Synaptotagmin-9 in mouse retina

Abstract

Synaptotagmin-9 (Syt9) is a Ca²⁺ sensor mediating fast synaptic release expressed in various parts of the brain. The presence and role of Syt9 in retina is unknown. We found evidence for Syt9 expression throughout the retina and created mice to conditionally eliminate Syt9 in a cre-dependent manner. We crossed Syt9^fl/fl mice with Rho-iCre, HRGP-Cre, and CMV-Cre mice to generate mice in which Syt9 was eliminated from rods (rod^Syt9CKO), cones (cone^Syt9CKO), or whole animals (CMV^Syt9). CMV^Syt9 mice showed an increase in scotopic electroretinogram (ERG) b-waves evoked by bright flashes with no change in a-waves. Cone-driven photopic ERG b-waves were not significantly different in CMV^Syt9 knockout mice and selective elimination of Syt9 from cones had no effect on ERGs. However, selective elimination from rods decreased scotopic and photopic b-waves as well as oscillatory potentials. These changes occurred only with bright flashes where cone responses contribute. Synaptic release was measured in individual rods by recording anion currents activated by glutamate binding to presynaptic glutamate transporters. Loss of Syt9 from rods had no effect on spontaneous or depolarization-evoked release. Our data show that Syt9 acts at multiple sites in the retina and suggest that it may play a role in regulating transmission of cone signals by rods.

Keywords: electroretinogram; retina; ribbon synapse; rod photoreceptor cell; synaptotagmin.

Figure 1.

Syt9 mRNA visualized using RNAscope techniques (red puncta) was seen in the outer nuclear layer (ONL), inner nuclear layer (INL), and ganglion cell layer (GCL). Panels A–D show tissue fixed with 4% paraformaldehyde (PFA). Panels E–H show tissue fixed with 10% neutral buffered formalin (NBF). Panel A shows a retinal section of control C57Bl6J retina fixed with 4% PFA. Panel B shows a magnified region of the same section. Syt9 labeling (red puncta) in rods is indicated by the arrows in panel B. Panel C and magnified region in panel D show that labeling for Syt9 mRNA was eliminated from rods in rod-specific conditional Syt9 knockout mice (Rod^Syt9CKO), but remained in inner retinal neurons and ganglion cells. Panel E shows a different control C57Bl6J retina fixed with 10% NBF. Labeling of neurons in the ONL and INL (F; arrows) was also seen with this fixative, but fewer cells were labeled. Labeling was abolished altogether in a whole animal knockout of Syt9 created by crossing floxed Syt9 mice with CMV-Cre mice that express cre-recombinase constitutively (CMV^Syt9).

Figure 2.

Global elimination of Syt9 increased scotopic b-wave responses at the highest flash intensities. (A) Representative scotopic ERG waveforms evoked from a high intensity flash (5 dB) for control (n = 6) and CMV^Syt9 (n = 7) mice are shown in black and red traces, respectively. (B) Scotopic a-wave amplitude as a function of flash intensity. (C) Plot of scotopic b-wave amplitude as a function of intensity. (D) Example photopic waveforms evoked from a high intensity flash (13 dB). (E) Plot of photopic b-wave amplitude as a function of intensity. *P < 0.01. Error bars show ±SD.

Figure 3.

Elimination of Syt9 selectively from cones had no effect on scotopic or photopic ERGs. (A) Representative scotopic ERG waveforms evoked from a high intensity flash (5 dB) for control (n = 6) and Cone^Syt9CKO (n = 7) mice are shown in black and red traces, respectively. (B) Scotopic a-wave amplitude as a function of flash intensity. (C) Plot of scotopic b-wave amplitude as a function of intensity. (D) Example photopic waveforms evoked from a high intensity flash (13 dB). (E) Plot of photopic b-wave amplitude as a function of intensity. Error bars show ±SD.

Figure 4.

Elimination of Syt9 selectively from rods decreased scotopic and photopic ERG b-waves. (A) Representative scotopic ERG waveforms (5 dB) for control and Rod^Syt9CKO (n = 4) mice are shown in black and red traces, respectively. (B) Scotopic a-wave amplitude as a function of flash intensity. (C) Plot of scotopic b-wave amplitude as a function of intensity. (D) Example photopic waveforms evoked from a high intensity flash (13 dB). (E) Plot of photopic b-wave amplitude as a function of flash intensity. *P < 0.01. Error bars show ±SD.

Figure 5.

Effects of Syt9 elimination from the whole retina on ERG oscillatory potentials (A, A’, A”). Scotopic and photopic OPs extracted from representative ERG waveforms (5 dB and 13 dB, respectively) by bandpass filtering to remove frequencies below 70 Hz and above 280 Hz in CMV^Syt9, Cone^Syt9CKO, and Rod^Syt9CKO mice, respectively (B, B’, B”). Scotopic OPs in CMV^Syt9, Cone^Syt9CKO, and Rod^Syt9CKO mice, respectively (C, C′, C”). *P < 0.01. Error bars show ±SD.

Figure 6.

Whole cell recordings of glutamate transporter anion currents (I _A(glu)) revealed no significant differences between control and Syt9 KO rods. (A) Example I _A(glu) recordings from control and Syt9 KO rods evoked by 25 ms steps from −70 to −10 mV. Passive capacitive and resistive currents were subtracted using P/6 protocols. (B) I _A(glu) amplitude from control (5 ms, n = 34; 25 ms, n = 27; 500 ms, n = 24) and Syt9 KO rods (5 ms, n = 15; 25 ms, n = 16; 500 ms, n = 11). Control data from Mesnard et al. (2022). Response amplitude was measured 2 ms after termination of the test step. (C) I _A(glu) charge transfer from control (5 ms, n = 37 rods; 25 ms, n = 26; 500 ms, n = 20; n = 15 mice) versus Syt9 KO rods (5 ms, n = 15; 25 ms, n = 16; 500 ms, n = 11; 4 mice). Charge transfer was measured beginning 2 ms after termination of the test step until current recovered to baseline. Baseline was defined as the current level at the end of the 2 s trial. (D) Example recordings of spontaneous I _A(glu) release events showing release of individual glutamate-filled synaptic vesicles. Rods were held at −70 mV when measuring spontaneous release. Error bars show 95% confidence intervals.