Quantitative Analysis of Mouse Retinal Layers Using Automated Segmentation of Spectral Domain Optical Coherence Tomography Images

Assessment of Global and Local Alterations in Retinal Layer Thickness in Ins2 (Akita) Diabetic Mice by Spectral Domain Optical Coherence Tomography

PURPOSE/AIM

The Ins2 (Akita) mouse is a spontaneous diabetic mouse model with a heterozygous mutation in the insulin 2 gene that results in sustained hyperglycemia. The purpose of the study was to assess global and local retinal layer thickness alterations in Akita mice by analysis of spectral domain optical coherence tomography (SD-OCT) images.

MATERIALS AND METHODS

SD-OCT imaging was performed in Akita and wild-type mice at 12 and 24 weeks of age. Inner retinal thickness (IRT), outer retinal thickness (ORT), total retinal thickness (TRT), and photoreceptor outer segment length (OSL) were measured. Mean global thickness values were compared between Akita and wild-type mice. Local thickness variations in Akita mice were assessed based on normative values in wild-type mice.

RESULTS

Akita mice had higher blood glucose levels and lower body weights (p < 0.001). On average, IRT, ORT, and TRT were approximately 2% lower in Akita mice than in wild-type mice (p ≤ 0.02). In Akita mice, the percent difference between retinal areas with thickness below and above normative values for IRT, ORT, and TRT was 22%, 32%, and 38%, respectively.

CONCLUSIONS

These findings support the use of the Akita mouse model to study the retinal neurodegenerative effects of hyperglycemia.

In Vivo Imaging of Cx3cr1gfp/gfp Reporter Mice with Spectral-domain Optical Coherence Tomography and Scanning Laser Ophthalmoscopy

Spectral domain optical coherence tomography (SD-OCT) and scanning laser ophthalmoscopy (SLO) are extensively used in experimental ophthalmology. In the present protocol, mice expressing green fluorescent protein (gfp) under the promoter of Cx3cr1 (BALB/c-Cx3cr1^gfp/gfp) were used to image microglia cells in vivo in the retina. Microglia are resident macrophages of the retina and have been implicated in several retinal diseases^1,2,3,4,5,6. This protocol provides a detailed approach for generation of retinal B-scans, with SD-OCT, and imaging of microglia cell distribution in Cx3cr1^gfp/gfp mice with SLO in vivo, using an ophthalmic imaging platform system. The protocol can be used in several reporter mouse lines. However, there are some limitations to the protocol presented here. First, both SLO and SD-OCT, when used in the high-resolution mode, collect data with high axial resolution but the lateral resolution is lower (3.5 µm and 6 µm, respectively). Moreover, the focus and saturation level in SLO is highly dependent on parameter selection and correct alignment of the eye. Additionally, using devices designed for human patients in mice is challenging due to the higher total optical power of the mouse eye compared to the human eye; this can lead to lateral magnification inaccuracies⁷, which are also dependent on the magnification by the mouse lens among others. However, despite that the axial scan position is dependent upon lateral magnification, the axial SD-OCT measurements are accurate⁸.

The effect of topical administration of cyclopentolate on ocular biometry: An analysis for mouse and human models

Mydriasis with muscarinic antagonists have been used routinely prior to retinal examination and sometimes prior to refractive measurements of the mouse eye. However, biometric changes during topical administration of muscarinic antagonists have not been fully investigated in mice and humans. We found that the mouse eyes treated with cyclopentolate developed a hyperopia with a reduction in both the vitreous chamber depth and axial length. In humans, prior to the cyclopentolate treatment, a 6D accommodative stimulus produced a myopic shift with a reduced anterior chamber depth, choroidal thickness and anterior lens radius of curvature and an increase in lens thickness. After the cyclopentolate treatment, human eyes developed a hyperopic shift with an increased anterior chamber depth and anterior lens radius of curvature and a reduced lens thickness. Therefore, the biometric changes associated with this hyperopic shift were mainly located in the posterior segment of the eye in mice. However, it is the anterior segment of the eye that plays a main role in the hyperopic shift in human subjects. These results further indicate that mouse eyes do not have accommodation which needs to be taken into account when they are used for the study of human refractive errors.

