Dopamine Modulates the Voltage Response of Human Rod Photoreceptors by Inhibiting the h Current

Abstract

Purpose.: The h current (I _h) is a hyperpolarization-activated current that plays important roles in the physiological functions of different types of cells. In the retina of lower vertebrates, I _h contributes to the rod responses to light stimuli by bringing the membrane potential back to the dark level in the presence of continuous light. The purpose of this study was to determine how dopamine modulates I _h in human rods and regulates voltage responses.

Methods.: A patch-clamp recording technique was used on surgically excised human retinas to investigate the effects of dopamine on the I _h of isolated rods. Dopamine was applied in the superfusate.

Results.: Dopamine reversibly decreased the amplitude of the I _h induced by hyperpolarizing voltage steps from a holding potential of −60 mV. At a voltage step of −100 mV, 20 μM dopamine decreased the amplitude of I _h. The D2 dopamine agonist quinpirole inhibited I _h, but the D1 agonist SKF-38393 had no effect. Dopamine-induced reduction of I _h amplitude was blocked by the D2 dopamine antagonist sulpiride. Under current-clamp conditions, an injection of hyperpolarizing current steps to rods produced voltage responses that exhibited a gradual decay. Adding dopamine to the superfusate inhibited the decay in the voltage responses. Quinpirole also inhibited the voltage decay, whereas SKF-38393 was ineffective.

Conclusions.: Dopamine reduced I _h through a D2 receptor and inhibited the gradual decay in the voltage response through a D2 receptor, indicating that dopamine slows the recovery phase of responses to light stimuli by inhibiting I _h in human rods.

The photocurrent of the photoreceptors primarily reflects the light-activated biochemical processes in the outer segments. ^{1 –4} Photoexcited visual pigments activate the GTP-binding protein transducin, which stimulates cGMP phosphodiesterase to hydrolyze cGMP, thus closing cGMP-activated cationic channels in the outer segments. ^5,6 The photovoltage can be altered by voltage-gated channels in the inner segment. ^{7 –11} These voltage-gated currents are of particular importance because the photovoltage, not the photocurrent, controls synaptic transmission. ¹¹ Light stimuli hyperpolarize the photoreceptors and decrease the release of neurotransmitters by decreasing the Ca²⁺ influx at the synaptic terminals. ^12,13  

The hyperpolarization-activated current, the h current (I _h), plays important roles in the physiological functions of different types of cells. In the heart, I _h is well known to contribute to the function of cardiac pacemaking. ¹⁴ In the retinas of different types of vertebrates, I _h contributes to the rod responses to light stimuli by bringing the membrane potential back toward the dark level in the presence of continuous light. ^{15 –18} In a previous paper, we showed that human rod photoreceptors also express I _h. ¹⁹  

Dopamine is thought to decrease the flow of visual information through rod-dependent circuits in amphibian retinas. ²⁰ We find that in human rods dopamine reduces I _h and slows the I _h-mediated decline of voltage responses to steady hyperpolarizing current injection. 

Materials and Methods

Preparations

A small piece of retina was excised from each of 15 adult patients (age range, 39–71 years) with a fresh retinal detachment during the surgical procedure to reattach the retina. ²¹ In the standard vitreous surgery to reattach a detached retina, the fluid between the detached retina and the retinal pigment epithelium is removed through an artificial retinal hole that is made during surgery. ²¹ To make the artificial hole, a small piece of retina (diameter <1 mm) had to be exercised, and this piece was used in our experiment as we have previously reported. ¹⁹ The estimated time between retina detachment and surgery was 2 days. In all cases, functional recovery of the retina occurred after surgery, indicating that there was no severe pathology of the retina. In addition, electron microscopy of the retinal pieces revealed no abnormality in the inner segment of photoreceptors 7 days after the onset of retinal detachment. ²²  

All experiments were performed in compliance with the guideline of the Society for Neuroscience and the Declaration of Helsinki, and the procedures were approved by the Ethics Committee of the School of Medicine of Fujita Health University. All patients were informed of the purpose of the surgery and the necessity of removing the retinal tissue. All patients signed informed consent to allow the use of the tissue for these experiments. 

