A low-cost, easy-to-use and powerful system is established to evaluate potential treatments that could ameliorate blood retinal barrier breach induced by histamine. Blood vessel leakage, Müller cell activation and the continuity of neuronal processes are utilized to assess the damage response and its reversal with a potential drug, lipoxin A4.
Cite this ArticleCopy Citation
Wu, H., Rodriguez, A. R., Spur, B. W., Venkataraman, V. An Acute Retinal Model for Evaluating Blood Retinal Barrier Breach and Potential Drugs for Treatment. J. Vis. Exp. (115), e54619, doi:10.3791/54619 (2016).
Translate text to:
A low-cost, easy-to-use and powerful model system is established to evaluate potential treatments that could ameliorate blood retinal barrier breach. An inflammatory factor, histamine, is demonstrated to compromise vessel integrity in the cultured retina through positive staining of IgG outside of the blood vessels. The effects of histamine itself and those of candidate drugs for potential treatments, such as lipoxin A4, are assessed using three parameters: blood vessel leakage via IgG immunostaining, activation of Müller cells via GFAP staining and change in neuronal dendrites through staining for MAP2. Furthermore, the layered organization of the retina allows a detailed analysis of the processes of Müller and ganglion cells, such as changes in width and continuity. While the data presented is with swine retinal culture, the system is applicable to multiple species. Thus, the model provides a reliable tool to investigate the early effects of compromised retinal vessel integrity on different cell types and also to evaluate potential drug candidates for treatment.
A growing body of evidence supports the existence of blood retinal barrier (BRB)1-5 and its similarity to the blood brain barrier (BBB)6,7. Compromise of the BBB has been tightly linked causally or as a diagnostic marker to chronic neurodegenerative diseases such as Alzheimer's disease (AD)8,9 and acute conditions such as delirium10. Mechanistic insights into these pathologies and discoveries for potential drug targets are generally hampered by the limited accessibility and network intricacy of the brain. Alternatives such as in vivo imaging11, brain organotypic culture12, primary cell cultures13,14 and co-culture systems15 have been generated. However, most of these models require special instruments, long experimental periods or multiple markers to identify cells. Functional and structural similarities between BBB and BRB as well as a correlation between the dysfunctions of the two have been argued16-19. In addition, easier access, well-defined cell types and a layered structure have allowed the well-characterized retina as a window to the brain. The structural and functional identities of the BBB and BRB remain to be compared in detail. However, retinal pathologies, especially the BRB breach, have also been tightly associated with the progression of various diseases, including diabetes18-19 and AD21,22. Thus, it is of interest to establish a BRB dysfunction system not only to delineate the mechanism but also to screen potential drugs. In this report, a protocol enabling BRB dysfunction using a simple acute retinal culture is developed and presented.
Increased BBB permeability and AD-like pathological changes have been established in a brain organotypic culture incubated with histamine, a pro-inflammatory mediator12. Therefore, in the presented system, histamine was applied to the ex vivo retinal culture to induce BRB dysfunction. Retinas from several species, such as Mus musculus and Bos Taurus, have been tested. Due to their commercial availability and resemblance to human tissue, fresh swine eyeballs were utilized to provide the data reported here. After incubating with histamine and/or other drugs, the retinas were processed for evaluation by immunostaining for several proteins12, such as Immunoglobulin G (IgG), one of the major components of the blood; glial fibrillary acidic protein (GFAP), a well-known marker for glial activation; and microtubule-associated protein 2 (MAP2), a neuron-specific cytoskeletal protein essential for microtubule assembly. Furthermore, the layered structure of the retina allows a detailed analysis of the processes of Müller cells and ganglion cells, such as changes in their width and continuity. Thus several additional parameters are available to assess the consequences of the BRB breach at an early stage and to evaluate the reversal effects of potential treatments as well.
In this protocol, potential reversal effects of screened drugs are evaluated from three perspectives: the leakage of blood vessels (BVs), the activation of glial cells and the damage-response of neuronal cells. Several quantification methods are utilized, for instance, the expression level shown by intensity of the immunostaining, width measurement of a process and continuity of neuronal processes shown by an enhancement filter. To better illustrate the method and to help interpret results, lipoxin A4 (LXA4), a compound endogenously synthesized in response to inflammatory injury and attenuating endothelial dysfunction23, has been chosen for demonstration purposes.
All protocols were carried out in compliance with the policies of the Institutional Animal Care and Use Committee where applicable.
