Subretinal injection damage has implication for experimental control and treatment in rat model of retinopathy of prematurity

In their 2017 article, “Effect of subretinal injection on retinal structure and function in a rat oxygen-induced retinopathy model,” Becker et al used the Phoenix MICRON IV fundus camera, Phoenix MICRON OCT2 and corresponding layer analysis software Insight 2D, and the Phoenix MICRON focal ERG to find that subretinal injection of saline or even introduction of a needle into the vitreal space has a detrimental effect on a rat model of retinopathy of prematurity. In normal adult rats, there is no effect of subretinal injections (results in other species vary) so these rats with oxygen-induced retinopathy may be more sensitive to damage.

Becker et al induced a retinopathy of prematurity model in rat pups by oxygen cycling from birth through P14. At P8, they either subretinally injected PBS, introduced a needle into the vitreous without injecting anything (pilot-treated eyes), or did not inject eyes at all. They examined the fundus appearance with the Phoenix MICRON IV and found no difference in appearance (Fig 1 shows the fundus appearance for oxygen-induced retinopathy rats as well as a GFP-expressing lentivirus for proof of concept). The Phoenix MICRON IV OCT2 revealed reduced photoreceptor outer segment and outer nuclear layer thickness in both the PBS-injected and pilot-injected eyes (Fig 2A). The thickness was measured with the semi-automated Insight 2D OCT analysis software (Fig 2B).

The Phoenix MICRON focal ERG measured the electric response of the retina at the injection spot; since the focal ERG allows precise selection of stimulation area (under red illumination to maintain dark adaptation), Becker et al could identify and stimulate the injection site. The PBS-injected and pilot-injected eyes had reduced a-waves but normal b-waves (Fig 3). This aligns with the Phoenix MICRON IV OCT findings that the photoreceptor layer was reduced.

Oxygen-induced retinopathy manifests as retinal thinning and reduced retinal function. Becker et al demonstrated that subretinal injection or even insertion of a needle into the vitreous cause more thinning and further reduced function. This has implications for considering treatment and when selecting control protocols in future studies.

Becker, S., Wang, H., Stoddard, G. J., & Hartnett, M. E. (2017). Effect of subretinal injection on retinal structure and function in a rat oxygen-induced retinopathy model. Molecular Vision, 23, 832–843.

fundus images of subretinal injections
Figure 1. Fundus images and localization of subretinal injections. A–B: The fundus images of the untreated and pilot- and subretinal PBS injection (sPBS)-treated eyes were normal. D–E: After subretinal injection of pFmCD44 green fluorescent protein (GFP) lentivirus, the fundus image was normal (D), and GFP expression was localized in the central retina (E).
OCT images in untreated eyes
Figure 2. OCT in untreated eyes, after sham treatment and subretinal PBS injection. A: Representative images of OCT images in untreated eyes and after sham injection and subretinal PBS. B: The thickness of the PRS+ONL was significantly reduced by sham treatment (**p<0.01, n=7) and subretinal PBS injection (*p<0.05, n=5) compared to untreated eyes (n=4). No significant differences were observed between any groups in the OPL+INL, IPL+GCL or total thickness.
Reduced A-wave focal ERG images
Figure 3. Reduced a-wave in focal ERGs in the central retina of pilot- and sPBS-treated eyes compared to untreated eyes. A: Representative recordings of scotopic focal electroretinograms (ERGs) in the untreated and pilot- and subretinal PBS injection (sPBS)-treated eyes. B: The amplitudes of the scotopic a- and b-waves in the pilot- (**p<0.01, n=10) and sPBS-treated eyes (†††p<0.001, n=8) were statistically significantly reduced compared to those in the untreated eyes (n=9). Note that a- amplitude is usually negative but is recorded as a positive absolute value for ease in comparison