The dissociation of human rods is similar to that used for monkey photoreceptors, which was reported in detail by Yagi and Macleish. ¹⁰ In brief, the small piece of human retina was incubated for 5 minutes at 37°C in a solution containing 7 U/mL papain (Sigma, St. Louis, MO) without Ca²⁺ and Mg²⁺. The tissue was then rinsed twice with Ames medium and triturated. Isolated cells were plated on a concanavalin A-coated glass coverslip, and all electrophysiological recordings were made within 7 hours of the surgical removal. 

Electrophysiology

Isolated human rod photoreceptors can be clearly distinguished from other types of retinal neurons by their shape. ^23,24 Membrane currents and voltages were recorded in the whole-cell configuration using a patch-clamp amplifier (Axopatch 200B; Axon Instruments, Sunnyvale, CA) linked to a computer. ^25,26 The voltage clamp procedures were controlled by data acquisition and analysis software (pCLAMP; Axon Instruments). The data were low-pass filtered (4-pole Bessel type) with a cutoff frequency of 5 kHz and then digitized at 10 kHz by an analog-to-digital interface. Isolated rod photoreceptors were perfused at 1 mL/min with Ames medium buffered by 22.6 mM bicarbonate, equilibrated with 95% O₂/5% CO₂, and maintained at 37°C. Tetrodotoxin (1 μM) was added to the bath to block a voltage-gated Na⁺ current and Na⁺ action potentials. The recording pipette was filled with a pseudointracellular solution: potassium gluconate, 125 mM; KCl, 15 mM; CaCl₂, 1 mM; EGTA, 5 mM; and HEPES, 10 mM. The solution was adjusted with KOH to pH 7.4. The pipette resistance was 5 to 7 MΩ. Series resistance was neutralized by 60% to 70% during the recording of the membrane currents. The voltage drop across the remaining series resistance was not corrected. 

Results

Dopamine Inhibits an h Current (I _h) in Human Rods

To examine the effects of dopamine on I _h, we recorded the membrane current of human rods under voltage-clamp conditions. At a holding potential (V_h) of −40 mV, no steady current was observed in the isolated human rods (Fig. 1A). Hyperpolarizing voltage steps from a V_h of −40 mV evoked a slow, inward current in the rod (Fig. 1A). This slow, inward current was identified as I _h in our earlier study. ^19,27 After the termination of the command pulse, an inward tail current was also recorded. 

Figure 1.

 

Dopamine inhibits the h current (I _h) in human rods. (A) Membrane currents (I _h) of a human rod induced by hyperpolarizing voltage steps from a holding potential (V_h) of −40 mV in control solution. Command voltages were increased in 10-mV steps from −100 mV to −40 mV. (B) I _h recorded from the same rod as in (A) in the bath solution containing 20 μM dopamine (DA). Conditions of test pulse voltage are also the same as in (A). (C) Recovered I _h recorded from the same rod as in (A) after the washout of dopamine. (D) I-V relationship of the rod shown in (A–C). Peak currents in control (filled circles), dopamine (open circles), and washout (filled triangles) were plotted against test-pulse voltage.

Adding 20 μM dopamine to the bath decreased both the peak amplitude of the slow, inward current during the hyperpolarizing voltage steps and that of the tail current (n = 11; Fig. 1B). These changes were reversible (Fig. 1C). In control solutions, the steady state I-V relationship during the voltage steps, plotted in Figure 1D (filled circles), had an inward rectification of that current. The peak amplitude at a voltage step of −100 mV was 306 ± 21 pA (mean ± SEM; n = 14) in control solution. Adding 20 μM dopamine to the bath decreased the peak amplitude at voltage steps between −100 mV and −50 mV (Fig. 1D, open circles). The peak amplitude at a voltage step of −100 mV was 213 ± 19 pA (n = 11). At the voltage step of −100 mV, dopamine decreased the peak amplitude of I _h by 33% ± 4%. Thus, our results showed that dopamine inhibits the amplitude of I _h in human rods. 