- Prepare the stabilization media with 75% Dulbecco's Modified Eagle Medium (DMEM), 25% Hanks' Balanced Salt Solution (HBSS). Mix well, aliquot and store at -20 °C until use.
- Prepare the Phosphate Buffered Saline (PBS) and sterilize by autoclaving Store the solution at room temperature. Warm the PBS to 37 °C before the experiment (5 ml per well).
- Prepare 10%, 20% and 30% sucrose in PBS. Filter all the solutions individually over separate 0.22 μm filters for sterilization. Save the solutions at 4 °C.
- Prepare the histamine stock solution in PBS to a concentration of 90 mM. Sterilize the solution by filtering over a 0.22 µM filter. Aliquot and save the solution at -80 °C. Avoid multiple freeze-and-thaw cycles.
- Make fresh 4% paraformaldehyde (PFA) in PBS before the experiment. Place it on ice.
Caution: Keep PFA in a fume hood and wear appropriate protective equipment.
- Thaw the stabilization medium (20 ml per retina). Add Penicillin-Streptomycin to a final concentration of 100 U/ml. Warm part of the medium (15 ml per retina) to 37 °C in incubator before the experiment. Leave the rest of the medium on ice.
- Place the HBSS (10 ml per retina) on ice.
2. Retinal Organotypic Culture
- Transfer the samples to a tissue culture hood Open the eyecup by cutting around the lens. Using a brush, gently remove the vitreous humor without pulling on the retina. Gently detach the retina from the cut edge with a brush to expose the optic disc. Severe the optic nerve with a razor and release the retina.
- Transfer the retina into a Petri dish and carefully rinse the retina once using cold HBSS. Place the sample in HBSS on ice. Dissect as many retinas out as needed by repeating Step 2.1 and this step.
- Cut the retina in half symmetrically with a razor. Gently transfer samples to a 6-well plate with 3 ml of stabilization media per well. Allow equilibration for 30 min at 37 °C in an incubator with the atmosphere containing 5% CO2.
- Prepare the following reagents during the incubation at Step 2.3: Dilute the histamine stock solution to the desired concentration (450 μM) in warm medium. Dissolve LXA4 (stored under Argon) into the medium to a final concentration of 250 ng/ml.
NOTE: When measuring the effect of LXA4, one half from a single retina is used for histamine alone, while the other half is used for histamine with LXA4.
- Aspirate the stabilization medium carefully and add the prepared medium to each well (3 ml per well). Incubate the samples at 37 °C for 1 hr in an incubator with the atmosphere containing 5% CO2.
- Rinse the samples once in warm, sterile PBS.
3. Prepare the Samples for Immunostaining
- Aspirate PBS from the plates. Add 4% PFA (3 ml per well) and incubate for 15 min at room temperature.
- Quickly aspirate the PFA. Rinse the samples once using PBS. Transfer the samples to 10% sucrose (5 ml per well) and incubate for 2 to 4 hr at 4 °C.
- Transfer the samples to a 30% sucrose (5 ml per well) and incubate overnight at 4 °C. Mix one part of the commercial freezing medium with two parts of 20% sucrose to make the working freezing medium. Leave it at 4 °C overnight or longer until the air bubbles disappear.
- Transfer the samples to the working freezing medium and allow equilibration for 5 min at room temperature. Trim each retina into rectangles (roughly 3 mm by 5 mm).
- Carefully transfer the samples into the working freezing media contained in a cylindrical container (about 1 cm radius x 2 cm height) devised from aluminum foil. Ensure that the retinal slices are vertical (oriented to provide a cross-section of the retina upon sectioning) and freeze them in liquid nitrogen Save the blocks at -80 °C until use.
- Section each block on a cryostat into 14 μm thick slices and mount the sections on glass slides. Store the slides at -80 °C until use.
- Wash the sections with 5% sucrose phosphate buffer (SPB, 5% sucrose in PBS, pH 7.4) once. Block the non-specific binding sites by incubating the sections in blocking buffer (5% SPB with 10% normal goat serum and 0.5% Triton X-100) for 1 hr at room temperature.
- Incubate the sections (on slides) with the appropriate primary antibody (GFAP or MAP2, diluted at 1:500 in blocking buffer) for 1 hr. Aspirate the solution. Wash three times in 5% SPB. For staining the endogenous IgG, skip this step.
- Incubate the sections with the corresponding secondary antibodies (Cyanine3/FITC conjugated donkey anti-mouse/anti-rabbit IgG, diluted at 1:800 in the blocking buffer) for 1 hr.