Segmented inner plexiform layer thickness as a potential biomarker to evaluate open-angle glaucoma: Dendritic degeneration of retinal ganglion cell

Purpose

To evaluate the changes of retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), and ganglion cell-inner plexiform layer (GCIPL) thicknesses and compare structure-function relationships of 4 retinal layers using spectral-domain optical coherence tomography (SD-OCT) in macular region of glaucoma patients.

Methods

In cross-sectional study, a total of 85 eyes with pre-perimetric to advanced glaucoma and 26 normal controls were enrolled. The glaucomatous eyes were subdivided into three groups according to the severity of visual field defect: a preperimetric glaucoma group, an early glaucoma group, and a moderate to advanced glaucoma group. RNFL, GCL, IPL, and GCIPL thicknesses were measured at the level of the macula by the Spectralis (Heidelberg Engineering, Heidelberg, Germany) SD-OCT with automated segmentation software. For functional evaluation, corresponding mean sensitivity (MS) values were measured using 24–2 standard automated perimetry (SAP).

Results

RNFL, GCL, IPL, and GCIPL thicknesses were significantly different among 4 groups (P < .001). Macular structure losses were positively correlated with the MS values of the 24–2 SAP for RNFL, GCL, IPL, and GCIPL (R = 0.553, 0.636, 0.648 and 0.646, respectively, P < .001). In regression analysis, IPL and GCIPL thicknesses showed stronger association with the corresponding MS values of 24–2 SAP compared with RNFL and GCL thicknesses (R² = 0.420, P < .001 for IPL; R² = 0.417, P< .001 for GCIPL thickness).

Conclusions

Segmented IPL thickness was significantly associated with the degree of glaucoma. Segmental analysis of the inner retinal layer including the IPL in macular region may provide valuable information for evaluating glaucoma.

Relationship between Retinal Inner Nuclear Layer Thickness and Severity of Visual Field Loss in Glaucoma

Glaucoma is a disease characterized by pathologic changes in inner retinal layers, which are comprised of retinal ganglion cells (RGCs). As retinal ganglion cells (RGCs) cross over other retinal neurons that are connected by synapses, it is meaningful to investigate the outer retinal changes in glaucoma. We evaluated the association between thicknesses of segmented retinal layers in macular region and severity of visual field loss in open-angle glaucoma (OAG). This study involved 103 glaucomatous eyes. Retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform layer (OPL), and outer nuclear layer (ONL) thicknesses were measured at the macular level using the Spectral-domain optical coherence tomography with segmentation software. The functional losses were measured using 24-2 standard automated perimetry. Macular structure losses were positively correlated with functional loss for RNFL, GCL, and IPL (R = 0.550, 0.637, and 0.649, respectively, P < 0.001) and negatively correlated with INL (R = −0.295, P = 0.041). By multivariate regression analysis, INL thickness was significantly associated with visual field mean deviation (dB) and optic disc hemorrhage. These finding carefully suggest reactive responses of neuronal or glial cells located in the INL occur during glaucoma progression.

Increased susceptibility to fundus camera-delivered light-induced retinal degeneration in mice deficient in oxidative stress response proteins