Dopamine Inhibits I _h through D2 Receptors

To identify which type of dopamine receptor is involved in the dopamine-mediated reduction in I _h in human rods, we applied various agonists and antagonists of the dopamine receptors. In control solution, hyperpolarization to −100 mV from a V_h of −40 mV induced I _h (Fig. 2A, thin line). Peak amplitude was −320 pA. Quinpirole (20 μM), a D2 dopamine agonist, reduced I _h by 34% (Fig. 2A, thick line). A similar reduction (32% ± 5%) was obtained in six rods. Quinpirole decreased the peak amplitude of I _h at all voltage steps between −100 mV and −50 mV (Fig. 2B, filled circles). 

Figure 2.

 

Dopamine inhibits I _h by D2 receptors. (A) I _h induced by hyperpolarization to −100 mV from a V_h of −40 mV in control (thin line) and 20 μM quinpirole, a D2 dopamine agonist (thick line). (B) I-V relationship of the rod shown in (A) in control (filled squares) and quinpirole (filled circles). (C) I _h induced by hyperpolarization in control (thin line) and 100 μM SKF-38393 (D1 dopamine agonist, thick line). (D) I-V relationship of the rod shown in (C) in control (filled squares) and SKF-38393 (filled circles). (E) I _h induced by hyperpolarizing voltage steps from a V_h of −60 mV in the solution containing 20 μM sulpiride, a D2 dopamine antagonist. Command voltages were increased in 10-mV steps from −100 mV to −60 mV. (F) I _h recorded from the same rod as in (E) in the bath solution containing 20 μM sulpiride and 20 μM dopamine. Conditions of test-pulse voltage were also the same as in (E).

We also examined the effects of the D1 agonist SKF-38393 on I _h. SKF-38393 (100 μM) changed neither the I _h induced by hyperpolarization to −100 mV (Fig. 2C) nor its I-V relationship (Fig. 2D). To confirm whether dopamine reduces I _h through D2 receptors, we tested the effects of dopamine on I _h by adding 20 μM sulpiride, a D2 dopamine antagonist, to the bath. The I _h induced by hyperpolarizing voltage steps from a V_h of −60 mV in the solution containing 20 μM sulpiride is shown in Figure 2E. Adding both 20 μM sulpiride and 20 μM dopamine did not affect I _h significantly (Fig. 2F), suggesting that sulpiride blocked the reduction of I _h by dopamine. These results suggest that the action of dopamine on I _h is mediated by D2 receptors and not by D1 receptors. 

cAMP Increases I _h in Human Rods

Dopamine receptors are known to modulate various currents through a cAMP cascade. ²⁸ Activation of D1 receptors increases intracellular cAMP by activating adenylyl cyclase, whereas activation of D2 receptors decreases intracellular cAMP by suppressing adenylyl cyclase. ²⁸ To investigate the mechanism underlying the modulation of I _h by D2 receptors, we tested the effects of intracellular cAMP on I _h in human rods. 

After an intracellular application of 100 μM cAMP by the whole-cell patch pipette, the peak amplitude of I _h increased by 28% at the hyperpolarizing voltage step to −100 mV (Fig. 3A). Similar values were obtained from six cells (26% ± 4%). The enhancement of I _h by the intracellular application cAMP was observed at voltage steps between −100 mV and −50 mV (Fig. 3B). This effect of cAMP on I _h was consistent with previous evidence obtained from different types of preparations. 

Figure 3.