- Wash the sections thoroughly with 5% SPB three times. Prepare the samples for microscopy by mounting them in mounting medium containing DAPI and covering with coverslips.
- Using a laser confocal microscope, capture images through a 20X lens under identical parameters such as laser intensity, gain value, dwell time, etc. across groups. Set the excitation/emission wavelengths for the blue channel (to detect DAPI) as 408/447 nm, for the red channel (to detect Cyanine3) as 561/785 nm and for the green channel (to detect FITC) as 488/525 nm.
5. Image Analysis
- Quantify percentage of leaky blood vessels according to the following criteria12: Include only blood vessels with endothelial cell nuclei within the plane of section in the count; consider a blood vessel leaky if it shows a gradient of immunostaining surrounding the vessel.
- To determine the expression level, select regions of interest and measure the mean grey value using the image analysis software.
- To measure the width of the process, choose an area of the section where the processes are distinct (the outer nuclear layer, ONL, for example). Then choose an optical field where such processes are visible and continuous. Set the scale bar according to the microscopic setting, draw a line across the process and perform the measurement.
- To analyze the continuity of the process, select regions of interest, apply the filter called variance which enhances the edges in the image by replacing each pixel with its neighborhood variance, automatically adjust the contrast and brightness, count the lines and normalize the count by the area.
- Calculate statistical significances using a two-tailed Student's t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Plot the result as Mean ± SEM.
We present a low-cost, time-efficient and easy-to-use system to evaluate potential treatments that could protect against the BRB breach induced by histamine. IgG is constrained within the vessels in control retina (Figure 1A), but leaks out of the blood vessels upon histamine exposure (Figure 1B), confirming that the model is successfully established.
LXA4 was chosen to demonstrate the screening for drugs against BRB breach in the presented system. BRB dysfunction is attenuated upon treatment with LXA4 (Figure 1D), which has no significant effect by itself (Figure 1C). The reversal is quantitated as a percentage of the leaked blood vessels and the result is significant (Figure 1E). Two other retinal cell types, Müller glial cells (Figure 2) and ganglion cells (Figure 3), are further assessed. LXA4 treatment reverses the decrease in the width of the Müller glial cell processes (Figure 2E) shown through GFAP staining (Figure 2, A-D) but has no significant effects on the continuity of the dendritic processes in ganglion cells (Figure 3G), shown by MAP2 staining (Figure 3, A-E). It is also noted that no significant change in the expression level of either proteins is detected.
Figure 1: The Blood Retinal Barrier is Compromised upon Histamine Exposure and Rescued by LXA4 Treatment. IgG immunostaining (red) is presented from retina which was untreated (A), damage-induced with histamine only (B), treated with LXA4 only (C), or exposed to both histamine and LXA4 (D). In control (A) and LXA4 treated (C) groups, IgG is restricted within the blood vessels (BVs) while in the histamine group (B), IgG is detected out of the BVs where it forms leakage clouds indicated by dotted circles. Further addition of LXA4 rescues the vessel dysfunction (D). Scale bar, 20 μm. (E) Quantitative measurement of the leaky BVs across the tested groups. Statistical significance is determined by a two-tailed Student's t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Mean ± SEM (n = 18) is plotted in the graph. Please click here to view a larger version of this figure.
Figure 2: Müller Cell Processes are Compromised by Histamine and Rescued by LXA4 Treatment. Immunostaining for GFAP is presented from retina which was untreated (A), damage-induced with histamine only (B) treated with LXA4 only (C) or exposed to both histamine and LXA4 (D). Positive staining is obtained in the processes of Müller cells across the retina and around the BVs. Scale bar, 20 μm. (E) The width of the Müller cell processes from all groups are presented. About 30 individual processes were measured per group. Statistical significance is determined by a two-tailed Student's t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Mean ± SEM (n = 30) is plotted in the graph. Please click here to view a larger version of this figure.
Figure 3: The Continuity of Ganglion Cell Processes is not Affected by LXA4. (A) Immunostaining against MAP2 is presented from retina which was untreated (A), damage-induced with histamine only (B) treated with LXA4 only (C) or exposed to both histamine and LXA4 (D). Positive staining is obtained in the processes and cell bodies of ganglion cells. A sample image of MAP2 staining (E) together with its corresponding filter-enhanced image (F) is presented. The continuous processes are indicated by arrowheads. Scale bar, 20 μm. (G) The density of the continuous, MAP2-positive ganglion cell processes from all groups are presented. Statistical significance is determined by a two-tailed Student's t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Mean ± SEM (n= 18) is plotted in the graph. Please click here to view a larger version of this figure.