Oxidative stress is an important contributor to the pathogenesis of many retinal diseases including age-related macular degeneration and retinal dystrophies. Light-induced retinal degeneration (LIRD) can serve as a model in which to study the response of the retina to stress. Of note, many genetic mutant mice are in a C57BL/6 J background and are thus resistant to the usual LIRD models. We recently developed a new model of fundus camera-delivered light-induced retinal degeneration (FCD-LIRD) which is effective in strains of mice expressing the light-resistant variant of RPE65 (450Met), including C57BL/6 J. In this work we investigated whether FCD-LIRD would be useful as a model in which to test the effect of genetic mutations on the response of the retina to stress. Furthermore, we tested whether oxidative stress plays an important role in the setting of this new FCD-LIRD model. FCD-LIRD was applied to C57BL/6 J mice and to mice simultaneously deficient in three proteins that are important in the response of the retina to oxidative stress (SOD1, DJ-1 and Parkin). Using fundus photography, we found that retinal damage was dramatically increased in the SOD1/DJ-1/Parkin deficient mice compared to C57BL/6 J. Outer retinal OCT volume and RPE cell morphology analysis in ZO-1-stained flat mounts added support to these findings. Gene expression analysis confirmed a strong oxidative stress response after FCD-LIRD, which was differentially altered in the SOD1/DJ1/Parkin deficient mice. We conclude that FCD-LIRD is useful to study the effect of genetic mutations on the response of the retina to light stress in light-resistant strains of mice. Furthermore, oxidative stress seems to be an important component of FCD-LIRD. Finally, we have established protocols to quantify the effect of FCD-LIRD on the retina and RPE which will be useful for future studies. Further dissection of the mechanisms by which the retina responds to light-induced oxidative stress may result in new strategies to modulate this response, which could lead to a reduction in retinal and RPE damage.

GARP2 accelerates retinal degeneration in rod cGMP-gated cation channel β-subunit knockout mice

The Cngb1 locus-encoded β-subunit of rod cGMP-gated cation channel and associated glutamic acid rich proteins (GARPs) are required for phototransduction, disk morphogenesis, and rod structural integrity. To probe individual protein structure/function of the GARPs, we have characterized several transgenic mouse lines selectively restoring GARPs on a Cngb1 knockout (X1^−/−) mouse background. Optical coherence tomography (OCT), light and transmission electron microscopy (TEM), and electroretinography (ERG) were used to analyze 6 genotypes including WT at three and ten weeks postnatal. Comparison of aligned histology/OCT images demonstrated that GARP2 accelerates the rate of degeneration. ERG results are consistent with the structural analyses showing the greatest attenuation of function when GARP2 is present. Even 100-fold or more overexpression of GARP1 could not accelerate degeneration as rapidly as GARP2, and when co-expressed GARP1 attenuated the structural and functional deficits elicited by GARP2. These results indicate that the GARPs are not fully interchangeable and thus, likely have separate and distinct functions in the photoreceptor. We also present a uniform murine OCT layer naming nomenclature system that is consistent with human retina layer designations to standardize murine OCT, which will facilitate data evaluation across different laboratories.

Effect of age and sex on retinal layer thickness and volume in normal eyes

The aim of the study was to evaluate the effect of sex and age on the thickness of the retinal layer in normal eyes using spectral-domain optical coherence tomography (SD-OCT).Fifty healthy subjects between the ages of 20 and 80 had their retinal layers measured using SD-OCT at Seoul St. Mary's Hospital. Mean thickness and volume were measured for 9 retinal layers in the fovea, the pericentral ring, and the peripheral ring. The differences of sex- and age-related thickness and volume in each retinal layer were analyzed.The retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), inner plexiform layer (IPL), inner nuclear layer (INL), and outer plexiform layer (OPL) were thinnest in the fovea area, whereas the outer nuclear layer (ONL), photoreceptor layer (PHL), and retinal pigment epithelium (RPE) were thickest at similar locations. Mean thickness of the RNFL, GCL, IPL, and OPL was significantly greater in men than women. However, mean thickness of the ONL was greater in women than in men. When compared between patients < 30 years and > 60 years of age, the thickness and volume of peripheral RNFL, GCL, and pericentral and peripheral IPL were significantly larger in the younger group than the older group. Conversely, the thickness and volume of foveal INL and IR were larger in the older group than in the younger group.The thickness and volume of the retinal layer in normal eyes significantly vary depending on age and sex. These results should be considered when evaluating layer analysis in retinal disease.

IRBP deficiency permits precocious ocular development and myopia

PURPOSE

Interphotoreceptor retinoid-binding protein (IRBP) is abundant in the subretinal space and binds retinoids and lipophilic molecules. The expression of IRBP begins precociously early in mouse eye development. IRBP-deficient (KO) mice show less cell death in the inner retinal layers of the retina before eyelid opening compared to wild-type C57BL/6J (WT) controls and eventually develop profound myopia. Thus, IRBP may play a role in eye development before visually-driven phenomena. We report comparative observations during the course of the natural development of eyes in WT and congenic IRBP KO mice that suggest IRBP is necessary at the early stages of mouse eye development for correct function and development to exist in later stages.