 

cAMP increases I_h in human rods. (A) I _h induced by hyperpolarization to −100 mV from a V_h of −40 mV. I _h was recorded just after (thin line) and 3 minutes after (thick line) rupture of the patch membrane. The recording pipette contained 100 μM cAMP. (B) I-V relationship of the rod shown in (A). Peak currents just after (filled squares) and 3 minutes after (filled circles) rupture of the patch membrane were plotted against test-pulse voltage.

The effect of cAMP (an increase in amplitude of I _h) was opposite to the effect of dopamine (a decrease in amplitude of I _h), which is consistent with the idea that dopamine acts by decreasing the concentration of cAMP. 

Dopamine Inhibits Decay in Rod Voltage Responses

To determine the mechanism underlying the modulation of voltage responses by dopamine through I _h, we examined the effects of the dopamine on rod voltage responses under current-clamp conditions. Isolated human rods that lacked outer segments had a resting membrane potential of −61 mV in normal Ames medium (Fig. 4A). Similar values were obtained from 14 cells (−58 ± 4 mV). The shape of the voltage response to light was mimicked by injecting a step of hyperpolarizing current. Injection of −60 pA current step produced a hyperpolarizing response that reached a maximum and then gradually declined to a lower polarized plateau (Fig. 4A, arrow). This time course of the decay of the voltage responses was similar to that of I _h activation during hyperpolarizing voltage steps (Figs. 1 2–3). 

Figure 4.

 

Dopamine inhibits the decay in the rod's voltage response. (A) Voltage responses to hyperpolarizing current steps of increasing intensities (−20, −40, and −60 pA) in control solution. (arrow) Decay in the voltage response. (B) Voltage responses recorded from the same rod as in (A) in the bath solution containing 20 μM dopamine (DA). Conditions of injected currents are also the same as in (A).

When 20 μM dopamine was added to the medium, the resting potential was depolarized by 8 mV (Fig. 4B). Similar values were obtained from seven rods (7 ± 4 mV). Adding dopamine to the bath eliminated the gradual decay of the voltage response in the rods (Fig. 4B). In addition (not shown), adding D2 agonist (20 μM quinpirole) to the bath depolarized the resting potential of rods by 9 ± 5 mV (n = 5) and eliminated the gradual decay of the voltage response, whereas adding D1 agonist (10 μM SKF-38393) to the bath had no effect on either resting potential (0 ± 1 mV; n = 5) or the decay of the voltage response. These results suggested that dopamine modulates the voltage response of human rods by inhibiting I _h by D2 receptors. 

Discussion

Types of Dopamine Receptor in Human Rods

Our results showed that dopamine reversibly decreased the amplitude of I _h induced by hyperpolarizing voltage steps. We have also identified the types of dopamine receptors on human rods. The D2 dopamine agonist quinpirole inhibited I _h, whereas the D1 agonist SKF-38393 had no effect (Fig. 2). The dopamine-induced reduction of I _h was blocked by the D2 dopamine antagonist sulpiride (Fig. 2). These findings suggest that, as in Xenopus rods, ¹⁷ human rods express D2 receptors rather than D1 receptors. 

Activation of D1 receptors increases the intracellular cAMP by activating adenylyl cyclase, whereas activation of D2 receptors decreases the intracellular cAMP by suppressing adenylyl cyclase. ²⁸ In human rods, the effect of cAMP (an increase in the amplitude of I _h) is opposite the effect of dopamine (a decrease in the amplitude of I _h), which is consistent with the idea that dopamine acts by decreasing the concentration of cAMP. Thus, it is highly likely that the activation of D2 receptors decreases the amplitude of I _h by reducing the intracellular cAMP concentration. 

Under current-clamping, injection of hyperpolarizing current steps to a human rod produced voltage responses that had a slow decay (Fig. 4). The addition of dopamine to the bath inhibited the decay in voltage responses. Quinpirole also inhibited voltage decay, whereas SKF-38393 did not change the decay significantly. Results of both the voltage- and current-clamp experiments suggest that dopamine reduces I _h in human rods through D2 receptors. 