In this report, we present a powerful ex vivo acute retinal model of BRB dysfunction using the swine retina. This model system does not require special instruments and can be easily adapted under most laboratory settings. However, to obtain a successful result, several steps require close attention. After obtaining the eyeballs from the source, they must be kept at 4 °C or on ice and processed as soon as possible. When the effect of a treatment is being analyzed, two halves of the same retina must be used - one half to detect changes solely due to experimental procedures (baseline) and the other to detect the effect of the treatment. This approach also allows corrections for animal variations such as age, sex and health.
This protocol can be successfully applied to retina from various species, such as M. musculus and B. Taurus (data not shown). Using commercial swine eyeballs is a significant cost-saving factor compared to maintaining the animals. Furthermore, in this study, the incubation time with histamine is only 60 minutes. This short incubation time provides the advantage of teasing out the early responses. It is noted that the study was carried out over different durations (20 min to 120 min) and different histamine concentrations (up to 700 μM). Effects are observable at 30 min and at 225 μM histamine. For illustrative purposes, the result obtained at 60 min with 450 μM histamine is presented. In its present form, the following are the limitations of the model: (1) use of commercially obtained eyeballs increases the variability (but significantly cuts down on the cost); (2) the short incubation time may not allow complete manifestation of the response to the treatment - for example, those that require de novo protein synthesis. Modifications may help including longer incubation periods that could enable monitoring delayed responses.
LXA4 is chosen as the drug to investigate possible protection against the histamine-induced BRB damage. LXA4 is an endogenous compound that plays a key role in inflammatory events. It is produced in response to stress in vivo and has a short half life23. To mimic these conditions, a single addition of LXA4 was carried out along with histamine. This model system could also be used to further investigate if a prior treatment with LXA4 (or any other drug) would yield protection as well as to evaluate the duration of the protective treatment.
To the best of our knowledge, this is the first report of a retinal culture system with an experimentally-induced BRB breach. Evidently, exogenous perturbations other than histamine, such as lipopolysaccharides (LPS)24 and vascular endothelial growth factor (VEGF)25, can also be utilized to induce BRB breach. The effects of those perturbations and their potential treatments can be investigated on different cell types such as endothelial, Müller glial and neuronal cells in the presented model. Analyses may be performed on the entire retina or on one or more specific layers. Available cell-specific markers for retinal cells and synaptic layers allow an efficient detection of the responses via a large combination of parameters. Thus, we have established a retinal culture system for the rapid screening of products that protect against BRB dysfunction.
The authors have nothing to disclose.
Bringhurst Meats (Berlin, NJ) is acknowledged for their genuine help in providing the swine eyeballs.
|PFA||Electron Microscopy Sciences||15710|
|Freezing Media||Triangle Biomedical Sciences||TFM-5|
|Normal Goat Serum||Rockland||D104-00-0050|
|MAP2 Antibody||EMD Millipore||MAB3418|
|FITC conjugated Donkey anti-rabbit IgG||Jackson ImmunoResearch Laboratories, Inc.||711-095-152|
|Cy3 conjugated Donkey anti-mouse IgG||Jackson ImmunoResearch Laboratories, Inc.||715-165-150|
|mounting medium containing DAPI||Vector Laboratories, Inc.||H-1200|
|Laser Confocal Microscope||Nikon||Eclipse Ti microscope|
|ImageJ||National Institutes of Health||1.45s|
- Cunha-Vaz, J., Bernardes, R., Lobo, C. Blood-retinal barrier. J Ophthalmol. 21, 3 (2011).
- Kim, J. H., et al. Blood-neural barrier: intercellular communication at glio-vascular interface. J Biochem Molec Biol. 39, 339-345 (2006).
- Kumagai, A. K. Glucose transport in brain and retina: implications in the management and complications of diabetes. Diabetes Metab Res Rev. 15, 261-273 (1999).
- Runkle, E. A., Antonetti, D. A. The blood-retinal barrier: structure and functional significance. Methods Molec Biol. 686, Clifton, N.J. 133-148 (2011).
- Takata, K., Hirano, H., Kasahara, M. Transport of glucose across the blood-tissue barriers. Int Rev Cytol. 172, 1-53 (1997).