METHODS

We observed the natural development of congenic WT and IRBP KO mice, monitoring several markers of eye size and development, including haze and clarity of optical components in the eye, eye size, axial length, immunohistological markers of differentiation and eye development, visually guided behavior, and levels of a putative eye growth stop signal, dopamine. We conducted these measurements at several ages. Slit-lamp examinations were conducted at post-natal day (P)21. Fundus and spectral domain optical coherence tomography (SD-OCT) images were compared at P15, P30, P45, and P80. Enucleated eyes from P5 to P10 were measured for weight, and ocular dimensions were measured with a noncontact light-emitting diode (LED) micrometer. We counted the cells that expressed tyrosine hydroxylase (TH-positive cells) at P23-P36 using immunohistochemistry on retinal flatmounts. High-performance liquid chromatography (HPLC) was used to analyze the amounts of dopamine (DA) and 3,4-dihydroxyphenylacetic acid (DOPAC) at P7-P60. Monocular form deprivation in the right eye was induced using head-mounted goggles from P28 to P56.

RESULTS

Eye elongation and eye size in the IRBP KO mice began to increase at P7 compared to the WT mice. This difference increased until P12, and the difference was maintained thereafter. SD-OCT images in live mice confirmed previously reported retinal thinning of the outer nuclear layer in the IRBP KO mice compared to the WT mice from P15 to P80. Slit-lamp and fundoscopy examination outcomes did not differ between the WT and KO mice. SD-OCT measurements of the optical axis components showed that the only factor contributing to excess optical axis length was the depth of the vitreous body. No other component of optical axis length (including corneal thickness, anterior chamber depth, and lens thickness) was different from that of the WT mouse. The refractive power of the IRBP KO mice did not change in response to form deprivation. The number of retinal TH-positive cells was 28% greater in the IRBP KO retinas compared to the WT mice at P30. No significant differences were observed in the steady-state retinal DA or DOPAC levels or in the DOPAC/DA ratios between the WT and IRBP KO mice.

CONCLUSIONS

The IRBP KO mouse eye underwent precocious development and rapid eye size growth temporally about a day sooner than the WT mouse eye. Eye size began to differ between the WT and KO mice before eyelid opening, indicating no requirement for focus-dependent vision, and suggesting a developmental abnormality in the IRBP KO mouse eye that precedes form vision-dependent emmetropization. Additionally, the profoundly myopic KO eye did not respond to form deprivation compared to the non-deprived contralateral eye. Too much growth occurred in some parts of the eye, possibly upsetting a balance among size, differentiation, and focus-dependent growth suppression. Thus, the loss of IRBP may simply cause growth that is too rapid, possibly due to a lack of sequestration or buffering of morphogens that normally would bind to IRBP but are unbound in the IRBP KO eye. Despite the development of profound myopia, the DA levels in the IRBP KO mice were not statistically different from those in the WT mice, even with the excess of TH-positive cells in the IRBP KO mice compared to the WT mice. Overall, these data suggest that abnormal eye elongation in the IRBP KO mouse is independent of, precedes, and is epistatic to the process(es) of visually-driven refractive development.

Effect of pharmacologically induced retinal degeneration on retinal autofluorescence lifetimes in mice

