Retinal expression of Fgf2 in RCS rats with subretinal microphotodiode array

Abstract

Purpose: To test the hypothesis that subretinal electrical stimulation from a microphotodiode array (MPA) exerts a neuroprotective effect in Royal College of Surgeons (RCS) rats through the induction of growth factors.

Methods: At postnatal day 21, RCS rats were divided into four groups in which one eye per rat received treatment: (A) active MPA, (M) minimally active MPA, (S) sham surgery, or (C) no surgery and the opposite eye was unoperated. Dark- and light-adapted ERGs were recorded 1 week after surgery. A second set of A-, M-, and C-treated RCS rats had weekly ERG recordings for 4 weeks. Real-time RT-PCR was used to measure relative expression of mRNAs (Bdnf, Fgf2, Fgf1, Cntf, Gdnf, and Igf1) in retina samples collected 2 days after the final ERG.

Results: One week after surgery, there was a slight difference in dark-adapted ERG b-wave at the brightest flash intensity. Mean retinal Fgf2 expression in the treated eye relative to the opposite eye was greater for the A group (4.67 +/- 0.72) than for the M group (2.80 +/- 0.45; P = 0.0501), S group (2.03 +/- 0.45; P < 0.01), and C group (1.30 +/- 0.22; P < 0.001). No significant change was detected for Bdnf, Cntf, Fgf1, Gdnf, and Igf1. Four weeks after surgery, the A group had significantly larger dark- and light-adapted ERG b-waves than for the M and C groups (P < 0.01). Simultaneously, mean relative Fgf2 expression was again greater for the A group (3.28 +/- 0.61) than for the M (1.28 +/- 0.32; P < 0.05) and C (1.05 +/- 0.04; P < 0.05) groups.

Conclusions: The results show subretinal implantation of an MPA induces selective expression of Fgf2 above that expected from a retina-piercing injury. Preservation of ERG b-wave amplitude 4 weeks after implantation is accompanied by elevated Fgf2 expression. These results suggest that Fgf2 may play a role in the neuroprotection provided by subretinal electrical stimulation.

Figure 1

In vitro testing of the active and minimally-active microphotodiode arrays. MPAs were placed in PBS and exposed to an 870 nm flash of 10.5 mW/cm² intensity and 1 msec duration. The graphs show representative current output of active (A) and minimally-active (B) MPAs from this testing. Note the change in current scale from microamps to nanoamps between A and B, respectively. The active MPA produces current that is more than two orders of magnitude above that from the minimally-active MPA (~8 μA versus ~30 nA, respesctively).

Figure 2

Retinal function as assessed by dark-adapted ERG b-wave at 1 and 4 weeks post-implantation. A – Representative waveforms from each treatment group in response to a bright flash stimulus (2.1 log cd·s/m²) at 1 week post-implantation. The eyes from the A and S groups have slightly larger b-wave amplitudes than the eyes from the C and M groups. The large negative spike at 0.05 msec reflects the electrical activity of the A and M devices. B –Representative waveforms from A, C, and M groups at 4 weeks post-implantation. The response from the A group is twice as large as the response from the C and M groups. C- Average (± sem) amplitude of a- and b-waves at 1 week after surgery from A (n=12), C (n=9), M (n=7), and S (n=7) groups. The b-wave amplitudes of the A and S groups are significantly larger than the M and S groups at the brightest flash intensity. No differences were found in a-wave amplitudes at this time. D- Average amplitude of a- and b-waves at 4 weeks after surgery from A (n=5), C (n=3), and M (n=4) groups. The b-wave amplitude from the A group is significantly larger than the M and C groups at most intensities. There were no differences in a-wave amplitudes between the groups.

Figure 3

Cone function assessed by light-adapted ERGs at 1 and 4 weeks post-implantation. A and B- Representative light-adapted waveforms from each treatment group at 1 and 4 weeks post-implantation, respectively. While no differences are observed in the 1 week waveforms, the response from the A group is ~45% larger than the C and M waveforms at 4 weeks after implantation. C- Average light-adapted b-wave amplitude at one week after implantation from A (n=12), C (n=9), M (n=7), and S (n=7) groups showed no differences. D- Average light-adapted b-wave amplitude from A (n=5), C (n=3), and M (n=4) groups at 4 weeks after implantation showed significantly increased amplitudes in the A group at mid to bright intensities.

Figure 4

Summary of growth factor expression analyses by RT-PCR done 9 days after surgery. Results are reported as growth factor expression in the treated eye relative to control eye. Fgf2 expression in the A group was greater than in all other groups, p < 0.05. Number of replicates for A, M, S, and C treatment groups were: Fgf2, Cntf, and Fgf1: 4, 3, 4, 6; Gdnf: 4, 3, 4, 5; Igf1: 8, 0, 3, 10, respectively. Error bars are +/− sem.

Figure 5

Fgf2 expression analyses by RT-PCR at 9 and 30 days after surgery. Results are reported as Fgf2 expression in the treated eye relative to opposite eye at 9 days (A) and 30 days (B) after surgery. A – Nine days after surgery, Fgf2 expression tended to increase as treatment progressed from C to S to M to A. B – Thirty days after surgery, Fgf2 expression remained elevated in the A group compared to the M and C groups. Post hoc Holm-Sidak analysis significance scores were: #, p = 0.0501; *, p < 0.05; **, p < 0.01, ***, p < 0.001. Error bars are +/− sem.