Physiological Functions of I _h in Human Rods

The physiological function of I _h in various cells is to depolarize cells after periods of hyperpolarization. In thalamic and cardiac cells, I _h contributes to rhythmic firing by depolarizing the membrane potential after the hyperpolarization of an action potential. ^29,30 In the photoreceptors of cold-blooded vertebrates, I _h is thought to initiate the recovery from a strong hyperpolarization arising from a bright flash of light. ^9,15,16  

In our experiments, the resting potential of isolated human rods was approximately −58 mV, which is approximately 20 to 30 mV more negative than in intact photoreceptors in monkeys ¹¹ and cold-blooded vertebrates. ¹⁶ This occurred most likely because the human rods in the present experiments lacked outer segments and therefore lacked the steady inward current through cGMP-gated channels in the outer segment. 

Under current-clamp conditions, the voltage responses of human rods to hyperpolarizing current steps showed a gradual decay (Fig. 4A). A similar decay of the voltage response of rods was reported in cold-blooded vertebrates. ^9,15,16 In control solutions, the time course of the gradual decay in the voltage responses during the injection of hyperpolarizing current steps (Fig. 4A) was similar to that of I _h activation (Fig. 1A). This suggested that the time course of I _h modulates that of the voltage responses of human rods. 

We previously reported that 3 mM Cs⁺, a blocker of I _h, reduced the gradual decay in the voltage responses. ²⁷ A similar effect of Cs⁺ on I _h was reported in the rods of cold-blooded vertebrates ^9,15,16 and guinea pigs. ¹⁸ The present results indicate that the mechanism underlying the modulation of the voltage response of rods in humans is similar to that in lower vertebrates. 

Physiological Role of Modulation of I _h by Dopamine in Human Rods

Dopamine, which is released by a unique set of amacrine cells, ²⁰ has been postulated to act as an intraretinal messenger for light adaptation by biasing retinal circuits to favor cone over rod inputs to second- and third-order neurons. ³¹ The amphibian cone photoreceptor synapses are influenced by dopamine through a D1 receptor, which acts to increase the effectiveness of glutamate at the postsynaptic membrane. ³² In contrast, amphibian rods have D2 dopamine receptors, suggesting the possibility of presynaptic modulation by D2 dopamine receptors. ³¹ It is unclear where D2 receptors are expressed in human rods. Further study would be required to investigate the location of the D2 receptor in human rods. 

In Xenopus rod photoreceptors, Akopian and Witkovsky ¹⁷ speculated that the reduction of I _h by dopamine may reduce the ability of rods to signal time-modulated light stimuli, although they recorded only the current responses of rods, not their voltage responses. In our study, we showed that an injection of hyperpolarizing current steps in human rods produced voltage responses that exhibited a gradual decay under current-clamp conditions. Application of dopamine in the bath inhibited the decay in the voltage responses (Fig. 4). The D2 agonist quinpirole also inhibited voltage decay, whereas the D1 agonist SKF-38393 did not change the decay significantly. These results suggest that dopamine inhibited the gradual decay in the voltage response by D2 receptors. 

In a previous study, we reported that the recovery by I _h from a strong hyperpolarization from bright flashes of light may enhance temporal resolution in human rods. ²⁷ Therefore, dopamine, which is released by a unique set of amacrine cells, ²⁰ may lower the temporal resolution of human rods because it reduces the amplitude of I _h in human rods. In other words, when the concentration of dopamine around human rods is low, the temporal resolution of the rods should be enhanced. 

Footnotes

 Supported by Japan Society of the Promotion of Science Grant 20500294 (EM) and by the Promotion and Mutual Aid Corporation for Private Schools of Japan.

Footnotes

 Disclosure: F. Kawai, None; M. Horiguchi, None; E. Miyachi, None

The authors thank Mahito Ohkuma for his comments.