Fluorescence lifetime imaging ophthalmoscopy (FLIO) was used to investigate retinal autofluorescence lifetimes in mouse models of pharmacologically induced retinal degeneration over time. Sodium iodate (NaIO₃, 35 mg/kg intravenously) was used to induce retinal pigment epithelium (RPE) degeneration with subsequent loss of photoreceptors (PR) whereas N-methyl-N-nitrosourea (MNU, 45 mg/kg intraperitoneally) was employed for degeneration of the photoreceptor cell layer alone. All mice were measured at day 3, 7, 14, and 28 after the respective injection of NaIO₃, MNU or NaCl (control). Fluorescence lifetime imaging was performed using a fluorescence lifetime imaging ophthalmoscope (Heidelberg Engineering, Heidelberg, Germany). Fluorescence was excited at 473 nm and fluorescence lifetimes were measured in a short and a long spectral channel (498-560 nm and 560-720 nm). Corresponding optical coherence tomography (OCT) images were consecutively acquired and histology was performed at the end of the experiments. Segmentation of OCT images and histology verified the cell type-specific degeneration process over time. Retinal autofluorescence lifetimes increased from day 3 to day 28 in mice after NaIO₃ treatment. Finally, at day 28, fluorescence lifetimes were prolonged by 8% in the short and 61% in the long spectral channel compared to control animals (p = 0.21 and p = 0.004, respectively). In mice after MNU treatment, the mean retinal autofluorescence lifetimes were already decreased at day 3 and retinal lifetimes were finally shortened by 27% in the short and 51% in the long spectral channel at day 28 (p = 0.0028). In conclusion, degeneration of the RPE with subsequent photoreceptor degeneration by NaIO₃ lead to longer mean fluorescence lifetimes of the retina compared to control mice, whereas during specific degeneration of the photoreceptor layer induced by MNU shorter lifetimes were measured. Therefore, short retinal fluorescence lifetimes may originate from the RPE and may be modified by the overlaying retinal layers.

Optical Coherence Tomography Angiography in Mice: Comparison with Confocal Scanning Laser Microscopy and Fluorescein Angiography

PURPOSE

Optical coherence tomography angiography (OCT-A) allows noninvasive visualization of retinal vessels in vivo. OCT-A was used to characterize the vascular network of the mouse retina and was compared with fluorescein angiography (FA) and histology.

METHODS

In the present study, OCT-A based on a Heidelberg Engineering Spectralis system was used to investigate the vascular network in mice. Data was compared with FA and confocal microscopy of flat-mount histology stained with isolectin IB4. For quantitative analysis the National Cancer Institute's AngioTool software was used. Vessel density, the number of vessel junctions, and endpoints were measured and compared between the imaging modalities.

RESULTS

The configuration of the superficial capillary network was comparable with OCT-A and flat-mount histology in BALBc mice. However, vessel density and the number of vessel junctions per region of interest (P = 0.0161 and P = 0.0015, respectively) in the deep vascular network of BALBc mice measured by OCT-A was significantly higher than with flat-mount histology. In C3A.Cg-Pde6b+Prph2Rd2/J mice, where the deep capillary plexus is absent, analysis of the superficial network provided similar results for all three imaging modalities.

CONCLUSION

OCT-A is a helpful imaging tool for noninvasive, in vivo imaging of the vascular plexus in mice. It may offer advantages over FA and confocal microscopy especially for imaging the deep vascular plexus.

TRANSLATIONAL RELEVANCE

The present study shows that OCT-A can be employed for small animal imaging to assess the vascular network and offers advantages over flat-mount histology and FA.

Retinal neurodegeneration may precede microvascular changes characteristic of diabetic retinopathy in diabetes mellitus

Diabetic retinopathy (DR) has long been recognized as a microvasculopathy, but retinal diabetic neuropathy (RDN), characterized by inner retinal neurodegeneration, also occurs in people with diabetes mellitus (DM). We report that in 45 people with DM and no to minimal DR there was significant, progressive loss of the nerve fiber layer (NFL) (0.25 μm/y) and the ganglion cell (GC)/inner plexiform layer (0.29 μm/y) on optical coherence tomography analysis (OCT) over a 4-y period, independent of glycated hemoglobin, age, and sex. The NFL was significantly thinner (17.3 μm) in the eyes of six donors with DM than in the eyes of six similarly aged control donors (30.4 μm), although retinal capillary density did not differ in the two groups. We confirmed significant, progressive inner retinal thinning in streptozotocin-induced "type 1" and B6.BKS(D)-Lepr(db)/J "type 2" diabetic mouse models on OCT; immunohistochemistry in type 1 mice showed GC loss but no difference in pericyte density or acellular capillaries. The results suggest that RDN may precede the established clinical and morphometric vascular changes caused by DM and represent a paradigm shift in our understanding of ocular diabetic complications.