in Vision and Ophthalmology, Inc. Downloaded from iovs.arvojournals.org on 09/27/2021 IOVS, September 2011, Vol. 52, No. 10 Gene Expression and Visual Function in Diabetes 7655 in diabetic Long Evans rats. Furthermore, decreased visual Whole Genome Microarray function is observed in these animals. Both changes appear early during the disease before significant cell death in the One retina from each of three rats was removed and immediately retina or photoreceptor damage occurs, suggesting that hyper- transferred to tissue storage reagent (RNAlater; Ambion, Austin, TX), glycemia in the retina affects the vision-supporting function of and the pooled sample was incubated overnight at 4°C. Excess reagent secondary neurons or glial cells. (RNAlater; Ambion) was then removed, and the samples were stored at Ϫ70°C. RNA isolation and microarray was performed by Genus Bio- systems (Northbrook, IL). Briefly, after RNA isolation (RiboPure; Am- MATERIALS AND METHODS bion), RNA concentration was measured by spectrophotometry, and quality was assessed using a bioanalyzer (Agilent Technologies, Santa Animals and Timeline of Experiments Clara, CA). RNA was labeled and amplified, and resulting Cy3-cRNA was applied to a one-color rat whole genome array (Agilent Technol- Male Long Evans rats (Charles River Laboratories, Wilmington, MA), ogies) and scanned on a microarray scanner (G2565; Agilent Technol- ϳ each weighing 250 g, were used in all experiments. The timeline is ogies). represented in Figure 1. For microarray experiments, rats were killed 7 days, 4 weeks, and 3 months after the induction of diabetes. Each Microarray Analysis microarray experiment was performed using one retina from each of three rats as a pooled sample on one array chip (n ϭ 3 rats). Each data Microarray data were analyzed using gene expression software (Gene- ϭ ϭ point consisted of three independent experiments (n 3 chips; n Spring X; Agilent Technologies). Data were log2 transformed, quantile 9 rats per group). For real-time PCR, rats were killed after 1 month normalized, and baseline transformed to the median of all samples. (retina and RPE) or 3 months (retina only), and samples from one retina Probes with Ͻ50% of samples in diabetic or control samples present or of each of three rats were analyzed separately (retina) or pooled (RPE) marginal were excluded. After calculation of the average of the data for diabetic and control groups. Data were from three independent from three independent experiments, P value and fold changes were experiments (n ϭ 9 animals per group). Determination of the spatial established, and Student’s t-test with a cutoff of Ͻ0.05 was used to frequency threshold was performed on four diabetic and five control determine significance. All genes with fold changes Ͼ1.1 were in- animals each in two independent experiments (n ϭ 8 diabetic and n ϭ cluded in the analysis. 10 control rats). Histopathology and immunohistochemistry were per- Gene ontology analysis