Fig 1.
Schematic diagram showing signaling between cone and rod photoreceptor cells and synaptically-connected second order neurons in the goldfish retina.
Signaling between a cone and a cone horizontal cell (cHC) and between a rod and a rod horizontal cell (rHC) is shown. Cones and rods use glutamate (Glu) to signal cHCs and rHCs, respectively. Also shown is a gap junction between a cone pedicle and rod spherule that is open at night in the dark and closed in the day.
Fig 2.
Light responses of dark-adapted cones, cHCs and rHCs during day and night.
(A) Representative examples of dark-adapted cone (left panel), cHC (middle panel) and rHC (right panel) responses to a full-field white light stimulus (500 ms; -5 log Io) recorded during the day (grey traces) and night (black traces). (B) Normalized intensity-response functions of dark-adapted cones (circles), cHCs (squares) and rHCs (diamonds) during the day (open symbols; left panel) and night (filled symbols; right panel).
Table 1.
Light response kinetics of goldfish cones, cHCs and rHCs under dark-adapted conditions in the day and night.
Fig 3.
Light response kinetics and spectral sensitivity of dark-adapted cones, cHCs and rHCs in the day and night.
(A-C) Normalized response latency (A), time-to-peak (B) and response duration (C) of dark-adapted cones (circles), cHCs (squares) and rHCs (diamonds) during the day (open symbols; left panel) and night (filled symbols; right panel). The responses of only 1 cell per retina to the full series of light stimuli were recorded because stimuli of intensities > -4.5 log Io light-adapted the cells. Note the increased response latency and response duration of cones and cHCs at night. (D) Spectral sensitivity of cones, cHCs and rHCs recorded under dark-adapted conditions during the day (left) and night (right). During the day, the spectral sensitivity of dark-adapted cones fit one of three spectral sensitivity curves corresponding to the three major known cone subtypes in goldfish: L (~ 608 nm, red cone nomogram), M (~ 539 nm, green cone nomogram), and S (~ 451 nm, blue cone nomogram). The spectral sensitivity of all cones (filled circles), cHCs (filled squares) and rHCs (filled diamonds) recorded at night peaked at ~ 525 nm. The spectral sensitivity of rHCs at night closely fit a rod nomogram (dashed curve). The spectral sensitivity of cones and cHCs fit a template that combined the rod nomogram and that of an L-cone visual pigment (dotted curve) [19].
Table 2.
Spectral characteristics of goldfish cones, cHCs and rHCs under dark-adapted conditions in the day and night.
Fig 4.
Threshold responses of dark-adapted cones, cHCs and rHCs to 500 nm and 700 nm light stimuli during day and night.
(A) Representative examples of dark-adapted cone, cHC and rHC threshold responses (0.5 mV) to a light stimulus (500 ms) at 500 nm (grey trace) and 700 nm (black trace) during the subjective day (left column) and night (right column). The amplitude of each trace has been normalized for better comparison. (B) Averaged response latency, time-to-peak, duration of the hyperpolarization, and quantum sensitivity measured in cones (left column), cHCs (center column) and rHCs (right column) under the conditions described in (A). Data in (B) were analyzed with a two-WAY ANOVA, the first factor was time of day (t) and the second factor was wavelength (w). Significant variations (α = 0.05) are indicated on the left side of each figure; i: w x t interaction.
Fig 5.
Characteristics of the cone-to-cHC synaptic transfer function during day and night under dark-adapted conditions.
(A) Intensity-response functions of dark-adapted cones (circles) and cHCs (squares) during the day (open symbols; left panel) and night (filled symbols; right panel). Data points were fit to an allometric-type function (see Methods). The responses of only 1 cell per retina to the full series of light stimuli were recorded. Note that the maximum response amplitude (Vmax) is much lower at night for both cones and cHCs. (B) The cone-cHC transfer function was calculated from the normalized intensity-response functions illustrated in Fig 2B. The deviation of the transfer function from the identity line suggests that signal transmission from cones to cHCs is non-linear. Note the increased non-linearity of the function at night. (C) Synaptic gain calculated from the curves illustrated in (A). Note the high gain for low intensities during the subjective day and the low gain (< 1) at all intensities at night. (D) Speed of synaptic transfer was calculated from the curves shown in Fig 3A. Note that synaptic transfer is significantly longer at night compared to the day.
Fig 6.
Light responses of bright light-adapted cones, cHCs and rHCs in the day and night.
(A) Representative examples of bright (photopic) light-adapted cone (left panel), cHC (middle panel) and rHC (right panel) responses to a full-field white light stimulus (500 ms; -5 log Io) recorded during the day (grey traces) and night (black traces). (B) Normalized intensity-response functions of light-adapted cones (circles), cHCs (squares) and rHCs (diamonds) during the day (open symbols; left panel) and night (filled symbols; right panel).
Table 3.
Light response kinetics of goldfish cones, cHCs and rHCs under light-adapted conditions in the day and night.
Fig 7.
Light response kinetics and spectral sensitivity of bright light-adapted cones, cHCs and rHCs in the day and night.
(A-C) Normalized response latency (A), time-to-peak (B) and response duration (C) of bright (photopic) light-adapted cones (circles), cHCs (squares) and rHCs (diamonds) during the day (open symbols; left panel) and night (filled symbols; right panel). Note that for each cell type, the daytime and nighttime intensity-response curves are similar and resemble those recorded during the day under dark-adapted conditions. (D) Spectral sensitivity of cones, cHCs and rHCs recorded under bright light-adapted conditions during the day (left) and night (right). Both during the day and night the spectral sensitivity of light-adapted cones fit one of three spectral sensitivity curves corresponding to the three major known cone subtypes in goldfish: L (~ 608 nm, red cone nomogram), M (~ 539 nm, green cone nomogram), and S (~ 451 nm, blue cone nomogram). The spectral sensitivity of cHCs (squares) matched that of L-cones in both day and night. We did not record any S-cones during the day and did not determine the spectral sensitivity properties of rHCs under light-adapted conditions.
Table 4.
Spectral characteristics of goldfish cones, cHCs and rHCs under light-adapted conditions in the day and night.
Fig 8.
Threshold responses of bright light-adapted cones and cHCs to 500 nm and 700 nm light stimuli during day and night.
(A) Representative examples of bright (photopic) light-adapted cone and cHC responses at threshold (0.5 mV) to a light stimulus (500 ms) at 500 nm (grey trace) and 700 nm (black trace) during the day (left column) and night (right column). The amplitude of each trace has been normalized for better comparison. (B) Averaged response latency, time-to-peak, duration of the hyperpolarization, and quantum sensitivity measured in cones (left column) and cHCs (right column) under the conditions described in (A). Data in (B) were analyzed with a two-WAY ANOVA, the first factor was time of day (t) and the second factor was wavelength (w). Significant variations (α = 0.05) are indicated on the left side of each figure.
Fig 9.
Characteristics of the cone-to-cHC synaptic transfer function during day and night under bright light-adapted conditions.
(A) Intensity-response functions of bright (photopic) light-adapted cones (circles) and cHCs (squares) during the day (open symbols; left panel) and night (filled symbols; right panel). Data points were fit to an allometric-type function (see Methods). (B) The cone-cHC transfer function was calculated from the normalized intensity-response functions illustrated in Fig 6B. The result that the transfer function follows the identity line indicates that signal transmission from cones to cHCs is linear. (C) Synaptic gain calculated from the curves illustrated in (A). (D) Speed of synaptic transfer in the day and night was calculated from the curves shown in Fig 7A. Note the similarities between the day and night curves.