- Goncalves, A., Ambrosio, A. F., Fernandes, R. Regulation of claudins in blood-tissue barriers under physiological and pathological states. Tissue Barriers. 1, 24782 (2013).
- Patton, N., et al. Retinal image analysis: concepts, applications and potential. Prog Ret Eye Res. 25, 99-127 (2006).
- Di Marco, L. Y., et al. Vascular dysfunction in the pathogenesis of Alzheimer's disease--A review of endothelium-mediated mechanisms and ensuing vicious circles. Neurobiol Dis. 82, 593-606 (2015).
- Nagele, R. G., et al. Brain-reactive autoantibodies prevalent in human sera increase intraneuronal amyloid-beta(1-42) deposition. J Alzheimer's Dis: JAD. 25, 605-622 (2011).
- Goldwaser, E., Acharya, N., Nagele, R. Cerebrovascular and blood-brain barrier compromise: A mechanistic link between vascular disease and Alzheimer's disease subtypes of neurocognitive disorders. J Parkinsons Dis Alzheimer's Dis. 2, 10 (2015).
- Horton, N. G., et al. In vivo three-photon microscopy of subcortical structures within an intact mouse brain. Nat Photonics. 7, (2013).
- Sedeyn, J. C., et al. Histamine Induces Alzheimer's Disease-Like Blood Brain Barrier Breach and Local Cellular Responses in Mouse Brain Organotypic Cultures. Biomed Res Int. 2015, 937148 (2015).
- Avdeef, A., Deli, M. A., Neuhaus, W. Blood-Brain Barrier in Drug Discovery: Optimizing Brain Exposure of CNS Drugs and Minimizing Brain Side Effects for Peripheral Drugs. Di, L., Kerns, E. H. John Wiley & Sons, Inc. Ch. 10 188 (2014).
- Eigenmann, D. E., et al. Comparative study of four immortalized human brain capillary endothelial cell lines, hCMEC/D3, hBMEC, TY10, and BB19, and optimization of culture conditions, for an in vitro blood-brain barrier model for drug permeability studies. Fluids Barriers CNS. 10, 33 (2013).
- Takeshita, Y., et al. An in vitro blood-brain barrier model combining shear stress and endothelial cell/astrocyte co-culture. J Neurosci Methods. 232, 165-172 (2014).
- Alberghina, M., Lupo, G., Anfuso, C. D., Moro, F. Palmitate transport through the blood-retina and blood-brain barrier of rat visual system during aging. Neurosci Lett. 150, 17-20 (1993).
- Hosoya, K., Yamamoto, A., Akanuma, S., Tachikawa, M. Lipophilicity and transporter influence on blood-retinal barrier permeability: a comparison with blood-brain barrier permeability. Pharm Res. 27, 2715-2724 (2010).
- Minamizono, A., Tomi, M., Hosoya, K. Inhibition of dehydroascorbic acid transport across the rat blood-retinal and -brain barriers in experimental diabetes. Biol Pharm Bull. 29, 2148-2150 (2006).
- Serlin, Y., Levy, J., Shalev, H. Vascular pathology and blood-brain barrier disruption in cognitive and psychiatric complications of type 2 diabetes mellitus. Cardiovasc Psychiatry Neurol. 2011, 609202 (2011).
- Kolb, H. How the retina works - Much of the construction of an image takes place in the retina itself through the use of specialized neural circuits. Am Sci. 91, 28-35 (2003).
- Ikram, M. K., Cheung, C. Y., Wong, T. Y., Chen, C. P. Retinal pathology as biomarker for cognitive impairment and Alzheimer's disease. J Neurol Neurosurgery Psychiatry. 83, 917-922 (2012).
- Tan, Z., Ge, J. Amyloid-beta, the retina, and mouse models of Alzheimer disease. Am J Pathol. 176, 2055 (2010).
- Serhan, C. N. Pro-resolving lipid mediators are leads for resolution physiology. Nature. 510, 92-101 (2014).
- Bucolo, C., et al. Effects of topical indomethacin, bromfenac and nepafenac on lipopolysaccharide-induced ocular inflammation. J Pharm Pharmacol. 66, 954-960 (2014).
- Edelman, J. L., Lutz, D., Castro, M. R. Corticosteroids inhibit VEGF-induced vascular leakage in a rabbit model of blood-retinal and blood-aqueous barrier breakdown. Exp Eye Res. 80, 249-258 (2005).